Documentos de Académico
Documentos de Profesional
Documentos de Cultura
Introduction
Welcome to the World of
Photovoltaics
Thank you for your interest in photovoltaic technology and system design, and
welcome to the world of solar electricity! Siemens Solar is proud to be recognized
as the world-wide leader in solar electric power generation, and we wish to support
you in your efforts to apply photovoltaic technology to solve electrical power
problems, whether you are an individual designing your own system, or a system
designer working on large applications.
1-1
Fundamentals Introduction
1-2
Fundamentals Introduction
Caution
Products specified for use in this manual might not conform to the National Electric
Code and may not conform to local requirements if the system is installed on a
building, movable structure or vehicle. Before assembly and installation of any
photovoltaic power system, you should consult with local authorities so that you may
be assured installation will safely conform to all local building code requirements. A
permit may be required.
Also, consult local codes before using products or installation procedures outlined in
this manual. Codes could possibly cover applicable inspections, permits for
electrical wiring, wire size, interconnections, grounding, enclosures, conduits, overcurrent protection, receptacles, load restrictions, disconnects and appliances.
Failure to follow applicable codes constitutes misuse of the products.
Do not attempt installation before reviewing all applicable instructions.
The technical information and suggestions for installation, operation, use and
maintenance made herein are based on Siemens Solar Industries knowledge and
experience and are believed to be reliable, but such information and suggestions do
not constitute a warranty, expressed or implied.
Since the conditions or methods of installation, operation, use and maintenance of
the equipment described in this manual are beyond Siemens Solar Industries
control, Siemens Solar Industries does not assume responsibility and expressly
disclaims liability for loss, damage or expense arising out of or in any way connected
with such installation, operation, use or maintenance.
1-3
Fundamentals Introduction
(End of Chapter)
1-4
Fundamentals Introduction
CHAPTER ONE
INTRODUCTION
1-1
1-1
1-1
1-2
Caution
1-3
1-5
Fundamentals Introduction
Chapter Two
The Basics of Photovoltaics
Photovoltaic refers to the creation of voltage from light, and is often abbreviated as
just PV." A more common term for photovoltaic cells is solar cells," although the
cells work with any kind of light and not just sunlight.
A solar cell is a converter it changes energy of light into electrical energy. A cell
does not store any energy, so when the source of light (typically the sun) is removed,
there is no electrical current from the cell. If electricity is needed during the night,
some form of electrical storage (typically a battery) must be included in the circuit.
In this chapter we will present some of the most fundamental concepts of energy
and power that are the basis for understanding photovoltaic power systems. We will
also discuss some of the common terms used in photovoltaic technology, and
present the prime benefits of using solar electricity for your power requirements.
2-1
2-2
2-3
Minimal Maintenance
Solar electric systems typically require very minimal maintenance because there are
so few moving parts. Contrast this with a diesel-powered system or even other
renewable sources such a wind generators or hydro generators, which often have
costly repairs or regular maintenance of moving parts. Very complex photovoltaic
systems do have more parts and may require some maintenance. But when looking
at small power requirements, such as for home lighting or remote
telecommunications systems, only occasional battery maintenance is required. It
would be a mistake to say that photovoltaic systems require NO maintenance, but
the absolute amount of time and money required for photovoltaic systems is quite
low.
2-4
Maximum Reliability
This is perhaps the primary advantage of photovoltaics when compared to any other
form of electrical power generation. Because there are typically few or no moving
parts and the complexity of the systems can be kept low, the ultimate reliability of
photovoltaic power systems in the real world is quite high. The photovoltaic
generator typically is not affected by environmental effects such as lightning strikes,
high winds or blowing sand, humidity and heat, or snow and ice. The key to reliability
is quality and simplicity. If high quality components are used with the solid-state
solar generators, and if the component count and complexity of the system design
are kept to a minimum, the chance of any failure occurring is remarkably low.
2-5
Reduced Vulnerability
Because you can avoid stringing long power lines for many miles or kilometers from
some central generation source, many of the problems with utility power losses can
be avoided. Ice storms or vehicle accidents can cause power lines to go down,
perhaps tens or hundreds of miles from where the power is actually needed. With a
reliable photovoltaic power system at your site you could still have power, while
others around you have none.
And if you have chosen to distribute the generation of power to various load sites at
your location, you can insure even more reliability and less vulnerability to each load
site. For example, if separate homes have their own lighting system, and one user
overdischarges their batteries, or damages their system, the other users will be
unaffected.
This also applies to deliberate vandalism or terrorism. For example, having electrical
generation for lighting and security distributed to each building would make loss of
security for the whole site more difficult or practically impossible.
2-6
Easily Expanded
Photovoltaic power generators are modular by design. More power can be added to
an existing array easily. Old modules can be added to new ones without any penalty
(if the voltages are properly matched). Just enough power can be purchased and
installed today to meet your current needs, and as demand grows more modules
can be added in later years. This also means that financially it is easy to start with a
minimal power system today, and then add to the power as your budget allows later.
2-7
2-8
photon
h+
internal
field
ee-
P/N junction
2-9
Water Analogy
It is often helpful to give an analogy to water flowing. Imagine a water pump
connected to a circuit of pipes that are already full of water. The pipe circuit also
includes some sort of load like a water wheel, and all the water returns back to the
pump through the pipes, so that no water is ever lost due to evaporation or
splashing. In the analogy the pump is the solar cell, the pipes are the wires
connecting the cell to an electrical load and back to the cell, and the water in the
pipes is like the electrons already in the wires.
When sunlight (or any light for that matter) shines on the cell it delivers the fuel that
is needed by the cell needs to operate. Electrons are freed and set in motion. The
internal electro-static field pushes the freed electrons out of the cell and into the
wire. The analogy would be as if the pump was turned on. The pump begins to
push water into the pipe. But the pipe is already full of water, so water flows almost
simultaneously throughout the whole system.
The water flows on to the load like a water wheel, where its pressure and flow allow
useful work to be done. All of the water is then captured and flows again though
pipes back to the pump. The pump continues to push new water to the load through
the pipes.
In the real case of the solar cell the electrons freed by the incoming sunlight photons
flow out of the cell and on to the electrical load. They give up their extra potential
energy or voltage there and allow useful work to be done. The electrons then
continue to flow to the back of the solar cell, where they become available once
again to be knocked loose and flow on to the load.
The electrical circuit is closed, just like the pipe system was closed, so no electrons
are ever used up in the process. The solar cell never runs out of electrons. It
only needs continuous input of fuel in the form of light energy to keep running.
2-10
But manufacturers could produce different module designs to better match other
loads, for example high voltage motors for water pumps or utility connected systems
that often operate at hundreds of volts.
If the voltage or current from one module is not enough to power the load, then
modules can also be connected together, just as the cells were. Manufactures
usually build modules with convenient junction boxes that allow interconnecting in
series or parallel.
2-11
The fundamental physics of the materials used determines the voltage of a cell, and
the size determines the current. Usually manufacturers settle on one or two basic
sizes and designs for their cells and then proceed to make millions of them. They
can be packaged and exported to other module manufacturing facilities, and built
into specialized products such as lanterns, radios and garden lights. Think of the
cell as the smallest unit to work with. But it is fragile, and its voltage is low (typically
1/2 volt). For use in the real world it must be protected and connected to other cells
to give useful voltage.
The MODULE is really the basic building block for real-world remote power
systems.
This would be the building block for larger power systems. Usually the modules are
wired together on the panel to give the final system voltage (for example 12, 24, 48
volts or higher) and the panels are wired together or individually through field
junction boxes and then on to the system controls and batteries. Individual panels
can be disassembled or maintained while the other panels are operational.
2-12
Terms Used
CELL
2-13
Voltage
We can draw an analogy between the term voltage and water pressure. Imagine
water held behind a large dam where there is tremendous pressure but no forward
movement of the water. An electrical system with high voltage potential means that
the electrons in the components and wires have stored pressure and is capable of
doing work if released to flow, just as the water could do work if released from the
dam.
The unit of measure used for electrical voltage is the volt.
It is important to realize that voltage does not flow. It is a measure of the difference
in force or pressure between two points in an electrical circuit.
Current
The analogy to electrical current is the rate of flow of water. Electrons actually flow
past any point in an electrical circuit, just as water actually flows past when released
from a dam (or any other source of pressure).
The unit of measure for electrical current is the ampere or just amp. The number
of electrons flowing is enormous, amounting to over 1,000,000,000,000,000,000
electrons per second in just one amp!
And dont refer to current flow as amps per second. It is simply amps -- the unit
already includes the notion of number of electrons per second.
2-14
Power
The concepts of power and energy are so close that people often think they are the
same things. But the difference is critical in understanding how photovoltaic
systems are designed and how they work.
Power is the rate of doing work, and is NOT the amount of work done.
If we use the water analogy, the power of flowing water is given by the rate of flow of
the water past a point multiplied by the pressure of that water flow.
In the electrical case power is the current flowing times the voltage of that flow.
Power (electricity) =
Current X Voltage
The units of measure for power are watts. It is incorrect to say watts per hour,
because the unit already has the measure of time included.
From the formula, you can see that increasing either the flow (current) or the
pressure (voltage) increases the power. Also, if there is no current or no voltage,
there is no power.
2-15
Energy
If power is the rate at which work is done, then energy is the AMOUNT of work that
is done during a specific period of time. The rate of doing work (power) must be
multiplied by the time to give the amount of work.
Amount
Rate X Time
Energy
Power X Time
So ENERGY is what you have to work with, while POWER is the rate at which you
convert or use it.
2-16
Use
Anytime
The importance of understanding the difference between energy and power can now
be presented in the context of a photovoltaic system. Most photovoltaic systems
use batteries to store the energy converted by the solar modules during a day into
chemical energy for use during the night or on stormy days. The battery acts as a
reservoir of ENERGY and mediates between the POWER that might be available at
any moment from the solar modules and the POWER that the loads might want to
draw at that instant. If the loads need more power than the modules can produce,
then the battery discharges a bit to supply the difference. During the night, for
example, the modules produce no power so the battery must discharge to supply all
the power needed by the loads. During a day, if the loads do not require all the
power available from the modules, then the extra power goes into recharging the
battery.
2-17
During this course, you will learn how to predict the energy demand of the loads and
determine how many solar modules are needed to meet that energy demand.
2-18
(End of Chapter)
2-19
CHAPTER TWO
THE BASICS OF PHOTOVOLTAICS
2-1
2-2
2-4
2-4
2-4
2-4
2-5
2-6
2-6
2-7
2-8
2-8
2-10
2-11
2-12
2-14
2-14
2-14
2-15
2-16
2-17
2-18
2-20
Chapter Three
Market Overview
In this section, we will examine some of the key players in the worldwide
photovoltaic market, including the major manufacturers and the common
technologies used today. We will look at some trends in application and market
growth that will give you a sense of the excitement we feel for the future. The main
purpose of this section is to give you an orientation to the current and future status
of this power generation technology and how Siemens Solar is positioned to stay the
world leader.
3-1
Major Manufacturers
The number of manufacturers of solar photovoltaic cells and modules peaked during
the late 1970s and early 1980s at almost 30, but has decreased during the 1980s
and 90s to about a dozen major contributors. The cumulative shipments of the top
dozen manufacturers during the last fifteen years are shown below. There is
manufacturing occurring in all regions of the world, including India, Europe, Japan as
well as the US.
Cumulative Shipments
1980-1996
ie
en
140
S
ar
ex
100
Ph
K
P
B
E
S
ur
E
le
So
p
Sh
ar
s
io
el
H
ot
os
40
ol
ow
ar
at
yo
60
20
ce
ra
ol
80
S h ip p e d
M e g a w a tts
120
0
Source: Strategies Unlimited
In 1996 Siemens Solar celebrated a milestone for the world photovoltaic industry by
becoming the first company to ship a cumulative total of 100 Megawatts of solar
cells and modules. The other major manufacturers are Solarex (U.S.), Kyocera
(Japan), and BP Solar (U.K.).
3-2
The overall worldwide market for photovoltaic power is shared by the major
manufacturers. The market share of the key manufacturers is shown below for
1995. Again Siemens Solar has the largest single portion of the worldwide market,
with about 23% of the total.
Market Share
Siemens Solar 21%
89.5 MW
1996
(by shipments)
Solarex 11%
Other 35%
Kyocera 11%
ASE 4%
Sharp 3%
Source: Strategies Unlimited
BP Solar 9%
Eurosolare 3%
Astropower 3%
3-3
3500
3200
3000
2500
Shipments
MWp
2000
Accelerated growth
~29% to 2000
~22% to 2010
1800
1500
1000
500
42.7
0
1990
1995
2000
Accelerated
2005
2010
As Usual
Source:
Strategies Unlimited Report PM-39 March 1993
3-4
6000
5000
4000
Sales Volume
$ Million
3000
2000
1000
271
0
1990
1995
2000
2005
2010
Source:
Strategies Unlimited Report PM-39 March 1993
3-5
Remote Industrial: This has been the major application area for 30 years, including
telecommunications, cathodic protection, telemetry, navigational systems and other
unmanned installations in harsh remote sites. The load demands are well known
and the requirements for reliable power are the highest.
Government Demonstration: This market segment has always been small, and
constitutes projects funded by government or military organizations. These are
typically funded early in a governments learning curve about photovoltaics, to
provide information on installed costs and reliability.
Consumer Indoor: These products use photovoltaic cells to provide the small
amount of power needed for small electronic devices such as watches and
calculators.
3-6
The latest estimate for the way modules are currently used worldwide is shown on
the next page. The remote habitation application area is about 50% of the total
worldwide market, with the traditional remote industrial applications making up the
next largest market segment.
Market Segments
Forcast 1997
Remote
Industrial
27%
50%
Remote
Habitation
10%
5%
8%
Grid
Connected
Consumer
Other
Source: Strategies Unlimited
3-7
Projections have been made to the year 2010 estimating the changes that might
occur in the relative size of different application groups. Once again, they present a
business as usual forecast and an accelerated forecast. The two scenarios are
similar in most respects. The remote industrial application group, the traditional
bread and butter of the industry, will continue to be a major segment, but will
continue to yield to remote habitation and grid connected systems. The major
difference in the two scenarios is the timing and the extent of the growth of the grid
connected application group.
Remote Industrial
80%
70%
Remote Habitation
Consumer Power
60%
Government
Demonstration
50%
40%
Consumer Indoor
30%
Grid Connected
20%
10%
0%
1990
1995
2000
2005
2010
3-8
Remote Industrial
80%
70%
Remote Habitation
Consumer Power
60%
Government
Demonstration
50%
40%
Consumer Indoor
30%
Grid Connected
20%
10%
0%
1990
1995
2000
2005
2010
The future will have photovoltaics used closer to where we live and work, as the cost
of delivered power comes down to compete with generators and utility power.
Remote habitation already constitutes the largest market segment. In developed
countries, this would constitute power for remote homes, cabins, and farms. For
developing regions, these applications would be home lighting, schools, clinics,
farms and village power.
The most important trend to understand is the impending growth of the grid
connected market. As the cost of photovoltaic power continues to decrease, and as
traditionally generated utility power costs continue to increase, the market
acceptance of photovoltaic power will accelerate. In these scenarios, the market
share due to grid connected systems will be near or greater than 50% by 2010. This
will indicate the maturation of the photovoltaic market, as what has been traditionally
considered a remote power solution becomes economic for the vast urban market.
3-9
The assumptions used in the business as usual and accelerated scenarios of the
Strategies Unlimited report need to be presented, so that you can judge the
probability of either case. The differences are presented next.
The major differences arise from how aggressively and how quickly industry,
government and utility organizations become involved with developing photovoltaic
technology.
Factor
Business As Usual
Accelerated
Industry
Investment
Utility Investment
Incentives
PV Infrastructure
Financial Sources
National Energy
Policy
Government
Support
Environment
Concern
Petroleum
Changes
Natural Gas
Changes
Global Economy
3-10
3-11
Ribbon Silicon
The slicing process in the previous two methods is wasteful, often converting 4050% of the material into dust. This is because the wafers are only approximately
0.015" thick, and the saw blade is about this thickness as well.
One method of producing wafers avoids most of this waste by growing a thin ribbon
from the melted silicon. The ribbon is either pulled sideways off the top of the melt,
or pulled up through a die.
Very fast growth rates are possible, but the speed results in polycrystalline
structures. If the pulling process is done very carefully, near single crystal structure
is possible. The ribbon thickness is approximately 0.010"-0.015", so no further
sawing is necessary. The ribbon is simply scribed and broken to produce
rectangular wafers.
However, the surface of the wafer is not typically flat and often bulges. Waviness in
the surface makes further manufacturing steps and interconnection difficult.
Efficiencies similar to polycrystalline silicon are typical.
Thin Films
All of the previous methods produce a single cell as the basic building block. This
means that many cells must be connected together to produce a module of useful
voltage, because each cell produces approximately 0.5 volts. The interconnecting
and subsequent lamination steps are costly.
In the past 10 years great progress has been made in manufacturing solar modules
by depositing extremely thin films of semiconductors onto glass or metal substrates.
This process has many advantages over the "traditional" methods mentioned above
that produce individual cells.
The semiconductor layers are only a few hundred atoms thick, so expensive material
costs are reduced. The entire module is made as a unit, so interconnecting
machinery and processing are eliminated. And the cell size can be modified easily,
so it becomes easy and cost effective to make modules of different power output for
different applications or products. A unique characteristic of some thin film solar
devices is that the light that does not interact to knock loose electrons can pass
through the device because the layers are so thin. This means semi-transparent
films are possible. Car sunroofs, boat hatches, and building glass, for example, can
be made that produce useful electric power and also serve as windows!
Several approaches are being pursued around the world to develop thin film solar
cells and modules. Some of these are described next.
3-12
3-13
Tandem TFS/CIS
An exciting prospect for high efficiency and low cost lies in combining thin film
devices of different spectral responses together in one module. One plan for such a
"tandem" module involves placing a module of thin film silicon:hydrogen alloy with a
transparent back conductor in front of a module made from CIS. The front thin film
silicon:hydrogen module absorbs the short wavelength blue, green and yellow light,
and lets the orange, red, and infrared light pass through to the CIS module.
Research devices achieving over 15.5% have been made using this method, and it
is anticipated that 18-20% is achievable. The combination of higher efficiencies and
lower costs than any traditional design makes this a very exciting near-term
technology.
Concentrator Cells
Covering large areas with solar cells and intercepting what light falls on the surface
is called a "flat plate" method. This is the common approach taken by
manufacturers of single and polycrystalline silicon and thin film devices. Another
approach is to focus the light onto a small area and have a specially made cell at the
focus. This seeks to reduce overall cost by reducing the amount of cell material
required and substituting inexpensive lens or mirror materials.
Concentrator cells must be specially made to handle the large currents produced.
From 10-500 times the normal sun intensity can be produced on the cell surface,
resulting in an equally large amount of current compared to what would be produced
by flat plate converters.
To keep the focus on the small cell, the entire module assembly must very
accurately track the sun. Since tracking is essential, concentrator devices are
appropriate only where clear skies are predominant.
Recent advances in concentrator cells have been achieved using a "point contact"
structure. Thousands of p/n junctions are made on the back of a single silicon chip.
Thousands of interconnections must be made to make a working cell, but some of
the losses that occur in single junction devices are avoided and efficiencies of
greater than 30% have been reported. The transfer from laboratory to large-scale
commercial production has yet to be achieved.
3-14
T h in F ilm s
13%
11.7 MW
48.3 MW
R ib b o n
89.5 MW
24 MW
P o ly c ry s ta llin e
O th e r
3%
Source: The Solar Letter, February 14, 1997
3%
CdTe,
Concentrators
Si
27%
The total worldwide production of photovoltaic power generators in 1995 was about
81 megawatts. This is large for the photovoltaic manufacturers, but is quite small
when compared to the generating capacity of conventional power plants. One large
power plant can produce 800-2000 megawatts, continuously day and night. This
shows how far the photovoltaic market has to grow to become a major contributor to
the worlds energy needs.
3-15
90
80
70
54 MWp
60
CZ
+145%
50
40
+15%
30
Poly
20
-15%
10
Thin Film
0
1991
1996
T h in F ilm
P o ly
CZ
3-16
Efficiency Defined
Efficiency
Output
Input
Module Efficiency
Efficiency
Solar devices work only during daylight, when there is solar energy to convert. The
best devices will capture and convert the most of the limited amount of solar
radiation that is made available during daylight hours. There are important
differences between the ability of the three major types of technologies to efficiently
convert sunlight into electricity.
3-17
In general, single crystal silicon devices are the most efficient, with polycrystalline
silicon cells slightly lower in efficiency, and thin film devices of amorphous silicon the
lowest in efficiency. Data that has been gathered over a period of seven years from
actual outdoor testing of these different technologies is shown. This data was
gathered as a part of a US government/utility program called Photovoltaics for Utility
Scale Applications (PVUSA), where multi-kilowatt arrays of various technologies
have been installed and are being monitored for performance in real world
conditions.
Single Crystal
DC Efficiency %
10
Polycrystalline
8
6
4
Amorphous
Jan-96
Jul
Jan-95
Jul
Jan-94
Jul
Jan-93
Jul
Jan-92
Jul
Jan-91
Jul
Jan-90
Jul
Jan-89
0
Source:
PVUSA Progress
Report March 1996
The PVUSA data shows clearly how the different technologies compare in efficiency.
The average efficiency of the single crystal silicon array is about 10%, while the
efficiency of the polycrystalline modules is about 8%. The thin film amorphous
silicon array averages only about 3% efficient. This comparison shows the normal
seasonal variations in efficiency due to ambient heat in the summer months for the
single and polycrystalline modules. One interesting feature of thin film amorphous
silicon devices is that they are less affected by temperature variations, as shown by
the nearly flat efficiency data throughout the hot and cold seasons.
3-18
Best Cell
Efficiency
Module Area
Efficiency
Advantages
Weaknesses
Silicon
22.7 %
12-15 %
Well understood;
Receiving
renewed attention
CdTe/CdS
15.8 %
6-8 %
Low cost;
High deposition
rates possible
Cd liability;
Needs more
development
Amorphous
Silicon
13.2 %
4-9 %
Low cost
Looses power
over time;
Low efficiency
CuInSe2
16.9 %
10%
23% potential;
Low cost
Manufacturing yields
are low; Needs more
development
Single Junction
Concentrator
28.7 %
NA
Hybrid PV / thermal;
central power
generation
Lacks production
economy of scale;
Complex BOS
Multijunction
Concentrator
35 %
NA
Hybrid PV/thermal
Space
Lacks production
economy of scale;
Complex BOS
3-19
The bulk silicon (both single crystal and polycrystalline) and thin film amorphous
silicon technologies have been described earlier. The cadmium telluride
(CdTe/CdS) material is another heterojunction thin film technology that shows much
promise and is being pursued by a variety of manufacturers. The multijunction
concentrators are more exotic devices that utilize different cell technologies
deposited on top of one another, to capture different regions of the solar spectrum.
These offer the highest potential efficiencies, but are by far the most complex and
expensive. Their best application would be where space is an expensive premium,
such as in large central power generation plants or in orbiting space facilities.
The primary trade-offs that are evident from this comparison are potentially high
device efficiency vs. reduced efficiency for real modules, and potential low costs vs.
process stability and manufacturing yield issues. There is not one clear choice for
the best technology, as all of them have potential improvements and difficulties to
overcome.
The module area efficiencies mentioned in this comparison are general values for
actual production volume power modules, and are contrasted with the best cell
efficiency values that are often reported in press releases and research papers. It
is very important to draw a distinction between the peak efficiency of a single
research device and the consistently attainable efficiency of a high volume
production module. A single research device may have a high efficiency, and show
the potential that production devices can attain, but actual full production modules
incorporate interconnect space, frame space, gridline coverage and other physical
features that decrease the efficiency of the overall device.
Another concept that is highlighted in this table comparison of technologies is that
concentrator technologies have not achieved mass production status yet. They
have the potential for the highest efficiencies of all the technologies presented, but
involve extensive balance-of-system (BOS) equipment, such as specially designed
sun-position tracking frameworks, concentrating lenses, computer controls, and
special heat sinks. Their ability to compete with flat-plate technologies will depend
on the cost-effectiveness of the entire package needed to create useable power,
and not just on the efficiency of the small individual photovoltaic cells.
3-20
(End of Chapter)
3-21
CHAPTER THREE
MARKET OVERVIEW
3-1
Major Manufacturers
3-2
3-4
3-4
3-6
3-11
3-11
3-11
3-12
3-12
3-14
3-15
3-17
3-19
3-22
Chapter Four
General Skills of System
Designers
Many people feel that photovoltaic system design involves using a computer or
calculator and solving some equations that lead to one unique solution. The truth is
that designing solar power systems involves a great deal of judgment before and
after any calculations are made, and any recommendation by a computer program
must be tempered by experience and a keen sense of unpredictable human and
environmental factors. Skilled system designers are critical of their own
assumptions and question their results to see if all contingencies have been
considered.
In this section, we will discuss some of the approaches that should be followed by
system designers to gathering the initial information, dealing with objections and
uncertainties, and thoroughly transferring the final system to the user. These
include:
Considering Tradeoffs
Gathering Information
Installation
4-1
Consider Tradeoffs
Most of the time, system design involves some degree of compromise between
competing and desirable qualities. There are many choices that have to be made:
type and size of equipment, location, amount of backup or redundancy, degree of
protection, level of safety, amount of complexity, initial costs or costs over time, and
so on. Factors that may influence the choices might include the clients budget, the
remoteness of the site, how critical the loads are, the sophistication of the actual
users, and future growth possibilities.
Efficiency
Initial
Cost
Lifetime
Cost
Centralized
Generation
Distributed
Generation
Discussed next is some fundamental tradeoffs that must be made constantly when
designing photovoltaic systems. Both of the extremes of the tradeoffs are desirable,
so the dilemma facing the designer is how far to go towards one side or the other.
4-2
4-3
On the other hand, if a distributed system is designed, then each site or user must
be responsible for their own energy generation and usage. If they abuse or overuse
their system, their system may fail or shut itself down. But this will not affect the
other users, and the abuser will quickly learn to be more conscious of their energy
usage. Also the geography of the overall site may favor a distributed approach. For
example, Navaho Indians in the central region of the US have their individual homes
located perhaps a kilometer or so from the nearest neighbor. Stringing power lines
to each home would prove uneconomical compared to installing separate and standalone power systems for each home.
4-4
4-5
4-6
Remoteness
Reliability
Vulnerability
Critical Loads
Low Voltage
Small Power
Noise
Pollution
There may be a desire for a modular approach to power generation, where some
power is needed now and more will be needed in the future. Instead of buying and
installing an oversized generator anticipating long term future power requirements,
you can install only the power that is needed today, and add modules at any time in
the future when budgets or load demands increase. Entire village power systems
can grow this way, starting small and growing as the community grows.
Individual modules are lightweight and can be easily transported if necessary.
Modules have been used for mobile radios, camera battery recharging, lighting for
temporary camps, as well as transported on trucks or by hand from one water pump
to another.
And certainly if there is abundant solar insolation, photovoltaics should be
considered. Why not use a resource that is plentiful, and delivered for free!
4-7
Handle Objections
Needs Maintenance
User Education
Breakable
Theft
Safety Concerns
Building Codes
Need the Sun
Initial Cost
Produces DC
Area Limitations
Certainly a common objection is the relatively high initial cost of a photovoltaic power
system compared to a generator only system. This must be handled by presenting
the true costs of operating a generator over time, including regular and unscheduled
maintenance, parts, labor, fuel (including delivery costs) and other invisible costs.
The problem of high initial cost can also be addressed by developing creative
financing arrangements, so that users pay for their energy over time just as they do
for conventional energy.
4-8
Another key objection is that photovoltaic devices make DC electricity, while most
common appliances and other loads are designed to operate from AC power. This
can be dealt with by either proposing the use of DC appliances, or by including a
small inverter that converts DC to AC. The extra cost of the inverter can be
compared to the usually higher prices of DC appliances, and a choice can be made
as to which way to go.
4-9
Load requirements
Load profile
Surges
Power quality
DC or AC
Critical loads
Ease of access to site
The individual loads must be well understood, including their current and voltage
requirements, and how many hours each day they will be on. In addition, the profile
of the load demand over a typical week or for each month must be specified. It will
be important to know if the load is greater in the summer or winter. This information
will help determine the proper tilt angle of the modules so that they intercept the
most solar radiation when the loads need it. Also if the load is used only a few days
each week, as in the case of a weekend cabin, can affect the balancing of modules
and battery storage and final system cost.
4-10
Certainly if AC loads are to be included, their requirements for power quality and
surge currents will affect the choice and sizing of a DC-to-AC inverter. The
inefficiency of the inverter must be added into the calculations, and the input voltage
of the inverter chosen will determine the voltage of the solar array and battery bank.
60
50
40
30
20
10
0
Jan Feb Mar Apr May Jun
Jul Aug Sep Oct Nov Dec
If there are any critical or essential loads this will influence the choice of sizing of the
array and battery to handle unexpected bad weather. And the ease or difficulty of
reaching the site may influence the reserve time built into the battery, as well as the
setting of any low voltage alarm signals to prompt maintenance actions.
4-11
Latitude, longitude
Insolation
Temperature
Variability of weather
Harshness
4-12
Temperature ranges as well must be well know, as modules loose voltage potential
and batteries loose life expectancy with higher temperatures. Temperatures will also
influence the choice of the module design, as 36 cells module circuits are typically
used for the hottest climates, while 33 cells in series are adequate for more
moderate climates.
Local knowledge about the variability or harshness of the weather is important to
gather. Statistical data on solar insolation is usually based on averaged data, so
particular variation patterns from day to day are lost. The severity and duration of
typical seasonal storms would be very helpful in deciding how much reserve capacity
to build into the battery bank, and whether backup generator capacity should be
considered.
4-13
4-14
Budget
Reliability of their data
Level of technical skill
Future growth possibilities
Aesthetics
You should keep in mind the probability of near-term future growth of demand, and
design your system to accommodate that growth. If the load is well known and
fixed, with no possibility of increase, then you can design your system to fit that load
value exactly. But if there is the real chance of load expansion, for example more
channels of radio being utilized, or more hours of lighting needed or more users in a
village, then expansion and increased power flows should be planned. Space
should be left in the module mounting structures, larger wires should be used,
regulators and inverters should be oversized, and so on. This will help to reduce the
need to purchase more components in the near future.
And aesthetics may have an influence on your design choices. Whether the user
wants the array to be visible or not will influence your location options and may affect
your wire lengths and sizes for example. Instead of mounting the array on the
ground at the proper tilt angle, the user may want the array to be mounted on a roof
with a tilt or azimuth angle that is not optimum. This will definitely affect your array
sizing calculations.
4-15
Array sizing
Battery bank sizing
Wiring
Safety components
User feedback
DC or AC or Hybrid
Mounting
Accomodate future growth
The number of modules in series and parallel for the array can now be calculated,
as well as the capacity and voltage of the battery bank. All of the system wiring
must be chosen, with attention to safety equipment such as circuit breakers, fuses,
disconnects and grounding. Electrical code considerations may dictate the use of
equipment that adds to the overall system complexity and cost, but also adds to
safety.
At this time you should consider the hardware needed to give proper user feedback.
Simple dials or fancy digital meters can be chosen, and even remote transmission of
system parameters can be designed.
Siemens Solar Basic PV Technology Course
Copyright 19978Siemens Solar Industries
4-16
Fundamental decisions need to be made here about whether the system is all DC,
or all AC, or some combination of the two. You have to decide whether the system
will be a pure stand-alone photovoltaic system or a hybrid design with a fuel
generator or wind or hydro assist.
The array mounting design must be made, perhaps including empty space for future
module additions. Components such as inverters charge regulators, and wiring
should be chosen to allow maximum growth, so that costly upgrades or changes can
be avoided.
4-17
Installation
You can perform all the calculations in the world, but if the equipment is not installed
correctly, your system could still fail. Bad communications between the designers
and the installers can lead to wasted time, money and faulty system performance.
As much of the system should be pre-assembled as possible and shipped to the site
ready for installation. This allows you to test the components as they work together
and make repairs or adjustments as needed at the home shop (where tools and
spare equipment is readily available). Charge controls, system disconnects,
inverters and feedback components can often be pre-mounted to a strong
backboard, transported to the site, and installed quickly.
Installation
4-18
The installation site should be visited before system design is completed so that
special circumstances can be taken into account. Perhaps there are trees or other
sources of shading at the site. This might require clever mounting solutions or
longer and larger wire than anticipated. Foundation design might depend on the
type of soil, the wind loading, the growth of vegetation and so on.
Safe practices should be followed at all times during the installation phase. All
appropriate local and regional electrical and building codes should be followed.
Each component and the overall system should be fully tested before the installation
is considered complete. Abnormal operations should be corrected, and all system
voltage drops should be carefully measured to insure that all connections are proper.
4-19
4-20
4-21
The Homeowners Manual presented on the following pages was prepared for the
owners of simple 12-volt DC home electrification systems installed on the Navajo
Indian Reservation in the United States.
4-22
(End of Chapter)
4-23
CHAPTER FOUR
GENERAL SKILLS OF SYSTEM DESIGNERS
4-1
Consider Tradeoffs
Efficiency vs. Simplicity and Reliability
Initial Cost vs. Lifetime Cost
Centralized vs. Distributed Generation
4-2
4-3
4-3
4-3
4-4
4-6
4-8
4-10
4-12
4-14
4-16
Installation
4-18
4-20
4-21
4-24
Chapter Five
Applications and Typical
Systems
In this chapter, we look at some of the most popular and economical types of
applications for solar photovoltaic power systems. There are many new applications
that are being developed, but we can discuss here those that have already proven to
be cost effective and reliable.
We can also look briefly at some common system configurations, to become familiar
with some of the equipment that is used in systems. Most all of the equipment
mentioned (such as inverters and charge controls) will be discussed in more detail in
later chapters.
5-1
Flexibility of Application
One of the blessings of photovoltaic technology is that there is a tremendous
flexibility in the kinds of designs and applications that solar power can
accommodate. There is a wide variation in the size of system, the complexity, and
the dependence or independence on natural cycles. There is not one best way to
design a photovoltaic system.
Application Flexibility
1 watt
Megawatts
Near existing
utility power
Remote
Daytime only
Purely PV
Hybrid system
Centralized
generation
Distributed
generation
5-2
Power Flexibility
The inherent modularity of solar photovoltaic systems means that there is
tremendous flexibility in the amount of power you can install. Wind generators begin
at about 500 watts of generating capacity, and fuel generators begin at a few
thousand watts. But you can install as little as a few watts of photovoltaic generating
capacity if that is all that is needed, for example to charge an electric fence energizer
or operate a small telemetry unit.
Hundreds or thousands of watts of power can be installed for homes, farms,
industrial or commercial uses, schools and clinics.
And hundreds of thousands or millions of watts of generating capacity can also be
installed economically as well. Large-scale utility systems serving thousands of
homes or businesses have been installed.
5-3
5-4
Centralized or Decentralized
Generation
Photovoltaics are inherently a decentralized generation technology. The power that
is needed can be installed where it is needed. Each house, schoolroom,
commercial site or transmitter can be outfitted with its own independent system. If
one system fails or needs maintenance, or if the user exceeds the energy available,
the other systems continue to operate without interruption.
But decentralized systems require vigilance by their owners. If the skills or
knowledge of how to operate or maintain the system were not available, then
perhaps a centralized approach would be best.
A centralized photovoltaic power system could have all the modules, batteries,
inverter capacity and controls needed for an entire village. Trained personnel could
maintain the equipment, and would be responsible for proper operation. The users
of the energy are not as directly connected to the source of their energy, but perhaps
this approach is best for their situation.
5-5
Roadside Flashers
5-6
Bus Shelters
Garden Lights
Portable Lanterns
5-7
Telecommunications Applications
Telecommunications has always been an important market for photovoltaics. The
equipment used in many telecom applications is operated on DC power, making the
match to DC generated solar power simple and economical. Reliability in severe
climates is perhaps the most important attribute of photovoltaics for this market.
Downtime is terribly expensive and may be life threatening, so reliability is critical.
And the modularity and flexibility of photovoltaic system design means that modules
can be used to power even the smallest telemetry station as well as very large
microwave repeaters.
Microwave, TV or
Radio Repeaters
Telemetry Stations
5-8
5-9
Navigation Applications
Similar to the telecom market, aides to navigation require the highest reliability.
Signals or lights must operate under all conditions, in all seasons. They are typically
located in remote sites, and traditional power sources such as utility line extensions
or diesel generators would be quite costly. Each navigational aid can have its own
photovoltaic power supply, making vulnerability to overall power loss extremely low.
And this market is a perfect example of how a relatively small amount of power can
produce a very large benefit.
Railroad Signals
Buoys
5-10
5-11
Well Heads
5-12
Livestock Watering
Irrigation
5-13
5-14
Home
Electrification
Schools
Hospitals
and Clinics
5-15
Recreational
Boats / Vehicles
5-16
5-17
5-18
Load (kW)
100
80
40
20
Computing
0
0
10
12
14
16
18
20
22
24
Hour of Day
PG&E R&D Building, June 21, 1990
5-19
With PV
90
Without PV
85
80
0
10
15
20
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
5-20
5-21
5-22
Energy:
Capacity:
Min. Load Savings:
QF Savings:
Loss Savings:
Voltage Support:
Substation:
Reliability:
Transmission:
Environment:
5-23
700
Transmission
600
Reliability
500
Substation
Voltage Support
400
Loss Savings
300
QF Savings
Min.Load
200
Capacity
100
Energy
0
PG&E
SECWA
Figure 5-4 Two utility analyses show real cost of delivered utility power
5-24
Large DC Systems
AC Power Systems
5-25
Simple DC System
Configuration
DC Loads
5-26
5-27
Simple DC System
Configuration
Floating pump: A floating pump is appropriate for canals, lakes, ponds and
even open wells.
5-28
Large DC Systems
DC
Loads
Load distribution box: If the load circuits were too numerous, a DC circuit
breaker distribution box could be used to allow connection of multiple load
circuits.
Multiple charge controllers: If the array is larger than the current capacity of
one single charge controller, multiple charge controllers can be connected in
parallel. Slightly different charge voltages can be set to allow the battery to
gradually reach full charge.
5-29
AC Power Systems
DC
Loads
Inverter
Inverter
AC
Loads
Inverter operates AC loads: AC appliances can be powered by adding an ACto-DC inverter. This complex component switches electronically the DC voltage
and current from the battery to produce alternating current (AC) and voltage,
commonly used by household and office equipment. Such a power system could
then operate any load that might be needed, including computers, fax machines,
radio, TV, VCR and CD players, refrigerators and freezers, power tools and
kitchen and bathroom appliances. DC loads could still be included in the system
as well.
5-30
Photovoltaic-Generator Hybrid
Systems
AC
Loads
Rectifier charges battery: The generator can also pass power through a
rectifier that changes the AC back to DC current and voltage. This DC power is
passed to the battery bank, to recharge the batteries after a long period of below
average weather. A smaller battery bank can be installed compared to standalone systems because system autonomy is now provided by the generator.
5-31
AC
Loads
!"
DC charging only: A hybrid system can be designed to have the generator act
only as a battery charger. No AC output from the generator is used to run loads.
Instead all the AC power for the loads is output only from the inverter. The
generator is turned on, either automatically by the charge control system or
manually by the user, whenever the battery voltage gets too low during bad
weather.
Avoid transfer switch: Operating all loads from the inverter means there is no
transfer switch glitches that could harm electronic equipment.
Operates generator at full power: The generator is sized so that it can operate
at its full rated output to charge the battery. When the batteries are sufficiently
recharged, the generator is turned off, and the finishing charge is supplied by the
solar modules. Operating at full output means maximum fuel efficiency and long
life.
Battery life extended: The generator is turned whenever needed to keep the
battery from staying discharged for more than a few days. Battery life is
maximized due to minimal sulfation.
5-32
Characteristic
PV only
Dependence on
natural cycles
Cost effective
size
highly
dependent
below 20-30
kwh/day
Reliability
excellent
Fuel
none
Maintenance
annually or
semi-annually
frequent
moderate
good
slow
fast
Initial cost
high
low
Operating cost
low
high
Handling load
peaks
Battery charge
rate
5-33
Diesel only
independent
above 150
kwh/day
depends on
maintenance
frequent
deliveries
!"# $!%&
Instead of having a system independent of the utility grid a system can be designed
to work with the grid. A specially designed utility-interactive inverter is needed, and
many models are available worldwide. The basic arrangement of a utilityinterconnected system is quite simple.
The solar array is connected to the inverter, as are the utility lines. The output is
connected to the normal distribution box for the house or business.
During the day power generated by the array is fed into the inverter and changed
into pure sinusoidal AC power that is synchronized with the grid frequency.
If that power is needed in the home it is passed on. If the load demand is less
than what the array is producing, the excess is fed into the utility grid system, and
energy is credited to the home.
If more power is needed than the array can produce at a particular moment then
power is drawn from the utility to add to the array power.
Typically there are no batteries in utility-interconnected systems, so at night all
the power needed flows from the utility.
5-34
Recently, new designs of bi-directional inverters have been created that allow utilityinterconnected systems to have battery backup as well. AC power from the utility
passes through the normal meter and into a standard distribution center.
Synchronized sinewave power from the inverter is connected to this distribution
center as well. Inverter power is also available from a second AC output connection
and can be sent to a dedicated distribution box for critical loads.
Critical Loads
(Operate from
battery)
Regulator
Battery
Inverter
Meter
Non-Critical Loads .
(Operate only
from utility
power)
!"%$&' $!%&
The inverter continuously shares between AC utility power and DC battery power.
AC power from the utility can be rectified into DC power and used to recharge the
battery bank during bad weather.
Battery power is drawn upon instantly if there is a utility power failure or
brownout, much like an uninteruptable power supply (UPS) system.
The non-critical load distribution center looses power when there is utility failure,
but the critical center continues to draw power from the inverter and battery.
5-35
(End of Chapter)
5-36
CHAPTER FIVE
APPLICATIONS AND TYPICAL SYSTEMS
5-1
Flexibility of Application
Power Flexibility
Near Utility Power or Remote
Daytime Only or Anytime
Photovoltaics Only or Hybrid Generation
Centralized or Decentralized Generation
5-2
5-3
5-3
5-4
5-4
5-5
5-6
5-6
5-6
5-7
5-7
5-7
Telecommunications Applications
Microwave, TV or Radio Repeaters
Telemetry Stations
Radio and Telephones
5-8
5-8
5-8
5-9
Navigation Applications
Railroad Signals
Buoys
Airport Approach Systems
Offshore Oil Platforms
5-10
5-10
5-10
5-11
5-11
5-12
5-12
5-13
5-13
5-13
5-13
5-37
5-14
5-14
5-15
5-15
5-15
5-15
5-16
5-16
5-16
5-16
5-17
5-17
5-17
5-19
5-20
5-20
5-22
5-22
5-23
5-25
5-26
5-28
5-29
5-30
5-31
5-34
5-38
Chapter Six
The Physics of
Solar Cells
The process of direct conversion of light into electricity seems almost magic. In this
chapter, you will learn about the basics of how this process occurs, without the use
of intimidating formulas or confusing jargon. The purpose of this explanation is to
make you more comfortable with the fundamentals of how silicon solar cells work.
Other semiconductor materials can be used to make PV devices, but describing a
silicon-based device will illuminate the general principles common to all.
6-1
Silicon Atom
nucleus
outer electron "cloud"
ee-
ee-
6-2
When single silicon atoms are combined, they connect together to form a solid.
Neighboring atoms share outer electrons, forming bonds. These bonds where
electrons are shared between atoms is what holds all matter together. Crystalline
silicon consists of orderly bonding of each silicon atom with 4 neighboring silicon
atoms. Such a highly ordered structure of atoms is also called a crystal lattice.
A model of crystalline silicon is shown below. Each atom is represented by a large
ball, and the interconnecting rods represent the shared electron bonds. Each
silicon atom is connected or bonded to four other silicon atoms, forming a beautiful
repeating lattice structure. The lattice is the same arrangement as found in a
diamond, except that silicon atoms are present instead of the carbon found in a
diamond.
6-3
At the atomic level, light acts as a flux of discrete particles called photons. Photons
carry momentum and energy but are electrically neutral. When semiconductor
material is illuminated by light photons of light actually penetrate into the material,
traversing deep into the solid.
Photons with enough energy that collide with electrons can dislodge them from their
bond. The photon disappears and its energy is transferred to the electron, which
becomes free to wander throughout the semiconductor material as a conduction
electron, carrying a negative charge and usable energy.
It is at the moment of releasing the electron that sunlight energy has been converted
into electrical energy.
e-
e-
e-
ee- ee-
ee- ee-
e-
ee-
e-
eee-
e-
eee-
e-
e-
ee- ee-
ee- ee-
e-
eee-
e-
eee-
e-
eee-
e-
e-
ee- ee-
ee- ee-
ee- ee-
ee- ee-
ee- ee-
e-
e-
e-
e-
e-
e-
e-
e-
e-
6-4
6-5
P-Type Dopants
In the absence of any external electric field electrons freed and energized by
photons will wander for a short time and then recombine with a wandering hole.
The energy originally transferred to the electron from the photon is simply lost to the
semiconductor lattice as heat. The key to producing usable output current is to
sweep the freed electrons out of the material before they recombine with a hole.
When there is just silicon atoms in the lattice, the material is called intrinsic or pure.
One way to alter the electrical properties of silicon is to introduce elements into the
intrinsic semiconductor that contribute excess holes and electrons. Materials that
significantly alter the properties of semiconductors are called dopants and the
process of placing them into the semiconductor is called doping.
Nearby electrons
can move in and
"fill" hole.
ee-
e-
e-
e-
e-
ee-
e-
e-
e-
e-
e-
e-
e-
ee-
ee-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e- B eee-
e-
e-
e-
ee-
e-
e-
e-
e-
e-
e-
e-
e-
6-6
One dopant used with silicon is boron, which has only three outer electrons, one less
bonding electron than silicon. Each boron atom can only bond with three
neighboring silicon atoms, leaving one bond half-complete. A nearby electron from
a silicon atom can vibrate into the hole site next to the boron, filling the hole. But it
leaves behind an absence of an electron, a hole, where it originally came from.
Another nearby electron can vibrate into that hole, but leaves behind a hole where it
came from. So there exists in the semiconductor structure a wandering absence of
an electron.
Wherever this absence of electron (hole) is there is one too few electrons to balance
the positive charges of the nucleus of that particular atom. This results in a net
positive charge at that atomic site. As the hole site moves around, so does the net
positive charge. Since the moving charges in this material have a positive charge,
the semiconductor material is called positive-type or P-type.
The concentration of boron is quite low, usually around one boron atom to every
10,000,000 silicon atoms.
It is important to understand that the overall net charge in the semiconductor is
neutral, but if you look at small regions, there will be net negative charges at the
boron atom sites and net positive charges moving around with the holes. The boron
atoms will have a negative charge because a neighboring electron has fallen into the
incomplete bond and brings with it a net negative charge. The local negative charge
at the boron atom is balanced by the local positive charge near the hole, this being
now a silicon atom that has temporarily lost one of its outer electrons.
6-7
N-Type Dopants
Another dopant used with silicon is phosphorous. Each atom of phosphorous has
one more outer bonding electron than silicon. The fifth electron breaks away from
the phosphorous atom easily because there is not bond to hold it. It becomes a
freely moving negatively charged particle. Since the moving charges in this type of
semiconductor carry a negative charge, this type of doped semiconductor is called
negative-type or N-type.
The concentration of phosphorous atoms is again quite low, but typically greater
than the boron concentration, usually around one phosphorous atom for every 1000
silicon atoms.
e-
e-
e-
e-
e-
e-
e-
e- ee- e- e- P eeee-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
ee-
e-
e-
e-
e-
ee-
e-
e-
e-
e-
e-
e-
e-
e-
Once again, the net charge of the entire semiconductor is neutral. But in small local
regions you can find fixed net positive charges where phosphorous atoms are
permanently missing their fifth outer electron. And there will be moving negative
charges that are those fifth outer electrons, freely wandering throughout the lattice.
6-8
e-
e-
e-
e-
e-
ee-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
ee- B
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
ee-
e-
e-
e-
ee-
e-
e-
e-
e-
e-
e- ee- P ee-
e-
ee-
ee-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
e-
6-9
The strength of the internal electric field is quite strong. The distance across the
region of the field is only about one micron (1/10,000 of a centimeter, or 1/25,000
inch), but the field potential is about one volt. This means that if the field were to
extend across one inch it would be about 25,000 volts! This strong field is
equivalent to an electronic broom that can sweep freed electrons out of the cell
and create the one-way flow of electrons that can be called electric current.
6-10
Lines of force of
permanent internal
electric field
h+
photon
eElectrons
return from
eexternal
Freed
electron
circuit to
fill holes.
Electrons
flow on to
next cell
or out of
module
and into
external
circuit.
In a module, a number of cells are connected together in series. The electrons flow
from one cell into conductors that carry them to the next cell. In the next cell they
are once again struck by photons, given more potential energy, and swept out of the
cell. Finally the electrons leave the last cell in the module and flow to the load.
6-11
For every electron that leaves a cell, there is another that is returning from the load
to replace it. The wire that is used to make the circuit from the module to the battery
or load and back to the module contains electrons, so as soon as an electron leaves
the last cell in a module and enters the wire, an electron at the other end of the wire
moves into the first cell in the module.
So the PV device cannot run down like a battery, nor can it run out of electrons.
It produces output (electrical energy) in response to input fuel (light energy). A PV
cell cannot store electrical energy, it can only convert light energy into electrical
energy.
6-12
6-13
6-14
Exercises
!
6-15
"
#
$
%
6-16
Spectral Response
Different sources of light may appear equal in brightness to the human eye but will
contain different amounts and intensities of colors. For example, fluorescent light is
typically stronger in the blue than incandescent lights. Also throughout a day there
is a difference in the spectral content of morning, noon, and evening sunlight, as can
be observed by looking at the sky.
Light is just a narrow range of all electromagnetic radiation that is emitted by the
sun. Radiation is a moving electric-magnetic field, and the field vibrates regularly at
a very rapid pace. The distance between the peaks of the vibrations of radiation is
called the wavelength of that radiation. Light is radiation between approximately
400 nanometers or nm (violet color) and 800 nm (red color).
Standardized Solar
Spectral Distribution
UV
Power
Density
Visible
Light
IR
1.2
( mw/cm2/nm)
0.8
0.6
0.4
0.2
200 400 600 800 1000 1200 1400
Wavelength ( nanometers, nm )
6-17
Spectral Response
Q
u
a
n
t
u
m
E
f
f
i
c
i
e
n
c
y
1.00
CZ
0.80
CIS
TFS:H
0.60
0.40
0.20
0.00
0
0.2
0.4
0.6
0.8
1.2
Wavelength (m)
Light of the same color or wavelength will produce different amounts of current in
different semiconductor devices. The spectral response of a typical CZ silicon cell
begins about 350 nm, peaks around 800 nm and falls off rapidly beyond 1100 nm.
The response range spans the entire visible spectrum and reaches into the near
infrared. The spectral response of a typical cell made from thin film
silicon:hydrogen, on the other hand, is narrower, beginning around 300 nm, peaking
around 500 nm, and falling off beyond 700 nm. And a third type of PV device
material, copper indium diselenide or CIS, has yet another spectral response range,
beginning just short of the other two but extending further into the invisible infrared
region.
Because a cells response to light depends on the wavelength of that light, just
knowing the total energy of the light is not enough information to predict cell output.
Two sources of light can have the same total energy density but one source could
have its energy in the form of a few high energy blue photons, while another could
6-18
contain many low energy red or infrared photons. Also, two light sources can
appear to be similar in brightness to the human eye, but one may emit a great deal
of extra radiation beyond the visible range where our eye will not notice but to which
the PV device will respond. A standard typical outdoor spectrum has been defined
as the spectrum from the sun that filters through 1.5 thickness of our atmosphere,
and is referred to as Air Mass 1.5. This serves as a common reference spectrum
for comparing device output.
Exercises
"
&'((#
)*
!+,-$./
!
0
+,-
$.
6-19
(End of Chapter)
6-20
CHAPTER SIX
THE PHYSICS OF SOLAR CELLS
6-1
6-2
P-Type Dopants
6-6
N-Type Dopants
6-8
6-9
6-11
6-13
6-14
Spectral Response
6-17
6-21
Chapter 6 Answers
The Physics of Solar Cells
The photovoltaic effect describes the property of certain materials to produce electrons
when light falls on (and is absorbed by) the material's surface.
Electron, positive (+)
b. An electron around a silicon atom
c. Get the electrons out before they recombine with a hole
N, negative, electron
P, positive, "hole" (absence of an electron)
Negative, positive
N, phosphorous
b. Convert energy
6-1
Photovoltaic Physics
c. Light particles (photons)
d. Receive electrons from the circuit as other electrons leave the cell and pass into the
circuit.
b. CZ Cell (refer to spectral response graph)
6-2
Photovoltaic Physics
Chapter Seven
Module Manufacturing
and Testing
The manufacturing process used by Siemens Solar involves many proprietary elements
and is quite involved. But it is important to discuss the manufacturing process so that
you can appreciate the technical features and benefits of the modules, and can have
confidence in the durability and longevity of the products.
The manufacturing process can be presented as three segments or phases: crystal and
wafer fabrication; cell manufacturing; and module assembly. Each of these processes
is discussed next.
Manufacturing Flow
Crystal and
Wafer Fabrication
Cell Manufacturing
Module Manufacturing
Change quartz
into pure silicon
Diffuse with
phosphorus
Solder cells
together
Laminate
Print contacts
Attach frame,
j-box, diodes
Etch surface
Electrical test
Final electrical
test
7-1
Wafer Manufacturing
Raw Materials
The process of making a solar module ultimately begins by mining the raw material.
This usually consists of high quality silica or quartz (SiO2) from mines around the world.
Silicon (Si) is an abundant element on the earth, and can be found in common sand,
but it is more economical today to begin the process with silica that has been somewhat
naturally purified. In large furnaces, the silica is reduced to silicon and then purified
until it becomes 99.9999% pure! Examples of these purified rocks of silicon are shown
below.
7-2
Crystal Growing
Chunks of purified silicon are carefully loaded into a crystal growing furnace, along with
a small amount of the element boron. The furnace is sealed and the chunks are heated
to greater than 1400 deg.C. (over 2500 deg.F.) until they melt (Figure 7-4). A small
seed crystal of pure silicon about the size of a pencil is lowered inside the furnace until
it touches the liquid. The cooler seed acts as a template and atoms of silicon and
boron freeze onto it, and the seed begins to grow (Figure 7-5). By carefully controlling
temperatures and the speed of growth, the diameter of the growing crystal is increased
to about 5" and then held steady by computerized controls. The crystal is then grown in
length until almost all of the original silicon material is exhausted.
This process of growing a single crystal structure from a small seed crystal is called the
Czochralski method, usually abbreviated as just CZ.
7-3
Cropping
After removal from the crystal growing furnace, the ingot top and bottom (crown and
tail) are cropped off and recycled (Figure 7-6). The ingot is cut into short sections for
easier handling. The sides of the section are cropped along its length to make a
squared-off block. The M line of cells is cut almost square for the greatest packing
density in a module. The Pro line of cells is cut less, leaving more of a rounded shape
(Figure 7-7).
7-4
7-5
Slicing Wafers
The section is mounted into a wire saw to cut individual wafers. The wire saw involves
a single long thin wire wrapped many hundreds of times around four rotating drums.
The ingot section is placed in the center of the four drums, and is slowly pushed up
through the web of the wires. Hundreds of thin silicon wafers are cut all at the same
time this way.
Surface Etching
The wafers are then etched to remove some surface damage caused by the wire saw
abrasion, and to create a surface that helps absorb light (Figure 7-9). The final wafer
surface is made of millions of tiny 4-sided pyramids, following the pure crystal structure
in the original seed. These pyramids reflect light amongst themselves and allow more
light to be absorbed (Figure 7-8).
7-6
Cell Manufacturing
Diffusion
Wafers are next loaded into diffusion tubes which are long tubes of silica glass,
surrounded by large resistive heaters, and that will contain a hot gas of phosphorous. A
computer controls the entire process, slowly moving wafers into the chamber, allowing
gases to flow for precise amounts of time, and then removing the wafers (Figure 7-10).
This process is conducted in clean room conditions similar to the semiconductor
industry. Clean room conditions are important in optimizing the performance of the
cells.
Inside the tube atoms of phosphorous penetrate or diffuse into the wafers a short
distance, only 12 millionths of an inch or 0.3 microns, and embed themselves in the
crystal structure. This embedding of select impurities into a semiconductor like silicon
is called doping.
The differences between phosphorous-doped silicon and the base material of borondoped silicon set up a permanent electric field in each wafer. This field sweeps out
electrons that are knocked loose by light, and makes the wafer into an active solar cell
when light shines on it.
7-7
7-8
The metallic gridlines are the final structure that makes up a complete cell.
Structure of CZ Cell
Metallic
Gridlines
Anti-reflective
Coating
150 nm
300 nm
N-type layer
Electric Field
at the p/n
junction
P-type layer
250000 nm
Metallic
Gridlines
Cell Operation
e-
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
7-9
Every cell is tested under simulated noonday sunlight to determine its electrical output,
and cells of similar output are sorted and grouped together (Figure 7-14).
7-10
Module Manufacturing
Interconnecting Cells
Cells are connected together in series to add voltage. This is done by an automated
soldering machine that connects the back of one cell to the front of the next with two
ribbons of tin-coated copper ribbon (Figure 7-15). The cells are soldered to the ribbons
at multiple points to allow for expansion and contraction of the copper against the
silicon. Two ribbons are used to add reliability to this critical interconnection.
Lamination
Strings of 10, 11, or 12 cells are soldered together to create final groups of 30, 33, or 36
cells, and then are laminated into a sandwich of support materials designed to insure
long operating life. The sandwich consists of tempered glass, then one layer of EVA
(ethylene vinyl acetate), the cell circuit, another layer of EVA, and finally a back cover of
multiple polymer sheets that prevent moisture penetration. The interconnected cell
circuit is therefore floating in a sea of plastic. The sandwich is first sealed in vacuum
laminators that remove moisture and air and soften the plastics, and is then thermally
set in large ovens.
7-11
Every completed module is tested again under simulated noonday sunlight to insure it
meets minimum power output specifications. The output is recorded on computer for
quality control and customer assurance. The modules also are tested for safe high
voltage operation by putting 3000 volts between the circuit and the metal frame. All
modules are also visually inspected for blemishes or flaws before being packaged 4 to
a box for shipment.
e-
CZ cell
Interconnecting
copper ribbons
7-12
Exercises
%
&'
(
))))))))))))))))))))))))))))))))))))))))))))
7-13
Qualification Testing of
Module Designs
PV power systems must be reliable even in the harshest environments and operate for
many years to be economically justified. Siemens Solar currently guarantees module
output for 10 years, and life cycle cost analyses are done based on module life of 20
years or greater. To be confident to make such long-range projections, module
manufacturers subject their modules to intense testing to qualify their electrical and
physical design. These tests are not normally done on every production module
because they are destructive tests and they can cause some discoloration or
degradation to the tested modules. These tests described below are qualification
tests which means that they are used to qualify an existing or new module design as
a truly robust, long-life design, suitable for installation in any climate for many years.
These tests are typically done on any new module design or when changes are made
to existing designs. Module manufacturers may have their Quality Assurance
Department perform some of these tests on an ongoing basis on samples of regular
production modules, to continually check that manufacturing processes are producing
quality products.
!"#
7-14
7-15
Damp Heat:
Modules subjected to 1000 hours continuous of 85% relative humidity at 85oC. This
simulates severely humid tropical environments and will reveal any weaknesses to
moisture penetration.
Thermal Cycling:
Modules are heated to 85oC then cooled to -40oC for 200 cycles. This is a greater
temperature swing than any real module on the earth would experience. This tests for
effects of expansion and contraction of contacts and lamination.
Humidity-Freezing:
Modules are thermal cycled from 85oC to -40oC in 85% relative humidity for 10 cycles.
This tests for ingress of moisture that could result in corrosion and deterioration of the
laminating plastics.
Twisting:
Modules are fixed at three corners and the forth is lifted approximately 1" (3 cm), to
again simulate wind conditions and torquing, to check for cell breakage and loss of
electrical contact.
7-16
Construction Integrity:
Insulating materials, current carrying parts, internal wiring wireways, assembly and
installation factors, connecting means, bonding, material compatibility, spacing, wiring
compartments, corrosion resistance, sharp edges, accessibility, fire resistance and
encapsulation.
Performance Tests:
Temperature, voltage and current measurement, leakage current, strain relief, dielectric
voltage withstand, inverse current overload, installation/maintenance, impact, fire,
exposure to water spray, accelerated aging of gaskets and seals, temperature cycling,
humidity cycling, metallic coating thickness, hot spot endurance, arcing, mechanical
loading, junction cover crush resistance.
Performance Tests
impact, inverse current, water,
crushing, gaskets
MAX. POWER
53 WATTS
30B9 LISTED
SHORT CKT.
3.35 A
RATED
3.05 A
OPEN CKT.
21.7 V
RATED
17.4 V
FIRE RATING
CLASS C
SERIES FUSE
5A
FIELD WIRING
BYPASS DIODE
COPPER ONLY, 14 AWG MIN. INSTALLATION GUIDE
INSULATED FOR 75 C MIN.
233-701500-20
MADE IN U.S.A.
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
7-17
Shock:
Modules are lifted at one edge to a height of 4" (10 cm) and dropped.
Twisting:
This twist test is similar to the CEC test where one corner is lifted approximately 1
(3cm).
Termination Robustness:
The module is suspended by its electrical output cable.
Temperature Shock:
Modules are cycled 3 times between 71oC. and -57oC. with a transition time of less than
5 minutes. This is more severe than the gradual temperature transitions of the thermal
cycling described earlier, and would simulate modules mounted on offshore buoys
being tipped into freezing arctic waters.
Vibration:
Modules are subjected to varying frequency of vibrations from 5 to 200 Hz for 84
minutes along each of the three axes.
7-18
7-19
6 or 12 volt Configuration
The new large single J-box allows for field adjustable operation at either 6- or 12-volts.
A wiring diagram is molded into the terminal cover for permanent reference.
$"#%
"
7-20
7-21
UL Approval
The Underwriter's Laboratory approval gives extra assurance to developers, installers,
lending institutions, and customers that the technology is proven and safe.
7-22
An analysis was done in 1994 for our popular SM50-H module. The total returns due to
module failure were accumulated for a ten year period, and compared to the total hours
of operation worldwide for all the SM50-H modules shipped and installed during that
time. With only 385 modules returned for failure during the ten years, and a total
estimated operating time of over 637,000 hours for all the 181,000 modules shipped,
the failure rate calculates to be only .069 failures in one million hours of operation! Or
about one failure in 14 million hours!
It must be emphasized that a complete power solution usually involves more than just
modules. Batteries, charge regulators and control systems, fuses and circuit breakers,
diodes, and of course load devices are involved. All of these other components have
failure rates higher than solar modules, and care must be taken when designing and
installing photovoltaic power systems to minimize their effect on overall reliability.
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
7-23
Exercises
*
!
!
+
!
%
,
-.
))))
/
))))
))))
(
012(
))))
3
))))
7-24
))))
))))
))))
))))
))))
7-25
(END OF CHAPTER)
7-26
CHAPTER SEVEN
MODULE MANUFACTURING AND TESTING
7-1
Wafer Manufacturing
7-2
Cell Manufacturing
7-7
Module Manufacturing
7-11
7-14
7-15
7-17
7-18
7-19
7-23
7-25
7-27
Chapter 7 Answers
Module Manufacturing & Testing
The incorrect steps are:
f. 20% of all modules made are tested for output power (100% of the modules are
tested for output power)
j. Boron is added to each wafer before printing (boron is added when the silicon is
melted to make crystals)
The proper order for the steps is:
l - d - b - m - g - i - a - e - c - k - h
Condition
A.
B.
C.
D.
E.
Test
h
e
g
a
b
7-1
Chapter Eight
Output Curves
The single most important technical aspect of photovoltaic cells and modules is the
current-voltage output curve. Understanding the curve allows a system designer to
anticipate how a module will be influenced by the environment, and how a particular
load will interact with the module. The output curve also shows how the module
output is "bounded" and therefore safer than battery or generator in the case of short
circuits. As you read this chapter sketch the curve often, label points on the curve,
and sketch how the curve might be influenced by the environment. This will help in
understanding sizing and wiring concepts presented later in this book.
8-1
60.00
Isc
50.00
3.00
Imp
2.50
40.00
2.00
30.00
Power
1.50
20.00
1.00
Voc
Vmp
0.50
0.00
0.00
5.00
10.00
15.00
20.00
Power (watts)
Current (amps)
3.50
Current
10.00
0.00
25.00
Voltage (volts)
8-2
8-3
SM Modules
5.00
4.50
Current (amps)
4.00
3.50
3.00
SM55
2.50
SM50-H
2.00
SM1
SM6
SM46
SM20
1.50
1.00
0.50
0.00
0.00
5.00
10.00
15.00
20.00
25.00
Voltage (volts)
It is the number of solar cells connected in SERIES that determines the maximum
voltage of a module. Each silicon solar cell can produce a maximum of about 0.6
volts at full sunlight. So 36 cells in series will result in a maximum voltage of about
21 volts. Some modules are made with 33 cells for a maximum of about 19 volts.
And modules made with only 30 series cells will produce a maximum voltage of only
18 volts.
And it is the SIZE (area) of a single cell that determines the maximum current
output. The Siemens Solar modules are made from two basic sizes of cell. The
square cell used in the SM-series of modules is about 104 mm in across, and
produces a maximum current of about 3.4 amps. The larger cell used in the SP
series of modules is about 125 mm across and produces about 4.8 amps maximum.
8-4
To create a module that produces less current than the standard cell, cells are cut
into half or quarter, to give 1/2 or 1/4 the standard current. This is the case for the
Siemens Solar SM20, SM10, and SM5 (with 1/8 cell size). And the Siemens Solar
SP36 and SP18 are made of 1/2 and 1/4 cells cut from the larger SP75 cells.
SP Modules
5.00
4.50
SP75
Current (amps)
4.00
3.50
3.00
SP36
2.50
2.00
1.50
1.00
SP18
0.50
0.00
0.00
5.00
10.00
15.00
20.00
25.00
Voltage (volts)
8-5
Exercises
Manufacturer
and model
Isc
Voc
Imp
Vmp
Pmax
# Series
Cells
Cell Size
8-6
Effect of Environment on
Module Output Potential
The IV curve is really just a "snapshot" view of the potential output of a photovoltaic
device under static environmental conditions. If the environmental conditions are
changed, the output potential of the device changes. The three main environmental
conditions that we will examine are:
Voc drops
slowly with
lower
irradiance
Lower irradiance
reduces current
8-7
The Isc is DIRECTLY PROPORTIONAL to the light intensity, and the Voc varies
more slowly in a logarithmic relationship. In other words, the ratio of the Isc to the
light intensity or irradiance will be the same. If the light intensity is halved, the Isc
will drop to half.
Isc1
Light intensity1
Example:
Isc2
Light intensity2
A module has a rated Isc of 3.4 amps at 1000 w/m2. We can calculate
the Isc at another light level by taking the proportion of the new light
level to the standard level of 1000. If the actual light intensity on a
module were 850 w/m2, then the actual Isc would be given by the
proportion:
Isc1
Light intensity1
Isc2
Light intensity2
3.4 amps
1000 w/m2
Isc2
850 w/m2
Isc2
2.9 amps
8-8
1000 watts/m2
Cloudy Day
Clear Day
1000 w/m
1000 w/m
8-9
Effect of Temperature
As the cell temperature rises, the main effect is to reduce the voltage available at
most currents. There is a slight rise in current at very low voltages. The change in
voltage is DIRECTLY PROPORTIONAL to the rise in temperature. It is important to
note that these factors refer to cell temperature, not just ambient or air temperature.
The relationship between air temperature and cell temperature depends on light
intensity, and is discussed later in this chapter.
Effect of Temperature
Isc rises slightly as temperature goes up
Higher temperature
reduces voltage
!
Manufacturers of modules anticipate the lost of voltage in real world hot conditions,
and compensate by building modules with enough cells in series so that even when
very hot, the module has enough voltage to charge batteries or operate the load
device.
The general formula for determining the change in voltage with temperature is given
on the next page. The effect of temperature on Voc for one cell is multiplied by the
amount of temperature change and the number of cells in series. This will give the
voltage change for the whole module. Subtracting this voltage change from the
voltage at one temperature gives the voltage at another temperature.
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
8-10
Siemens Solar
103 mm cell (104 cm2)
Parameter
Change
per oC
% Change
per oC
Change
per oC
% Change
per oC
Voc
-2.15 mV
-0.36 %
-2.15 mV
-0.36 %
Vmp
-2.18 mV
-0.44%
-2.19 mV
-0.45 %
Isc
2.06 mA
-0.04%
1.20 mA
-0.04 %
Imp
-4.37 mA
-0.10%
-3.23 mA
-0.10 %
Pmax
-9.53 mW
-0.45%
-7.08 mW
-0.47 %
"#
SM55
SP75
SM50-H30
(33 cells)
Voc
-0.077 volt / oC
-0.077 volt / oC
-0.071 volt / oC
Vmp
-0.079 volt / oC
-0.078 volt / oC
-0.072 volt / oC
Isc
1.20 mA / oC
2.06 mA / oC
1.20 mA / oC
Imp
-3.22 mA / oC
-4.42 mA / oC
-3.25 mA / oC
Pmax
-0.255 watt / oC
-0.345 watt / oC
-0.234 watt / oC
"#
8-11
Voltage Change
Example:
A module has 36 cells in series for a Vmp of 17.3 volts at 25 oC. In real
world conditions, the cells will easily heat up to 50oC. Using the factors
presented, and multiplying by 36 cells and 50 - 25 = 25 oC.
temperature difference, we get
Vmp Change
Vmp2
which is still enough to fully charge a typical "12 volt" battery that
actually needs up to 15 volts to reach full charge.
8-12
- 0.45 % / oC
The formula for determining the effect of temperature on overall device power is
given below. Use the overall reduction in Pmax factor given above and multiply by
the amount of temperature change. This will give the percentage of power change.
Example:
= -0.45% / oC
= -.45% / oC
X
X
(50 oC - 25 oC)
25 oC
= -11.25% change
Pmax2
= Pmax1
(1 - % Power Change)
100
8-13
( 1 - 11.25 )
100
25 oC
8-14
SM55
36 cells
Current (amps)
2 5 d e g .C .
2 .0 0
47 deg. C.
1 .0 0
65 deg C.
0 .0 0
SM50-H
33 cells
Current (amps)
0 .0 0
5 .0 0
1 0.00
2 0.00
2 5.00
Volta ge
4.00
3.00
2.00
1.00
0.00
0.00
1 5.00
25 deg. C.
47 deg. C.
65 deg. C.
5.00
10.00 15.00
Voltage
20.00
25.00
$
We can see in the figures that the 36-cell module (Siemens SM55 in this example)
are less affected by high operating temperatures than the 33-cell module (Siemens
SM50-H). The 36-cell module drops from 3.3 amps at 25 oC to only 3.0 amps at 65
o
C, while the 33-cell module drops from 3.25 amps at 25 oC to less than 2.5 amps at
65 oC. However the 33-cell module drops to only 3.0 amps at a more moderate 47
o
C. Thus we see that a 36-cell module is designed to operate well in the hottest of
climates, while a 33-cell module performs well in more moderate temperatures.
8-15
Exercises
2
%
% ,$
#$$$- .
2
%!$- //////////
0
*1
#
,
% *
2
2- 1
,
*
2
3$-
% ,#
%
2
#$$$-
#
,
2
*%#$$$-
8-16
Sun at
mid-morning
8-17
A standard set of atmospheric conditions has been established for the photovoltaics
industry by NREL, the US National Renewable Energy Laboratory (formerly SERI,
Solar Energy Research Institute) in Golden, Colorado, for the U.S. Department of
Energy. This standard is listed as ASTM #E892, and among other factors is a
spectral distribution that is equivalent to light passing through 1.5 thickness of
atmosphere, or Air Mass 1.5 (AM 1.5).
8-18
1000 watts/m2
Cell Temperature
25 oC
ASTM #E892 or
Air Mass 1.5
8-19
Exercises
)
:
;
9
:;
//////
<
//////
&
//////
=
//////
>
//////
5
//////
//////
@
//////
,&
3&
#$$$-
5$28-&
#9
1
#9
1
#$$$
#$$$-
A(9
8-20
Irradiance
Ambient Temperature
Wind Speed
Module at Open Circuit
800 w/m2
20 oC
1.0 m/s maximum
1000 w/m2
NOCT
800 w/m2
NOCT
%
8-21
Current (amps)
4
3.5
STC
SOC
2.5
NOC
2
1.5
1
0.5
0
0
10
15
20
25
Voltage (volts)
8-22
Efficiency
Output
Input
Example:
Pmax
Total Device Area X Input Light Power
Pmax =
55 watts
Length
Width
=
=
1.293 m
0.330 m
55 watts
1.293m X 0.33 m X 1000 w/m2
0.129 or 12.9%
8-23
From the example, we see that the overall efficiency of the example module is
12.9% under full sun conditions. In other words, of the total solar radiation that falls
on the total module are (including all the inactive areas like frames, inter-cell spaces,
and gridlines) 12.9% of that energy is output as electrical power. This may seem
like a low figure, with about 87% of the incident power not being converted, but
remember that the fuel for the electrical power is free sunlight. This efficiency ranks
amongst the highest module efficiencies in the world, and is the direct result of the
single crystal technology used in the Siemens Solar cells.
Pmax
Active Area Only X Input Light Power
8-24
Exercises
7
.
*
E
8-25
8-26
3.4 amps
...36 cells in series...
0.6 volts
each cell
21 volts
If the main application for the modules were not 12-volt battery charging, then some
other number of cells might be more appropriate. For example, if the ultimate
application was to be direct connection to utility power at voltages of 120 or 240
volts, then there would be no reason to keep the voltage of the modules around 12
volts. Modules with a Vmp of 40 or 60 volts could be designed. The ultimate
application voltage determines the number of cells needed in series. The most
common market for large power photovoltaic modules is still 12 volt battery charging,
so that is why most manufacturers produce modules with 30-36 CZ cells in series.
8-27
If more voltage or current than one module can produce is needed, modules can
also be connected in series and parallel to achieve practically any final voltage and
current. Systems producing 600 volts DC and hundreds of amps have been
installed, and successfully operated for years.
84 volts
For example, in the drawing above there are 4 modules connected in series to make
a nominal 48-volt string, and two strings are connected in parallel to increase current
output. The composite curve for the entire 8-module array can be constructed by
either adding voltage-wise first and then adding current-wise, or visa versa.
8-28
If we add voltage first, the 4 modules in series are added voltage-wise to reach out
to beyond 80 volts at Voc, and then the 2 parallel strings are added current-wise to
show that the current adds.
If we use the values calculated previously for a 36-cell module, the resulting curve
for the entire array has values given below:
X 4 series modules
X 4 series modules
8-29
Exercise
9?
* !
*2
9
.
*:)
;
1
1 E///////////////////////////////////////////
%//////
//////
% //////
10
16
10
14
16
14
12
//////
12
10
10
2
1
0
0
20
40
60
0
0
80
a.
20
40
60
0
0
b.
20
40
60
80
c.
20
40
60
d.
0%
1
9
*
16
14
(1 )
12
10
(3 )
8
6
4
s in g le c e ll
(4 )
(2 )
0
0
10
20
30
40
50
8-30
/////
Cell Temperature
Irradiance
Spectral Distribution
Load Interaction
How these different types of loads interact with a solar module and determine the
actual output are discussed next.
8-31
40
30
20
10
0
0
200
400
600
800
1000
1200
Irradiance (Watts/m2)
8-32
Knowing the air temperature and the irradiance level, the cell temperature can be
determined, as given below.
Example:
Cell Temperature
Once the actual cell temperature has been calculated, it can be used to predict how
the IV curve voltage potential is reduced. All computer programs used for array
sizing use some sort of adjustment for cell temperature to better predict actual
module performance in the real world.
It is important to see that the irradiance and ambient temperature work together to
predict the actual cell temperature in a module. For example, we just calculated that
if the ambient air temperature were 35 oC. and the irradiance was 900 w/m2, then
the cell temperature would be 60 oC. But what if the same module was located in a
cold alpine climate with an ambient temperature of only 5 oC. The same irradiance
level of 900 w/m2 would result in the same temperature rise value of 25 oC., but the
final cell temperature would be less.
Example:
The same solar module in a cold alpine climate with only 5oC ambient
in 900 w/m2
Cell temperature
The cell temperature is much lower, and therefore the module voltage would be
much higher in the alpine climate than in the hot climate. So just knowing the
irradiance on the module is not enough to predict module performance. The
ambient air temperature must be known as well.
8-33
Exercise
%
:;
4, &
2
!$$- .
//////////&
445
)
&
E
F#
F
F
#"!
#"
% F////////-
8-34
8-35
12:00 noon
(B)
2.8
3:00 p.m.
(C)
8:00 a.m.
0
10
(A)
12
14
16
18
40
35
30
25
20
15
10
5
20 0 4
Voltage (volts)
(B)
(C)
Total Energy
253 Wh
(A)
6
10
12
14
16
18
20
Time of Day
8-36
Exercise
Hour
of Day
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
Watts
Feb
Watts
June
0.48
5.60
10.79
14.09
15.18
14.00
10.67
5.51
0.47
0.81
2.04
3.91
9.28
16.27
22.27
26.31
28.65
29.36
28.51
26.06
21.96
15.98
9.08
3.8
2.0
.8
8-37
Graph the watts against time to show the output over a typical day.
Watts
35
30
25
20
15
10
0
4
10
11
12
Time of Day
8-38
8-39
Current (amps)
2.8
Power (watts)
40
35
12:00 noon
30
(B)
25
2
3:00 p.m.
20
(C)
15
8:00 a.m.
247 wh
10
1
(A)
5
0
0
0
10
12
14
16
18
20
Voltage (volts)
10
12
14
16
18
Time of Day
!
8-40
20
Exercise
Hour
of Day
6
7
8
9
10
11
12
1
2
3
4
5
6
Watts
April
Watts
July
7.4
20.3
32.2
39.0
41.7
41.6
40.1
37.2
31.0
19.9
7.1
1.3
8.1
18.9
28.0
32.1
33.1
32.7
31.4
29.3
25.7
17.7
7.6
1.0
8-41
Graph the watts against time to show the output over a typical day
Watts
45
40
35
30
25
20
15
10
0
4
10
11
12
Time of Day
8-42
8-43
2.8
(B)
3:00 p.m.
(C)
8:00 a.m.
1
(A)
0
0
10
12
14
16
18
20
Voltage (volts)
$
8-44
Current
Module
or
Resistance
(ohms)
array
Voltage
of system .
Using a resistor in the field to test the curve shape of a PV array is another useful
application of a pure resistive load. A resistor also has a load curve or an "I-V
curve". It is a straight line. This defines a single intersection point with the module
or array I-V Curve.
8-45
Current (amps)
3
load line
for low R
1
0
8
12
16
Voltage (volts)
20
%
A resistance can be calculated that will operate a module or array at or very near its
maximum power point. This can be a useful tool in the field. Using standard digital
volt meters in the field, it is easy to measure Isc and Voc, but you are not really
looking at the output of a module or array anywhere near its maximum output point.
By using a resistor to operate the module or array somewhere near its knee, you
can get a complete picture of the curve shape.
8-46
In real outdoor conditions, the light level would probably be lower than the standard
level of 1000 w/m2, and the cell temperature would probably be higher than the
standard temperature of 25 oC. A 36 cell 36-watt module at outdoor conditions of for
example 800 watts/m2 and 50 oC cell temperature has Vmp = 17 volts and Imp = 2.1
amps (approximate). Dividing these values will give a resistance that will operate
the module near its maximum power point in typical field conditions:
Resistance needed to operate at Pmax
(35 Pmax, typical outdoors)
Vmp
Imp
17 volts
2.1 amps
8.1ohms
So connecting an 8.1-ohm resistor to a 36-watt module in the field will operate the
module very near its maximum power point, to allow checking of output other than
just Voc and Isc.
3
R = Vmp/Imp
2
1
0
12
16
20
Voltage (volts)
8-47
To calculate the size of resistor needed for an array of modules, multiply the above
resistance value by the number of modules in series, and divide by the number of
modules in parallel. The final resistance value will operate the array near its
maximum power point.
Resistance array =
Example:
Resistance module
# Modules in Series
# Modules in Parallel
10 ohms
20 ohms
4 in series
2 in parallel
The example above shows that to operate an array of four 36-watt modules in series
by two in parallel, you would need a resistor of approximately 20 ohms. This is
shown graphically below.
8-48
Exercise:
9
*
* 5
*2
0
*
'
*
*
D
+
*
*
*
G7
%
/////////////
8-49
(End of Chapter)
8-50
CHAPTER EIGHT
OUTPUT CURVES
8-1
8-2
8-3
8-4
8-7
8-7
8-8
8-9
8-10
8-14
8-14
8-17
8-19
8-21
8-23
8-23
8-24
8-24
8-26
8-31
8-32
8-35
8-39
8-43
8-45
8-51
Chapter 8 Answers
Output Curves
SM55
SM50-H
SM46
SM20
SM10
SM6
SP75 (12V)
SP36 (12V)
SP18 (12V)
Isc
Voc
Imp
Vmp
Pmax
(Amps)
3.45
3.35
3.35
1.6
0.71
0.42
4.8
2.4
1.2
(Volts)
21.7
19.8
18.0
18.0
19.9
19.5
21.7
21.7
21.7
(Amps)
3.15
3.15
3.15
1.38
0.61
0.39
4.4
2.1
1.1
(Volts)
17.4
15.9
14.6
14.5
16.3
15.0
17.0
17.0
17.0
(Watts)
55
50
46
20
10
6
75
36
18
c. 0.8
d. 0.9
b. No
b., c.
8-1
Output Curves
Isc1
Light intensity1
Isc2
Light intensity2
2.0 amps
1000 w/m2
Isc2
850 w/m2
Isc2
1.7 amps
If the two days are the same temperature, then we can calculate the Isc at standard
irradiance for each module.
Module 1:
2.5 amps
745 w/m2
Isc2
1000 w/m2
Isc2
2.1 amps
650 w/m2
Isc2
1000 w/m2
Isc2
Module 2:
Under standard irradiance, Module 1 has an Isc of 3.36 and Module 2 has an Isc of
3.23. Therefore the correct answer is:
a. Module 1
Note: If the temperatures were significantly different on the two days that the
measurements were taken, we could not answer the question without knowing the
actual temperatures.
8-2
Output Curves
Using the cell temperature coefficients for either the SP cell or the SM cell, we see that
the change in Voc per oC is
-2.15 mV/ oC
or
-0.36%/ oC
Voc2
Voc2
10.8 )
100
= 19.9 X (1-0.108)
= 19.9 X (0.892)
= 17.75 Voc (at 55 oC)
Siemens Solar Basic Photovoltaic Technology
8-3
Output Curves
The voltage loss factor for the 33-cell module was calculated in the previous problem to
be:
Voltage loss factor = -0.071 V/oC
So for every 10 oC, the voltage change is:
Voc Change
Standard Value
j. N/A
j. N/A
c.
a.
j. N/A
j. N/A
f.
j. N/A
8-4
Output Curves
The work below shows the calculations and results for the SM10, SM50, SM55, SP75,
and SR100 modules. The method for other modules will be the same.
The datasheet for the SM10 module gives the following information:
Pmax =
10 watts
Length
Width
=
=
0.360 m
0.330 m
10 watts
0.360 m X 0.330 m X 1000 w/m2
0.084 or 8.4%
50 watts
Length
Width
=
=
1.293 m
0.329 m
50 watts
1.293 m X 0.329 m X 1000 w/m2
0.118 or 11.8%
55 watts
Length
Width
=
=
1.293 m
0.329 m
55 watts
1.293 m X 0.329 m X 1000 w/m2
0.129 or 12.9%
75 watts
Length
Width
=
=
1.200 m
0.527 m
75 watts
1.200 m X 0.527 m X 1000 w/m2
0.119 or 11.9%
8-5
Output Curves
100 watts
Length
Width
=
=
1.498 m
0.594 m
100 watts
1.498 m X 0.594 m X 1000 w/m2
0.112 or 11.2%
The data sheet for the SM55 module gives the following information:
Voc: 21.7 Volts
Vmp: 17.4 Volts
Isc:
Imp:
3.45 Amps
3.15 Amps
Reading the graph of cell temperature rise vs. irradiance, we estimate that at 800 watts/
m2 the cell temperature rise will be approximately 22 C. Then the cell temperature will
be:
Cell Temperature = 32 C (ambient air) + 22 C (rise)
= 52 C
8-6
Output Curves
The thermocouple reports the cell temperature to be 68 C. The rise in cell temperature
can then be calculated as:
68 C (Cell )
Rise in Temperature = 68 C - 40 C = 28 C
Looking at the graph of cell temperature rise vs. irradiance, we estimate that a rise in
temperature of 28 C results when the irradiance is equal to 950 watts/m2.
a. Voc = 17.5 volts
Ambient temp. = 40 oC
From the module literature, we know the following data for the SM-50H module.
Voc: 19.8 Volts
The Voc change is therefore:
Voc Change
The temperature coefficient for an SM50-H module is given in the chapter as -0.071
Volts / C. We can calculate the temperature change of the cell as:
2.3 Volts
Change in T
Ambient temp. = 35 oC
From the module literature, we know the following data for the SM-50H module.
Vmp: 15.9 Volts
8-7
Output Curves
Looking at the graph of cell Temperature Rise vs. Irradiance, we determine that an
irradiance of 700 watts/m2 would result in a temperature rise of about 20 oC. If the
ambient temperature is 35 oC, then the cell temperature is 35 oC + 20 oC = 55 oC. The
temperature coefficient stated in the chapter for Vmp is -0.072 V / oC. We calculate the
change in Vmp as:
Vmp change
By summing the each hour's energy, we get the output for each month:
Output in February =
Output in June
=
76.8 Wh
247.1 Wh
Watts
35
30
25
Feb
Jun
20
15
10
5
0
4
10
12
Time of Day
8-8
Output Curves
By summing the each hour's energy, we get the output for each month:
Output in April
Output in July
=
=
317.5 Wh
266.9 Wh
Watts
45
40
35
30
25
Apr
Jul
20
15
10
5
0
6
9 10 11 12 1
Time of Day
8-9
Output Curves
The module literature gives the following values for the SP75 module:
Vmp = 17.0 volts
Imp = 4.4 amps
The resistance that will operate a single SP75 module at the maximum power point
under standard conditions is then:
Resistance
Vmp =
Imp
17.0
4.4
3.86 Ohms
Therefore, connecting a 3.86-ohm resistor to the 75 watt module in the field will operate
the module very near its maximum power point. To operate the whole array at the
maximum power point, we calculate the resistance as:
Resistance (array) =
3.86 ohms
5.79 ohms
6 in series
4 in parallel
Note that under real field conditions, the values of Vmp and Imp may be slightly
different. This is because the irradiance may be less than 1000 watts/m2 and the cell
temperature will likely be hotter than 25 C.
8-10
Output Curves
Chapter Nine
Load Estimation
The loads cannot be taken as given while all other calculations are carefully
scrutinized. The loads influence every aspect of system design, and must be as
efficient and reliable as possible. As you read, think about what influences load
efficiency, and how you might have to argue with a client to replace their existing or
proposed load with a more efficient one.
The entire system design is based on the size of the load. If the information is
inaccurate the initial costs will be too high or the array and battery could be too small
and the system will eventually fail. It is, therefore, essential that time be taken to
look carefully at the load requirements and the expected usage pattern.
Using literature values for load consumption is common, but it is more accurate to
have the load demand measured to be sure. Often nominal numbers are presented
in literature, and a particular piece of equipment may require more or less power
than stated.
If an existing application is being retrofitted with a PV power system it is very
important to not just look at the old generator capacity and try to recreate it with PV.
Very often an oversized diesel generator was installed, perhaps because that size
was used elsewhere or to allow for future growth. A PV power system can be
designed to accurately match the current load requirement without limiting the ability
to expand in the future to meet greater demand.
The load profile throughout the year must be accurately determined. Any seasonal
variation might influence the choice of tilt angle or battery size for autonomy. The
"duty cycle" or hours of operation for intermittent loads must be estimated carefully.
In the case of telecommunications equipment, not only the hours of transmitting but
also the hours of standby or quiescent operation need to be included in the load
calculations.
Improving load efficiency is the quickest way to reduce PV power system cost. A
more efficient load device may even be slightly more expensive than an existing or
conventional load device. But if the power consumption of the efficient load is
significantly lower than the other, the cost savings in the array and battery may offset
the higher load cost in many cases, and result in a total system cost (modules +
battery + loads) that is lower than if inefficient loads are used.
9-1
Using oversized diesel generators may have allowed load device efficiency to be
neglected in the past. But when a small load is operated by a large capacity
generator, fuel efficiency is low so fuel is wasted. Simply retrofitting a new PV power
system to an existing inefficient load is a mistake if a more efficient load device can
be installed.
9-2
Qty.
Amps
Hours/day
DC Loads (Ah) =
AC Loads
Qty.
Watts
Hours/day
AC Sub-Total (Wh)
Continuous Watts
Surge Est.
Inverter Choice:
[
][
][
]
+
AC Sub-Total Efficiency Input Voltage
9-3
[
] =
DC Loads
____________
Daily Load (Ah)
Qty.
:
Amps
Hours
X
X
X
X
X
X
______
______
______
______
______
______
Days/wk
:
:
:
:
:
X ______
X ______
X ______
X ______
X ______
X ______
X
X
X
X
X
X
=
=
=
=
=
=
X ______ X ______ X
X ______ X ______ X
AC Loads
Qty.
:
:
:
:
:
:
:
:
Watts
X ______
X ______
X ______
X ______
X ______
X ______
X ______
X ______
Hours
X
X
X
X
X
X
X
X
______
______
______
______
______
______
______
______
Days/wk
= ______
Surge Est.
= ______
=
=
=
=
=
=
=
=
=
Inverter Choice:
[
][
][
]
+
AC Sub-Total Efficiency Input Voltage
[
] =
DC Loads
[
] 7 days
Total Weekly
Demand (Ah)
X
X
X
X
X
X
X
X
9-4
Total Weekly
Demand (Ah)
=
Week-Averaged
Daily Load (Ah)
DC Load Demand
AC Load Demand
X Hours of Operation
If DC loads are given in watts instead of amps, then you can easily convert by
dividing by the nominal operating voltage. This is usually 12 volts, and sometimes
24 volts.
Example:
Example:
40 watts
12 volts
3.3 amps
200 watts
24 volts
8.3 amps
9-5
quiescent
current
=
=
10 amps
50 Ah
5 hours
=
=
.5 amps
9.5 Ah
(24 - 5 hours)
The standby daily load is almost 1/5 of the transmit daily load, even though
the instantaneous current draw is only 1/20th.
9-6
60
50
40
30
20
10
0
Jan Feb Mar Apr May Jun
Jul Aug Sep Oct Nov Dec
9-7
Insolation (Langleys)
600
500
400
300
200
100
0 deg.
15 deg.
30 deg.
45 deg.
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
60 deg.
If the load is relatively constant every month of the year, as might be the case for a
navigational aid or a constantly transmitting repeater, then a different angle is better.
Tilting the array up increases the insolation intercepted during the winter months and
sacrifices some during the summer months, with a resulting profile that is more
constant throughout the year to better match the requirements of a constant load.
If a tilt is chosen that does not match the insolation profile to the load profile, then
the array will have to be quite large to produce enough output when insolation is low
and may produce excess power that is just wasted when insolation is high. By
choosing a tilt angle for the array that gives the best match, array size is minimized,
and a reliable fixed angle mounting structure can be designed for the system.
Computer models or repetitive hand calculations can predict the best angle to give a
match between load demand and insolation.
9-8
Example:
DC Loads
Lights
PL lights
TV
Fan
Qty.
:
:
:
:
2
3
1
1
Amps
X
X
X
X
3.3
0.92
3.3
2
Hours
X
X
X
X
4
5
4
8
DC Loads (Ah)
9-9
Weekly
Demand (Wh)
=
26.4
13.8
13.2
16
=
=
=
=
69.4
For practice at adding AC loads, lets estimate the demand for a remote school with
a variety of AC appliances, lights and other loads.
Example:
AC Loads
Qty.
Watts
Lights
PL lights
Computer
Projector
:
:
:
:
8
2
2
1
Microwave
Refrigerator
:
:
1
1
X
X
X
Hours
40
11
200
300
800
200
X
X
X
8
2
4
3
2
12
AC Sub-Total (Wh)
9-10
Weekly
Demand (Wh)
2,560
44
1,600
900
=
=
=
1,600
2,400
=
=
9,104
Exercise
!"#
$
%
& "#
%
' $"#
(
"
%
,
&
-
% * ./ %*0
,
!
%
* . %*0 &
(
,
"
%
.1 %*0
&
,
+
(
2
3
& 4
- (
, 4
"/ (
& &
4
%
2
3
& 4
, 4
& &
4
/ (
"5 (
- %
6
,
+
,
9-11
Choosing An Inverter
Before you can complete the sizing forms for AC loads, you must choose an
inverter. This choice will determine the average inverter efficiency and the nominal
DC voltage of the array and battery, both of which are needed to finish the
calculations indicated at the bottom of the sizing forms.
Estimate of Continuos Watts: Add all AC loads that might be on at same time
9-12
A more practical approach to estimating total surge power is to identify the load with
the largest surge power and add this value to the continuos power demand of all the
other AC loads that might be on at the moment that the large load is turned on.
Thus inverter would have to operate all the loads (except the one with the largest
surge) and be able to turn on the load with the largest surge as well.
If you are given no information on the surge demand of a particular inductive load, it
is safe to assume that six times the continuous current will be drawn as a surge to
start the load.
An example of a residential refrigerator load profile is shown. As time passes
(moving from right to left) the compressor turns on and then off, trying to keep the
refrigerator box cool. A refrigerators efficiency is measured by not just the
continuous power level, but also by how many hours in a typical day the compressor
actually operates. Each time it turns on, there is a spike or surge in the current that
is drawn. The continuous running current of this particular refrigerator is about 2
amps, but the surge current needed is about 12 amps, a factor of 6 times greater!
Refrigerator Load Profile -- Continuous and Surge Levels
9-13
1135 kWh/year
190
39
363
106
103
1350
60
200
144
108
100
108
320
35
4361 kWh/year
4361 kWh/year
365 days/year
12 kWh/day
The system designer and the client could look at this list and see what items contribute most
to the energy burden, and perhaps discuss alternative or more efficient loads to reduce the
overall energy and therefore the solar array and battery size as well.
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
9-14
9-15
Using another statistical tool presented next, you can estimate the continuous and
surge power that might be required from an inverter. A chart has been prepared
showing how much continuous power and surge power might be needed to serve a
range of daily loads.
Estimating Surge
Estimated Power
Requirement (kW)
24
22
20
18
16
14
12
10
8
6
4
2
0
surge
range
continuous
range
12
16
20
24
Read up from the estimated daily load requirement to find the approximate
continuous and surge demand. For example, if the total daily load was about 20
kWh/day, you would read up from 20 to find that the continuous power output of the
inverter may be between 4-7 kW, and that the surge demand may be between 10-20
kW. There is quite a range allowed, and this method is not precise. But it is a quick
way to estimate values that otherwise could be tedious to calculate and which might
be ultimately inaccurate anyway.
Example:
Using the daily load demand of 12 kWh/day already estimated for the
remote home, the inverter chosen should be able to supply
Continuous Watts: 2-4 kW
Surge Watts:
6-12 kW
9-16
Exercise
9-17
AC Load (Ah)
AC Subtotal (Wh)
Efficiency X Voltage
Why do we divide by the efficiency? We want to determine the amount of DC amphours that the solar array must produce. What we have been calculating so far in
the Load Sizing Forms is the amount of AC watt-hours that the loads need. Since
the inverter is not 100% efficient, more energy must be put into the inverter than is
output to the AC loads. How much more is determined by the efficiency.
Inverter Efficiency
DC power in
AC power out
Inverter
9-18
Efficiency is defined as the ratio of output power or energy divided by the input
power or energy. Therefore the energy or power in is given by the energy or power
out divided by the efficiency.
So dividing the AC Subtotal (Wh) by the efficiency CONVERTS the watt-hours out to
the AC loads into the watt-hours needed to be put into the inverter on the DC side.
AC Subtotal (Wh)
Efficiency
Then to complete the calculation, divide the DC watt-hours into the inverter by the
DC voltage to get the DC amp-hours into the inverter. This is the Ah load demand
that we need to deliver from the batteries and array.
If the DC voltage of the inverter and any DC loads are the same, then both can be
operated from the same battery bank and the same array. The Ah values for the DC
loads and AC loads can be added to give the grand total for the whole system. If the
voltages are different, then either a DC-DC converter must be used, or separate
systems can be designed for each.
9-19
Exercise:
9-20
Load
Battery Net
At End Of
Day
9-21
If the load is the same every day, then the array and battery are sized to meet that
constant daily demand, using the Daily Load Sizing Form.
Load
Battery
Net At End
Of Day
To see the impact of approaching load estimation from an averaging method, on the
next pages we redo the examples for the small remote cabin and school, this time
estimating what the averaged load would be. This might lead to a substantially
smaller load value and therefore a smaller array.
9-22
Example:
The small remote cabin that was previously used (page 11) will only be
occupied on the weekends. Recalculate the Ah demand using the WeekAveraged Load Sizing Form. Compare the result to the daily usage value
originally calculated.
DC Loads
Lights
PL lights
TV
Fan
Qty.
:
:
:
:
2
3
1
1
Amps
X
X
X
X
Hours
X
X
X
X
3.3
0.92
3.3
2
4
5
4
8
Days/wk
X
X
X
X
=
=
=
=
2
2
2
2
Weekly
Demand
(Wh)
52.8
27.6
26.4
32
138.8
19.8
Week-Averaged
Daily Load (Ah)
On the next page an example should be worked out for the remote school discussed
earlier. Recalculating the loads using the Week-Average Load Sizing method, as if
they were spread out over all seven days of the week, can be done to see if the load
is substantially smaller taking this approach.
Array and battery sizing calculations will be discussed in the chapter System Sizing.
But you can see that by averaging the load demand over the entire week that the
energy demand from the array is reduced. The difference between the values for
daily loading and week averaging is greater for the remote cabin than for the school.
This is because the cabin changes from a regular daily loading to only a two day per
week loading. The change for the school design from daily loading to having the
weekends off is not so great.
If the load distribution during a typical week is fairly even, then use the Daily Load
Sizing Form value for both array and battery sizing. Use the Week-Averaged Load
Sizing Form value for array sizing only if the distribution during a 7-day week is really
skewed to a few days. Then you might see some economics in sizing the array for
the average. In all cases, used the Daily Load Sizing Form value for battery sizing.
The battery has to be sized to handle the actual heavy daily loads.
9-23
Example:
The school also previously used (page 12) is to be occupied only during
the week (Monday-Friday). Recalculate the Ah load demand if it is
spread out over the full seven days of the week using the WeekAverage Load Sizing Form. Note that the refrigerator is not turned off
and must be kept operating during all seven days. How does this value
compare to the previously calculated load demand?
AC Loads
Qty.
Watts
Hours
Days/wk
Weekly
Demand
(Wh)
8
2
2
1
X
X
X
X
40
11
200
300
X
X
X
X
8
2
4
3
X
X
X
X
5
5
5
5
=
=
=
=
12,800
222
8,000
4,500
Microwave : 1
Refrigera : 1
tor
X
X
800
200
X
X
2
12
X
X
5
7
=
=
8,000
16,800
Lights
PL lights
Computer
Projector
:
:
:
:
9-24
50,320
7,188
Week-Averaged
Daily Load (Wh)
Exercise
9:
:;
91
:=
#
?;;
@=
*
2;
9-25
12<
><
%/ )*'
2<
%<
'
><.
%/ )*'
><.
%/ )* '
Types of Lamps
Incandescent
This is the common "light bulb". The incandescent and quartz-halogen lights
develop their light by passing current through a thin high resistance tungsten wire
which gets hot and radiates light. Along with the visible light comes a great deal of
heat, so the lights are the lowest in efficiency. Only 5% of the input energy is
converted to visible light. The tungsten filament gradually evaporates and weakens,
and the metal that evaporates gradually coats the inside of the bulb, reducing
brightness. Incandescents are available in a wide range of output powers and are
available for 12-volt DC operation.
Quartz-Halogen
The quartz-halogen light differs from the standard incandescent by using a tungsten
filament surrounded by a halogen gas (bromine). The wire glows hotter than the
standard incandescent, so high temperature resistant quartz is used for the bulb.
The halogen gas aids in re-depositing evaporated tungsten back onto the filament,
so the bulb life is longer, and little coating occurs on the inside of the bulb,
increasing end-of-life brightness. The lamps last almost twice as long as
conventional incandescents, and the overall efficiency is almost double.
Quartz-halogen lamps are available in a wide range of power, and often have an
integral reflector attached to the bulb to enhance output, and can operate from 12
volt DC power. Many automobile headlights are quartz halogen, and can be used
for home lighting applications. Bulbs used for slide projectors can easily be used for
household lighting, for example in track lighting.
9-26
Fluorescent
In fluorescent lamps mercury vapor is excited inside a tube by an alternating current,
and light is emitted by the gas. The light strikes the inner coating of the tube, which
is some type of fluorescing material, producing a soft glow. The light produced falls
well within the visible range, so the efficiency is good.
Traditionally fluorescent bulbs have been long cylinders ranging from a few watts to
40 watts. New "parallel length" or PL lamps are now available that are much more
compact and can often times replace incandescent bulbs in standard light fixtures.
The PL lamps fold back the long cylinder to make a compact "H" shape. Double "H"
or "quad" PL lamps are also available. PL type lamps are available in 5, 7, 9, 13, 18,
24 and 36-watt models. To operate from DC power a ballast is needed to produce
high frequency AC current. These are discussed further in this section.
New types of traditionally shaped fluorescent lamps are available that produce more
lumens/watt, better color, and longer life (up to 24,000 hours). These lamps are a
smaller tube diameter (1 5/16") and are called "T-10" tubes.
9-27
Lamp Efficiency
However, if a lamp produces light beyond the visible range, that radiation does not
contribute to apparent brightness of the lamp. A better measure of the ability of a
lamp to convert electrical power into visible light is called efficacy, which compares
the APPARENT BRIGHTNESS OF A LAMP, measured in lumens, against the
electrical power input, measured in watts.
Lamp Efficacy
A lamp with twice the efficacy of another will appear twice as bright with the same
input power, or can use half as much power to produce the same apparent
brightness.
Fluorescents are an excellent choice compared to conventional incandescents, and
typically give 4-5 times the brightness compared to incandescents for the same input
power. Bulbs that use less power can be used to achieve the same lighting, so a PV
power system can be smaller and less costly.
Low-pressure sodium lamps are the winners regarding efficacy, but the amber color
is too harsh for indoor lighting. They are chosen for general outdoor area lighting,
where color may not be as important as area coverage. High-pressure sodium lamps
would still have very high efficacy and would deliver a much more tolerable light.
9-28
Light Efficacy
Efficacy (lumens/watt)
100
50
0
Incandescent
High Pressre
Low Pressure
Sodium
Sodium
Fluorescent
Mercury Vapor
Tungsten Halogen
9-29
The life of the lamp must also be a consideration in system design. It is encouraging
that the more efficient or efficace lamps are also the more long life products as well.
Fluorescents can last 8,000-12,000 hours (some even up to 24,000 hours)
compared to typical incandescents that last only about 500-1500 hours. The entire
family of high intensity discharge lamps (HID) typically lasts about 15,000-25,000
hours.
Lamp Life
25000
Life (hours)
20000
Hours of Life
15000
10000
5000
0
Incandescent
Tungsten
Halogen
Mercury Vapor
Fluorescent
High Pressure
Sodium
Low Pressure
Sodium
9-30
Fluorescent Ballasts
Incandescents need only to have adequate voltage and current to operate. They will
glow immediately. All the other forms of lamps need some form of ballast to give an
initial pulse to start the lamp, or to convert the power to an alternating current to
keep the gas glowing.
Traditional ballasts have been made using a "core and coil" technology. These
operate using the available 60 Hz utility power, and are not particularly efficient.
There is a slightly noticeable flicker to the light, and sometimes a delay or pulsing
when starting lamps. Recently available improved high frequency electronic ballasts
(20,000 Hz output) help to reduce these problems. They are almost twice as
efficient, turn on the lamps instantly, operate the lamps with no flicker or electronic
interference, and some even offer a degree of dimming to fluorescent lamps.
The overall efficiency of the lamp must have the efficiency of the ballast included to
determine final power requirements. The small electronic ballasts for PL lamps
consume approximately 2 watts, while the traditional ballasts use almost 4 watts.
The cost of the ballast must also enter into the cost effectiveness equation. A small
PL lamp may cost $5 and the ballast may add $15-25 to that. When compared to a
$1-3 incandescent lamp, this looks bad. But the PL is expected to last 10 times as
long as the incandescent and consumes 4-5 times less power to give the same
brightness. Over the life of the PL lamp and ballast, the total cost of the
incandescent choice becomes much greater.
The ballasts typically do not operate more than one PL lamp however. Each PL
lamp for example must have its own ballast, so usage in track lighting is difficult. An
option is to use small 12-volt Quartz-Halogen bulbs, although they are not as
efficient.
Ballast and fluorescent lamp design may have to be modified to allow starting in cold
climates. The ballast may not be able to pulse strongly or long enough to start a
cold lamp. Special pre-heaters can be installed in the bulb to allow for starting cold
lamps.
9-31
Color Quality
One measurement that relates to color quality is the color temperature of the light.
At high temperatures, matter emits a spectrum of light, and the spectrum changes
with higher temperature. The surface of the sun is approximately 5900 deg. C. or
6200 deg.Kelvin, another temperature scale similar to the Celsius (C) scale but
beginning at "absolute zero" degrees. Our sunlight spectrum is very close to the
spectrum emitted by a body at that temperature. The hotter the temperature of a
radiating source of light, the higher the percentage of high-energy blue light is
emitted. The colder the temperature, the higher the percentage of low energy red
light is present in the spectrum. So the spectrum emitted from lamps can be
compared to the spectrum that would be emitted by a perfect radiator. A color
temperature is given to a lamp to give a measure of how "red" or "blue" the lamp
spectrum appears. The small energy efficient PL lamps are usually available with
color temperature of 2700 deg.K, but are also available at 3000, 3500, and 4100
deg.K.
Another measure of the quality of light is the color rendition index (CRI). This is a
figure of merit that compares how well lamp light matches the balance of colors in
daylight. Two lamps may be compared only if they share similar color temperatures.
Incandescents can give fairly good color rendition. Fluorescents have traditionally
been more blue than incandescents, but now "warm" and "full spectrum"
fluorescents are available. The mercury and sodium lamps do not produce a broad
gradually changing spectrum of wavelengths and so their color is quite harsh.
9-32
Exercise
!
8
A
@=
(PL)
:=
B;=;
B:=;;
::;;
+@=
:;;;
>;;;
7
B;:;<.-
B;:;<.-
>;;;
"
9-33
Types of Refrigerators
Refrigeration units operate from two basic sources of power, electricity or liquid fuel.
Electric units can be DC or AC or switchable between the two. Fuel types can use
kerosene or propane.
There are units that automatically switch from DC to AC to gas power, most
commonly used on recreational vehicles (RV's). These units are typically very
inefficient and use a great deal of power (up to 400 watts) on DC power. They are
therefore not a good choice for a remote PV powered site.
It is not uncommon for remote sites using PV for lighting; communications and tools
to use a gas powered refrigeration unit. Gas may already be on site for cooking, and
avoiding module and battery costs for refrigeration keeps the overall system more
cost effective.
However there are many manufacturers of DC powered refrigeration units. These
vary from small portable units of only 0.4 cubic feet (10 liters) capacity, to full sized
17 cubic feet (480 liters) capacity units for a home or village.
9-34
The condenser coils should be located to allow maximum escape of heat without
heating the walls of the unit. Many common AC units are inefficient because they
locate the condenser coils underneath the refrigerator, forcing all the escaping heat
to pass up and around the refrigerator walls! Coils can be located on the top of the
unit, or ideally they can be located away from the unit, perhaps outside where the
ambient temperature may be lower than inside the home.
There are available on the market super efficient refrigerator designs. A particular
model of PV powered refrigerator, manufactured by SunFrost, needs only five 50watt solar modules to operate a 17 cubic foot capacity unit, even at 90o F ambient air
temperature. The unit alone costs more than double what a conventional AC
powered design costs, but the total cost of the modules, batteries, and inverter
required for the inefficient conventional AC unit far surpass the total cost of the
efficient DC unit plus its 5 modules and small battery bank. The SunFrost product
also comes ready for AC power and is only slightly less efficient. It can be
incorporated into a conventional all-AC residential system design if so desired.
9-35
Exercise
B1=;;
A
11;
&;
:1<
:?<
7 6 /
:&;-<
:&;-<
/
B?1;
B?1;
9-36
9-37
9-38
(End of Chapter)
9-39
CHAPTER NINE
LOAD ESTIMATION
9-1
9-2
9-5
9-6
9-7
9-9
Choosing An Inverter
Meeting Continuous Power Demand
Meeting Surge Power Demand
Estimating Continuous and Surge Power for Complex Systems
Reading Inverter Literature
Calculate Ampere-Hours For AC Loads
9-12
9-12
9-12
9-14
9-17
9-18
9-21
9-26
9-26
9-28
9-31
9-32
9-32
9-34
9-34
9-34
9-37
9-37
9-37
9-38
9-38
9-40
Chapter 9 Answers
Load Estimation
We first calculate the current for each load type:
Fluorescent lights
40 Watts
12 Volts
3.3 amps
Ceiling Fans
20 Watts
12 Volts
1.7 amps
Vaccine refrigerator
60 Watts
12 Volts
5.0 amps
DC Loads
Fluorescent Lights
Ceiling Fans
Vaccine Refrigerator
:
:
:
Qty.
6
3
1
X
X
X
Amps
3.3
1.7
5
X
X
X
Hours
/day
6
12
10
DC Loads (Ah)
9-1
=
=
=
Daily
Demand (Ah)
118.8
61.2
50.0
230.0
Load Estimation
System Description:
DC Loads
Qty.
Amps
Hours
Weekly
Demand (Ah)
Days/W
k
6
3
X
X
3.3
1.7
X
X
8
8
X
X
5
5
=
=
792.0
204.0
6
3
X
X
3.3
1.7
X
X
4
4
X
X
2
2
=
=
158.4
40.8
10
350.0
1545.2
AC Subtotal
Effic.
9-2
+ 1545.2
Input
DC
Voltage
Loads
1545.2
Total Weekly
Demand (Ah)
1545.2
Weekly
Ah/day
220.7
Avg Load
(Ah/day)
7
Days
Load Estimation
To clarify the power requirements, look at the amount of time that the loads are in
transmit and standby. The combined time on and off must equal 24 hrs (1 day).
Radio transmitter #1
Transmit
Standby
12 hrs / day
12 hrs / day
Radio transmitter #2
Transmit
Standby
8 hrs / day
16 hrs /day
DC Loads
Trans. #1 Transmit
Trans. #1 Standby
Trans. #2 Transmit
Trans. #2 Standby
Qty.
1
1
1
1
:
:
:
:
X
X
X
X
Amps
8
0.5
5
0.3
X
X
X
X
Hours
/day
12
12
8
16
=
=
=
=
Daily
Demand (Ah)
96.0
6.0
40.0
4.8
146.8
DC Loads (Ah)
During the winter months, the stated transmit times are reduced:
Radio transmitter #1
Transmit
Standby
8 hrs / day
16 hrs / day
Radio transmitter #2
Transmit
Standby
5 hrs / day
19 hrs /day
DC Loads
Transm. #1 Transmit
Transm. #1 Standby
Transm. #2 Transmit
Transm. #2 Standby
:
:
:
:
Qty.
1
1
1
1
X
X
X
X
Amps
8
0.5
5
0.3
=
=
=
=
Daily
Demand (Ah)
64.0
8.0
25.0
5.7
102.7
DC Loads (Ah)
X
X
X
X
Hours
/day
8
16
5
19
9-3
Load Estimation
Refer to the manufacturer's literature.
The refrigerator will be the largest surge load. We will first take all the other AC load
watts, then add six times the refrigerator load. As a worst case, we assume that all of
the loads can be on at the same time.
Continuous AC Loads
Load
Lights
PL lights
Computer
Projector
Microwave
Qty
8
2
2
1
1
Watts
40
11
200
300
800
Total
320
22
400
300
800
1842
The refrigerator is 200 Watts continuous. At 6 times the rating, this gives a surge
power of 1200 Watts for the refrigerator.
The total surge requirement is 1842 + 1200 = 3042 Watts.
Select an inverter with a surge capacity of at least 3042 Watts.
Results will vary depending on student input.
9-4
Load Estimation
We first calculate the load size during the week:
System Description: Telecom site - load during the week
DC Loads
Trans #1 - Transmit
Trans #2 - Transmit
Trans #2 - Standby
:
:
:
Qty.
1
1
1
X
X
X
Amps
10
15
0.8
X
X
X
Hours
/day
24
8
16
=
=
=
372.8
DC Loads (Ah)
AC Loads
Computers - On
Computers - Off
Lights
:
:
:
Qty.
2
2
4
X
X
X
Watts
300
75
40
X
X
X
Hours
/day
8
16
8
=
=
=
8,480
AC Sub-total
90%
Effic
24
Input
Daily
Demand (Wh)
4,800
2,400
1,280
8,480
AC Loads (Wh)
Continuous Watts =
Surge Est. =
Daily
Demand (Ah)
240.0
120.0
12.8
760
1560
372.8
DC
Loads
765.4
Daily
(Ah/Day)
Note: We calculated the Continuous Watt load to be two computers (on) plus all four
lights. So, 2 X 300 + 4 X 40 = 760. The Surge Estimate was calculated by taking
6 times the power for the lights and adding it to the computer loads: 2 X 300 + 6 X
4 40 = 1560.
9-5
Load Estimation
DC Loads
Repeater #1
Repeater #2:
Transmit (M-F)
Standby (M-F)
Transmit
(Weekend)
Standby
(Weekend)
Qty.
Amps
Hours
10
24
Days/W
k
7
X
:
:
:
1
1
1
X
X
X
15
0.8
15
X
X
X
8
16
4
X
X
X
0.8
20
Weekly
Demand (Ah)
=
1680.0
5
5
2
=
=
=
600.0
64.0
120.0
32.0
2496.0
AC Loads
Computer Use:
On - (M-F)
Off - (M-F)
Off - Weekend
Fluorescent Lights
Qty.
:
:
:
:
2
2
2
4
Watts
X
X
X
X
300
75
75
40
Hours
X
X
X
X
8
16
24
8
Weekly
Demand (Wh)
Days/W
k
X
X
X
X
5
5
2
5
24,000
12,000
7,200
6,400
=
=
=
=
49,600
0.85
Effic
24
Input
760
1560
+ 2496.0
DC
Load
4927.4
Weekly
Ah
7
Days
4927.4
Weekly Ah
703.9
Avg Load
(Ah/day)
The Continous Watt load and Surge Estimates are the same as above. Note that the
load averaged out over the week is slightly less than the load from Monday to Friday,
704 Ah/day vs. 765 Ah/day. This is because the weekend loads are lower, which
reduces the average.
Siemens Solar Basic Photovoltaic Technology
9-6
Load Estimation
When sizing a battery for this system, we should use the higher load (765 Ah/day) for
Monday through Friday. This is because the higher load during the week represents a
significant portion of the amount of energy storage. In other words, a 5-day battery
needs to be sized for the worst 5 days of load. We might size the array based on the
weekly load, however, since over a 7-day period, we only need to replace the average
load each day (704 Ah/day).
We first consider the costs of the incandescent bulbs. Over the total time period of
8000 hours, the total amount of electricity used by the bulbs is:
8000 hours X 75 Watts = 600,000 Watt-hrs = 600 kilowatt-hours
The cost of this energy is:
600 kilowatt-hours X $0.10 / kWh = $60.
The lifetime of a single incandescent bulb is 1000 hours, so we will need to purchase 8
total bulbs for the 8000 hour period. The cost of the bulbs and the energy is:
8 X $0.50 + $60 = $64.
Now we calculate the same costs for the compact fluorescent. The electricity used is:
8000 hours X 15 Watts = 120,000 Watt-hrs = 120 kilowatt-hours
120 kilowatt-hours X $0.10 / kWh = $12.
One compact fluorescent should last for 8000 hours, so we only need a single bulb.
The cost of the bulbs and energy is then:
$15 + $12 = $27.
We see that over the lifetime of the compact fluorescent, the total cost is less than half
of the using incandescent bulbs.
9-7
Load Estimation
We first calculate the costs of using the conventional refrigerator. The energy used in
one day by the conventional refrigerator is:
220 Watts X 12 hrs / day = 2640 Watt-hrs
The number of modules required to produce this energy is:
2640 Watt-hrs / 160 Wh per day = 16.5, rounded up to 17 modules.
17 modules X $320 = $5440
The energy required for the super-efficient refrigerator is found the same way:
60 Watts X 13 hrs / day = 780 Watt-hrs
And we need to buy a number of modules equal to:
780 Watt-hrs / 160 Wh per day = 4.9, rounded up to 5 modules.
5 modules X $320 = $1600
We can now compare the total costs:
Conventional
Refrigerator
Refrigerator $0 (already purchased)
Modules
$5440
Total $5440
Super-efficient
Refrigerator
$2500
$1600
$4100
So even with the purchase of a new refrigerator, the cost savings in the energy result in
a lower overall cost.
9-8
Load Estimation
Chapter Ten
Battery Technology
In stand-alone photovoltaic power systems, the most important and most poorly
understood component next to the PV modules is the battery. There is no one
perfect type of battery for all remote photovoltaic power systems. There are many
factors that influence the choice and performance of a battery in a photovoltaic
system. And battery technology is changing, with new constructions and
performance levels available to choose from. The following discussion is intended to
give an overview of battery technology and to equip a photovoltaic system designer
with the questions needed to determine which battery is best suited for a particular
application.
10-1
10-2
Autonomy or the days of storage are often referred to when speaking about the
battery storage capacity of a stand-alone PV system. A stand-alone PV system is
described as having "autonomy" if sufficient battery storage capacity is available to
operate the electrical loads directly from the battery, without any energy input from
the PV array. The greater the design autonomy period the larger the battery
capacity required for a given load demand. For common, less critical PV
applications, autonomy periods are typically designed for between two and six days.
For critical applications involving essential loads or public safety or where weather
patterns dictate, autonomy periods may be greater than ten days. Longer autonomy
periods result in a lower average daily depth of discharge, and lower the probability
that the allowable depth of discharge, or minimum load voltage is reached.
10-3
10-4
10-5
Cell
The cell is the basic electrochemical unit in a battery consisting of a set of positive
and negative plates divided by separators, immersed in an electrolyte solution and
enclosed in a case. In a typical lead-acid battery, each cell has a nominal voltage of
about 2.1 volts, so there are 6 series cells in a nominal 12-volt battery. Figure 10-1
shows a diagram of a basic battery cell.
Active Material
The active materials in a battery are the raw composition materials that form the
positive and negative plates, and are reactants in the electrochemical cell. The
amount of active material in a battery is proportional to the capacity a battery can
deliver. In lead-acid batteries, the active materials are lead dioxide (PbO2) in the
positive plates and metallic sponge lead (Pb) in the negative plates, which react with
a sulfuric acid (H2SO4) solution during battery operation.
10-6
Negative plate
Positive plate
Grid
Grid
Separator
Active material
Active material
Case
Electrolyte
Electrolyte
The electrolyte is a conducting medium that allows the flow of current through ionic
transfer or the transfer of electrons between the plates in a battery. In a lead-acid
battery the electrolyte is a diluted sulfuric acid solution, either in liquid (flooded) form,
gelled or absorbed in glass mats. In flooded nickel-cadmium cells, the electrolyte is
an alkaline solution of potassium hydroxide and water. In most flooded battery
types, periodic water additions are required to replenish the electrolyte lost through
gassing. When adding water to batteries it is very important to use distilled or demineralized water, as even the impurities in normal tap water can poison the battery
and result in premature failure.
10-7
Grid
In a lead-acid battery the grid is typically a lead alloy framework that supports the
active material on a battery plate, and which also conducts current. Alloying
elements such as antimony and calcium are often used to strengthen the lead grids,
and have characteristic effects on battery performance such as cycle performance
and gassing. Some grids are made by expanding a thin lead alloy sheet into a flat
plate web. Others are made of long spines of lead with the active material plated
around them forming tubes, or what are referred to as tubular plates.
Plate
A plate is a basic battery component, consisting of a grid and active material,
sometimes called an electrode. There are generally a number of positive and
negative plates in each battery cell, typically connected in parallel at a bus bar or
inter-cell connector at the top of the plates. A pasted plate is manufactured by
applying a mixture of lead oxide, sulfuric acid, fibers and water on to the grid.
The thickness of the grid and plate affect the deep cycle performance of a battery.
In automotive starting or SLI type batteries many thin plates are used per cell. This
results in maximum surface area for delivering high currents, but not much thickness
and mechanical durability for deep and prolonged discharges. Thick plates are used
for deep cycling applications such as for forklifts, golf carts and other electric
vehicles. The thick plates permit deep discharges over long periods, while
maintaining good adhesion of the active material to the grid, resulting in longer life.
Separator
A separator is a porous, insulating divider between the positive and negative plates
in a battery, used to keep the plates from coming into electrical contact and shortcircuiting, and which also allows the flow of electrolyte and ions between the positive
and negative plates. Separators are made from microporous rubber, plastic or
glass-wool mats. In some cases, the separators may be like an envelope, enclosing
the entire plate and preventing shed materials from creating short circuits at the
bottom of the plates.
10-8
Element
An element is defined as a stack of positive and negative plate groups and
separators, assembled together with plate straps interconnecting the positive and
negative plates.
Terminal Posts
Terminal posts are the external positive and negative electrical connections to a
battery. A battery is connected in a PV system and to electrical loads at the
terminal posts. In a lead-acid battery the posts are generally lead or a lead alloy, or
possibly stainless steel or copper-plated steel for greater corrosion resistance.
Battery terminals may require periodic cleaning, particularly for flooded designs. It is
also recommended that the clamps or connections to battery terminals be secured
occasionally as they may loosen over time.
Cell Vents
During battery charging, gasses are produced within a battery that may be vented to
the atmosphere. In flooded designs the loss of electrolyte through gas escape from
the cell vents it a normal occurrence, and requires the periodic addition of water to
maintain proper electrolyte levels. In sealed or valve-regulated batteries the vents
are designed with a pressure relief mechanism, remaining closed under normal
conditions, but opening during higher than normal battery pressures, often the result
of overcharging or high temperature operation. Each cell of a complete battery unit
has some type of cell vent.
Flame arrestor vent caps are commonly supplied component on larger, industrial
battery systems. The venting occurs through a charcoal filter, designed to contain a
cell explosion to one cell, minimizing the potential for a catastrophic explosion of the
entire battery bank.
Case
Commonly made from a hard rubber or plastic, the case contains the plates,
separators and electrolyte in a battery. The case is typically enclosed, with the
exception of inter-cell connectors which attach the plate assembly from one cell to
the next, terminal posts, and vents or caps which allow gassing products to escape
and to permit water additions if required. Clear battery cases or containers allow for
easy monitoring of electrolyte levels and battery plate condition. For very large or
tall batteries, plastic cases are often supported with an external metal or rigid plastic
casing.
10-9
Primary Batteries
Primary batteries can store and deliver electrical energy, but cannot be recharged.
Typical carbon-zinc and lithium batteries commonly used in consumer electronic
devices are primary batteries. Primary batteries are not used in PV systems
because they can not be recharged.
Secondary Batteries
A secondary battery can store and deliver electrical energy, and can also be
recharged by passing a current through it in an opposite direction to the discharge
current. Common lead-acid batteries used in automobiles and PV systems are
secondary batteries. The table lists common secondary battery types and their
characteristics that are important to PV system designers. A detailed discussion of
each battery type follows.
10-10
Cost
Deep Cycle
Performance
Maintenance
Flooded Lead-Acid
Lead-Antimony
low
good
high
low
poor
medium
low
poor
low
medium
good
medium
medium
fair
low
medium
fair
low
Nickel-Cadmium
Sintered-Plate
high
good
none
Pocket-Plate
high
good
medium
10-11
SLI Batteries
Starting, lighting and ignition (SLI) batteries are types of lead-acid batteries designed
primarily for shallow cycle service, most often used to power automobile starters.
These batteries have a number of thin positive and negative plates per cell designed
to increase the total plate active surface area. The large number of plates per cell
allows the battery to deliver high discharge currents for short periods. While they
are not designed for long life under deep cycle service, SLI batteries are sometimes
used for PV systems in developing countries where they are the only types of battery
locally manufactured. Although not recommended for most PV applications, SLI
batteries may provide up to two years of useful service in small stand-alone PV
systems where the average daily depth of discharge is limited to 10-20%, and the
maximum allowable depth of discharge is limited to 40-60%.
Stationary Batteries
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
10-12
Lead-Antimony Batteries
Lead-antimony batteries are a type of lead-acid battery which use antimony (Sb) as
the primary alloying element with lead in the plate grids. The use of lead-antimony
alloys in the grids has both advantages and disadvantages. Advantages include
providing greater mechanical strength than pure lead grids, and excellent deep
discharge and high discharge rate performance. Lead-antimony grids also limit the
shedding of active material and have better lifetime than lead-calcium batteries when
operated at higher temperatures. Disadvantages of lead-antimony batteries are a
high self-discharge rate, and as the result of necessary overcharge, require frequent
water additions depending on the temperature and amount of overcharge.
Most lead-antimony batteries are flooded, open vent types with removable caps to
permit water additions. They are well suited to application in PV systems due to
their deep cycle capability and ability to take abuse, however they do require
periodic water additions. The frequency of water additions can be minimized by the
use of catalytic recombination caps or battery designs with excess electrolyte
reservoirs. The health of flooded, open vent lead-antimony batteries can easily be
checked by measuring the specific gravity of the electrolyte with a hydrometer.
Lead-antimony batteries with thick plates and robust design are generally classified
as motive power or traction type batteries, are widely available and are typically used
in electrically operated vehicles where deep cycle long-life performance is required.
10-13
Lead-Calcium Batteries
Lead-calcium batteries are a type of lead-acid battery which use calcium (Ca) as the
primary alloying element with lead in the plate grids. Like lead-antimony, the use of
lead-calcium alloys in the grids has both advantages and disadvantages.
Advantages include providing greater mechanical strength than pure lead grids, a
low self-discharge rate, and reduced gassing resulting in lower water loss and lower
maintenance requirements than for lead-antimony batteries. Disadvantages of leadcalcium batteries include poor charge acceptance after deep discharges and
shortened battery life at higher operating temperatures and if discharged to greater
than 25% depth of discharge repeatedly.
Lead-Antimony/Lead-Calcium Hybrid
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
10-14
These are typically flooded batteries, with capacity ratings of over 200 amperehours. A common design for this battery type uses lead-calcium tubular positive
electrodes and pasted lead-antimony negative plates. This design combines the
advantages of both lead-calcium and lead-antimony design, including good deep
cycle performance, low water loss and long life. Stratification and sulfation can also
be a problem with these batteries, and must be treated accordingly. These batteries
are sometimes used in PV systems with larger capacity and deep cycle
requirements. A common hybrid battery using tubular plates is the Exide Solar
battery line manufactured in the United States.
10-15
10-16
Gelled Batteries
Initially designed for electronic instruments and consumer devices, gelled lead-acid
batteries typically use lead-calcium grids. The electrolyte is 'gelled' by the addition of
silicon dioxide to the electrolyte, which is then added to the battery in a warm liquid
form and gels as it cools. Gelled batteries use an internal recombinant process to
limit gas escape from the battery, reducing water loss. Cracks and voids develop
within the gelled electrolyte during the first few cycles, providing paths for gas
transport between the positive and negative plates, facilitating the recombinant
process.
Some gelled batteries have a small amount of phosphoric acid added to the
electrolyte to improve the deep discharge cycle performance of the battery. The
phosphoric acid is similar to the common commercial corrosion inhibitors and metal
preservers, and minimizes grid oxidation at low states of charge. Gelled batteries
represent over 90% of the captive electrolyte batteries used in small PV systems in
the United States.
10-17
10-18
Formation
Forming is the process of initial battery charging during manufacture. Formation of a
lead-acid battery changes the lead oxide (PbO) on the positive plate grids to lead
dioxide (PbO2), and to metallic sponge lead (Pb) on the negative plates. The extent
to which a battery has been formed during manufacture dictates the need for
additional cycles in the field to achieve rated capacity.
Stratification
Stratification is a condition that can occur in flooded lead-acid batteries in which the
concentration or specific gravity of the electrolyte increases from the bottom to top of
a cell. Stratification is generally the result of undercharging, or not providing enough
overcharge to gas and agitate the electrolyte during finish charging. Prolonged
stratification can result in the bottom of the plates being consumed, while the upper
portions remaining in relatively good shape, reducing battery life and capacity. Tall
stationary cells, typically of large capacity, are particularly prone to stratification
when charged at low rates. Periodic equalization charges thoroughly mix the
electrolyte and can prevent stratification problems.
10-19
Specific Gravity
Specific gravity is defined as the ratio of the density of a solution to the density of
water, typically measured with a hydrometer. By definition, water has a specific
gravity of one. In a lead-acid battery, the electrolyte is a diluted solution of sulfuric
acid and water. In a fully charged battery, the electrolyte is approximately 36%
sulfuric acid by weight, or 25% by volume, with the remainder water. The specific
gravity of the electrolyte is related to the battery state of charge, depending on the
design electrolyte concentration and temperature.
In a fully charged flooded lead-acid battery, the specific gravity of the electrolyte is
typically in the range of 1.250 to 1.280 at a temperature of 27o C, meaning that the
density of the electrolyte is between 1.25 and 1.28 times that of pure water. When
the battery is discharged, the hydrogen (H+) and sulfate (SO42-) ions from the sulfuric
acid solution combine with the active materials in the positive and negative plates to
form lead sulfate (PbSO4), decreasing the specific gravity of the electrolyte. As the
battery is discharged to greater depths, the sulfuric acid solution becomes diluted
until there are no ions left in solution. At this point the battery is fully discharged,
and the electrolyte is essentially water with a specific gravity of one.
Concentrated sulfuric acid has a very low freezing point (less than -50o C) while
water has a much higher freezing point of 0o C. This has important implications in
that the freezing point of the electrolyte in a lead-acid battery varies with the
concentration or specific gravity of the electrolyte. As the battery becomes
discharged, the specific gravity decreases resulting in a higher freezing point for the
electrolyte.
Lead-acid batteries used in PV systems may be susceptible to freezing in some
applications, particularly during cold winters when the batteries may not be fully
charged during below average insolation periods. The PV system designer must
carefully consider the temperature extremes of the application along with the
anticipated battery state of charge during the winter months to ensure that lead-acid
batteries are not subjected to freezing. Table 10-2 shows the properties and
freezing points for sulfuric acid solutions.
10-20
1.000
1.050
1.100
1.150
1.200
1.250
1.300
H2SO4 (Wt%)
H2SO4 (Vol%)
Freezing Point
(oC)
0.0
7.3
14.3
20.9
27.2
33.4
39.1
0.0
4.2
8.5
13.0
17.1
22.6
27.6
0
-3.3
-7.8
-15
-27
-52
-71
10-21
Sulfation
Sulfation is a normal process that occurs in lead-acid batteries resulting from
prolonged operation at partial states of charge. Even batteries that are frequently
fully charged suffer from the effects of sulfation as the battery ages. The sulfation
process involves the growth of lead sulfate crystals on the positive plate, decreasing
the active area and capacity of the cell. During normal battery discharge, the active
materials of the plates are converted to lead sulfate. The deeper the discharge, the
greater the amount of active material that is converted to lead sulfate. During
recharge the lead sulfate is converted back into lead dioxide and sponge lead on the
positive and negative plates, respectively. If the battery is recharged soon after
being discharged the lead sulfate converts easily back into the active materials.
However, if a lead-acid battery is left at less than full state of charge for prolonged
periods (days or weeks), the lead sulfate crystallizes on the plate and inhibits the
conversion back to the active materials during recharge. The crystals essentially
lock away active material and prevent it from reforming into lead and lead dioxide,
effectively reducing the capacity of the battery. If the lead sulfate crystals grow too
large they can cause physical damage to the plates. Sulfation also leads to higher
internal resistance within the battery, making it more difficult to recharge.
Sulfation is a common problem experienced with lead-acid batteries in many PV
applications. As the PV array is sized to meet the load under average conditions,
the battery must sometimes be used to supply reserve energy during periods of
excessive load usage or below average insolation. As a consequence, batteries in
most PV systems normally operate for some length of time over the course of a year
at partial states of charge, resulting in some degree of sulfation. The longer the
period and greater the depth of discharge, the greater the extent of sulfation.
To minimize sulfation of lead acid batteries in photovoltaic systems the PV array is
generally designed to recharge the battery on the average daily conditions during the
worst insolation month of the year. By sizing for the worst months weather, the PV
array has the best chance of minimizing the seasonal battery depth of discharge. In
hybrid systems using a backup source such as a generator or wind turbine, the
backup source can be effectively used to keep the batteries fully charged even if the
PV array can not. In general, proper battery and array sizing, as well as periodic
equalization charges can minimize the onset of sulfation.
10-22
Nickel-Cadmium Batteries
Nickel-cadmium (Ni-Cad) batteries are secondary, or rechargeable batteries and
have several advantages over lead-acid batteries that make them attractive for use
in stand-alone PV systems. These advantages include long life, low maintenance,
survivability from excessive discharges, excellent low temperature capacity
retention, and non-critical voltage regulation requirements. The main disadvantages
of nickel-cadmium batteries are their high cost and limited availability compared to
lead-acid designs.
A typical nickel-cadmium cell consists of positive electrodes made from nickelhydroxide (NiO(OH))and negative electrodes made from cadmium (Cd) and
immersed in an alkaline potassium hydroxide (KOH) electrolyte solution. When a
nickel-cadmium cell is discharged, the nickel hydroxide changes form (Ni(OH)2) and
the cadmium becomes cadmium hydroxide (Cd(OH)2). The concentration of the
electrolyte does not change during the reaction so the freezing point stays very low.
10-23
Notice these reactions are reversible and that the elements and charge are balanced
on both sides of the equations. The discharge reactions occur from left to right,
while the charge reactions are reversed.
The nominal voltage for a nickel-cadmium cell is 1.2 volts, compared to about 2.1
volts for a lead-acid cell, requiring 10 nickel-cadmium cells to be configured in series
for a nominal 12-volt battery. The voltage of a nickel-cadmium cell remains relatively
stable until the cell is almost completely discharged, where the voltage drops off
dramatically. Nickel-cadmium batteries can accept charge rates as high as C/1, and
are tolerant of continuous overcharge up to a C/15 rate. Nickel-cadmium batteries
are commonly subdivided in to two primary types; sintered plate and pocket plate.
10-24
Exercises
!
!
"
#
#
# "
!
!
! !
$
"
'+,
! !
-+,
10-25
.
/
0
%
1
2$
#
$
$
.
10-26
Advantages
Disadvantages
Lead Antimony/
Calcium Hybrid
Nickel-Cadmium
Sealed Sintered-Plate
Flooded Pocket-Plate
10-27
Battery Cycling
One of the main parameters that distinguishes battery types is their ability to cycle.
A battery cycle refers to the process of charging and discharging a battery. Battery
discharge is the process that occurs when a battery delivers current, quantified by
the discharge current or rate. Charging is the process when a battery receives or
accepts current, quantified by the charge current or rate. A discharge followed by a
recharge is considered one cycle.
The discharge can be very small or shallow, or it can be very severe or deep. A 100
percent depth of discharge cycle provides a measure of the total battery capacity at
a given discharge rate. All batteries can be cycled, but the question is how deeply
and how many times before a permanent loss of capacity occurs. Batteries used in
photovoltaic applications will definitely be subjected to cycling on a daily basis, and
perhaps deeply cycled occasionally.
We can think of a battery as being full of charge, even though it is actually full of
chemicals that hold potential energy. We use the capacity of a battery bank in a
photovoltaic power system to operate the loads during each night, and during
periods of heavy load use or below average insolation. If a series of below average
weather days occurs in a row, then the battery is not fully recharged at the end of
each daily cycle, and the capacity and state of charge of the battery reduces daily.
B attery C ycling
B attery capacity
O ne discharge and
recharge is one cycle
B attery life depends on
how deep and how
m any tim es the battery
is cycled
B atteries designed for
shallow cycling w ill
w ork, but for short tim e
Solar system s need
deep cycling batteries
+
available
for use
on e d ays use
leave unused
10-28
Battery Discharging
When a battery is discharged the chemical reaction on the plates proceeds inward
toward the grid. The deeper the discharge, the deeper the chemical reaction occurs.
In lead-acid batteries the lead-sulfate molecules that are formed are larger than the
lead or lead-oxide molecules, and the bonding of the active material to the plates is
gradually weakened due to grid growth. Figure 10-3 illustrates the discharge
process.
Discharge Process
Discharge reaction
proceeds inward toward
grid
100% discharge
weakens adhesion
Increased resistance
produces heat
Degradation accelerates
grid
active material
coated onto grid
10-29
Rate in Amperes =
This notation is helpful because it allows us to talk about relative rates of battery
charge and discharge, without referring to the exact size of a battery. For example,
most manufacturers recommend charging their batteries no faster than the C/5 rate
to limit gassing and overcharge. This means 20 amps for a 100 Ah battery, and 100
amps for a 500 Ah battery. Moderate charge rates are around C/20 or C/30, while
trickle charging at C/100 will hardly produce any gassing at all in most batteries.
This notation is used for discussing discharge rates as well. For example, batteries
used for UPS systems generally have their capacity measured at the C/5 or C/2 hour
rate, because in the event of a utility power failure, the UPS system is expected to
operate for only 2 or 5 hours. Batteries used in electric forklift operations will have
their capacity rated at the C/8 hour rate, because it is anticipated that they will be
discharged during a typical 8-hour shift.
Batteries used in typical PV systems experience very low rates of charge and
discharge compared to these common industrial applications. For example, the
maximum charge rates from the PV array to battery are commonly about C/40, and
typical discharge rates supplied to the loads may be as low as C/100 to C/200.
10-30
10-31
Autonomy (Reserve)
Determines Daily Discharge
Depth of
Discharge (%)
80
70
60
80% Maximum D.O.D.
50
40
30
20
10
0
11
13
15
17
19
10-32
Example:
Since most remote PV power systems are designed with at least 4 days of
autonomy, batteries in remote photovoltaic systems will be shallow cycled on an
average daily basis, whether deep or shallow cycling type batteries are used.
However, we recommend using deep cycling batteries in remote photovoltaic power
systems, even though they will typically be shallow cycled, perhaps 15-10% daily or
even less. By using deep cycling type batteries, the system can withstand the
expected seasonal drops in capacity due to below average insolation. Shallow
cycling batteries can be used as well, but shorter life should be expected.
10-33
Exercises
!4
10-34
10-35
-60
-40
-20
10-36
The maximum depth of discharge must not be allowed to exceed the point that
would allow freezing. The photovoltaic power system must be designed with enough
battery capacity so that after the maximum expected days of autonomous operation,
the depth of discharge does not exceed the danger point. If the coldest expected
battery temperatures are not known, the coldest 24-hour average temperature for
the location may be used.
You can see that if the coldest temperature does not go below -8 oC, then there is no
effect on the allowable maximum depth of discharge to prevent freezing. This factor
is only applicable in very cold climates.
Example:
10-37
Exercises
(
)))))))))) ,
"
'+ 6
$
$
7(
)))))))))) ,
6 #
#
!
10-38
Deep cycle batteries are often traction of motive power types and can handle greater
discharges for longer times than typical shallow cycle batteries. Even so they should
not be cycled to 100% discharge. Most manufacturers recommend that deep cycling
batteries be discharged to no more than 80% of the rated capacity, to prevent
reactions from occurring close to the grid. The expansion from lead and lead dioxide
molecules into large lead sulfate molecules can severely weaken the bonding of the
active materials to the grid, causing increased internal resistance and heating.
Typically deep cycle batteries can deliver 1500-1800 cycles to 80% depth of
discharge before needing replacement, and will deliver 3000-4000 cycles if
discharged more moderately to 25% depth of discharge or less.
10-39
The difference in cycle life between deep and shallow cycle batteries is shown in the
figure below. The general trend for both types of batteries is that a greater depth of
discharge yields a fewer number of cycles. All batteries deliver fewer cycles at
greater depths of discharge.
3000
2000
1000
0
30
50
70
10-40
Battery Capacity
Capacity is a measure of a battery's ability to store or deliver electrical energy,
commonly expressed in units of ampere-hours.
Ampere-Hour Definition
The ampere-hour (Ah) is the common unit of measure for a battery's electrical
storage capacity, obtained by integrating the discharge or charge current in amperes
over a specific time period. An ampere-hour is equal to the transfer of one ampere
over one hour, equal to 3600 coulombs of charge. For example, a battery that
delivers 5 amps for 20 hours is said to have delivered 100 ampere-hours.
10-41
25 oC
10-42
Example:
120
110
100
90
80
70
60
C/500
C/50
C/0.5
50
40
C/120
C/5
30
-30
-20
-10
10
20
30
40
10-43
10-44
10 hours
Photovoltaic applications
100-300 hours
The time we need to specify is the time to TOTAL discharge (100% DOD). But we
do not allow batteries to fully discharge every cycle. In fact, most manufacturers
recommend that shallow cycling type batteries only be discharged to a maximum of
50% of their full capacity, and that deep cycling batteries be discharged to a
maximum of 80% of their capacity. This is to prevent weakening of the bond
between the active materials and the grids.
50%
80%
Thus we need to take into account the maximum allowable depth of discharge in
calculating the time it would take to fully discharge a battery. The time (hours) to
discharge to the maximum DOD limit is simply the number of days of autonomy
reserve times the number of hours each day that the loads operate.
Time to discharge to max. DOD = Days of Reserve X Load Operating Time
For continuous loads, such a microwave repeaters or navigational systems, the
operating time might be 24 hours/day. For simple home lighting systems, the load (a
single fluorescent light for example) might be specified as 4 hours/day. For systems
with a variety of intermittent loads, a weighted average for the load operating time
can be calculated by summing each load (amps DC, or watts AC) multiplied times its
operating time and dividing this by the sum of the load values alone. This weights
the average load operating time for a collection of different loads based on the size
of the load and its operating time.
10-45
Continuous Loads:
Single Load Systems:
Multiple Load Systems:
use 24 hours
use load operating time
use weighted average load operating time
load time
loads
The formula for the time to total discharge (not just to max. DOD as described
above) is time to discharge to the maximum DOD divided by the maximum allowed
discharge.
Example:
The small remote cabin used in the Load Estimation Chapter has a
variety of loads. The weighted average load operating time would be
given by dividing the sum of the load X time by the sum of the loads
only:
Weighted Average =
Load Time
=
69.4 Ah
(2 x 3.3) + (3 x .92) + (1 x 3.3) + (1 x 2)
4.7 hours
37.6 hours
10-46
Example:
5 days X 24 hours/day
0.80
150 hours
10-47
1 2 0
State
of
Charge
(%)
#Days X 24 hrs
Max % Discharge
8 0
# Days X 24 hours
6 0
4 0
Limit of discharge
recommended by manufacturer
2 0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
1 4 0
1 6 0
10-48
rate. So the manufacturers nominal capacity for the 3-35A27 is 470 Ah at the 10
hour rate, but the capacity obtained at the slower 100 hour rate would be 624 Ah. If
a range of rates is given, use the formula presented previously for calculating the
time for full discharge to determine the proper rate to use. If the exact discharge
time is not given by the manufacturer just use the time that is closest or less than the
actual time to conservatively size the battery.
Example:
10-49
Type
Volts
Nominal
AH
Capacity
Width
Height
Depth
Cell Type
8h
24
h
48
h
100
h
35A05
72
86
94
96
in
mm
in
mm
in
mm
6-35A05
12
75
17.23
438
8.60
218
14.2
361
6-35A07
12
110
21.67
550
35A07
108
129
141
144
6-35A09
12
145
26.17
665
35A09
144
172
188
192
6-35A11
12
180
30.67
779
35A11
180
215
235
240
6-35A13
12
215
35.17
893
35A13
216
258
282
288
6-35A15
12
250
39.67
1008
35A15
252
301
329
336
3-35A17
290
24.52
623
35A17
288
344
376
384
3-35A19
325
26.77
680
35A19
324
387
423
432
3-35A21
360
29.02
737
35A21
360
430
470
480
3-35A23
395
31.27
794
35A23
396
473
517
528
3-35A25
430
33.52
851
35A25
432
516
563
576
3-35A27
470
35.77
909
35A27
468
559
610
624
!"#$%&!''$$(
)'#*
10-50
Voltage (volts)
20
40
60
80
100
Time (hours)
C/100
10-51
C/10
20
15
10
5
0
-50
-25
25
50
75
10-52
Exercises
3
*
&5 6
4
+ 6 68-&+(
))))))))))
+ 6 685+(
))))))))))
))))))))))
))))))))))
-+ 6 68-&+(
))))))))))
"
!
#
-5++ " &' !
! -' !
685++ 3
-5 6
*
$
(
)))))))))) " )))))))))) 6
"
685+ 3
5 6
#
(
)))))))))) "
10-53
10-54
10-55
13
Open Circuit Voltage
Voltage
(volts)
Specific
Gravity
12
1.200
11
1.100
Specific
Gravity
1.000
10
100
80
60
40
20
Note that there is little variation in specific gravity with state of charge for nickelcadmium cells and voltage trends are different than lead-acid batteries.
10-56
10-57
In addition to estimating the state of charge from the battery terminal voltages at
different rates of charge and discharge, these curves can also be used to determine
the voltage set point for the charge controller low voltage load disconnect set point
or alarm settings. For example, if it is desired to limit the maximum battery depth of
discharge to 80% at the 40 hour rate, the corresponding load disconnect voltage
should be about 11.5 volts.
17
20 hour rate
16
50 hour rate
15
14
40 hour rate
13
5 hour rate
12
11
cut-off voltage
10
0
20
40
60
80
20
40
60
80
100
10-58
Battery Charging
Methods and procedures for battery charging vary considerably. In a stand-alone
PV system the ways in which a battery is charged are generally much different from
the charging methods battery manufacturers use to rate battery performance. The
various methods and considerations for battery charging in PV systems are
discussed in the next chapter on battery charge controllers.
Battery manufacturers often refer to three modes of battery charging: normal or bulk
charge, finishing or float charge and equalizing charge.
10-59
4H 2 O + 4e 2H 2 + 4OH
At the positive plate or electrode:
4OH 2H 2 O + O2 + 4e
Overall nickel-cadmium cell overcharge reaction:
2H 2 O 2H 2 + O2
10-60
10-61
10-62
3.0
2.9
Lead-Antimony Grids
Charge Rate
2.8
C/2.5
2.7
2.6
Gassing Voltage at 0
C/5
2.5
C/20
Gassing Voltage at 27 oC
2.4
Gassing Voltage at 50 oC
2.3
2.2
2.1
2.0
0
20
40
60
80
100
By examining the figure, one can see that at 27o C and at a charge rate of C/20, the
gassing voltage of about 2.35 volts per cell is reached at about 90% state of charge.
At a charge rate of C/5 at 27o C, the gassing voltage is reached at about 75% state
of charge. At a battery temperature of 0o C the gassing voltage increases to about
2.5 volt per cell, or 15 volts for a nominal 12-volt battery. The effects of temperature
on the gassing voltage is the reason the charge regulation voltage is sometimes
temperature compensated - to fully charge batteries in cold weather and to limit
overcharge during warm weather. This type of information is needed to properly
select battery charge controller voltage regulation set points in order to limit the
amount of gassing for a specific battery design and operational conditions.
Some recommended ranges for charge regulation voltages (at 25o C) for different
battery types used in PV systems are presented in Table 10-4 below. These values
are typical of voltage regulation set points for battery charge controllers used in
small PV systems. These recommendations are meant to be only general in nature,
and specific battery manufacturers should be consulted for their suggested values.
10-63
Battery Type
Charge
Regulation
Voltage at 25 oC
Flooded
LeadAntimony
Flooded
LeadCalcium
Sealed, Valve
Regulated
Lead-Acid
Flooded
Pocket Plate
NickelCadmium
Per nominal 12
volt battery
14.4 - 14.8
14.0 - 14.4
14.0 - 14.4
14.5 - 15.0
Per Cell
2.40 - 2.47
2.33 - 2.40
2.33 - 2.40
1.45 - 1.50
The charge regulation voltage ranges presented in Table 10-4 are much higher than
the typical charge regulation values often presented in manufacturers literature.
This is because battery manufacturers often speak of regulation voltage in terms of
the float voltage, or the voltage limit suggested for when batteries are float charged
for extended periods (for example, in non-interruptible power supply (UPS) systems).
In these and many other commercial battery applications, batteries can be trickle or
float charged for extended period, requiring a voltage low enough to limit gassing.
Typical float voltages are between 13.5 and 13.8 volts for a nominal 12-volt battery,
or between 2.25 and 2.30 volts for a single cell.
In a PV system however, the battery must be recharged within a limited time (usually
during sunlight hours), requiring that the regulation voltage be much higher than the
manufacturers float voltage to ensure that the battery is fully recharged. If charge
regulation voltages in a typical PV system were set at the manufacturers
recommended float voltage, the batteries would never be fully charged.
10-64
Exercises
10-65
Battery Trade-Off
Pay Now or Pay Later
Pay less initially for starting type batteries, but get short life (1-2 years)
Pay more initially for deep discharge batteries, and get longer life (4-8
years)
Starting types may be more available, but may not be the best value in
the long run
For low-income rural villagers, the high cost of a better longer life deep cycling
battery may be prohibitive. But they may be able to afford a lower cost, more readily
available and replaceable shallow cycling, starting type battery. In a year or two,
they can afford to buy a replacement. For them, the more affordable choice may be
to not pay now for the best (longest life) but to pay later (for replacements).
On the other hand, a remote telecom site in a difficult or dangerous location would
be best designed with the longest life battery possible. The cost of poor
performance would be much greater than the small savings of a cheaper battery.
The budget would probably be present for purchasing the better longer life battery
initially. The cost savings in reduced maintenance and replacement costs over time
would be of primary consideration in this case.
10-66
10-67
Battery Life
(% life at 25 oC)
1000
Lead-Antimony Grids
Lead-Calcium Grids
Nickel-Cadmium
100
10
5
10
15
20
25
30
35
40
45
10-68
As sown in the figure, the grid alloying elements play an important part in battery life.
The lifetime of batteries with lead-calcium grids is more affected by temperature than
lead-antimony grids or nickel-cadmium batteries.
The importance of temperature can be seen by looking at an example of a
manufacturers warranty statement. A portion of the warranty for GNB Absolyte IIP
batteries is presented on the next page. The portion of interest to us now is
highlighted below.
The wording of the warranty statement has been condensed to focus attention on
the temperature factor.
GNB also warrants ...that GNB ABSOLYTE IIP batteries used in photovoltaic
service will furnish... 80% of the specified capacity...for m years...provided that the
following conditions are satisfied.
.....The average daily ambient temperature in the area of use in any year is
established:
not to exceed 25 degrees C
m equal to 10 years
not to exceed 30 degrees C
m equal to 7 years
not to exceed 35 degrees C
m equal to 5 years
Notice that the warranty period decreases substantially as the average operating
temperature of the battery increases. Notice that with a temperature change from
25 to 35 oC the warranty period decreases by 50%, as per the general rule stated
previously.
10-69
+ ,!"#$%-,!$'$(,!
10-70
Battery Life
(Cycles)
10000
1000
100
Motive Power Battery (deep cycle)
Automotive (SLI) Battery (shallow cycle)
10
0
20
40
60
80
100
10-71
10-72
Battery Efficiency
Battery efficiency is dependent on the type of battery, charging methods, rates of
charge and discharge, depth of discharge and temperature. In general, the
efficiency of a battery is much greater at lower states of charge than when the
battery is nearly fully charged. The total or round-trip battery energy efficiency is
composed of two types of efficiencies. There is a voltage or voltaic efficiency, and a
charge or coulombic efficiency. They measure different characteristics of batteries,
and are often confused.
Recall how battery voltage increases during charging and decreases during
discharging as compared to the open-circuit voltage. Battery voltage in a PV system
may vary considerably, depending on state of charge and rate of charge or
discharge. An overall average voltaic efficiency can be calculated by assuming that
on the average, a battery is charged at about 14 volts and discharged at about 12
volts at the given charge and discharge rates in the system. This approximation
yields an overall voltaic efficiency of about 85%.
10-73
12 volts
= 0.85 = 85%
14 volts
The charge and discharge curves presented below illustrate battery voltaic
efficiency. During charging the voltage is elevated above the open circuit voltage,
and during discharging the voltage is depressed below it. Since the rate of charge
or discharge determines the battery voltage at a certain state of charge, the voltaic
efficiency is higher at lower rates of charge and discharge.
C/5
Battery Voltage
15.0
C/20
14.0
13.0
C/20
Open-Circuit Voltage
12.0
11.0
Discharge Rates
C/5
10.0
0
20
40
60
80
100
10-74
Typical industrial applications using motive power or traction batteries must fully
recharge deeply discharged batteries in a limited time, usually less than 8 to 10
hours. To accomplish this, high charge rates (C/5 to C/10) are required resulting in
low coulombic efficiencies. In PV systems, the coulombic efficiency of batteries is
generally very good due to the relatively low charge rates used. Typical charge rates
in PV systems are often C/20 or lower, because the amount of battery storage
required for autonomy is relatively large with respect to the PV array charging
currents. While an accurate estimation of coulombic efficiency is difficult to
determine, 90% is a typical value for most batteries used in small stand-alone PV
systems.
10-75
Example:
222 Ah
Battery Energy
Efficiency
.85
.80 or 80%
.95
The round-trip battery energy efficiency is often used when discussing batteries.
However it is the coulombic efficiency that is most commonly used in photovoltaic
system sizing methods using the amp-hour approach. In cases where energy or
watt-hour approach is used in system sizing, the energy or round-trip battery
efficiency is sometimes used to account for the voltage as well as the charge
efficiency of the battery, especially for quantifying losses due to gassing. For the
sizing calculations presented in the chapter on System Sizing, the only battery
efficiency that needs to be considered will be the charge or coulombic efficiency,
which is usually about 90% for most batteries.
10-76
Exercise
-&!
4
3
&+
-++
>
8
-&
8
10-77
10-78
Low-antimony, thick plates for reducing water loss and providing adequate
mechanical strength for long cycle life.
Separator envelopes around the positive plates, and appropriate plate edge
protection to minimize internal short-circuit potential.
Large electrolyte volume below plates to allow for shed materials to accumulate
without causing short circuits.
Availability
Due to the weight and hazardous nature of batteries, shipping costs are high, and
results in most batteries produced at regional plants being used to supply to local
markets. In developing regions, the optimal battery for a given application may not
be locally available. In these cases, the best battery available should be used, with
special attention paid to properly sizing and treating the battery for maximum life.
10-79
Ventilation
Batteries often produce toxic and explosive mixtures of gasses, namely hydrogen,
and adequate ventilation of the battery enclosure is required. In most cases,
passive ventilation techniques such as vents or ducts may be sufficient. In some
cases, fans may be required to provide mechanical ventilation. Required air change
rates are based on maintaining minimum levels of hazardous gasses in the
enclosure. Under no circumstances should batteries be kept in an unventilated area
or located in an area frequented by personnel.
10-80
Enclosures
Batteries are generally required by local electrical codes and safety standards to be
installed in an enclosure separated from controls or other PV system components.
The enclosure may also be insulated, or may have active or passive cooling/heating
mechanisms to protect the batteries from excessive temperatures. Battery
enclosures must be of sufficient size and strength hold the batteries, and can be
located below ground if needed to prevent freezing. If the enclosure is located
above ground, care should be taken to limit the direct exposure to sunlight, or some
type of shading or reflective coating should be provided.
10-81
",!'$$
'$
The performance of the Cool CellTM depends on the ambient temperature and the
extent and type of cloud cover. In areas with exceedingly high ambient
temperatures, this enclosure can provide a significant reduction in battery
temperatures compared to a standard box or insulated enclosure. Cloud cover plays
an important effect on the performance of the Cool CellTM as this affects the
radiation heat transfer from the lid. Clear skies generally have an effective
temperature (for purposes of radiation calculations) of below -40 oC, however cloudy
skies have a much higher effective temperature. Since the radiation heat transfer is
proportional to the effective temperature raised to the fourth power, overcast skies
can reduce the cooling effectiveness of this type of battery enclosure.
10-82
The relative performance of the Cool Cell is illustrated below. Four conditions are
compared: (a) the widely varying case of an uninsulated steel box; (b) the ambient
air; (c) a concrete vault with an insulated lid; and (d) the moderated temperature
swings of the passively cooled enclosure.
10-83
IMPORTANT:
Exercise extreme caution and follow recommended practices when
working with batteries!
Handling Electrolyte
The caustic sulfuric acid solution contained in lead-acid batteries can destroy
clothing and burn the skin. For these reasons protective clothing such as aprons
and face shields should be worn by personnel working with batteries. To neutralize
sulfuric acid spills or splashes on clothing, the spill should be rinsed immediately
with a solution of baking soda or household ammonia and water. For nickelcadmium batteries, the potassium hydroxide electrolyte can be neutralized with a
vinegar and water solution. If electrolyte is accidentally splashed in the eyes, the
eyes should be forced open and flooded with cool clean water for fifteen minutes. If
acid electrolyte is taken internally, drink large quantities of water or milk, followed by
milk of magnesia, beaten eggs or vegetable oil. Call a physician immediately.
If it is required that the electrolyte solution be prepared from concentrated acid and
water, the acid should be poured slowly into the water while mixing. The water
should never be poured into the acid. Appropriate non-metallic funnels and
containers should be used when mixing and transferring electrolyte solutions.
10-84
Personnel Protection
When performing battery maintenance, personnel should wear protective clothing
such as aprons, ventilation masks, goggles or face shields and gloves to protect
from acid spills or splashes and fumes. If sulfuric acid comes into contact with skin
or clothing, immediately flush the area with a solution of baking soda or ammonia
and water. Safety showers and eye washes may be required where batteries are
located in close access to personnel. As a good practice, some type of fire
extinguisher should be located in close proximity to the battery area if possible. In
some critical applications, automated fire sprinkler systems may be required to
protect facilities and expensive load equipment. Jewelry on the hands and wrists
should be removed, and properly insulated tools should be used to protect against
inadvertent battery short-circuits.
Dangers of Explosion
During operation, batteries may produce explosive mixtures of hydrogen and oxygen
gasses. Keep spark, flames, burning cigarettes, or other ignition sources away from
batteries at all times. Explosive gasses may be present for several hours after a
battery has been charged. Active or passive ventilation techniques are suggested
and often required, depending on the number of batteries located in an enclosure
and their gassing characteristics. The use of battery vent caps with a flame arrestor
feature lowers the possibility of a catastrophic battery explosion. Improper charging
and excessive overcharging may increase the possibility of battery explosions.
When making and breaking connections to a battery from a charging source or
electrical load, ensure that the charger or load is switched off as to not create sparks
or arcing during the connection.
10-85
Selected References
Institute of Electrical and Electronics Engineers, "IEEE Recommended Practice for Installation and
Operation of Lead-Acid Batteries for Photovoltaic (PV) Systems", ANSI/IEEE Std. 937-1987, New
York, NY, 1987.
Institute of Electrical and Electronics Engineers, "IEEE Recommended Practice for Installation and
Operation of Nickel-Cadmium Batteries for Photovoltaic (PV) Systems", ANSI/IEEE Std. 11451990, New York, NY, 1990.
Stand-Alone Photovoltaic Systems - A Handbook of Recommended Design Practices, Sandia National
Laboratories, SAND87-7023, revised November 1991.
Naval Facilities Engineering Command, Maintenance and Operation of Photovoltaic Power Systems,
NAVFAC MO-405.1, December 1989.
Exide Management and Technology Company, Handbook of Secondary Storage Batteries and Charge
Regulators in Photovoltaic Systems - Final Report, for Sandia National Laboratories,
SAND81-7135, August 1981.
Bechtel National, Inc., Handbook for Battery Energy Storage in Photovoltaic Power Systems, Final Report,
SAND80-7022, February 1980.
S. Harrington and J. Dunlop, "Battery Charge Controller Characteristics in Photovoltaic Systems",
Proceedings of the 7th Annual Battery Conference on Advances and Applications, Long Beach,
California, January 21, 1992.
H.A. Kiehne, Battery Technology Handbook, Marcel Dekker, Inc., 1989.
G.W. Vinal, Storage Batteries, John Wiley & Sons, Inc., Fourth Edition, 1954.
D. Linden, Handbook of Batteries and Fuel Cells, McGraw Hill, Inc., 1984.
10-86
(End of Chapter)
10-87
CHAPTER TEN
BATTERY TECHNOLOGY
10-1
10-2
Systems
Energy Storage Capacity and Autonomy
Voltage and Current Stabilization
Supply Surge Currents
10-2
10-3
10-4
10-5
10-6
10-6
10-6
10-7
10-8
10-8
10-8
10-9
10-9
10-9
10-9
10-10
10-10
10-10
10-12
Lead-Acid Batteries
SLI Batteries
Motive Power or Traction Batteries
Stationary Batteries
10-12
10-12
10-12
10-12
10-13
10-13
10-14
10-14
10-16
10-88
10-18
10-18
10-19
10-19
10-20
10-22
Nickel-Cadmium Batteries
Sintered Plate Ni-Cads
Pocket Plate Ni-Cads
Nickel-Cadmium Battery Chemistry
10-23
10-23
10-23
10-24
10-27
Battery Cycling
Battery Discharging
Charge/Discharge Rates Expressed as Time
Depth of Discharge (DOD)
Autonomy Determines Capacity and Depth of Discharge
State of Charge (SOC)
Temperature Limits Discharge Depth
Shallow and Deep Cycle Batteries
10-28
10-29
10-30
10-31
10-32
10-35
10-36
10-39
Battery Capacity
Ampere-Hour Definition
Factors Affecting Battery Capacity
Effects of Temperature on Capacity
Discharge Rate Affects Capacity
Cut-Off Voltage Affects Capacity
Effects of Self Discharge Rate on Capacity
10-41
10-41
10-41
10-42
10-44
10-51
10-52
10-54
10-55
10-57
Battery Charging
Battery Gassing and Overcharge
Charge Regulation Voltage Affects Gassing
10-59
10-59
10-62
10-66
10-89
10-67
10-67
10-68
10-71
10-72
10-72
10-72
Battery Efficiency
Battery Voltage (Voltaic) Efficiency
Battery Charge (Coulombic) Efficiency
Battery Round-Trip or Energy Efficiency
10-73
10-73
10-75
10-76
10-78
10-80
10-80
10-80
10-80
10-81
10-84
10-84
10-85
10-85
10-85
Selected References
10-86
10-90
Chapter 10 Answers
Battery Technology
a. Positive Plate
PbO2 + 4H+ + 2e- Pb2+ + 2H2O
Element / Ions
Lead (Pb)
Oxygen (O)
Hydrogen (H)
Ion Charges
Qty on Left
1
2
4
+4 -2 = +2
Qty on Right
1
2
4
+2
Qty on Left
1
4
1
+2 -2 = 0
Qty on Right
1
4
1
0
Qty on Left
1
0
Qty on Right
1
+2 -2 = 0
Qty on Left
1
4
1
+2 -2 = 0
Qty on Right
1
4
1
0
b. Negative Plate
Pb Pb2+ + 2eElement / Ions
Lead (Pb)
Ion Charges
Qty on Left
2
10
2
4
10-1
Qty on Right
2
10
2
4
Battery Technology
For a 24-volt (nominal) battery, we need
24 volts / 2.1 volts per cell = 11.4 = 12 lead acid cells
24 volts / 1.2 volts per cell = 20 nickel cadmium cells
Note: in most cases a lead-acid cell is assumed to have a nominal voltage of 2.0 volts.
We assume the nominal voltage of a lead-acid cell is 2 volts. Then 24 cells will have a
total nominal voltage of 24 X 2 Volts = 48 Volts.
b. Lower than
10-2
Battery Technology
Referring to the chart of Temperature Limits for Maximum Depth of Discharge, we see
that at a temperature of -40 C, the maximum recommended depth of discharge is
approximately 25%.
At first, we might be tempted to answer that we need a battery that is four times as
large. However, this would only be true if we assumed that we could use 100% of the
original battery. Since the maximum recommended depth of discharge is 80% at
normal temperature, the battery for the mountain application needs to be larger by a
factor of
80% = 3.2 times larger
25%
An example helps make this clearer. Assume we need 800 Ahr of storage. A 1000 Ahr
battery designed for use at normal temperatures can provide 800 Ahr storage to 80%.
Now, if we can only discharge the battery to 25% DOD, the battery must provide the
same 800 Ahr capacity. Using the ratio of 3.2 that we calculated above, we choose a
3,200 Ahr battery. Notice that 25% discharge = 25% X 3200 = 800 Ahrs as is required.
Since low temperatures can have a significant impact on the required battery size, we
would be very interested in any methods that might keep the battery from getting cold.
Some possible techniques to do this might include:
Keeping the batteries in a heated area (e.g. an equipment shelter)
Locating the batteries in a buried vault
Providing insulation to minimize the temperature changes
Refer to the manufacturer's literature.
a. 0 oC and C/120?
92%
b. 0 oC and C/50?
85%
90%
d. 25 oC and C/500?
100%
e. 10 oC and C/120?
95%
These values are taken from the chart showing Temperature and Discharge effects on
Lead Acid Capacity.
Siemens Solar Basic Photovoltaic Technology
10-3
Battery Technology
Referring to the chart, we see that at -15 C and C/500, the battery will have about 90%
of its standard rated capacity (at 25 C). The increased capacity required is
1500 Ahr
0.90
1667 Ahr at 25 C
At 5 C and C/50, the battery will have about 87% of its standard rated capacity (at 25
C). The increased capacity required is
200 Ahr
0.87
230 Ahr at 25 C
Refer to the manufacturer's literature.
At 12 volts, the current required by the loads is:
Lights
20 watts / 12 volts = 1.7 amps
Refrigerator 100 watts / 12 volts = 8.3 amps
System Description:
DC Loads
Lights
Refrigerator
:
:
Qty.
2
1
X
X
Amps
1.7
8.3
X
X
Hours
/day
6
12
DC Loads (Ah)
Ah From Array
133.8 Ah
10-4
=
=
Daily
Demand (Ah)
20.4
100.0
120.4
Battery Technology
Chapter Eleven
Inverter Technology
The photovoltaic array and battery produce DC current and voltage. If the loads require
AC power an inverter can be used to convert from DC to AC. Commonly available
inverters can output in 1- or 3-phase, 50 or 60 hertz, and 117 or 220 volts and can
range in continuous output power from a few hundred watts to thousands of kilowatts.
Large utility scale inverters are made to output at 480 volts AC or higher and have
capacities exceeding 1000 kilowatts.
In this section we will examine some of the technical features and characteristics of
inverter technology commonly used in photovoltaic systems. The three most important
characteristics of inverters that we will discuss will be:
Output efficiency
Output waveform
11-1
11-2
Surge Power
Notice that the output power of the inverter in the figure above can exceed the nominal
2000 watts for a few minutes. Often this is referred to as the surge capability of the
inverter, although strictly surge power may be defined as the power that the inverter can
output for less than a second to start large inductive (motor) loads. Recall from our
discussions in the chapter on Load Estimation that many inductive loads (motors,
compressors, refrigerators, washing machines, and large water pumps) can draw 5-6
times their nominal running power when they start!
For example, a remote home might have non-inductive loads of 1000 watts and a
3/4-hp water pump motor. Normally the water pump might draw about 700 watts, so
the total continuous power needed from the inverter would be about 1700 watts. But
when the pump turns on it draws about 3500 watts. If it turned on when the other 1000
watts of load were operating, the inverter would have to output about 4500 watts (1000
non-inductive + 3500 inductive surge) for a few moments to start the motor. The
inverter example used above in the figure could output about 6000 watts for a few
seconds, so this would be adequate. An inverter with a lower surge capacity might not
be able to start the pump.
11-3
Power vs Time
Output Power (watts)
6000
5000
4000
3000
2000
1000
0
20
40
60
Time (minutes)
The surge power of an inverter is again not a single fixed number, but varies with the
amount of time. A good inverter can typically output 200-300% of its nominal power for
a few seconds. Some low cost inverters can only output 110% of their output for a
short time. These would have limited surge capacity and would not be useful for
systems where inductive loads would operate.
11-4
11-5
Exercise
!
"
#
" $#%
11-6
Inverter Efficiency
An inverter consumes some power itself. This adds to the total load that the
photovoltaic array must operate and increases array size and initial cost. The inverter
should be as efficient as possible, certainly above 90% over most of its normal
operating range. Many moderately priced inverters can achieve 94% efficiency. The
efficiency may depend on the nature of the load. Pure resistive loads may operate the
inverter at a higher efficiency than inductive loads that absorb the power differently.
11-7
Inverter Efficiency
Efficiency
100
new Transistor type
90
80
old SCR type
70
60
50
40
30
20
10
0
0
20
40
60
80
100
11-8
(+)
12 v DC
(+)
120 v AC
Inverter
battery
variable
load
(ex: lights) .
(-)
(-)
shunt #2
shunt #1
11-9
Exercise
& &
&
&'
'
&
!
&
&(
'
'%
)
&
*
++,-
./0/
,
%
1 &2
3333333333/
45&2
3333333333/
&
&
&
6
7
&./
&
11-10
Squarewave
Generally the most inexpensive inverters are "squarewave" types. The input DC power
is chopped and boosted in voltage, with little filtering or modulation of the output. The
resulting output contains many unwanted harmonics, or waves of various frequencies
that fight against each other. The measure of this unwanted waveform content, the
total harmonic distortion, can be 40% or greater. Typically squarewave inverters cannot
surge significantly, perhaps only 10-20 % above the maximum continuous power. The
efficiency can be as low as 50-60%. And they have little output voltage regulation.
However these inverters may be useful for small inductive loads or resistive loads.
Squarewave Inverters
11-11
Modified Squarewave or
Modified Sinewave
Another type of inverter output is available that produces a "modified squarewave"
output. Sometimes these are referred to as modified sinewave in manufacturers
literature, although the form more closely resembles a squarewave with a delay through
zero. The peak voltage of the waveform is only about 150 volts, while utility sinewave
power peaks at about 175 volts. The total harmonic distortion is significantly less than a
squarewave, perhaps less than 20%. Also they can surge 300-400% above the
continuous power, and have good voltage regulation. Efficiencies of greater than 90%
are common. These types of inverters are quite popular in remote homes, and can
operate a wide variety of common appliances and electronic loads including home
computers and stereos. However there are reported problems with operating some
specific loads. Some models of laser printers malfunction or are damaged; microwave
ovens may not heat as much and may have to be operated longer; and small electronic
clocks may run twice as fast. You should check with the manufacturer to see if any
incompatibility with your planned loads have been reported.
11-12
Sinewave
Finally there are types of inverters that output a nearly pure "sinewave. These types of
inverters may involve extensive and carefully tuned filtering or digital synthesis. In the
past these types of inverters tended to have efficiencies and surge capabilities slightly
below the modified sinewave types, but recently models have been introduced with
efficiencies above 90% and costs below some modified squarewave models. In
general, it is always best to choose a sinewave inverter and supply a sinewave output
to your AC loads. This is what they were designed for, and little or no problems will
occur with respect to harmonic distortions or inadequate peak voltages. Some classes
of electronic loads, such as telecommunications or delicate instrumentation may require
a sinewave power waveform.
Sinewave Inverters
11-13
Exercise
8$
&
5& &
11-14
First convert to the final output frequency of 60/50Hz AC and then transform voltage
Upconvert the voltage first and then convert to 60/50 Hz AC for final output.
11-15
Multiple Transformers
Another newer topology (used in the sinewave utility interactive Trace Engineering
inverters for example) involves not one but three transformers. One transformer ratio is
3:1, another is 9:1 and another is 27:1. These are switched on and off in an additive or
subtractive manner to create a variable ratio configuration producing many small steps
(around 50 for the Trace model, depending on battery voltage). This allows for a
stepped approximation to a sinewave output waveform.
11-16
11-17
11-18
Exercise
11-19
Explanation
Considerations in PV System
Design
Output
Voltage and
Current
Waveform
Type
11-20
No Load
Consumption
Total
Harmonic
Distortion
Output Power
Surge Power
11-21
RMS Voltage
and Peak
Voltage
Power Factor
Protective
Circuitry
Battery
Charger
Three-stage charging is
preferred; Stage One: Bulk
Charge with maximum charge
amps to maximum voltage.
Stage Two: Constant
Voltage/Absorption, where
current tapers off at rate required
to hold voltage constant. Stage
Three: Float Voltage, which
holds full charge without
gassing. In addition an
equalization stage should be
included which allows charging
to a voltage high enough to gas
batteries in order to remove
sulfation.
11-22
Utility
Transfer
Relay
UL Listing
Grounding
Efficiency
11-23
DC Input Voltage
The input DC voltage tends to be a function of the size of the inverter. As the power
through the inverter increases, more current flows and there is greater internal heating.
Manufacturers tend to increase the DC voltage for larger inverters to keep the currents
at a manageable level, around 100 amps or so. So small inverters from 100-2500 watts
continuous tend to be available in 12 volts, while inverters in the 2000-3500 watt range
are made to operate at 24 volts, and models from about 2400-5000 watts are designed
with 48 volt input. Inverters with output greater than 5 kW can have input DC voltages
of 120 to 240 volts and even higher.
If all the loads in a system will be AC then the choice of input voltage is fairly free to
make. The higher the DC voltage the lower the currents into the inverter. This means
that smaller wire can be used, as well as smaller and less expensive disconnect
switches or circuit breakers and fuses. The DC voltage of the inverter will set the
voltage of the battery bank and the array as well.
Do not make the common mistake of thinking that a 24-volt inverter will need twice as
many modules as a 12-volt inverter. The 24-volt choice will use just as many total
modules, but they will be configured differently. Instead of all the modules being
connected in parallel for 12 volts, half as many will be connected in parallel but each
module will be connected to another in series. The total will be the same. And
comparing a 48-volt choice to a 24-volt choice would be similar. Strings of four
modules in series would be used to develop the 48 volts instead of strings of two in
series for 24 volts, but there would be half as many strings in parallel compared to a 24volt case. If DC and AC loads are to be operated in the same system, then the choice
of input voltage is pretty well restricted to 12, or perhaps 24 volts in some cases. Most
all DC loads are designed to operate at 12 volts. Some ballasts for fluorescent lights
and some refrigerator models come in a 24 volt version.
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
11-24
Voltage Regulation
At high power levels, the inverter draws larger currents from the battery. This causes
the battery voltage to fall. The inverter should be able to compensate for this voltage
drop and maintain output AC voltage fairly well. This information may not be presented
with the usual efficiency curves and is quite important. If output voltage drops too
much, then loads may not operate correctly.
Serviceability
The inverter design should allow easy servicing in the field, or allow for cards to be
swapped and exchanged to minimize down time.
Adjustable Threshold
Most inverters have some threshold of load power requirement before they actually
"turn on" and commutate to produce AC power. This may be a fixed power level, or
may be adjustable by the user. If the load threshold is higher than some small loads in
a house, for example a small electronic clock, the inverter may not sense the load and it
will not operate alone. Some other loads must be on so that the total power is greater
than the threshold value. Some inverter models offer an adjustable threshold level.
However, the lower the threshold level the slightly lower the efficiency of the inverter.
11-25
(End Of Chapter)
11-26
CHAPTER ELEVEN
INVERTER TECHNOLOGY
11-1
11-2
11-2
11-3
11-5
Inverter Efficiency
Efficiency Depends On Load
Inductive vs. Resistive Loads
Test Set-up For Measuring Efficiency
11-7
11-7
11-9
11-9
11-11
11-11
11-12
11-13
11-15
11-15
11-16
11-16
11-16
11-17
11-18
11-20
11-24
11-24
11-25
11-25
11-25
11-25
11-27
Chapter 11 Answers
Inverter Technology
Refer to the manufacturer's literature.
Refer to the manufacturer's literature.
The average efficiency of the transistor based inverter is about 92%. The average
efficiency of the SCR based inverter is about 75%.
Refer to the manufacturer's literature.
Compare waveforms to those shown in the text.
Refer to the manufacturer's literature.
11-1
Inverter Technology
Chapter Twelve
Charge Regulators and
System Controls
Photovoltaic modules are highly reliable and virtually maintenance-free; but,
modules alone do not solve a customers power problem. Other components are
generally required to properly control, distribute and store the energy produced by
the array to power an electrical load.
In stand-alone PV systems the array is usually connected to batteries that store the
array energy and supply power to the electrical loads on demand. In most cases,
when batteries are included in a PV system, a charge regulator or controller is
required to protect the batteries from being overcharged by the array and
overdischarged by the system loads. In some designs the charge regulator or
system controller may also provide status information to the user/operator on system
performance and battery state of charge.
In this chapter we will discuss the purpose of battery charge regulators and controls
in PV systems, their important features, and how different types operate to control
the flow of energy in a remote PV power system.
12-1
12-2
Charge regulation is the primary function of a battery charge controller and perhaps
the single most important issue related to battery performance and life. The purpose
of a charge controller is to supply power to the battery in a manner that fully
recharges the battery without overcharging. Without charge control the current from
the array will flow into a battery proportional to the irradiance, whether the battery
needs charging or not. If the battery is fully charged unregulated charging will cause
the battery voltage to reach exceedingly high levels, causing severe gassing,
electrolyte loss, internal heating and accelerated grid corrosion. In most cases if a
battery is not protected from overcharge in PV system, premature failure of the
battery and loss of load are likely to occur.
Charge regulators prevent excessive battery overcharge by interrupting or limiting
the current flow from the array to the battery when the battery becomes fully
charged. Charge regulation is most often accomplished by limiting the battery
voltage to a maximum value, often referred to as the voltage regulation (VR) set
point. Sometimes other methods such as integrating the ampere-hours into and out
of the battery are used. Depending on the regulation method the current may be
limited while maintaining the regulation voltage, or remain disconnected until the
battery voltage drops to the array reconnect voltage (ARV) set point. A further
discussion of charge regulation strategy set points is contained later in this chapter.
12-3
Non-critical system loads are generally always protected from overdischarging the
battery by connection to the low voltage load disconnect circuitry of the charge
controller. If the battery voltage falls to a low but safe level, a relay can open and
disconnect the load, preventing further battery discharge. Critical loads can be
connected directly to the battery, so that they are not automatically disconnected by
the charge regulator. However, the danger exists that these critical loads might
overdischarge the battery. An alarm or other method of user feedback should be
included to give information on the battery status if critical loads are connected
directly to the battery.
12-4
12-5
12-6
12-7
Terminology
While the specific regulation method or algorithm may vary among different charge
regulators, all have basic functions and characteristics. Charge regulator
manufacturer's data generally provides information about these functions and their
specifications.
12-8
Battery Voltage
Charging
Discharging
Time
12-9
12-10
12-11
12-12
The proper LVD set point will maintain a healthy battery while providing the
maximum battery capacity and load availability. To determine the proper load
disconnect voltage, the designer must consider the rate at which the battery is
discharged. Because the battery voltage is affected by the rate of discharge, a lower
load disconnect voltage set point is needed for high discharge rates to achieve the
same depth of discharge limit. In general, the low discharge rates in most small
stand-alone PV systems do not have a significant effect on the battery voltage.
Typical LVD values used are between 11.0 and 11.5 volts, which corresponds to
about 75-90% depth of discharge for most nominal 12 volt lead-acid batteries at
discharge rates lower than C/30.
A word of caution is in order when selecting the low voltage load disconnect set
point. Battery manufacturers rate discharge capacity to a specified cut-off voltage
that corresponds to 100% depth of discharge for the battery. For lead-acid batteries,
this cut-off voltage is typically 10.5 volts for a nominal 12-volt battery (1.75 volts per
cell). In PV systems, we never want to allow a battery to be completely discharged,
as this will shorten its service life. In general, the low voltage load disconnect set
point in PV systems is selected to discharge the battery to no greater than 75-80%
depth of discharge.
In cases where starting (SLI) batteries are used or it is otherwise desired to limit the
battery depth of discharge to prevent freezing or prolong cycle life, a higher LVD set
point may be desired. To protect the battery from freezing, the LVD set point may
be temperature compensated in some cases to increase the load disconnect voltage
automatically with decreasing battery temperature.
To properly specify the LVD set point in PV systems the designer must know how
the battery voltage is affected at different states of charge and discharge rates. In a
few designs, current compensation may be included in the LVD circuitry to lower the
LVD set point with increasing discharge rates to effectively keep a consistent depth
of discharge limit at which the LVD occurs.
12-13
12-14
12-15
12-16
12-17
Operational Limits
The environmental and mechanical design limits of charge regulators are an
important consideration for most PV applications. Generally, PV systems and
components are installed in remote areas, in unconditioned spaces, and subject to
the extremes of the weather. For these reasons most regulators have minimum and
maximum ratings for ambient temperature, battery temperature and relative
humidity. Where extreme environmental conditions exist the designer should
consider these specifications when selecting charge regulators.
The packaging and physical characteristics are another important characteristic of
charge regulators. In general, the regulator circuitry should be sealed from the
environment, either by conformably coating or potting the circuitry. A rigid case
should protect the regulator or the controller may be installed in a weatherproof
enclosure.
Terminations used to connect wiring to the charge regulator should be corrosion
resistant, and be large and sturdy enough to accept the conductor sizes that may be
used in the system.
12-18
12-19
Standard Configurations Of
Charge Regulation and Control
Systems
There are a number of ways that battery charge regulation and system controls are
configured photovoltaic systems. Depending on the type and size of the system, the
cost and availability of hardware, and designer choice, certain regulation and control
configurations may be more appropriate than others. Some of the more common
configurations of charge regulation and system control are described below. The
electronic design of charge regulators is presented in the next section.
12-20
Charge
Regulator
Battery
Load
Charge
Regulator
Load
Battery
12-21
The figure below shows a basic configuration for auxiliary load connection in PV
systems with similar battery overdischarge protection as described for the series
path configuration. In this case when the battery reaches full state of charge, the
charge regulator diverts the array power to an auxiliary load while the primary load
continues to operate from the battery. In most cases these regulators connect the
array directly to the auxiliary load, and bypass the circuit to the battery. In this
design it should be noted that the array power and voltage delivered to the load will
not be regulated, so the load must be compatible with the array output. Electrical
loads that can be directly connected to the array output include auxiliary batteries,
DC motor appliances such a pumps or fans, or resistive elements used for water or
space heating. Whenever excess energy is available from the array, these types of
loads can absorb it. When the primary battery voltage drops and array power is
once again needed to recharge it the auxiliary loads power is cut off.
During the summer months charge regulation limits the array energy utilization by as
much as 50% in most small stand-alone PV applications. The use of an auxiliary
load can make full use of all the array energy throughout the year that would
otherwise be wasted. Array energy diversion to auxiliary loads is usually used in
remote homes or cabins, where the primary load usage is intermittent or infrequent,
and where a non-essential load can be utilized. Generally, commercial and
industrial PV applications have little need for the common types of auxiliary loads.
PV Array
Charge
Regulator
Battery
Auxiliary
Load
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
12-22
PV Array
Battery
Charge
Regulator
Auxiliary
Load
12-23
Charge
Regulator
Sub-Array #2
Charge
Regulator
Sub-Array #3
Charge
Regulator
Battery
Optional LVD
Sub-Array #3
One sub-array may be connected directly to the
battery for finishing charge
Load
12-24
For this configuration, a common design approach is that several regulators are
used in parallel, each with its own sub-array connected. Another approach is one
that a master controller operates relays switching each sub-array, each set to a
slightly different voltage point. For example, a 12 volt array could have one subarray disconnected at 13.8 volts, a second at 14.0 volts, a third at 14.2 volts, and so
on. In this way the charge current is reduced in steps, and the battery gradually
approaches full charge. This approach allows for the use of simple on-off type
controls and yet results in the same gentle approach to full charge as complex multistage and constant-voltage regulators.
The switching elements used in sub-array switching control designs can either be
electro-mechanical relays or long-life mercury displacement relays. By separating
the control mechanism (relay) from the charge control circuitry greater flexibility in
design can be enjoyed, and very large array currents can be regulated with simple
single stage on-off type control elements (relays).
In systems with many sub-arrays one small sub-array may be left connected directly
to the battery (with appropriate fusing and disconnect switch for wire protection and
code compliance). This final sub-array current provides a gentle finishing charge at
the end of the day. The general requirement is that this one sub array contributes
only low charge rates (for example C/90 or less) to the battery to limit the battery
voltage.
The number of parallel sub-arrays determines the number of regulators or relays
required. To determine the number of regulators or sub-array relays required, divide
the maximum possible array current by the current capacity of one regulator or relay.
The maximum array current is the short circuit current of the entire array increased
by a safety factor for possible cloud and ground reflectance, usually a factor of 130%
is used.
Number of Sub-Arrays
and Regulators
12-25
Example:
35 watts/ module
114 modules
Since the system requires 24 volts, this means that two modules are
wired in series and 57 strings of two-in-series are wired in parallel. So
the number of parallel modules is 57.
114 modules total
2 in series
57 in parallel
12-26
Exercise
!
"#
$%!
&'
((((((((((
12-27
12-28
12-29
B a s ic S h u n t R e g u la to r D e s ig n
L o a d S w itc h in g E le m e n t
B lo c k in g D io d e
+
PV
A rra y
R e g u la tio n
C o n tro l
LVD
C o n tro l
DC
Load
B a tte ry
S h u n t E le m e n t
Shunt-Interrupting Design
The shunt-interrupting regulator completely disconnects the array current in an
interrupting or on-off fashion when the battery reaches the voltage regulation set
point. When the battery decreases to the array reconnect voltage, the regulator
connects the array to resume charging the battery. This cycling between the
regulation voltage and array reconnect voltage is why these regulators are often
called on-off or pulsing controllers. These are NOT to be confused with a true
pulsing or PWM type control, discussed later. Shunt-interrupting regulators are
widely available and are low cost, however they are generally limited to use in
systems with array currents less than 20 amps due to heat dissipation requirements.
In general, on-off shunt regulators consume less power than series type regulators
that use relays (discussed later), so they are best suited for small systems where
even minor parasitic losses become a significant part of the system load.
Shunt-interrupting charge regulators can be used on all battery types, however the
way in which they apply power to the battery may not be optimal for all battery
designs. In general, constant-voltage, PWM or linear regulator designs are
recommended by manufacturers of gelled and AGM lead-acid batteries. However,
shunt-interrupting regulators are simple, low cost and perform well in most small
stand-alone PV systems.
12-30
Shunt-Linear Design
Once a battery becomes nearly fully charged a shunt-linear regulator maintains the
battery at near a fixed voltage by gradually shunting the array through a
semiconductor regulation element. In some designs a comparator circuit in the
regulator senses the battery voltage, and makes corresponding adjustments to the
impedance of the shunt element, thus regulating the array current. In other designs
simple Zener power diodes are used, which are the limiting factor in the cost and
power ratings for these regulators. There is generally more heat dissipation in a
shunt-linear regulator than in shunt-interrupting types.
Shunt-linear regulators are popular for use with sealed VRLA batteries. This
algorithm applies power to the battery in a preferential method for these types of
batteries, by limiting the current while holding the battery at the regulation voltage.
12-31
The PWM design allows greater control over exactly how a battery approaches full
charge, and the high speed switching elements generate little heat. PWM type
regulators can be used with all battery types, however the controlled manner in
which power is applied to the battery makes them preferential for use with sealed
VRLA types batteries over on-off type controls. To limit overcharge and gassing, the
voltage regulation set points for PWM and constant-voltage controllers are generally
specified lower than those for on-off type regulators. For example, a PWM regulator
operating with a nominal 12 volt flooded lead-antimony battery might use a VR set
point of 14.4 to 14.6 volts at 25o C, while an on-off regulator used with the same
battery might require a VR set point of between 14.6 and 14.8 volts to fully recharge
the battery on a typical day.
P u ls e -W id th -M o d u la te d (P W M )
R e g u la to r D e s ig n
B a tte ry a t L o w S ta te o f C h a rg e
U n til th e b a tte ry is fu lly c h a rg e d ,
th e c u rre n t p u ls e s to th e b a tte ry
a re w id e .
B a tte ry a t H ig h S ta te o f C h a rg e
O n c e th e b a tte ry b e c o m e s fu lly
c h a rg e d , th e c u rre n t p u ls e s to
th e b a tte ry b e c o m e n a rro w e r.
!
12-32
A relay or solid-state switch can open the circuit between the array and the
battery to discontinue charging.
Array current is pulsed rapidly, and the pulse time or width is varied to keep the
battery voltage constant (PWM type).
A control circuit limits the current in a series-linear manner to hold the battery
voltage at a high value.
Because the series regulator open-circuits rather than short-circuits the array as in
shunt-controllers, no blocking diode is needed to prevent the battery from shortcircuiting when the controller regulates.
L o a d S w itc h in g E le m e n t
+
PV
A rray
R e g u latio n
C o n tro l
LVD
C o n tro l
DC
Load
B a tte ry
12-33
12-34
Once the array is disconnected from the battery at the higher regulation set point,
the voltage drops to the array reconnect voltage and the array is again connected to
the battery. However, on the second and subsequent cycles of the day, a lower
regulation voltage set point is used to limit battery overcharge and gassing.
This type of regulation strategy can be effective at maintaining high battery state of
charge while minimizing battery gassing and water loss for flooded lead-acid types.
The designer must make sure that the dual regulation set points are properly
adjusted for the battery type used. For example, typical set point values (at 25o C)
for this type of controller used with a flooded lead-antimony battery might be 15.0 to
15.3 volts for the higher regulation voltage, and between 14.2 and 14.4 volts for the
lower regulation voltage.
12-35
Exercise
.
#
/
0
#
*
**1+2#
03
# #
4
* #
# #
12-36
12-37
The sizing of the battery, PV array and load profile in the test systems was
configured to typify commercially available PV lighting systems. The different
charge controllers were selected from those commonly used in these type and sizes
of systems. The following table lists the nominal specifications for the FSEC test
systems.
Location:
Design Insolation:
PV Array:
Battery:
Load:
A final word of caution when examining the following daily operational profiles for the
different charge controllers. Since these were test systems designed to investigate
not only the behavior of the different controllers, but the effects the regulation set
points had on maintaining battery state of charge, the set points were not always
optimized for the specific system design. In some cases this was intentional while in
other cases was the result of the controller operating characteristics. The main point
to emphasize however is that the daily profiles presented here show how the charge
controllers typically operate in PV systems.
12-38
12-39
12-40
15
25
Vbat, max
20
14
Vbat, avg
15
13
Vbat, min
10
12
5
Vpv, avg
11
0
4
12
16
20
24
Irradiance (W/m2)
1000
10
Sun
7.5
Ipv, avg
800
5
600
2.5
400
0
Ibat, avg
-2.5
200
-5
0
4
12
16
20
24
#
12-41
12-42
15
25
Vpv, avg
20
Vbat, max
14
Vbat, avg
15
13
10
Vbat, min
12
11
0
4
12
16
20
24
Irradiance (W/m2)
1000
10
Sun
7.5
800
Ipv, avg
600
2.5
400
0
Ibat, avg
-2.5
200
-5
0
4
12
16
20
24
$
12-43
12-44
15
25
Vbat, max
Vbat, avg
14
20
Vbat, min
15
13
10
Vpv, avg
12
11
0
4
12
16
20
24
Irradiance (W/m2)
1000
10
Sun
7.5
800
5
600
Ipv, avg
2.5
400
0
Ibat, avg
-2.5
200
-5
0
4
12
16
20
24
12-45
12-46
15
25
Vbat, avg
Vbat, min
Vbat, max
14
Vbat, max
Vbat, avg
Vbat, min
20
Vpv, avg
15
13
Vpv, avg
10
12
5
11
0
4
12
16
20
24
Irradiance (W/m2)
1000
10
Sun
7.5
800
Ipv, avg
600
2.5
400
0
Ibat, avg
-2.5
200
-5
0
4
12
16
20
24
12-47
12-48
15
25
Vbat, max
Vbat, avg
Vbat, min
14
20
15
13
Vpv, avg
10
12
5
11
0
4
12
16
20
24
Irradiance (W/m2)
1000
10
Sun
7.5
Ipv, avg
800
5
600
2.5
400
0
Ibat, avg
-2.5
200
-5
0
4
12
16
20
24
12-49
Exercises
12-50
12-51
Some degree of gassing and overcharge is required for flooded lead-acid batteries,
but can be harmful to sealed VRLA captive electrolyte type batteries for which lost
electrolyte can not be replaced. In general, sealed maintenance free valveregulated batteries (using lead-calcium grids) require lower charge regulation voltage
set points than flooded deep cycling batteries (using lead-antimony grids). Battery
manufacturers should be consulted to determine the gassing voltages for specific
designs.
The figure shows the relationships between cell voltage, state of charge, charge rate
and temperature for a typical lead-acid cell with lead-antimony grids. At 27o C and at
a charge rate of C/20, the gassing voltage of about 2.35 volts per cell is reached at
about 90% state of charge (point A). At a charge rate five times faster (C/5) at 27o
C, the gassing voltage is reached at only 75% state of charge (point B). At a low
battery temperature of 0o C gassing does not occur until about 2.5 volt per cell, or 15
volts for a nominal 12-volt battery.
3.0
2.9
Lead-Antimony Grids
Charge Rate
2.8
C/2.5
2.7
2.6
Gassing Voltage at 0
C/5
2.5
C/20
Gassing Voltage at 27 oC
2.4
Gassing Voltage at 50 oC
2.3
2.2
2.1
2.0
0
20
40
60
80
100
12-52
Regulator Parameters
Battery Parameters
Regulator
Design Type
Charge
Regulation
Voltage at
25o C
Flooded
LeadAntimony
Flooded
LeadCalcium
Sealed,
Valve
Regulated
Lead-Acid
Flooded
Pocket Plate
NickelCadmium
On-Off,
Interrupting
Per 12 volt
battery
Per Cell
14.6 - 14.8
14.2 - 14.4
14.2 - 14.4
14.5 - 15.0
2.44 - 2.47
2.37 - 2.40
2.37 - 2.40
1.45 - 1.50
Per 12 volt
battery
14.4 - 14.6
14.0 - 14.2
14.0 - 14.2
14.5 - 15.0
Per Cell
2.40 - 2.44
2.33 - 2.37
2.33 - 2.37
1.45 - 1.50
ConstantVoltage,
PWM, Linear
12-53
Temperature Compensation
As discussed previously, the electrochemical reaction and gassing in a battery is
highly dependent on temperature. Lower battery temperatures slow down the
reaction, reduce capacity and raise the voltage required for gassing. Conversely,
higher temperatures accelerate the reaction, increase grid corrosion, and lower the
gassing voltage.
For these reasons temperature compensation (TC) of the VR set point is
recommended in PV systems where battery temperature might vary more than 5 o C
from 25o C. Temperature compensation is also required for all type of sealed VRLA
captive electrolyte batteries where any gassing means permanent loss of electrolyte
and life. By using TC a battery can be fully charged during cold weather, and
protected from overcharge during hot weather.
A widely accepted value of temperature compensation for lead-acid batteries is -5
mV/ o C /cell. For a nominal 12 volt battery, this amounts to -30 mV /o C. Where
battery temperatures vary by as much as 30o C, temperature compensation may
result in the regulation set point varying by as much as 1.0 volt in a 12 volt system.
It is important to notice that the TC coefficient is negative, meaning that increases in
temperature require a reduction in the charge regulation voltage. If battery
temperatures are lower than the design condition, the regulation voltage is increased
to allow the battery to reach a moderate gassing level and fully recharge.
Conversely, the regulation set point is reduced if battery temperatures are greater
than design conditions.
Battery temperatures may be sensed with an external probe connected to the
regulator, or approximated with an on-board sensor in regulator circuitry.
-.030 volts/o C
12-54
If the electrolyte concentration has been adjusted for local ambient temperature
(increase in specific gravity for cold environments, decrease in specific gravity for
warm environments) and temperature variation of the batteries is minimal,
compensation may not be as critical. Typically, the LVD set point is not temperature
compensated unless the batteries operate below 0o C on a frequent basis.
A graphical representation of the change in the charge regulation voltage with
temperature for a typical 12-volt lead acid battery is shown in the figure below. At
25o C the charge regulation about 14.5 volts. As the battery temperature decreases,
the charge regulation voltage increases. Conversely, as the battery temperature
increases, the charge regulation voltage must be reduced.
Temperature Compensation
Final Charging Voltage
for a 12 volt battery (volts)
16
15
14
13
10
30
20
40
Battery Temperature (deg.C)
50
12-55
Example:
Final voltage
+.3 volts
14.5 + .3
14.8 volts
With temperature compensation built into the regulator, batteries in cold climates will
receive the full charging they require, and batteries in hot climates will not be
overcharged and gassed excessively. It is often considered an option, but should be
incorporated into every system. It should always be included when sealed batteries
are used, as they are especially sensitive to temperature effects. The battery is too
important and expensive a component to not include such a feature.
12-56
Exercises
*
#
8! 4
$ #
#
# )
9.0) #1+2#
((((((((((
:
0
#*
#
#+ #
#
#
#*
-
12-57
12-58
The size of a regulator is determined by multiplying the peak rated current from an
array times this enhancement safety factor. The total current from an array is
given by the number of modules or strings in parallel, multiplied by the module
current. To be conservative, the short-circuit current (Isc) is generally used instead
of the maximum power current (Imp). In this way, shunt type regulators that operate
the array at short-circuit current conditions are covered safely.
Example:
Regulator size
28 amps
In the example, the array would produce 21 amps of current, but the regulator
should be sized to control up to 28 amps of current. A single regulator rated at 30
amps would be adequate, or two regulators rated at 15 or 20 amps could be used in
parallel. If two controllers were chosen the array would be split into two sub-arrays
of five modules each, and the output from the two regulators would be combined into
one battery. The two regulators could be set to slightly different final charge
voltages, and a two-step charging system would be the result.
Consult with regulator manufacturers to determine if they have already built a safety
factor into their rating value. Oversizing the regulator by 130% may not be
necessary if you are sure that the regulator design can handle any possible high
currents for short periods.
12-59
Exercises
)
### 3
;
&'+ ,#
-"#
#
##
#
'!
'
%!
%
)
'
#
<1%=
###
*
&%!>
# #
#
/
#
?
@
&!
12-60
12-61
12-62
Self-Regulation Using
Low-Voltage Module
Phoenix in July at 3:00 p.m.
Current (amps)
2
M55
1
M65
8
10 12
Voltage (volts)
14
16
18
20
!
The figure shows a comparison of operating points in the afternoon in a hot climate
between a 36-cell (M55) module and a "self-regulating" 30-cell (M65) module. As
the battery voltage rises there is a more dramatic reduction in current from the 30cell module. In the afternoon, in this example, the battery voltage has risen to about
14.4 volts, and the current from the 30-cell module is almost one third that from the
36-cell module. The battery is reaching full charge, and the current should be
reduced. The reduction in charging current that would be accomplished by a charge
regulator for the 36-cell module is performed automatically with the low-voltage 30cell module.
Self-regulating modules are popularly used in the recreational vehicle (RV) market.
A single 30-cell module can be safely matched to a single large 120-150 Ah battery
without needing a regulator. The current from the module in the afternoon is low
enough so that the battery is not overcharging.
12-63
However, there are conditions that must be met, even with RVs, for a selfregulating design to work properly. If more than one module were connected to
only one 120-150 Ah battery, or if the battery were too small (less than 120 Ah), then
overcharging could occur each day, and a small charge regulator would be needed
in the system. Other factors must also be considered before using a self regulating
module. These are discussed below.
12-64
12-65
Example:
480 Ah total
17.6 amps total
Because the 27-hour rate is faster than the general guideline of 100hour rate, this system would indeed need a charge regulator.
But if more autonomy were built into the battery bank, the ratio of charging current to
capacity would be reduced.
Example:
1800 Ah total
17.6 amps total
The ratio of module charge rate to battery capacity depends on the local weather
data and the desired reserve or autonomy capacity built into the battery bank, so it
varies for each location and application. The 100-hour rate does not occur until
there is about 15 days of autonomy or more designed into the battery. This amount
of autonomy is unusual, and would occur only for systems that are designed to
operate in climates with extremely bad seasonal weather. This might be the case for
remote telecom repeaters in mountainous regions. In that type of application, the
cost savings of not including a regulator are minor compared to the cost of the
overall system. A charge regulator would probably be included anyway, for added
system control.
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
12-66
Exercises
<#
# :#
=
$
#
0
#
"
*
,
3#
&!!
(((((((((()
; 3
##
#
#
12-67
System voltage
PV array and load currents
Battery type and size
Regulation algorithm and switching element design
Regulation and load disconnect set points
Environmental operating conditions
Mechanical design and packaging
System indicators, alarms, and meters
Overcurrent, disconnects and surge protection devices
Costs, warranty and availability
12-68
Selected References
Stand-Alone Photovoltaic Systems - A Handbook of Recommended Design Practices, Sandia National
Laboratories, SAND87-7023, revised November 1991.
Naval Facilities Engineering Command, Maintenance and Operation of Photovoltaic Power Systems,
NAVFAC MO-405.1, December 1989.
Exide Management and Technology Company, Handbook of Secondary Storage Batteries and Charge
Regulators in Photovoltaic Systems - Final Report, for Sandia National Laboratories,
SAND81-7135, August 1981.
Bechtel National, Inc., Handbook for Battery Energy Storage in Photovoltaic Power Systems, Final Report,
SAND80-7022, February 1980.
S. Harrington and J. Dunlop, "Battery Charge Controller Characteristics in Photovoltaic Systems",
Proceedings of the 7th Annual Battery Conference on Advances and Applications, Long Beach,
California, January 21, 1992.
12-69
(End of Chapter)
12-70
CHAPTER TWELVE
CHARGE REGULATORS AND SYSTEM CONTROLS
12-1
12-2
12-3
12-4
12-5
12-5
12-6
12-6
12-7
Terminology
Nominal System Voltage
Nominal Load and PV Array Current
Charge Regulator Set Points
12-8
12-8
12-8
12-9
12-16
12-16
12-16
12-16
12-17
12-17
12-18
12-18
12-19
12-20
12-20
12-21
12-23
12-24
12-28
12-29
12-33
12-71
12-37
12-37
12-39
12-39
12-42
12-42
12-51
12-51
12-53
12-54
12-58
12-61
12-62
12-65
12-68
12-72
12-44
12-46
12-48
Chapter 12 Answers
Charge Regulators & System Controls
According to the data sheet, an SP75 module has a peak power of 75 Watts and an Isc
of 4.8 amps (when configured at 12 Volts).
Since the array is 750 Watts total, there must be 750 75 = 10 modules.
With a 12-volt system, the array will have all 10 modules in parallel. The maximum
amount of current that the array can produce is 10 times the Isc of one module. We
can therefore calculate the number of regulators required as
Number of Regulators
The array of 1200 Watts is split into two sub-arrays. Therefore, each sub-array is 600
Watts. Using 75-Watt modules, the number of modules in one array is
600 watts 75 watts/ module = 8 modules
At 12 volts, the array is configured as 8 modules in parallel. We use 4.8 amps as the
Isc for the SP75 modules. Since the sub-array is supposed to have a single regulator it
must handle the full amount of current for the 8 modules (including the safety factor).
Size of Regulator
49.92 amps
This is greater than 30 amps, so we can not use a 30-amp regulator. However, we can
use a 50-amp regulator since the maximum expected current is slightly less than this
value. The correct answer is
d. 50 amps.
Refer to the manufacturer's literature.
12-1
The shaded area shows indicating greater total charge being put into the battery over
the period of the day.
Irradiance (W/m2)
1000
10
Constant
voltage
7.5
800
5
600
2.5
400
0
Modified
series
-2.5
200
-5
0
4
12
16
20
24
According to the Voltage Regulation Setpoint Table, the proper ending voltage for a
sealed VRLA battery is 2.33 - 2.37 volts when using a PWM controller. We will use the
average of this range, 2.35 volts, as our starting point. For a nominal 12-volt system,
the ending voltage will be equal to 6 times this (since there are 6 two-volt cells).
6 X 2.35 volts = 14.1 volts.
So, under normal circumstances (25 C), we would stop charging when the battery
reached 14.1 volts. However, this needs to be reduced because of the high
temperatures.
Voltage Change
Final voltage
-0.45 volts
12-2
Refer to the manufacturer's literature.
At 12 volts, all four modules will be in parallel. The maximum array current will be four
times the Isc of the module.
Total array current =
Regulator size
19.2 amps
24.96 amps
The necessary size is greater than the smallest regulator size available (20 amps). We
need to use a regulator that is greater than 24.96 amps, so the correct answer is
b. 25-amp regulator
The shunt regulator is rated for 25 amps. However, this needs to include the 1.30
safety factor. The amount of array current that is allowed is:
Array Current
25 amps
1.30
19.23 amps
Using an Isc of 2.4 for the SP36 module, the maximum number of modules is
Maximum Modules
19.23
2.4
8.01
We round down and can use a maximum of 8 modules. The correct answer is:
b. 8 modules
12-3
According to the module literature, a 6-Watt module has a Imp of 0.39 amps. The peak
charge rate is determined as
Peak charge rate
Battery capacity
Array peak amps
Battery capacity
39 Ah
So,
Remember, if the battery size is greater than 39 Ah, the charge rate will be slower
than a C/100 peak charge rate. The decision to use a charge regulator would depend
on the available types of regulators, the pattern of load use and the environment. Since
the application is located in the desert (hot conditions) this might not be a good situation
to use a self-regulating module.
The required battery capacity is given as 250 Ah battery capacity. The peak charge
rate is:
Peak charge rate
250 Ah
2.5 amps
This is just at the edge of the recommended peak charge rate for self-regulated system.
A regulator is probably not necessary from a charging standpoint. However, the
decision whether or not to use a regulator requires good judgement of the advantages
and disadvantages, as well as attention to overall system needs.
12-4
Chapter Thirteen
Energy Enhancement
Through Tracking
Technology
There are two methods for enhancing the output of a photovoltaic power system that
are commonly employed by system designers. Both of these approaches improve
the match between the potential output of solar modules and the load. Both
methods involve the word tracking, although one is electronic and the other is
mechanical. In this chapter we will discuss these methods for getting more out of
you solar generators.
The first method involves physically tracking the sun as it moves throughout the day
and the seasons. The second method involves electronically monitoring the
maximum power point of an array and transferring this maximum power to the loads.
13-1
Components Tracking
13-2
Components Tracking
Insolation (Langleys)
600
500
400
300
200
100
0 deg.
15 deg.
30 deg.
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
45 deg.
60 deg.
Insolation (Langleys)
600
500
400
300
200
100
0 deg.
10 deg.
20 deg.
30 deg.
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
40 deg.
13-3
Components Tracking
Exercise
Trivandrum
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
-15 deg.
0 deg.
15 deg.
30 deg.
45 deg.
404
452
495
474
454
382
388
428
433
389
359
366
461
494
515
471
436
362
371
419
441
413
399
422
519
533
528
461
414
342
352
406
444
436
442
481
546
540
509
424
367
302
313
370
421
433
458
511
540
516
464
366
305
252
262
317
378
408
447
509
Leh
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Latitude: 8 deg. N
Latitude: 34 deg. N
0 deg.
15 deg.
30 deg.
45 deg.
60 deg.
75 deg.
207
262
332
438
476
503
494
457
422
340
262
209
268
317
375
464
484
502
497
476
467
404
336
278
316
355
394
461
462
471
470
465
481
445
393
335
343
369
388
434
419
420
423
431
467
456
423
369
349
363
364
388
361
356
361
380
430
442
427
381
334
336
323
327
293
284
290
314
373
404
406
369
13-4
Components Tracking
400
300
200
100
Jan
Feb
Mar
Apr
May
Jun
July
Aug
Sep
Oct
Nov
Dec
Apr
May
Jun
July
Aug
Sep
Oct
Nov
Dec
400
300
200
100
Jan
Feb
Mar
13-5
Components Tracking
13-6
Components Tracking
Amps
1.5
1
0.5
0
5
11
13
15
17
19
Ho ur
Fix ed
Single A x is
Double A x is
Amps
1.5
1
0.5
0
5
11
13
15
17
19
Ho u r
Fix ed
Single A x is
Double A x is
Ah/day
One S L-35 in Le h
18
16
14
12
10
8
6
4
2
0
Jan
Mar
Fix ed
May
Jul
Sep
Single A x is
Nov
Double A x is
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
13-7
Components Tracking
The enhancement of sun position tracking will be different depending on the latitude.
The performance curves of a module in Minicoy are shown below. The latitude of
the site is only 8 deg. north of the equator. The enhancement from tracking is more
uniform throughout the year, compared to the heavier gain only in the summer for
Leh at 34-deg. latitude. Winter and summer daily enhancements are about the
same, because the suns path is always a large arc in the sky.
Amps
1.5
1
0.5
0
5
11
13
15
17
19
Hour
Fixed
Single Axis
Double Axis
Ah/day
Mar
May
Jul
Single Axis
Sep
Nov
Double Axis
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
13-8
Components Tracking
Exercise
%
$
%
%
&%
'
%
%
13-9
Components Tracking
At very high latitudes, the tilt angle of the structure is great, and the difference in
weight between the two sides becomes less. Therefore, it may be more difficult for
solar powered trackers to adequately perform at high latitudes.
Shadow Masks
Freon Tubes
!
13-10
Components Tracking
'()%&
"#$%
&
13-11
Components Tracking
*+),-%./
Trackers can be designed to hold from 4-24 modules or even more. Wind loading of
large tracker structures should be a concern. But most designs can handle winds up
to 40 mph or greater.
Because the tracking is driven by electronics, it can operate equally well in all
latitudes. There is no greater difficulty tracking at high latitudes as might occur with
solar powered trackers.
Exercise
*
13-12
Components Tracking
Pmax (array)
Imp (array) X Vmp (array)
Power (load)
I (load) X V (load)
An IV curve is shown with the point of maximum power indicated. If this power is
translated to a lower voltage, the current must rise to keep the value of power
constant. Thus a MPT effectively changes the shape of an IV curve from having a
flat region where current is almost constant into a descending geometric curve of
constant power. Thus the current available at voltages less than the Vmp of the
module actually increases. Dont think that energy or power has somehow been
magically created. The total power available stays the same. It is just the current
and voltage values that have been changed. Their product, the power, stays the
same.
This substantially increases efficiency for direct coupled DC motors and water
pumps and can even improve battery charging efficiency under certain
circumstances. The effect for a DC motor is shown below. During the middle of the
day, the module and motor should be matched in voltage by design. But in the early
13-13
Components Tracking
morning and late afternoon, or during overcast conditions, the module would
normally supply very little current directly to the DC motor. The motor would operate
the module at point A, well away from its maximum possible power output at point C.
By using the MPT, the motor now interacts with the power curve at point B,
substantially increasing the current and the voltage. The water pump begins to pump
earlier in the day, and continues to pump later in the afternoon than would be
possible without the MPT.
M a x im u m P o w e r T ra c k in g
Current (amps)
4
curve of constant power
3.5
3
2.5
2
1.5
1
0.5
0
0
12
16
Voltage (volts)
20
24
W ith o u t M P T
m o to r cu rve
3 .5
A
3
2 .5
m o to r
m o d u le
1 .5
1
0 .5
0
W ith M P T
C
A
M 5 5 m o d u le cu rve
P h o e n ix, A rizo n a , 8 :0 0 a .m .
4
12
16
V o lta g e (vo lts)
C
m o d u le
20
B
m o to r
24
13-14
Components Tracking
13-15
Components Tracking
Watts
25
20
15
10
5
0
5
9 10 11 12 13 14 15 16 17 18 19
Ho u r
Battery
Watts
25
20
15
10
5
0
5
9 10 11 12 13 14 15 16 17 18 19
Ho u r
Battery
13-16
Components Tracking
O ne S L-35 in Ra ipur
200
Wh/day
150
100
50
0
Jan
Mar
May
Jul
Sep
Nov
Battery
13-17
Components Tracking
Another example is shown, this time in Leh. The temperatures here will be lower
than in the case of Raipur, and the difference in benefit of MPT with battery charging
will show more clearly. The enhancement is positive throughout the year, higher in
the cool month of January and less in the hot month of August. But here again, the
relative heat in all months makes the gain small. The annual total charge put into a
battery directly is 50.3 kwh/year while the input through a MPT is 53.6 kwh/year.
Watts
20
15
10
5
0
5
11
13
15
17
Hour
MPT @95% ef ficiency
19
Battery
Watts
20
15
10
5
0
5
11
13
15
17
19
Hour
MPT @ 95% efficiency
Battery
!
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
13-18
Components Tracking
The annual performance profile shows that there is a greater contribution from MPT
in a cold climate throughout the year, again more so in the winter than in the warmer
summer.
Wh/day
150
100
50
0
Jan
Mar
May
Jul
Sep
Nov
Battery
'
The negative trade-off that must be considered is the cost of the device, and the
increased risk of failure. Also the MPT must overcome its own energy inefficiency
(usually about 5%) to prove economical.
13-19
Components Tracking
Exercise
(
)
!
!
13-20
Components Tracking
(End of Chapter)
13-21
Components Tracking
CHAPTER THIRTEEN
ENERGY ENHANCEMENT THROUGH TRACKING
TECHNOLOGY
13-1
13-2
13-6
13-10
13-11
13-13
13-13
13-15
13-20
13-22
Components Tracking
Chapter 13 Answers
Energy Enhancement through Tracking Technology
Location: Trivadrum
Trivandrum
600
500
-15
0
15
30
45
400
300
200
100
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Annual
13-1
Location: Leh
Leh
600
500
400
0
15
30
45
60
75
300
200
100
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Annual
Location: Leh, Latitude 34 N
Month
Largest June
Smallest Jan.
Fixed Axis
Ah/day
9
8
Double Axis
Ah/day
16
10
Percent
Increase
77%
25%
Fixed Axis
Ah/day
8
13
Double Axis
Ah/day
13
17
Percent
Increase
62%
31%
13-2
13-3
Chapter Fourteen
Diodes in Photovoltaic
Systems
This chapter discusses the function of a very specific piece of equipment used in
photovoltaic systems, the solid state diode. This device performs two different
functions, depending on where it is installed, and those functions are often confused
and misunderstood. The purpose of this chapter is to clarify the differences in
performance and to make clear why diodes are used in photovoltaic systems.
A diode is a solid state device made of P-type and N-type silicon. In fact a silicon
solar cell is a diode. Solar cells use the internal electrostatic field of their P/N
junction to prevent electrons from flowing back into the cell after they have been
knocked loose by light. Diodes use their internal field to allow electric current to flow
one way in a circuit and prevent it from flowing back.
When diodes are installed in series with a string of modules they perform a blocking
function, preventing backflow down the module string. When diodes are installed in
parallel with modules they perform a bypass function allowing current to pass around
a shaded area of a module. These two functions will be explained next.
14-1
series combination
14-2
Current 3
25%
25%
2
50%
50%
75%
-10
75%
100%
10
100%
20
-10
Voltage
10
20
Voltage
14-3
14-4
A
3 cells 50% shaded
-10
-5
10
15
20
(-)
Voltage
(+)
Current
Current
Voltage
(-)
(+)
14-5
However when the shading in a module is severe enough so that the new Isc of the
shaded module falls below the operating current of the other series modules in the
group, then the shaded module as a whole is forced to operate at a negative voltage
to pass that level of current (point B in the figure). The entire module subtracts
voltage from the group of modules. The voltage out of the module is actually lower
going into the module. It would be better if the entire module were taken out of the
circuit temporarily. It is also important to limit the amount of local heating that result
from this negative voltage to prevent damage to the plastic and stress to the cells.
This can be done by installing diodes in parallel with groups of cells in each module.
These are called bypass diodes, because they allow current to pass around
shaded cells and thereby reduce the voltage losses through the module.
A diode passes current when the voltage is higher on one side, but blocks any
current flow when the voltage is higher on the other side. In condition A, the diode
sees the same voltage polarity as the module, with the voltage going out of the
module higher than going in. Current that tries to flow from the high voltage side of
the module to the lower voltage side through the diode is blocked. The diode is
"reverse biased" and does not allow any current to backflow around the module. The
current produced by the shaded module passes on to the next module. It is as if the
diode were not in the circuit at all.
However under condition B, where enough cells become shaded so that the module
is forced to operate at a negative voltage, the final voltage going out of the module is
actually lower than the voltage going in. The voltage polarity shown is positive at
the beginning of the module and negative at the end, the opposite of the normal
case. The voltage across the diode becomes reversed ("forward biased") and it
conducts current through itself. All the current greater than the shaded cell's new Isc
is "bypassed" through the diode, thus reducing drastically the amount of local
heating at the shaded area. The diode also holds the entire shaded module or
group of cells to a small negative voltage of approximately -0.7 volts, thus limiting the
reduction in array output.
14-6
Shaded
Module
With
Bypass
Diode
-10
10
20
Current at
Current above
passes through
passes through diode
module
No current
passes
backwards
through bypass
diode
(-)
All current
passes through
module and
(+)
on to loads.
(+)
(-)
14-7
14-8
Another typical circuit design has four strings of cells, with the circuit beginning and
ending at the same end of the module. Only one junction box is used, and both
diodes are placed inside that box. Looking at the cell circuit stretched out, each
diode is connected around 18 cells and there is no overlap. In the case of severe
shading, current can pass through an unshaded group of 18 cells and then into a
bypass diode. Thus only 18 cells or 1/2 of the module is bypassed. This design
allows for even less voltage loss than the three-string configuration.
14-9
-20
20
40
60
80
100
14-10
For this example, assume that three cells in one module become 75% shaded. We
first show the result without bypass diodes, and then on the next page show the
effect with bypass diodes in place.
The entire array output potential is dragged down. The battery as always
determines the operating point for the array, and now results in an array current of
about 1.3 amps. The new Isc for the shaded module is only 1.1 amps, so the
shaded module is forced to operate at a negative voltage of -8 volts. The negative
voltage times the current gives the net power dissipated by the shaded module:
-8 volts X 1.3 amps = -10.4 watts
So about 10 watts of power is lost in shaded cells, in the form of heat. Remember
that an M55 module produces only 53 watts peak at standard conditions, and about
45 watts in field conditions when it is heated. So this potential power loss is quite
high.
Shaded
module
is forced
to operate
at a
negative
voltage
-20
20
40
60
80
100
Voltage (volts)
Battery determines voltage
for entire array.
14-11
(For simplicity, we show the benefit of one single bypass diode around the entire
module). With a bypass diode installed in parallel with the shaded module, the
result is quite different. There are two regions to examine one below the Isc of the
shaded module and the other above it.
Looking at current levels below the new Isc of the shaded module, everything is left
as is. The bypass diode can do nothing here.
Looking above the new Isc of the shaded module however, the shaded module
cannot operate beyond approximately -0.7 volts, as this is the maximum negative
voltage allowed by the diode. So the output potential of the array is not dragged
down as severely (-0.7 volts instead of -8 volts!). For current levels above the new
Isc of the shaded module, the shaded module is completely bypassed, so the
composite curve for the entire array is that of only four unshaded modules instead of
five.
The battery voltage now operates the entire array at a current of almost 2.0 amps,
nearly back to the full-unshaded potential (instead of only 1.3 amps as before with
no diode).
The shaded module operates at a net power loss of only:
-0.7 volt X 2.0 amps = -1.4 watts
Thus the total dissipated power in shaded cells and bypass diode is only about
-1.4 watts instead of -10 watts as before. This can be broken down into the power
dropped in the shaded cells and the power loss through the diode. The module
passes slightly more than its Isc of 1.1 amps at -0.7 volts so it is dissipating only a
small amount of power itself:
-0.7 volts X 1.1 amps = -0.77 watts
And the current going through the diode is just the difference between the shaded
module current the array current, given as 2.0 amps - 1.1 amps = 0.9 amps. At a
voltage of -0.7 volts across the diode, this means that the power lost in the diode is
given by:
-0.7 volts X 0.9 amps = -0.63 watts
In summary, we have reduced the power lost from 10.4 watts to only 1.4 watts, with
0.77 watts dissipated in the cells and 0.63 watts dissipated across the diode. The
two main benefits of bypass diodes are quite evident: (1) array output is preserved;
and (2) local heating is minimized.
14-12
3
Voltage
loss
2
through
shaded
module is
1
limited to
-0.7 volts
-20
20
40
60
80
100
14-13
14-14
If modules in a parallel group are shaded, then the group cannot produce the current
of the other groups in series with it. The unshaded modules in the group can
compensate by trying to produce more current. They do this by operating at lower
voltage, which moves their operating point up in current along their IV curve.
If the shading is bad enough, and the current compensation needs to be very large,
then the entire group will continue to compensate by operating further and further
down in voltage. The group can actually go past zero volts and operate in the region
of negative voltage to try to get the current level up to that of the other unshaded
groups in series with it. If this happens, all the bypass diodes in the separate
modules will begin to pass current. Now the current is not just that of one module,
but could be the current of the entire array. The small diodes for each module would
be overloaded and could fail.
The solution is to install an external diode large enough to handle the current of the
entire array. If a shaded group compensates by going into negative voltage, then
the current from the other groups bypasses the entire group through the large diode,
and no damage occurs. This can involve extra wiring and cost.
Current Compensation of
Parallel Wired Modules
current level of other modules
-Voltage
+Voltage
14-15
14-16
Blocking reverse flow down damaged modules from parallel modules during day.
Blocking diodes placed at the head of separate series wired strings in high
voltage systems can perform yet another function during daylight conditions. If
one string becomes severely shaded, or if there is a short circuit in one of the
modules, the blocking diode prevents the other strings from loosing current
backwards down the shaded or damaged string. The shaded or damaged string
is "isolated" from the others, and more current is sent on the load. In this
configuration, the blocking diodes are sometimes called "isolation diodes".
14-17
(+) Current
(+)
(+)
(-)
( -)
Leakage flow of
current at night
(-) Current
(+)
(+)
( -)
(-)
14-18
This can be prevented by installing a diode in series between the array and battery.
The diode allows current to flow into the battery, but "blocks" reverse flow at night.
So in this configuration, the diode is called a "blocking" diode. There is a small
penalty to pay in the form of a 0.6-1.0 volt drop across the diode.
12 Volt System
Blocking Diode
+
PV
Module
-
Battery
Load
14-19
Parallel-Wired Strings
This is not the preferred way to connect modules in a higher voltage array though. If
one or more modules in a group are damaged, or if maintenance work must be done
on a group, the entire array will not be able to produce power at the required voltage.
It is recommended to connect modules in series strings first, and then connect those
groups in parallel. This is described next.
14-20
Series-Wired Strings
14-21
Typically in this configuration all the blocking or isolation diodes are grouped
together in a field combiner box located at each structure. All the positive wires from
the series strings are fed separately into the box, and the diodes are mounted on
some type of wiring block. The separate outputs from the diodes are combined into
a large diameter wire, and all the current from the strings flows on to the controls
and battery. The negative wires are combined into a large diameter wire also.
Single J-box
style modules
Other functions that can be performed at the field combiner box include a disconnect
switch or circuit breaker and lightning protection devices. These could be
conveniently built in to the box and would allow further protection of the system from
short circuits and lightning strikes.
14-22
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
14-23
Double Isc
Double Voc
Most power modules produce current in the range of 3-4 amperes, so this means
that the current rating of bypass diodes should be at least 6-8 amps, a relatively
small value. Since the manufacturer does not know what the voltage of the final
array will be, the voltage rating of bypass diodes should be the highest reasonable
commercial value, usually 600 volts.
In the case of a single blocking diode for a 12-volt parallel string, the current rating
must be the Isc for the entire array. For example, if 8 35-watt modules were
connected in parallel for a 12-volt system, the single blocking diode needed should
have a current rating of at least
2 X 8 parallel modules X 2.15 amps (Isc) = 34.4 amps.
The voltage rating of the blocking or isolation diodes should be twice the full array
open circuit voltage (Voc). Do not use just the nominal system voltage. For the
simple 12-volt array with 8 modules in parallel, the number of series modules is only
one, so the voltage rating should be
2 X 1 series module X 20 volts (Voc) = 40 volts.
For a 48-volt nominal array, with 4 modules in series, the voltage should be at least
2 X 4 series modules X 20 volts (Voc) = 160 volts.
14-24
Exercise
14-25
(End of Chapter)
14-26
CHAPTER FOURTEEN
DIODES IN PHOTOVOLTAIC SYSTEMS
14-1
14-2
14-2
14-3
14-4
14-8
14-10
14-14
14-17
14-18
14-21
14-24
14-27
Chapter 14 Answers
Diodes in Photovoltaic Systems
First Array:
Blocking
Bypass
Bypass
Bypass
Bypass
A diode should be rated for twice the expected Isc and twice the full array Voc. The
bypass diodes are connected to 4 modules in parallel, so their rating should be:
Current Rating 8 X Isc
Voltage Rating 8 X Voc
The blocking diode must handle all 4 parallel strings, so its rating should be:
Current Rating 8 X Isc
Voltage Rating 8 X Voc
14-1
Diodes
Second Array:
Blocking
Blocking
Bypass
Bypass
Bypass
Bypass
The bypass diodes are now connected to a single module, so the rating should be:
Current Rating 2 X Isc
Voltage Rating 8 X Voc
Each blocking diode is connected to a single parallel string, so each will be rated at:
Current Rating 2 X Isc
Voltage Rating 8 X Voc
14-2
Diodes
Chapter Fifteen
System Sizing
The purpose of photovoltaic system sizing is to calculate the number of solar
modules and batteries needed to reliably operate the load throughout a typical year.
This involves balancing the often-opposing goals of maximum reliability and
minimum cost.
This section presents simple methods for calculating array and battery size for
standalone systems. While reading this chapter keep in mind that not all choices in
system sizing are based on calculations. There are decisions that require judgment
on the part of the designer and the user. The mechanics of the calculations are
quite simple. It is the judgment of the designer about the efficiency and
appropriateness of the loads that makes a system well designed and cost effective.
15-1
Basic Principles
The solar array is sized to replace the load on a daily basis based on average
weather conditions. The average is made up of below average days and above
average days so the array and battery must work together.
The array is NOT sized based on how quickly it can recharge the array after a
few days of below average weather. This would result in a large array most of
which is not needed or used during most of the year.
The battery has the job of supplying energy to the loads when the insolation is
below average. The array will replenish the battery during subsequent days of
above average insolation.
If there are concerns about quickly recovering from a storm or prolonged period of
below average weather then the designer should be considering a hybrid system
design. The backup generator (usually a diesel, gas or propane generator) would be
used to bring the batteries back to near full charge every few days during the winter
or periods of prolonged bad weather. The generator might not even be needed
during the summer months.
The sizing calculations presented in this chapter are based on designing a
standalone photovoltaic, non-hybrid and system. Hybrids are discussed in the
chapter on Hybrid Systems
15-2
15-3
Battery Sizing
Days of Autonomy or Reserve
The battery bank is sized to operate the loads during a long sequence of below
average insolation days. We can think of the battery as being full of charge.
During a below average day the array cannot supply all the amp-hours of charge
needed to replace what the load draws from the battery. So the battery ends up
being discharged at the end of the day. If the next day is again below average then
the battery again discharges some to operate the loads. This process can go on for
only so long before the battery is discharged to a point that is may become
damaged. The system designer must build into the battery capacity enough
equivalent days of charge to operate the loads autonomously, meaning without any
input of energy from the solar array. We refer to these equivalent days of reserve as
days of autonomy.
The usual rough values that are used in most system sizing calculations are about 35 days for non-critical applications and 7-14 days or even more for more critical
applications. Non-critical situations usually involve occupied systems, where the
users can adjust their load demand a little to accommodate the bad weather.
Critical situations might involve commercial or governmental systems for
communications or navigation, or important health needs such as hospitals and
clinics. And very remote systems must have a large capacity in their battery bank to
allow for the time it would take a maintenance crew to arrive at the site.
3-5 days
Typical 4 days
Critical Applications
7-14 days
Typical 10 days
15-4
Up to 80%
Up to 50%
The basic version of the formula is given below for calculating the battery capacity
that must be installed.
Battery Capacity =
Each battery has a nominal voltage, depending on how many single cells are
connected by the manufacturer. Some large cells are only two volts; some smaller
units are six or twelve volts. The formula for the number of batteries to connect in
series to give the voltage for the loads is simply the nominal system voltage divided
by the nominal battery voltage.
15-5
To illustrate these basic equations lets use the remote school system that was sized
on Page 11 in the Load Estimation chapter as an example.
Example:
The AC load demand for the school summed to 9104 Wh/day. If we use the
Trace 4024 sinewave inverter, and assume that the efficiency is about 90%
and the input voltage is 24 volts, then the DC load will be 421 Ah/day (9104
Wh 0.9 24 V = 421 Ah).
We shall consider this a non-critical system because the users can be a bit
flexible in their usage depending on the weather, so we will design for 5 days
of autonomy. We will assume that deep cycling industrial batteries are used.
The maximum % usable will therefore be 80%.
Battery Capacity
2631 Ah of capacity
2105
0.8
If they were to use Trojan L16 deep cycling flooded batteries, rated at 350 Ah
and 6 volts then they would need about eight in parallel and four in series:
(2631 Ah 350 Ah = 7.5, round up to 8) and (24 volts 6 volts = 4).
As another example, look again at the small remote cabin system sized on Page 10
of the Load Estimation chapter. Assume that it will be used as only a weekend
cabin and that cost is a problem so low cost batteries will be used. Recall that it is a
pure DC system with no AC loads.
Example:
2 days X 70 Ah/day =
0.5
280 Ah of capacity
140
0 .5
If they were to use Delco 2000 shallow cycling batteries, rated at 105 Ah at
12 volts each, they would need three batteries in parallel (280 105 = 2.7)
15-6
Exercise
15-7
load time
loads
C/500
C/50
C/0.5
-30
-20
-10
10
C/120
C/5
20
30
40
50
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
15-8
Example:
Reading between the 120 hour and 500 hour rate curves, the loss of
capacity factor reading up from -10 is about 0.91 or 91%. This means
that at this temperature, there would be only 91% of the capacity
available at 25o C. To have the capacity you really need for your
autonomy requirements, you would need to install 9% more.
For this example, the daily load of the data acquisition system is
2 amps X 24 hours/day = 48 Ah/day. Using the basic formula
presented earlier (not accounting for cold temperature) the capacity
would be given by:
Simple Capacity Calc
10 days X 48 Ah
0.8
600 Ah
But this is the capacity that we want, even during the coldest period
when it gets down to -10o C. To account for the loss of capacity at 10oC we need to divide by the correction factor to calculate the amount
to install when rated at 25o C:
Temperature Adjusted
Capacity Calc
660 Ah at 25o C
15-9
15-10
The average load operating time is either 24 hours for a continuous load, or the
actual load operating time if a single load or the weighted average time if several
intermittent loads are present.
Continuous Loads:
Single Load Systems:
Multiple Load Systems:
use 24 hours
use load operating time
use weighted average load operating time
load time
loads
Some examples of industrial battery manufacturers literature are shown on the next
page.
15-11
AH Capacity @ 25 deg. C.
1.75
vpc
Overall Dimensions
to 1.9 vpc
Length
Width
Weight
Height
8 h
72 h
120h
240h
480h
in.
mm
in.
mm
in.
mm
lbs.
kgs.
3DCPSA- 3
31
44
46
48
49
5.28
134
7.38
187
10.31
262
27.7
12.6
3DCPSA- 5
62
72
73
74
75
5.28
134
7.38
187
10.31
262
33.6
15.2
3DCPSA- 7
94
133
139
145
150
9.47
241
7.38
187
10.31
262
54.3
24.6
DCPSA-11
156
222
230
240
249
6.38
162
7.38
187
10.75
273
37.0
16.8
DCPSA-13
188
267
277
290
300
6.38
162
7.38
187
10.75
273
39.0
17.7
DCPSA-15
219
295
300
306
310
6.38
162
7.38
187
10.75
273
41.0
18.6
3KCPSA- 5
225
315
324
333
340
8.53
217
10.44
265
18.25
464
131.0
59.4
KCPSA- 7
337
384
389
394
400
3.62
92
10.44
265
18.25
464
58.0
26.3
KCPSA- 9
450
506
513
520
524
4.62
117
10.44
265
18.25
464
79.0
35.8
KCPSA-11
562
618
628
638
644
5.59
142
10.44
265
18.25
464
96.0
43.5
KCPSA-13
675
743
755
766
773
6.59
167
10.44
265
8.25
464
113.0
51.3
KCPSA-15
787
1011
1031
1047
1060
8.53
217
10.44
265
18.25
464
139.0
63.0
Vb2306
Vb2308
Vb2310
Vb2312
Vb2407
Vb2408
Vb2409
Vb2410
Vb2412
Vb2414
Vb2416
Vb2418
Vb2420
10 hour
24 hour
48 hour
72 hour
120 hour
240 hour
270
360
450
540
630
720
810
900
1080
1260
1440
1620
1800
300
400
500
600
700
800
900
1000
1200
1400
1600
1800
2000
360
480
600
720
840
960
1080
1200
1440
1680
1920
2160
2400
408
544
680
816
952
1088
1224
1360
1632
1904
2176
2448
2720
432
576
720
864
1008
1152
1296
1440
1728
2016
2304
2592
2880
450
600
750
900
1050
1200
1350
1500
1800
2100
2400
2700
3000
468
624
780
936
1092
1248
1404
1560
1872
2184
2496
2808
3120
15-12
Example:
A remote telecom site will use deep cycling industrial batteries (maximum
DOD of 80%). The system will have 10 days of autonomy designed into the
battery. Two loads are present:
Load 1:
Load 2:
Microwave Repeater
Radio
Weighted Average
Load Operating Time
10 amps
7 amps
24 hours/day
12 hours/day
19 hours
10 days X 19 hours/day
0.8 max. discharge
The nearest rate used by either manufacturer shown on the previous page is
the 240 hour rate, so use values from this column for this system. If your
calculated average discharge rate is between rates used by a manufacturer,
use the next faster rate (shorter time) to be conservative.
--------------------------Compare the capacity available at the 240-hour rate to the nominal rates of
the manufacturers:
C&D Literature: use as an example the KCPSA-15 battery. Capacity at the
slow 240 hour photovoltaic rate is 1047 Ah, and capacity at their standard 8
hour rate is 787 Ah. So a typical C&D battery will deliver about
1047 Ah @ 240 hr / 787 Ah @ 8 hr = 0.33 or 33% more capacity at the slow
photovoltaic rate compared to their standard rate.
Varta literature: use as an example the Vb2410 battery. Capacity at the slow
240 hour photovoltaic rate is about 1560 Ah, and capacity at their standard
10 hour rate is 1000 Ah. So a typical Varta battery will deliver 1560 Ah @
240 hr / 1000 Ah @ 10 hr = 0.56 or 56% more capacity at the slow
photovoltaic rate compared to their standard rate.
Rule of Thumb: You will have about 30% more capacity available from a battery at
typically slow photovoltaic discharge rates (usually C/100 to C/300) compared to
manufacturers standard or nominal rates (usually C/8 or C/10). This is only a rough
rule of thumb to use when no other capacity vs. rate information is available.
15-13
-40
-20
15-14
Battery Capacity
(@ specified rate)
(A) The Maximum Percent Usable is either the standard value of 50% for shallow
cycling batteries, 80% for deep cycling batteries, or a reduced value based on
freezing concerns. The designer can also reduce this value simply as a way to build
in more life to the system. For example, a designer might use a shallow cycling
battery, but size based on using only 30% of the capacity and not the usual 50%.
This might not impact the cost of the system too much, and would mean the battery
life would be extended.
(B) The Temperature Derating Factor is included to make sure that more capacity is
installed at 25o C so that when the battery gets cold and loses some capacity, there
will still be the required capacity present.
(C) The specified rate insures that you are taking into account the change of
capacity with slow discharge rate. Use manufacturers literature to choose a battery
that will give you the capacity you need at the average discharge rate of your
system.
An Array/Battery Sizing Form is presented next. Use this form for your battery
calculations. All the factors are indicated. An example is worked for your reference.
15-15
Example:
Calculate the capacity needed for the remote school example again, this time
with more information. The local meteorologist reports that it gets to an
o
average of -20 C in the winter at the site. This will mean that we
compensate for both loss of capacity with cold temperature, and limit the
maximum depth of discharge to prevent freezing.
(A) Using the freezing phenomenon chart in Figure 15-2, we determine that
the maximum allowable depth of discharge is about 50% or 0.5 , even though
we are using deep cycling type batteries that usually can be discharged to
80%.
(B) Calculate average rate of discharge. The weighted average load
operating time for the loads on p. 9-11 is given by:
Weighted Average
Load Operating Time =
9104 Wh/day
(8X40)+(2x11)+(2x200)+ 300+800+200 watts
4.5 hours
= 45 hr rate
(C) Use Figure 15-1 to determine the Temperature Derating Factor. Use the
50-hour curve (closest to our 45-hour rate), and determine that the
o
Temperature Derating Factor at -20 C is 70% or 0.70.
(D) Calculate the capacity. We will use Varta batteries for this example.
Battery Capacity
6014 Ah
@ 45 hour rate
Looking at the Varta capacity Table 15-2, use the 48 hour column (closest to
our 45 hour rate) to select a particular model to use. We could choose 4 of
the Vb2412 units, as each gives about 1/4 of the capacity we need at that
rate.
Since they are 2-volt cells, we would need strings of 12 cells connected in
series to give the 24 volts for the inverter.
15-16
System Description:
Max. Daily Load
421 Ah
System Voltage
24 V
Average Discharge
Rate
-20 C
Battery Capacity
(@ 45 hour rate)
Chosen Battery
Capacity =
5
0.5
] X [ 421 ]
] X [ 0.70 ]
[
[
Varta Vb 2412
Voltage
Battery Capacity
6014
0.5
45
(hours)
Temperature Derating =
Parallel Batteries
0.85
6014 Ah
2-volt cell
= 3.7
round to
4
[ 1632 ]
System Voltage
= [ 24 ]
= 12
Chosen Battery V.
[ 2 ]
----------------------------------------------------------------------------------------------------------------Daily Load = ________ Module : _______ Module Output = ______
Series Batteries
Parallel Modules
Daily Load
.
Module X Derating X Charge
Output
Factor
Efficiency
[
[
Series Modules
= System Voltage
12
]
]X[
]X[
=
[________]
[
]
15-17
__________
__________
Exercise
'
()*
+
!
"
,
"-
!
"%#
)
,
$&
"%
.
%
"#
15-18
If you find that your choice of battery means you have to connect 20 batteries in
parallel, for example, you should look at choosing a larger capacity unit so you need
fewer in parallel!
Many system designers prefer to design their system with two parallel strings. In this
way, if there is a problem with a battery cell, part of the battery bank can be removed
and worked on, while the system still has nominal voltage and can operate.
Different designers will have different preferences.
Exercise
15-19
Array Sizing
In the previous discussion we showed how battery sizing was based on operating
the loads during a period of below average insolation days, or days of autonomy.
Now we look at how to size the array based on replacing the daily load on the
average every day.
15-20
15-21
15-22
15-23
15-24
Choosing A Module
36 cells:
hot climates
direct couple to DC motors
utility interconnection with MPT
33 cells:
moderate climates
general battery charging
15-25
1kw/m2
6.6 hours
peak
irradiance
peak
hours
peak hours
6.6 hours
= 13.2 Ah
15-26
For example, the average daily irradiance profile is shown for a surface tilted to 20o
for Madras. The total insolation falling on that surface in March is 6.6 kw/m2. This
can be rewritten as 6.6 hours X 1 kw/m2. In the figure we show the equivalent
irradiance profile with the sun suddenly turning on to the peak irradiance level of
1000 w/m2 and staying at that level for 6.6 hours. Module output would be given by
multiplying this value for equivalent peak hours times the module Imp. For a typical
35 watt module with Imp of 2.0 amps, this would be 6.6 X 2 amps = 13.2 Ah/day.
For a typical 75-watt module with Imp of 4.4 amps, this would be 6.6 X 4.4 amps =
29 Ah/day.
If we look at the average daily irradiance profile for Madras in July, we see that the
total insolation that fell on a surface tilted to 20o would be only 4.5 kw/m2. This
would equate to 4.5 peak hours of irradiance.
Calculate array size based on the month with the lowest insolation for a load profile
that is constant throughout the year, or for each month and compare to the load
required for that month and choose the largest number of modules needed in any
month.
Peak Hours
o
1200
Irradiance (w/m )
1000
800
6.6 kwh/m2
600
4.5 kwh/m2
400
200
0
4
10
12
14
16
18
20
Hour of Day
March
July
15-27
Unit of Insolation
kwh / m2 / day
1.0
Langley / day
0.01162
MJ / m2 / day
0.2777
0.003155
15-28
There are some simplifications in the peak hour approach that should be
understood. For example, the effect of temperature on module output is
neglected in the peak hour approach. As a means to compensate the module
Imp is used in calculations. Since battery loads usually operate the module at
voltages slightly lower than the Vmp and therefore at currents slightly higher than
the Imp, using Imp for module output is being conservative. The effect of
temperature on module output is greater for modules with fewer cells in series.
So the peak hour method is more accurate for 36 cell modules, and less
accurate for 33 cell modules, especially in hot climates. Predictions in cold
climates will be more accurate for all modules.
In the peak hour method, the total measured solar insolation is translated into
peak hours of operation. Actually, at the beginning and ending of each day,
there will be some time when the irradiance will be too low and the module or
array voltage will not be sufficient to charge a battery. This error is usually quite
small, but is more pronounced for modules with fewer cells in series. Predictions
of Ah/day/module will be more accurate with this method for 36 cell modules than
for 33 cell modules.
The peak hour method assumes that module output is completely linear with
irradiance. It assumes that all modules will convert solar irradiance into electrical
power the same. But this is not the case. For example, high efficiency single
crystal solar cells can convert at low light levels more efficiently than some other
technologies. So this peak hour approach of multiplying hours times rated
current can overestimate the output of certain technologies.
In general however, the peak hour approach is a useful method for quickly
approximating module output given local measured insolation data on a tilted
surface.
15-29
Exercise
.
69
:7
;<&
=
9
5$4
>!#
;?
;
@@@@@@@@@
5$4
>!
A@@@@@@@@@@
A@@@@@@@@@@
<?
;
@@@@@@@@@
5$4
>!
15-30
A@@@@@@@@@@
A@@@@@@@@@@
Leh
Latitude: 34 oN
Best Tilt: 50 oS
Jan
4.0
Feb
4.3
Mar
4.4
Apr
4.9
May
4.7
Jun
4.6
Jul
4.7
Aug
4.8
Sep
5.3
Oct
5.3
Nov
5.0
Dec
4.4
Oct
5.4
Nov
5.2
Dec
5.4
Madras
Latitude: 13 oN
Best Tilt: 15 oS
15-31
Two examples of how to use the data are presented next. The first example is the
remote school, and the second is the remote cabin.
Example:
The lowest insolation during the year at 50 tilt angle is 4.0 kwh/m during
January. This translates into 4.0 peak hours. Calculating the output from a
typical 75 watt module:
Module Output
17.6 Ah/day
2 in series
The work for this example is worked out on the next page, using the lower portion of
the Array/Battery Sizing Form.
15-32
School
month: Jan)
Max. Daily Load
_____
_____
System Voltage
_____
_____
Average Discharge
Rate
_____
Battery Capacity
(@ ___ hour rate)
Chosen Battery
Capacity =
Temperature Derating
[_____ ] X [_____ ]
[
] X [
_____
_____ Ah
____________________________
Parallel Batteries
= _____
(hours)
Voltage
Battery Capacity
=
Chosen Battery Cap.
[_____]
[
_____
_____
System Voltage
= [_____]
= _____
Chosen Battery V.
[
]
----------------------------------------------------------------------------------------------------------------Week-Averaged = 421 Module : 75 watt Module Output = 17.6 Ah/day
Daily Load
Series Batteries
Parallel Modules
Daily Load
.
Module X Derating X Charge
Output
Factor
Efficiency
421 ]
[ 17.6 ] X [ 0.9 ] X [0.9 ]
Series Modules
= System Voltage
12
[24] =
12
15-33
29.5 -> 30
Example:
The weekend cabin system was determined to have a daily load demand of
69.7 Ah/day. But if the system is really only going to be for weekend use,
then the array can be sized based on the Week Averaged Load Demand .
The array can be sized to replace the 2-day weekend demand over the entire
7-day week. Recalculating the 69.4 Ah/day daily load spread over the week:
69.4 Ah/day X 2 days / 7 days = 19.8 Ah/day week averaged.
Now we size the array to replace this amount on a daily basis.
Let the location for this system be near Madras. In this system we will use a
35-watt module, with a typical Imp of 2.0 amps. The best tilt angle for a
o
constant load demand throughout the year is 15 S.
o
The lowest insolation during the year at 15 tilt angle is 4.5 kwh/m during
July. This translates into 4.5 peak hours. Calculating the output from a
typical 35 watt module:
Module Output
9.0 Ah/day
2.0 Imp
Assume the same conservative battery coulombic efficiency of 90%, and the
usual 10% module output derating factor.
The number of 35 watt modules needed for this application is given by
Number of Parallel Modules
1 module in series
The work for this example is again worked out on the next page, using the lower
portion of the Array/Battery Sizing Form.
15-34
_____
_____
System Voltage
_____
_____
Average Discharge
Rate
_____
Battery Capacity
(@ ___ hour rate)
Chosen Battery
Capacity =
_____
Temperature Derating
_____
_____ Ah
_____
_____
_____
Parallel Batteries
(hours)
Voltage
Battery Capacity
=
Chosen Battery Cap.
[_____]
[
System Voltage
= [_____]
= _____
Chosen Battery V.
[
]
----------------------------------------------------------------------------------------------------------------Week-Averaged = 19.8 Module : 35 watt Module Output = 9.0 Ah/day
Daily Load
Series Batteries
Parallel Modules
Daily Load
.
Module X Derating X Charge
Output
Factor
Efficiency
[ 19.8 ]
[9.0 ] X [ 0.9 ] X [0.9 ]
Series Modules
= System Voltage
12
[_12_]
12
15-35
2.7
-> 3
1
Assumptions
For each location, the data includes the latitude and longitude, the Siemens Solar
map code identifier number, and then twelve output values, one for each month.
The values are calculated by the program PVSYSTEM, and are based on the
following assumptions:
(1) Constant load every month. This simplifies the automatic calculation process
and is applicable to a wide variety of situations.
(2) Best tilt angle for constant load. The computer chooses the array tilt angle that
best matches a constant load. This tilt angle is presented at the end of each
locations data.
(3) Module operating voltage of 14.3 volts. This is based on an assumed average
battery voltage of about 13.7 volts plus about .6 volts added for system losses
(4) Ground reflectance of 0.20 from dry grass.
15-36
The first page of the North American Region data table is presented to illustrate the
information presented. There are five regions of data: North America; Latin and
South America; Europe; Asia; and Africa. The most appropriate regional data for
you has been included.
NORTH AMERICAN
REGION
Module = SM55
Ah/day/module
Location
Map
Code
Lat
Long :
:
:
UNITED STATES
Alabama
BIRMINGHAM
MOBILE
MONTGOMERY
Alaska
ADAK
ANNETTE
BARROW
BETHEL
BETTLES
BIG DELTA
FAIRBANKS
GULKANA
HOMER
JUNEAU
KING SALMON
KODIAK
KOTZEBUE
MATANUSKA
MCGRATH
NOME
SUMMIT
YAKUTAT
Arizona
PHOENIX
PRESCOTT
TUCSON
WINSLOW
YUMA
Arkansas
FORT SMITH
LITTLE ROCK
California
ALPINE
ARCATA
ARROWHEAD
BAKERSFIELD
BLYTHE
BUTLER VALLEY RANCH
CARLSBAD
CARRISA PLAIN
CHULA VISTA
DAGGETT
DAVIS
867
868
224
34
31
32
87
88
86
900
901
902
136
904
905
158
907
908
909
910
911
912
185
913
914
915
916
52
55
71
61
67
64
65
62
60
58
59
58
67
62
63
65
63
60
177
132
157
162
152
146
148
145
152
135
157
152
163
149
156
165
149
140
197
871
220
873
874
33
35
32
35
33
112
112
111
111
115
875
180
35
35
94
92
692
881
693
882
695
1115
696
1223
697
884
152
33
41
34
35
34
41
33
35
33
35
39
117
124
117
119
115
124
117
120
117
117
122
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
DEC
Best
Tilt
11.38 13.12 14.14 14.75 13.98 13.39 13.14 14.20 14.64 15.48 13.41 11.34
12.73 14.37 14.82 14.63 13.55 12.63 12.10 13.09 13.98 15.86 14.14 12.36
11.84 13.49 14.39 15.00 14.02 13.49 13.13 14.14 14.50 15.82 14.01 12.12
55
55
55
5.49
4.67
0.00
3.45
0.02
1.61
0.79
2.67
4.17
3.24
4.95
4.57
0.01
3.62
2.04
0.62
2.02
2.94
8.03
7.46
2.24
8.09
5.56
7.40
6.79
7.78
8.07
5.89
9.04
7.96
5.25
9.12
7.15
6.66
7.07
5.74
9.48
10.07
11.21
12.27
12.58
13.34
12.88
13.41
12.21
9.04
12.60
12.24
12.14
16.20
12.31
11.58
12.62
9.66
10.03
11.49
15.56
13.00
15.49
14.65
14.22
14.83
13.26
11.02
12.45
12.88
15.32
14.72
13.52
13.97
14.34
11.35
9.26
11.55
12.09
11.67
15.30
14.32
13.89
13.37
12.54
10.55
11.51
11.20
15.58
13.17
12.41
13.52
13.81
10.50
8.54
10.33
15.13
10.80
14.47
13.35
13.26
12.82
12.26
10.42
10.62
11.21
15.23
12.30
11.67
13.22
12.08
10.03
8.39
10.67
15.22
9.61
12.89
12.84
12.37
12.47
11.85
9.85
10.04
10.83
13.35
11.50
10.68
11.21
11.01
9.39
8.27
10.41
10.36
8.28
11.22
12.46
11.30
12.25
10.96
9.10
9.27
10.95
11.34
11.05
9.76
9.73
10.11
8.81
8.74
9.88
6.47
9.20
10.62
11.46
10.61
11.31
10.41
7.93
9.91
10.30
10.36
9.92
9.72
9.77
9.98
8.10
8.70
7.10
2.79
7.50
6.34
7.56
6.79
8.78
8.99
5.53
9.59
9.75
6.53
7.75
6.77
7.11
7.86
6.46
6.96
5.21
0.03
4.10
1.28
3.29
2.66
3.83
5.51
3.72
6.33
5.93
0.71
4.76
3.41
1.99
3.72
3.56
4.90
3.34
0.00
1.62
0.00
0.00
0.00
0.72
2.18
1.63
3.46
3.26
0.00
1.75
0.33
0.00
0.20
1.37
70
75
55
80
80
80
80
80
80
75
80
75
75
80
80
80
80
75
17.62
17.95
18.38
17.46
18.56
19.61
19.60
20.12
19.62
20.43
20.31
20.84
21.00
21.00
21.69
20.48
21.31
21.20
21.47
21.59
19.07
20.72
20.05
20.51
20.37
17.59
19.95
18.72
19.64
19.16
16.73
17.58
17.02
17.81
17.43
18.01
18.21
18.14
18.60
18.81
19.81
20.72
19.98
20.53
20.35
20.32
20.84
20.53
20.55
20.74
18.76
19.14
19.14
18.93
19.42
17.05
17.31
17.60
16.76
17.92
55
50
50
50
50
12.76 14.02 14.34 13.87 13.73 13.67 14.00 14.87 14.98 15.62 13.93 12.45
12.28 13.88 14.20 13.71 13.72 13.64 13.71 14.64 15.02 15.83 13.60 12.04
60
60
13.55
10.07
13.08
13.28
17.02
9.59
13.34
16.38
14.65
17.08
11.62
55
65
55
60
50
65
55
60
55
60
60
17.49
12.19
16.73
15.79
19.47
8.90
17.29
16.23
18.11
18.53
13.20
15.57
13.11
16.96
17.86
20.59
10.91
17.90
15.93
17.63
19.78
16.69
16.90
14.06
18.50
18.03
20.40
13.02
16.85
16.45
17.36
19.28
17.51
14.40
13.40
18.38
17.46
19.01
12.72
14.55
16.89
14.86
17.72
17.18
16.62
13.03
18.69
17.09
18.19
13.51
14.59
15.95
14.56
16.88
16.82
17.05
12.59
18.64
17.26
17.20
14.98
14.86
16.17
15.37
16.53
17.50
17.65
12.93
19.69
18.64
18.69
15.89
15.59
17.29
16.02
18.04
18.64
19.18
14.12
21.39
19.84
19.70
16.43
16.58
17.58
17.66
19.60
19.56
19.56
13.07
21.18
19.34
20.45
11.57
15.67
18.96
17.43
19.86
17.98
19.03
10.66
19.76
15.77
19.71
9.17
17.84
14.60
18.00
18.27
14.57
16.76
9.55
17.06
12.40
17.83
8.71
15.92
14.04
16.02
16.63
11.18
"# $$ %$&
15-37
Please find the Data Tables appropriate for your region in the Appendix section.
15-38
Two examples of how to use the data are presented next. The first example is the
remote school, and the second is the remote cabin.
15-39
Example:
29.8
round up to
30 modules in parallel
2 in series
The work for this example is worked out on the next page, using the lower portion of
the Array/Battery Sizing Form.
15-40
Remote School
_____
_____
System Voltage
_____
_____
Average Discharge
Rate
_____
_____
(hours)
Temperature Derating =
_____
_____
Ah
[
Chosen Battery
Capacity =
] X [
_____________
Parallel Batteries
Voltage
Battery Capacity
=
Chosen Battery Cap.
[_____]
[
_____
=
_____
System Voltage
= [_____]
= _____
Chosen Battery V.
[
]
----------------------------------------------------------------------------------------------------------------Week-Averaged = 421 Module : SM55 Module Output = 17.43 Ah/day
Daily Load
Series Batteries
Parallel Modules
Daily Load
.
Module X Derating X Charge
Output
Factor
Efficiency
[ 421 ]
[17.43 ] X [ 0.9 ] X [0.9 ]
Series Modules
= System Voltage
12
[_24_]
[ 12 ]
15-41
29.8 -> 30
=
Example:
=
=
=
1.3 modules
round up to
2 modules in parallel
1 in series
The work for this example is again worked out on the next page, using the lower
portion of the Array/Battery Sizing Form.
15-42
Weekend Cabin
_____
_____
System Voltage
_____
_____
Average Discharge
Rate
_____
_____
(hours)
Temperature Derating =
_____
_____
Ah
[
Chosen Battery
Capacity =
] X [
_____________
Parallel Batteries
Voltage
Battery Capacity
=
Chosen Battery Cap.
[_____]
[
_____
=
_____
System Voltage
= [_____]
= _____
Chosen Battery V.
[
]
----------------------------------------------------------------------------------------------------------------Week-Averaged = 19.8 Module : SP75 Module Output = 18.84
Series Batteries
Ah/day
Daily Load
Parallel Modules
Series Modules
Daily Load
.
Module X Derating X Charge
Output
Factor
Efficiency
[ 19.8 ]
[ 18.84 ] X [ 0.9 ] X [0.9 ]
= System Voltage
12
[ 12 ]
[12 ]
15-43
1.3
-> 2
Exercise
5
$
!4 B
5
5>!0
5<!!
5>!
%%
0
"%
15-44
Daily Depth of
Discharge
Example:
15-45
C/25
C/15
C/5
The formula for checking the maximum charge rate of your installed system is given
below. You simply divide the total peak current that can come from the installed
parallel modules into the total installed battery capacity.
Maximum Rate
Of Charge
Example:
The remote school problem worked on pages 16 and 32 has been designed
with 30 75-watt modules in parallel by 2 modules in series for 24 volt
operation. There will be 3175 Ah of installed capacity. A check on the
maximum rate of charge shows:
Max. Rate
24 hours
This is slower than the typical maximum of five hours for deep cycling
batteries, so the array is not too big for the battery.
15-46
=
+
+
15-47
Direct
beam
radiation
Horizontal
terrestrial
surface
'( $)
*+
Tilted Insolation
Sun
Direct
beam
Diffuse whole-sky radiation
Reflected
radiation
Tilted
terrestrial
surface
,( $
15-48
Summary of Steps
Summary of steps to calculate insolation on a fixed tilted surface using monthly
average insolation data measured on a horizontal surface:
Step
Procedure
Variable
15-49
s, L
H
, hs
Ho
Kt
H diffuse
h' s
Rb
g
H tilted
Where n = julian day number for the representative day in each month.
The following table gives the recommended average day for each month of the
year:
Date
January 17
February 16
March 16
April 15
May 15
June 11
July 17
August 16
September 15
October 15
November 14
December 10
,.$/-
Next, the sunset hour angle, hs, is determined for the representative day for each
month:
hs = cos 1 ( tan L tan )
15-50
Ho =
24
360n
2hs
Hsc 1 + 0.033 cos
cos L cos sin hs +
sin L sin
365
360
H
Ho
Then, the diffuse to total radiation ratio, H diffuse H , is given by Collares-Pereira &
Rabl:
] (
H diffuse
= 0.775 + 0.00653( hs 90) 0.505 + 0.00455(hs 90) cos 115 K T 103
H
The sunset hour angle, h's, on tilted surfaces facing the equator is determined by:
15-51
cos( L s) cos sin hs +
h sin( L s ) sin
180 s
RB =
cos L cos sin hs +
h sin L sin
180 s
H tilted
H diffuse
H diffuse 1 + cos s
1 cos s
= 1
RB +
+ g
2
2
H
H
H
Where g = ground reflectivity, generally use 0.2 as default value. A table of ground
reflectivity factors is presented below for various common surfaces:
Ground Condition
Earth roads
Water surfaces from above
Coniferous forest (winter)
Weathered blacktop
Gravel, bituminous roof
Soils (clay, loam)
Dry grass
Crushed rock
Weathered concrete
Green grass
Forest in autumn, ripe field
Dark building surfaces
Dead leaves
Light building surfaces
Snow
.04
.07
.07
.10
.13
.14
.20
.20
.22
.26
.26
.27
.30
.60
.75
0)
15-52
Finally, the monthly average daily total solar radiation , H t , on tilted surfaces is
approximated by:
H tilted = R H
This final equation can be re-written in another form to more clearly show the three
components adding up to give the total insolation:
We can rewrite
H H diffuse as the direct component
equation can look like the following:
2
2
15-53
Tracking Surfaces
For tracking surfaces, the monthly average insolation must be estimated using longterm hourly radiation data. First, a more comprehensive form of the equation for the
monthly average beam radiation tilt factor, R B , must be used to account for the daily
variation in surface tilt (s) and azimuth angles () for the tracking surface. An
equation for R B for surfaces of any orientation has been developed by Klien:
180
Where,
(hs hr )
180
15-54
The hourly average radiation on a tilted surface can be approximated from monthly
average daily insolation by the following analytical expression suggested by
Collares-Pereira and Rabl:
r=
cos h cos hs
( a + b cos h)
24
sin hs (2hs 360) cos hs
The hourly totals for insolation on the tracking surface must then be integrated over
the entire day, over the range of surface tilt and azimuth angles to determine the
daily total radiation on tilted surfaces.
References
Hsieh, J.S., Solar Energy Engineering, Prentice-Hall, 1986.
15-55
(End of Chapter)
15-56
CHAPTER FIFTEEN
System Sizing
15-1
Basic Principles
15-2
15-3
Battery Sizing
Days of Autonomy or Reserve
Basic Battery Sizing Calculation
Correcting For Cold Temperature
Cold Temperature Data
Specifying Battery Capacity At Proper Discharge Rate
Limiting Maximum Depth of Discharge To Prevent Freezing
Complete Battery Sizing Calculation
Battery Cell Size
15-4
15-4
15-5
15-8
15-10
15-11
15-14
15-15
15-19
Array Sizing
Basic Array Sizing Calculation
Modifications To Array Sizing Formula
Complete Array Sizing Calculation
Choose A Module With The Proper Number Of Cells In Series
Peak Hour Method Of Estimating Module Daily Output
Siemens Solar Module Output Tables
Checking Battery Sizing
Calculating Tilted Insolation
15-20
15-20
15-21
15-22
15-24
15-26
15-36
15-45
15-47
15-57
Chapter 15 Answers
System Sizing
A photovoltaic system needs to be sized for the worst case to insure that it will operate
reliably year round. If the system were only sized to meet the average requirements,
then the system would be undersized for half the time. This could result in a lack of
sufficient energy to support the load during the periods of poor weather. Even if the
system could support the load in these conditions, the battery would remain at a low
state of charge for long periods of time. This would damage the battery and reduce its
lifetime. Sizing for the worst case results in a PV system that will operate reliably
throughout the year and extend the lifetime of the battery.
The load demand calculated in Problem 9-1 is 230 Ah/day. Since low cost batteries will
be used, the maximum % useable capacity will be 50%.
Battery capacity
2,300 Ah
Since the system is 12 volts, we will need a 12-volt battery of this capacity or sufficient
smaller voltage cells in series.
The load demand calculated in Problem 9-3 is 146.8 Ah/day. Using deep cycle
industrial batteries, we have 80% useable capacity.
Battery capacity
2,569 Ah
The overall battery bank voltage must be configured at 24 volts. This could be two 12volt batteries in series, four 6-volt batteries, or twelve 2-volt cells.
15-1
System Sizing
Lead-acid cells should not be completely discharged, even for high quality, deep cycle
batteries. Discharging lead-acid batteries beyond the 80% DOD limit can damage the
plates, cause the plates to break apart, or reverse the cell polarity. Any of these could
significantly reduce the battery lifetime.
The only capacity information we have for the batteries is the "rated capacity" of 80 Ah.
We do not know how this varies with rate. To get an assumption of the capacity at a
reasonable discharge rate, we will use the "rule of thumb" indicated in the text. We will
multiply the rated capacity by 130%. So the capacity will be:
80 Ah X 130%
80 X 1.30
104 Ah
The load demand is 30 Watts / 12 volts = 2.5 amps. The total daily Ah load is 2.5 amps
X 12 hours = 30 Ah/day. The battery autonomy is specified to be 5 days. Low cost
batteries will result in a maximum % useable capacity of 50%. The load operating time
is 12.
The average rate of discharge is
Avg Rate of Discharge
5 days X 12 hours
50% useable capacity
Battery Capacity
(@ 77 hour rate)
5 days X 30 Ah
.50 X .92
326 Ah
326 Ah
104 Ah per battery
3.13 batteries
15-2
rounded to 4 batteries.
System Sizing
Since this is a 12 V system the number of batteries in series is 1. Note that we required
4 batteries in parallel for a very small load. A larger load would have required more
than 4 batteries in parallel, but this is not good practice. This illustrates one of the
limitations of using automotive batteries for solar applications.
We calculated the load demand in Problem 9-3 to be 146.8 Ah/day. The number of
days of autonomy is given as 14. The deep cycling batteries will allow a maximum
charge of 80%. However, we have an additional limitation due to concerns about
freezing, which limits the maximum % capacity to 75% (refer to Figure 15-2).
For the weighted average load operating time, we should only take the peak loads that
will be on at the same time (the transmit loads for both transmitter #1 and #2):
Weighted operating time
8 X 12 + 5 X 8
8+5
136 / 13
10.5 hours
14 X 10.5
0.75
196 hours
At - 30 C, we need to derate the battery capacity 88% for temperature (using the
C/120 curve of Figure 15-1). The battery capacity is then:
Battery Capacity
@ C/196
Looking at Table 15-1, we see columns for 120 hours and 240 hours. To be
conservative, we will use the 120-hour column (since this gives the smaller rating for the
same cell). There are no cells with a capacity of 3114 Ah, so we will need to use
multiple cells in parallel. The capacity of KCPSA-15 cells is 1031. Three of these cells
in parallel give 3093 Ah, slightly more than we need. To get a 24-volt string of 2-volt
cells, we need 12 cells in series. Multiplying by 3 cells in parallel, we will need 36 cells
total.
15-3
System Sizing
Our requirement from Problem 15-6 is for 3079 Ah at a discharge rate of C/120. There
are a number of possible configurations:
Cells in
Paralle
l
2
3
4
Cell
Model
Cell
Capacity Total Ah
Vb2412
Vb2407
Vb2312
1800
1050
900
3600
3150
3600
Although 5 strings of the Vb2310 (750 Ah) would also give sufficient capacity, we rule
this out because it exceeds the Rule of Thumb for using no more than 4 parallel strings.
Since the Vb2407 is the smallest size that meets our requirements, it would probably be
the best choice. However, we would need to look at overall price, cell size and weight,
battery layout and similar details.
Location: Leh
The month with the lowest insolation is January, with 4 kWh/m2/day. To convert to
peak hours, we multiply by the conversion factor (in this case just 1).
4 kWh/m2/day X 1 = 4 peak sun hours
Looking on the data sheet, we find the following information:
Module
SP36
SP75
Pmax
36
75
Imp
2.1 Amps
4.4 Amps
The output for each module is the Imp multiplied by the peak sun hours
SP36:
SP75:
15-4
System Sizing
Location: Madras
The month with the lowest insolation is July, with 4.5 kWh/m2/day. The peak sun hours
are:
4.5 kWh/m2/day X 1 = 4.5 peak sun hours
Again, we take the Imp and multiply by the peak sun hours:
SP36:
SP75:
The load demand in Problem 9-1 is 230 Ah/day. We assume a nominal 12-volt system.
Looking at the data for Montgomery, Alabama, we find that the lowest output during the
year for an SM55 module is 11.84 Ah / day in January. We use the same derating
factors in the example: a battery coulombic efficiency of 90% and a module output
derating of 10%.
Number of Parallel Modules
230
9.59
1 in series
15-5
23.98 modules
System Sizing
The load demand in Problem 9-3 is 146.8 Ah/day. According to the data for Yuma,
Arizona, the lowest output during the year for an SM55 module is 17.43 Ah / day in July.
Since we are using the SP75 module, we adjust the output by a factor of 1.44.
SP75 output
25.1 Ah/day
Assume the same conservative battery coulombic efficiency of 90% and a module
output derating of 10%.
Number of Parallel Modules
146.8
20.33
2 in series
15-6
7.22 modules
System Sizing
Chapter Sixteen
System Wiring
All the precise calculations in the world can be sabotaged by improper wiring of the
components. Poor choice of wire size can restrict battery charging and eventually
cause system failure even with an adequately sized array. Imagine someone working
on the electrical system. Anticipate problems or dangers and how you could eliminate
the dangers with safety disconnect switches and fuses. Thinking about safety and
wiring requires that you consider what can possibly go wrong not assuming that
everything will be safe and that people will not do stupid things. Design the system as if
you were living with it and consider how you would want yourself and your family
protected from the danger of fire and system failure.
Electrical wiring is a specialized field and this chapter cannot possibly cover all the
issues completely. In the United States the National Electric Code (NEC) covers all
issues of wiring in depth and specifically covers solar photovoltaic systems in Article
690. A professional electrician or electrical contractor should be involved in the wiring
of photovoltaic power systems.
The entire photovoltaic system should be diagrammed before costing or installation.
This should include the wiring of modules in the array or sub-arrays, controllers, AC and
DC load centers, batteries, inverter if present, grounding and circuit protection. The
wiring process is the time to consider safety features such as fuses, circuit breakers,
ground fault interrupters, and grounding rods and wires. Also efficiency and cost
concerns will affect many specific choices such as the sizes and types of wires to use.
In this chapter we will look at three aspects of photovoltaic system wiring: (1) proper
wire sizing and insulation; (2) array wiring in series and parallel groups; and (3) safety
equipment wiring.
Proper wiring and the design of safe, user-friendly photovoltaic systems are the
most overlooked aspects of PV system design. Adherence to the National
Electric Code and safe practice will result in reduced hazards associated with
electrical installations. Careful design of the wiring subsystem will result in
efficient and reliable PV systems that are safely and easily serviced.
16-1
Exercise
16-2
Letter Designation
Type of Insulation
T
TW
THW
Thermoplastic
Thermoplastic Wet
Thermoplastic, High
temperature, Wet
Thermoplastic, High,
High temperature, Wet
Thermoplastic High,
High temperature,
Nylon jacket
Rubber insulation
Rubber, High
temperature, Wet
Rubber, High, High
temperature
Rubber, High, High
temperature, Wet
THHW
THHN
R
RHW
RHH
RHW-2
16-3
Wire
Neutral or Ground
White or Gray
White
Equipment Ground or
Grounding
Conductor Types
A cable is a connection of two or more wires combined in a common outer covering. It
is pre-assembled with wires of various sizes and insulation covers. All of the wires are
protected with an outer sheath made of plastic, rubber, or metal. Most of the wire you
will be working with comes in cable form, which is easier to work with than multiple
individual wires.
The most commonly used cable for electrical wiring of AC systems is nonmetallic
sheathed cable or type NM. This is commonly referred to as Romex, which is used in
house and building wiring in dry locations. Non-metallic sheathed cable is available in
different versions for a variety of other applications including areas of moisture ( NMC)
and wet areas buried in the ground (UF) commonly referred to as underground feeder
cable.
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
16-4
Wires and cables used in PV systems are typically designed for outdoor situations with
moisture and sunlight resistance or are placed in conduit to protect against weather.
The following cable types are the most commonly used in PV systems.
Wire/Cable Type
USE or USE-2
(Underground Service
Entrance)
or
Design Use
When made to UL
Standard, it has a 90C
rating and is sunlight
resistant. Moisture
proof, resistant to acids
and chemicals.
Applications in
Photovoltaic
Systems
The -2 designation
means that the
insulation is rated for
90C. wet or dry, and is
especially suited for the
hot temperatures that
module wiring would
endure.
TC is sunlight resistant
and generally marked as
such.
Not recommended
within battery
enclosures, Should
specify sunlight resistant
coating if used outside.
SE (Service Entrance)
TC (Tray Cable)
UF (Underground
Feeder)
Multi-stranded flexible
UL listed cable available
in large sizes
Multi-stranded flexible
cable available in large
sizes
Interior house wiring, dry
locations
and
USE/RHH/RHW
Welding Cable
NM (Romex)
Not UL Listed
16-5
Ampacity
Voltage drop
Ampacity is the current carrying capacity of the wire, or its ability to carry current safely
without overheating. Ampacity is based upon the wire size or cross-sectional area, the
type of materials its constructed of (e.g., copper, thermoplastic insulation), its length,
and its temperature. For these different conditions, the National Electric Code has
determined a maximum ampacity for each wire type (Romex, USE, etc.) for various
temperatures and lengths. This is the same standard for which PV systems need to be
designed.
The second consideration in wiring is voltage drop. Unlike higher voltage AC systems
where a small voltage drop will not have measurable impact on the efficiency and
deliverability of energy, PV systems are very sensitive to voltage changes. These
voltage drops must be considered in the design of a photovoltaic system.
16-6
16-7
UF
USE,RHW,
THHW,THWN
60 o
75 o
THHN,THHW
RHW-2,USE-2
90 o
14
12
20 25
25 30
20 30
25 35
25 35
30 40
10
8
30 40
40 60
35 50
50 70
40 55
55 80
6
4
55 80
70 105
65 95
85 125
75 105
95 140
2
1
95 140
110 165
115 170
130 195
130 190
150 220
125 195
150 230
170 260
00
000
145 225
165 260
175 265
200 310
195 300
225 350
0000
195 300
230 360
260 405
'
Regular Font = NEC 1993, Table 310-16. Ampacities of Insulated Conductors Rated 0-2000 Volts.
Not More Than Three Conductors in Raceway, Cable, or Earth (Directly Buried), Ambient Temp. of 30 oC.
Italic Font = NEC 1993, Table 310-17. Ampacities of Single Insulated Conductors Rated 0-2000 Volts
in Free Air Based on Ambient Temp. of 30 oC.
Am pacity
(am ps)
2.5
4
6
10
16
25
35
50
70
95
120
150
32
42
54
73
98
129
158
198
245
292
344
391
W ire am pacity rated at am bient tem perature of 30 deg.C., m axim um con ductor tem perature of 60
deg.C ., a nd assu m ing cable touches surfaces of w alls, etc. V oltage loss fa ctor rated at 20 deg.C.
(
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
16-8
16-9
Ampacity of
Inverter Wire
Maximum
Continuous
Input Current
Nominal Voltage
12
24
48
Example:
4000 watts
.87 X 44 volts
=
104 amps
Applying the 125% safety factor, the wire chosen should have an
ampacity of at least:
Ampacity value
104 X 125% =
130 amps
104 X 1.25
16-10
Derate the ampacity of the wire by 20% (apply as 125% factor) to insure that the
wire is not operated continuously at more than 80% of its rated value.
The combination of the two factors means that a factor of 156% should be applied to
the published Isc value of a module for ampacity sizing purposes.
156%
This factor should be applied to all short lengths of wire interconnecting array power to
switches, controls and other equipment.
16-11
75 oC
90 oC
1.00
.94
.88
.82
.75
.67
.58
.33
---
1.00
.96
.91
.87
.82
.76
.71
.58
.41
26-30
1.00
31-35
.91
36-40
.82
41-45
.71
46-50
.58
51-55
.41
56-60
--61-70
--71-80
--
16-12
Example:
Example:
16-13
Exercise
!
//////////
;<
16-14
12
10
no wire loss
AWG #4
Current 8
(amps)
6
AWG #10
AWG #14
4
2
-4
0
4
Voltage (volts)
12
16
Range of
battery
voltage
)
16-15
The batteries operate in a narrow range of voltage close to the "knee" of the array IV
curve if the knee is brought in due to small wire. The operating point on the curve
then drops down the IV curve and less current flows to the battery. The net effect is to
operate the array at a lower current level. The figure shows that if there was no
voltage loss in the wire and the batteries were operating at about 13.5 volts, then the
current from the array would be about 11 amps. With the losses expected using AWG
#10 wire the battery would get only about 8 amps. By using common residential wire
size of AWG #14 the current would be only about 6 amps, close to half of what is
expected. The current would not be what the designer anticipated and over time the
battery would never be fully recharged and the system would fail.
16-16
The voltage loss in high voltage AC circuits is usually quite small whereas the losses in
low voltage DC systems can be quite high. This is not because the current is DC or AC,
but rather because the voltages in AC systems are usually higher (typically 120 or 240
volts) while the voltages in DC systems are usually low (typically 12 or 24 volts). Some
examples will illustrate this.
Example:
A 1000 watt load (ex: heater, microwave oven, hair dryer) is located 25
feet from a load center. The total length of wire for the circuit is therefore
50 feet. The amps of current to the load would be given by
Current
Power Volts
8.3 amps
According to the ampacity charts, AWG #14 wire would be more than
sufficient to handle 8 amps.
If the wire used was standard household AWG #14 (voltage loss factor of
.002823 volts/amp/foot), the voltage loss in the circuit would be:
Voltage Drop =
(#14 wire)
(120 volts)
The recommended maximum of 2% voltage loss for a 120-volt circuit would be a drop
of 2.4 volts. The 1.18-volt drop in this circuit is less than 2.4, so in this case the wire
chosen based on ampacity is also acceptable in terms of voltage loss.
16-17
14
12
10
8
6
4
2
1
0
00
000
0000
.002823
.001775
.001117
.0007023
.0004416
.0002778
.0001747
.0001385
.0001099
.0000871
.0000691
.0000548
*
+
16-18
If the same load were operating at nominal 12 volts, however, the current would be
much higher to produce the same power and the wire size needed would be much
greater.
Example:
Power Volts
83.3 amps
1000 watts 12
According to the ampacity charts, the AWG #14 wire used in the
previous example would be insufficient to handle this high current.
We would need at least AWG #4 wire if we used 75 C rated
conductor.
16-19
Another example illustrates that using wire with enough ampacity does not necessarily
give the acceptable voltage drop of 2%.
Example:
19.2 amps
30 amps
The nominal system voltage is only 12 volts. This voltage drop of 2.05 volts is more
than 17% of that nominal voltage! The acceptable voltage drop for a 12-volt system is
only 2% of 12, or only .24 volts.
16-20
This illustrates that choosing wire based on ampacity for the long run from the array into
the main system components is unacceptable. We must size that run of wire based on
voltage drop. The wire chosen will be much larger than dictated by ampacity alone. To
size for acceptable voltage drop, we must use the equation on Page 14 in reverse.
Insert the wire length, the array current, and the acceptable voltage drop and solve for
the voltage loss factor. An example will illustrate.
Example:
Using the above array again, the array current is 19.2 amps, the
total wire length is 60 feet. The acceptable voltage drop is 2% of
12 volts or .02 X 12 = .24 volts.
Voltage Loss Factor
.24 volts
19.2 amps X 60 feet
.000208
Referring to the Voltage Loss Factor chart of Table 16-7, the factor for #4 wire
(.0002778 volt/amp/foot) is not small enough, so go to the next larger wire. This means
that #2 wire would be the size needed to keep the voltage drop less than 2%. Compare
this to the #12 that was acceptable with regard to ampacity only.
16-21
Tables have been created for nominal 12, 24, 36, 48 and 120-volt systems.
Distances indicated are for the one-way length of the conductor. The total round-trip
length of conductor has been taken into account in the calculations.
American Wire Gauge (AWG) tables indicate lengths in feet and metric (mm2) tables
indicate in meters.
If your current is greater than indicated in the tables then use the formula on Page 17
and the Voltage Loss Factors on Page 18 to determine the acceptable wire size. When
using the formula be sure to use the total length of the conductor, not just the one-way
distance that is used in the tables.
(Ampacity limits have not been indicated in these charts. Be sure to check that the
ampacity of the wire is not exceeded especially in the case of high current values).
16-22
12
10
00
000
0000
43
21
14
11
9
7
6
5
5
4
3
2
2
1
1
1
1
1
1
1
1
0
0
68
34
23
17
14
11
10
8
8
7
5
3
3
2
2
2
2
1
1
1
1
1
1
107
54
36
27
21
18
15
13
12
11
7
5
4
4
3
3
2
2
2
2
1
1
1
171
85
57
43
34
28
24
21
19
17
11
9
7
6
5
4
4
3
3
2
2
2
2
272
136
91
68
54
45
39
34
30
27
18
14
11
9
8
7
6
5
5
4
3
3
3
432
216
144
108
86
72
62
54
48
43
29
22
17
14
12
11
10
9
7
6
5
5
4
687
343
229
172
137
114
98
86
76
69
46
34
27
23
20
17
15
14
11
10
9
8
7
866
433
289
217
173
144
124
108
96
87
58
43
35
29
25
22
19
17
14
12
11
10
9
1092
546
364
273
218
182
156
136
121
109
73
55
44
36
31
27
24
22
18
16
14
12
11
1377
689
459
344
275
230
197
172
153
138
92
69
55
46
39
34
31
28
23
20
17
15
14
1737
868
579
434
347
289
248
217
193
174
116
87
69
58
50
43
39
35
29
25
22
19
17
2190
1095
730
548
438
365
313
274
243
219
146
110
88
73
63
55
49
44
37
31
27
24
22
[distances in feet]
14
[distances in meters]
2.5
10
16
25
35
50
70
95
120
150
15
7
5
4
3
2
2
2
2
1
1
1
1
0
0
0
0
0
0
0
0
0
0
24
12
8
6
5
4
3
3
3
2
2
1
1
1
1
1
1
0
0
0
0
0
0
35
18
12
9
7
6
5
4
4
4
2
2
1
1
1
1
1
1
1
1
0
0
0
63
31
21
16
13
10
9
8
7
6
4
3
3
2
2
2
1
1
1
1
1
1
1
99
50
33
25
20
17
14
12
11
10
7
5
4
3
3
2
2
2
2
1
1
1
1
154
77
51
38
31
26
22
19
17
15
10
8
6
5
4
4
3
3
3
2
2
2
2
217
108
72
54
43
36
31
27
24
22
14
11
9
7
6
5
5
4
4
3
3
2
2
311
155
104
78
62
52
44
39
35
31
21
16
12
10
9
8
7
6
5
4
4
3
3
441
221
147
110
88
74
63
55
49
44
29
22
18
15
13
11
10
9
7
6
6
5
4
583
291
194
146
117
97
83
73
65
58
39
29
23
19
17
15
13
12
10
8
7
6
6
745
373
248
186
149
124
106
93
83
75
50
37
30
25
21
19
17
15
12
11
9
8
7
930
465
310
233
186
155
133
116
103
93
62
47
37
31
27
23
21
19
16
13
12
10
9
,
16-23
12
10
00
000
0000
85
43
28
21
17
14
12
11
9
9
6
4
3
3
2
2
2
2
1
1
1
1
1
135
68
45
34
27
23
19
17
15
14
9
7
5
5
4
3
3
3
2
2
2
2
1
215
107
72
54
43
36
31
27
24
21
14
11
9
7
6
5
5
4
4
3
3
2
2
342
171
114
85
68
57
49
43
38
34
23
17
14
11
10
9
8
7
6
5
4
4
3
543
272
181
136
109
91
78
68
60
54
36
27
22
18
16
14
12
11
9
8
7
6
5
864
432
288
216
173
144
123
108
96
86
58
43
35
29
25
22
19
17
14
12
11
10
9
1374
687
458
343
275
229
196
172
153
137
92
69
55
46
39
34
31
27
23
20
17
15
14
1733
866
578
433
347
289
248
217
193
173
116
87
69
58
50
43
39
35
29
25
22
19
17
2184
1092
728
546
437
364
312
273
243
218
146
109
87
73
62
55
49
44
36
31
27
24
22
2755
1377
918
689
551
459
394
344
306
275
184
138
110
92
79
69
61
55
46
39
34
31
28
3474
1737
1158
868
695
579
496
434
386
347
232
174
139
116
99
87
77
69
58
50
43
39
35
4380
2190
1460
1095
876
730
626
548
487
438
292
219
175
146
125
110
97
88
73
63
55
49
44
[distances in feet]
14
[distances in meters]
2.5
10
16
25
35
50
70
95
120
150
29
15
10
7
6
5
4
4
3
3
2
1
1
1
-
47
24
16
12
9
8
7
6
5
5
3
2
2
2
1
1
-
71
35
24
18
14
12
10
9
8
7
5
4
3
2
2
2
2
1
1
1
1
1
1
126
63
42
31
25
21
18
16
14
13
8
6
5
4
4
3
3
3
2
2
2
1
1
198
99
66
50
40
33
28
25
22
20
13
10
8
7
6
5
4
4
3
3
2
2
2
308
154
103
77
62
51
44
38
34
31
21
15
12
10
9
8
7
6
5
4
4
3
3
433
217
144
108
87
72
62
54
48
43
29
22
17
14
12
11
10
9
7
6
5
5
4
622
311
207
155
124
104
89
78
69
62
41
31
25
21
18
16
14
12
10
9
8
7
6
882
441
294
221
176
147
126
110
98
88
59
44
35
29
25
22
20
18
15
13
11
10
9
1165
583
388
291
233
194
166
146
129
117
78
58
47
39
33
29
26
23
19
17
15
13
12
1491
745
497
373
298
248
213
186
166
149
99
75
60
50
43
37
33
30
25
21
19
17
15
1860
930
620
465
372
310
266
233
207
186
124
93
74
62
53
47
41
37
31
27
23
21
19
-
16-24
12
10
00
000
0000
128
64
43
32
26
21
18
16
14
13
9
6
5
4
4
3
3
3
2
2
2
1
1
203
101
68
51
41
34
29
25
23
20
14
10
8
7
6
5
5
4
3
3
3
2
2
322
161
107
81
64
54
46
40
36
32
21
16
13
11
9
8
7
6
5
5
4
4
3
513
256
171
128
103
85
73
64
57
51
34
26
21
17
15
13
11
10
9
7
6
6
5
815
408
272
204
163
136
116
102
91
82
54
41
33
27
23
20
18
16
14
12
10
9
8
1296
648
432
324
259
216
185
162
144
130
86
65
52
43
37
32
29
26
22
19
16
14
13
2061
1030
687
515
412
343
294
258
229
206
137
103
82
69
59
52
46
41
34
29
26
23
21
2599
1300
866
650
520
433
371
325
289
260
173
130
104
87
74
65
58
52
43
37
32
29
26
3276
1638
1092
819
655
546
468
409
364
328
218
164
131
109
94
82
73
66
55
47
41
36
33
4132
2066
1377
1033
826
689
590
517
459
413
275
207
165
138
118
103
92
83
69
59
52
46
41
5211
2605
1737
1303
1042
868
744
651
579
521
347
261
208
174
149
130
116
104
87
74
65
58
52
6571
3285
2190
1643
1314
1095
939
821
730
657
438
329
263
219
188
164
146
131
110
94
82
73
66
[distances in feet]
14
[distances in meters]
2.5
10
16
25
35
50
70
95
120
150
44
22
15
11
9
7
6
5
5
4
3
2
2
1
1
1
1
1
1
1
1
0
0
71
35
24
18
14
12
10
9
8
7
5
4
3
2
2
2
2
1
1
1
1
1
1
106
53
35
27
21
18
15
13
12
11
7
5
4
4
3
3
2
2
2
2
1
1
1
188
94
63
47
38
31
27
24
21
19
13
9
8
6
5
5
4
4
3
3
2
2
2
298
149
99
74
60
50
43
37
33
30
20
15
12
10
9
7
7
6
5
4
4
3
3
462
231
154
115
92
77
66
58
51
46
31
23
18
15
13
12
10
9
8
7
6
5
5
650
325
217
162
130
108
93
81
72
65
43
32
26
22
19
16
14
13
11
9
8
7
6
933
466
311
233
187
155
133
117
104
93
62
47
37
31
27
23
21
19
16
13
12
10
9
1324
662
441
331
265
221
189
165
147
132
88
66
53
44
38
33
29
26
22
19
17
15
13
1748
874
583
437
350
291
250
218
194
175
117
87
70
58
50
44
39
35
29
25
22
19
17
2236
1118
745
559
447
373
319
280
248
224
149
112
89
75
64
56
50
45
37
32
28
25
22
2791
1395
930
698
558
465
399
349
310
279
186
140
112
93
80
70
62
56
47
40
35
31
28
16-25
12
10
00
000
0000
170
85
57
43
34
28
24
21
19
17
11
9
7
6
5
4
4
3
3
2
2
2
2
270
135
90
68
54
45
39
34
30
27
18
14
11
9
8
7
6
5
5
4
3
3
3
430
215
143
107
86
72
61
54
48
43
29
21
17
14
12
11
10
9
7
6
5
5
4
683
342
228
171
137
114
98
85
76
68
46
34
27
23
20
17
15
14
11
10
9
8
7
1087
543
362
272
217
181
155
136
121
109
72
54
43
36
31
27
24
22
18
16
14
12
11
1728
864
576
432
346
288
247
216
192
173
115
86
69
58
49
43
38
35
29
25
22
19
17
2748
1374
916
687
550
458
393
343
305
275
183
137
110
92
79
69
61
55
46
39
34
31
27
3466
1733
1155
866
693
578
495
433
385
347
231
173
139
116
99
87
77
69
58
50
43
39
35
4368
2184
1456
1092
874
728
624
546
485
437
291
218
175
146
125
109
97
87
73
62
55
49
44
5510
2755
1837
1377
1102
918
787
689
612
551
367
275
220
184
157
138
122
110
92
79
69
61
55
6947
3474
2316
1737
1389
1158
992
868
772
695
463
347
278
232
198
174
154
139
116
99
87
77
69
8761
4380
2920
2190
1752
1460
1252
1095
973
876
584
438
350
292
250
219
195
175
146
125
110
97
88
[distances in feet]
14
[distances in meters]
2.5
10
16
25
35
50
70
95
120
150
58
29
19
15
12
10
8
7
6
6
4
3
2
2
2
1
1
1
1
1
1
1
1
94
47
31
24
19
16
13
12
10
9
6
5
4
3
3
2
2
2
2
1
1
1
1
142
71
47
35
28
24
20
18
16
14
9
7
6
5
4
4
3
3
2
2
2
2
1
251
126
84
63
50
42
36
31
28
25
17
13
10
8
7
6
6
5
4
4
3
3
3
397
198
132
99
79
66
57
50
44
40
26
20
16
13
11
10
9
8
7
6
5
4
4
615
308
205
154
123
103
88
77
68
62
41
31
25
21
18
15
14
12
10
9
8
7
6
866
433
289
217
173
144
124
108
96
87
58
43
35
29
25
22
19
17
14
12
11
10
9
1244
622
415
311
249
207
178
155
138
124
83
62
50
41
36
31
28
25
21
18
16
14
12
1765
882
588
441
353
294
252
221
196
176
118
88
71
59
50
44
39
35
29
25
22
20
18
2330
1165
777
583
466
388
333
291
259
233
155
117
93
78
67
58
52
47
39
33
29
26
23
2981
1491
994
745
596
497
426
373
331
298
199
149
119
99
85
75
66
60
50
43
37
33
30
3721
1860
1240
930
744
620
532
465
413
372
248
186
149
124
106
93
83
74
62
53
47
41
37
16-26
12
10
00
000
0000
425
213
142
106
85
71
61
53
47
43
28
21
17
14
12
11
9
9
7
6
5
5
4
676
338
225
169
135
113
97
85
75
68
45
34
27
23
19
17
15
14
11
10
8
8
7
1074
537
358
269
215
179
153
134
119
107
72
54
43
36
31
27
24
21
18
15
13
12
11
1709
854
570
427
342
285
244
214
190
171
114
85
68
57
49
43
38
34
28
24
21
19
17
2717
1359
906
679
543
453
388
340
302
272
181
136
109
91
78
68
60
54
45
39
34
30
27
4320
2160
1440
1080
864
720
617
540
480
432
288
216
173
144
123
108
96
86
72
62
54
48
43
6869
3434
2290
1717
1374
1145
981
859
763
687
458
343
275
229
196
172
153
137
114
98
86
76
69
8664
4332
2888
2166
1733
1444
1238
1083
963
866
578
433
347
289
248
217
193
173
144
124
108
96
87
10919
5460
3640
2730
2184
1820
1560
1365
1213
1092
728
546
437
364
312
273
243
218
182
156
136
121
109
13774
6887
4591
3444
2755
2296
1968
1722
1530
1377
918
689
551
459
394
344
306
275
230
197
172
153
138
17369
8684
5790
4342
3474
2895
2481
2171
1930
1737
1158
868
695
579
496
434
386
347
289
248
217
193
174
21902
10951
7301
5475
4380
3650
3129
2738
2434
2190
1460
1095
876
730
626
548
487
438
365
313
274
243
219
[distances in feet]
14
[distances in meters]
2.5
10
16
25
35
50
70
95
120
150
146
73
49
37
29
24
21
18
16
15
10
7
6
5
4
4
3
3
2
2
2
2
1
236
118
79
59
47
39
34
29
26
24
16
12
9
8
7
6
5
5
4
3
3
3
2
354
177
118
88
71
59
51
44
39
35
24
18
14
12
10
9
8
7
6
5
4
4
4
628
314
209
157
126
105
90
79
70
63
42
31
25
21
18
16
14
13
10
9
8
7
6
992
496
331
248
198
165
142
124
110
99
66
50
40
33
28
25
22
20
17
14
12
11
10
1538
769
513
385
308
256
220
192
171
154
103
77
62
51
44
38
34
31
26
22
19
17
15
2166
1083
722
542
433
361
309
271
241
217
144
108
87
72
62
54
48
43
36
31
27
24
22
3109
1554
1036
777
622
518
444
389
345
311
207
155
124
104
89
78
69
62
52
44
39
35
31
4412
2206
1471
1103
882
735
630
551
490
441
294
221
176
147
126
110
98
88
74
63
55
49
44
5825
2913
1942
1456
1165
971
832
728
647
583
388
291
233
194
166
146
129
117
97
83
73
65
58
7453
3727
2484
1863
1491
1242
1065
932
828
745
497
373
298
248
213
186
166
149
124
106
93
83
75
9302
4651
3101
2326
1860
1550
1329
1163
1034
930
620
465
372
310
266
233
207
186
155
133
116
103
93
"
16-27
Exercise
+
0
,
>
!"3445!7
!
//////////"+1
"
&'
%&'
A) %&' $!+
0
,
//////////"+1
+
0
,
>
!"3445!7
!
//////////"+1
+
0
,
>
!"3445!7
!D
-E !
//////////"+1
16-28
16-29
12 Volt A rray
S chem atic
)
16-30
The actual installed wiring of a 12-volt array using dual J-box modules would also
have all the positive connections along one side and all the negative ones along the
other.
12 V o lt A rra y
A s In stalle d
D u al J-b o x
)
"
The installed wiring of a 12-volt array for single J-box modules would also have all the
positives connected together in parallel, and all the negatives connected together.
12 Volt Array
As Installed
Single J-box
)
'
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
16-31
Parallel Groups or
Series Groups
Buss bar
Buss bar
)
(
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
16-32
2 4 V o lt A r r a y W i r i n g
S c h e m a t ic
)
16-33
The installed wiring for 24 volts using modules with separate J-boxes is shown below.
The two-module strings would be flip-flopped alternating positive and negative Jboxes. The intermediate wires for each series string, connecting the positive of the
bottom module to the negative of the top module, are shown to the right of the array.
The positive outputs from the two-module strings would all combine, and the negative
inputs would also combine. Typical ground mount structures hold up to eight modules,
so four 24-volt strings could be mounted on one structure.
2 4 V o lt A r r a y
A s In s t a lle d
D u a l J -b o x
)
*
The installed wiring for single J-box modules is shown again with all the J-boxes aligned
along one side. The intermediate connection within each string, from the negative of
one module to the positive of the second module, is shown.
2 4 V o lt A r r a y
A s In s ta lle d
S in g le J -b o x
)
+
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
16-34
48 V o lt A rra y
S c h e m a tic
+
) ,
16-35
Installed field wiring for a 48-volt system with dual J-box modules would again have the
modules flipped. Typical ground mount structures hold up to eight modules, so two
48-volt strings could fit on one structure.
48 Volt Array
As Installed
Dual J-box
)
Installed wiring for single J-box modules would also have groups of four modules
connected in series.
48 Volt Array
As Installed
Single J-box
)
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
16-36
Exercise
!3
F0 G
!
3
!>
!
#
H
I
//////////
//////////
//////////
//////////
//////////
//////////
&
//////////
//////////
//////////
//////////
//////////
//////////
//////////
//////////
//////////
//////////
//////////
//////////
.
//////////
//////////
..
//////////
//////////
16-37
10
11
Load
6
zero volts
3
)
./...0.
16-38
(+)
+
(+)
(+)
(+)
+
+
+
-
(-)
-
(-)
(-)
(-)
)
"
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
16-39
16-40
Overcurrent Protection
Overcurrent devices must be able to handle the rated current that will open circuit.
All fuses and circuit breakers should be in enclosures.
Circuit breakers or fuses should not be paralleled in the same circuit.
Protection ratings of the overcurrent protection device must never exceed the
current ratings of the wire.
Grounded conductors should never be fused or switched.
16-41
Manufacturer
Type
Voltage
(volts)
AIR
(Amps)
Comments
5,000-25,000
Heinemann
Circuit Breakers
65-250
Continuous
Rating
(Amps)
15-700
Square D
Circuit Breakers
48
15 - 70
5,000
Bussmann
Gould
Littlefuse
125 - 300
1 - 1200
20,000
Bussmann
Gould
Littlefuse
125-600
0.1 -600
20,000
'
1)
..
16-42
Fuse Example
Presented below is the current vs. time characteristics of FLNR series time-delay
current-limiting fuses, DC rated, Class RK5. The time-delay models are really
designed to allow surge currents to flow through to loads such as motors and
compressors. They will have a slower response time curve than fast acting fuses
designed for sensitive electronic equipment. The continuous current range is shown
from 15 amps to 600 amps. For each value of fuse the amount of current that it will
pass is shown compared to the time before the fuse will interrupt the current.
Example:
The 200-amp rated fuse curve is shown (as the dark line). The
curve indicates that the fuse will pass 600 amps for about 60
seconds, and 1000 amps for about 10 seconds and so on. It will
allow 2000 amps for only about 0.2 seconds, and 4000 amps for
only about 0.02 seconds. So at higher currents, the time to
interrupt is quite short.
)
'
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
16-43
Example:
) (
16-44
Current-Limiting Devices
Batteries can produce thousands of DC amps under short circuit conditions. A 100-Ah
battery could release up to 2000 amps of short circuit current, while a 1000-Ah cell
might be able to produce up to 20,000 amps! This level of current can overwhelm and
damage downstream overcurrent protection. This downstream overcurrent equipment
needs protection too! Specially designed fuses, called current-limiting fuses, can
interrupt large currents. These are generally classified as RK5 or RK1 (DC rated Class
J and T fuses can be used as well).
16-45
A typical high interrupt rating for a current limiting fuse is 200,000 amps of AC current.
Under DC conditions this is derated to 20,000 amps. So a current limiting fuse should
be connected to each battery string up to about 600-1000 Ah capacity. In this way
downstream overcurrent equipment is protected from excessively high currents. This
means that designers should plan to break very large battery banks into units of
maximum 600-1000 Ah. Single banks of greater than 1000 Ah will need specially
designed overcurrent protection systems.
(-)
)
16-46
Exercise
D
'
HED
!.
,
//////////
D
41J$
.
$ =
&
,
///// /////
>
,
///// /////
"%'
$
!+
A* ) ,5
7!
<B<
16-47
Safety Disconnects
Safety disconnects or switches are placed into power systems to allow equipment to be
safely installed and maintained. The NEC requires that all source circuits or voltage
sources in photovoltaic systems such as photovoltaic arrays, battery banks, and
generators, must be able to be isolated in case there is a problem. Circuit breakers and
fused disconnect switches provide a means to disconnect safely, as well as providing
for overcurrent protection. The code limits the total number of switches or circuit
breakers in a PV system to six. They are further required to be grouped and marked
and readily accessible to be pulled.
All safety disconnects in photovoltaic systems must be DC rated for voltage as well as
current. This may involve specifying a more expensive or heavy duty version of an AC
rated model. DC rated switches are marked to indicate added protection against
arcing.
Typically there are four locations where disconnect devices are needed in photovoltaic
systems. They are: (1) between the array and the charge regulator; (2) between the
regulator and the battery; (3) between the battery and any DC loads or load center; and
(4) between the battery and the inverter. Disconnect means may be built into small
charge regulators or system controllers. For example, the load terminals on charge
regulators may have circuit breaker protection built -in.
Safety Disconnects
Modules
Regulator
Battery
DC Loads
Inverter & AC Loads
) *
16-48
system ground
RegArrayBattery-
Inv-
DC LoadA
Array+
Battery+
Reg Array+
Reg Battery+
Inv+
DC Load +
)
+
The charging circuit passes current from the Array + through circuit breaker A and on to
the charge regulator (terminal designated as Reg Array +). Current returns back from
the regulator (from terminal designated as Reg Battery +) and through circuit breaker
B, and then on to the Battery +. The negative current from the battery enters from the
Battery -, and passes through the central negative wiring block E, where it continues on
to the Array -. This completes the charging circuit.
16-49
The load circuit is adjacent and uses common components. The current begins at the
Battery +, flows through a jumper to circuit breaker C and then on to the Inverter +
terminal. The current can also flow to one of possibly many DC load breakers D where
it then flows on to the DC Load + terminal of various DC loads. Current would return
from the Inverter -, or from the DC Load -, and connect at the central negative wiring
block E, where it would then flow back to the Battery -. This would complete the load
circuit.
The flexibility of this method is that the charging circuit breakers and the load breakers
do not have to be of the same size. For example, the array might need only a 30-amp
breaker while the inverter might need a 200-amp breaker and the DC loads might use a
100-amp breaker.
The dashed lines represent the equipment grounding wires. All equipment grounds are
collected at one central point, the central negative wiring block E. This includes the
grounding wire from the array ground. This grounding wire must come all the way in
from the metal mounting structure and the module frames in the field. All negative
electrical conductors also connect at E. Thus this point becomes the one central point
in the circuit where electrical and equipment grounds are bonded. A main system
ground wire then connects from this central connection point to a ground rod that
makes the firm ground to earth.
16-50
) ,
16-51
16-52
ArrayBattery Array+
Battery+
inverter
ground
A
Reg Array+
system
ground
regulator
ground
DC
load
ground
Inv+
DC Load+
Reg Battery+
Reg-
Inv-
DC Load-
) -
16-53
16-54
(+)
Controller #1
Controller #2
(-)
DC
load
center
AC
load
center
Inverter
(-)
(+)
)
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
16-55
Exercise
16-56
16-57
Grounding Details
If the solar array will be a great distance away from the central ground rod for the
system a separate ground rod for the array structure can be driven close to the array.
However there should be a bonding wire that connects this remote ground rod to the
main system ground rod. This bonding wire keeps both ground rods at the same
voltage potential. If the two rods were not connected a nearby lightning strike would
result in a momentary surge of current into the earth that would cause large differences
in ground potential due to the resistance of the earth. The two ground rods would not
both be a "zero" potential and large induced currents will be created in the ground
system. By having the wire connecting the two rods the resistance of the earth is
overcome, and the two rods stay at the same potential and no induced current result.
16-58
Array disconnect
Bonding wire
Equipment
ground
System
grounding
electrode
Array frame
grounding
electrode
16-59
to load
circuits
from charging
circuits
to load
circuits
current limiting
fuses
disconnect
switches
Battery
Battery
Grounded Systems
Ungrounded Systems
) "
16-60
From array
SOV's
Bonding wire
to system electrode
16-61
An example of a field junction box with surge protection is presented below. Allowance
for up to eight separate array positive wires is made. Each positive connection is made
to a fuse and blocking diode and then combined into one positive buss bar. Current
flow through a field disconnect circuit breaker and then on to the positive output
terminal block. The array negative(s) would connect to the negative terminal block.
The SOV surge arrestor has connections to both the positive buss bar and to the
negative terminal block. It is also connected to the ground block. This is where the
high lightning induced surge current would flow.
The array grounding electrode would be connected to the ground block as would the
bonding wire to the system grounding electrode.
16-62
Array Size:
Batteries:
Load:
Description:
Module Interconnects:
=
=
=
16-63
49.5 amps
16-64
6 X 4.4 amps
26.4 amps
Using the Voltage Drop Table we see that for 25 Amps, 1/0 AWG
is required for a 2% drop with a maximum one-way distance of 55
feet. So we must use 1/0 size wire for this part of the circuit.
Connecting 1/0 AWG to the #8 AWG from the modules will require
a power distribution block in the field junction box.
16-65
Grounding Conductors:
Equipment grounding conductor sizing requires a number of
calculations and tables. (See Section 250 in NEC). #6 AWG is a
sufficient size to handle the grounding for this system. The
grounding electrode conductor should be # 1/0 AWG, as it is
required to be as large as the largest conductor in the system is.
Overcurrent Protection and Disconnect:
Each source circuit must be protected by overcurrent devices. A
current-limiting fuse in a fused disconnect switch is required by
code to protect the load and the array circuit from the batteries.
Array Circuit
The design current of the array circuit fuse must be greater than
the array short circuit current:
28.8 amps X 1.25 NEC X 1.25 UL = 44 amps
A 50 amp RK-5 type fuse is chosen to meet this requirement.
Each component must be able to be disconnected from all sources
of power. The charge control is between the array and the battery.
Hence a 50-amp single-pole circuit breaker (cheaper than a fused
disconnect switch) is specified to isolate the controller from the
array.
Load Circuit
To determine the protection on the load circuit, the maximum load
power of 75 watts at 12 volts is converted to
75 12 = 6.25 amps. The protection rating in this circuit must only
meet the standard NEC overrating factor:
6.25 amps X 1.25 NEC = 7.8 amps.
A 30-amp fuse will meet this requirement easily.
Both fuses should have an amperes interrupting rating (AIR) of
around 20,000 amps sufficient to withstand the short circuit
currents from the battery under fault conditions.
16-66
Voltage Ratings:
Disconnects, fuses, and circuit breakers must have a voltage rating
of at least 1.25 times the system open circuit voltage. Standard
module Voc is 22 volts.
22 volts X 1.25 NEC = 27.5 volts.
16-67
16-68
= [
Charge Controller Model :
] X [
] x 1.3
______________ amps
__________________________________________________________________________
Number of Controllers
and Sub-arrays
________________
Sub-Array :
Number of Parallel Modules =
amps
__________________
________________
Wiring :
Wire Segments Endpoints
Voltage
% loss Current
One-way Wire
Distance Gauge
Wire
Type
Ampacity Fuse
or CB
Rating
16-69
(End of Chapter)
16-70
CHAPTER SIXTEEN
SYSTEM WIRING
16-1
16-2
16-3
16-4
16-4
16-6
16-7
16-7
16-7
16-9
16-11
16-12
16-15
16-16
16-17
16-22
16-23
16-24
16-25
16-26
16-27
16-29
16-30
16-32
16-33
16-35
16-71
16-39
16-39
16-41
16-48
16-57
16-60
16-61
16-63
16-68
16-72
Chapter 16 Answers
System Wiring
Refer to the National Electric Code or the Photovoltaic Power Systems and the National
Electric Code, Suggested Practices.
The inverter has a continuous output of 2400 watts. A nominal voltage of 24 volts
means that the input voltage could be as low as 22 volts (see table in the text). Given
an efficiency of 85%, we calculate the maximum input current as:
Max Inverter Input Current
2400 Watts
.85 X 22 volts
128.3 amps
Using the 125% safety factor from the NEC, our wire needs to have an ampacity of at
least:
Ampacity
128.3 X 125%
160.4 amps
128.3 X 1.25
Per Table 16-4, a conductor made of 14 AWG wire and type THHN (75 C) insulation is
rated for not more than 20 amps. At an ambient temperature of 40 C, this value needs
to be derated by 0.88 as shown in Table 16-6. We also need to divide by the NEC
safety factor:
Derated Ampacity
20 X 0.88
=
1.25 (NEC)
16-1
14.1 amps
Wiring
Qty X Watts
Volts
13.3 amps
4 X 40
12
Since the ampacity under these conditions is greater than the load current, the
conductor can be used to carry the load current. The answer is then "Yes."
The temperature derate for 90 C conductors at 40 C is .91. Using the array Isc, we
calculate the minimum ampacity of the wire as:
Minimum Ampacity =
We first evaluate the voltage drop only, using a current of 6 amps and a total distance
of 20 X 2 = 40 feet. Based on a 12-volt system, the acceptable voltage drop is 2% of
12 volts or .02 X 12 = .24 volts.
Voltage Loss Factor
.24 volts
6 amps X 40 feet
0.001
Referring to Table 16-7, we see that the first conductor size with a voltage loss factor
less than this number is 8 AWG.
16-2
Wiring
Next looking at the ampacity, we calculate the minimum required ampacity as:
Minimum Ampacity
7.5 amps
Using THHN (90 C) in conduit, we find using Table 16-4 that 14 AWG has an ampacity
of 25 amps, easily accommodating this current. This is a case where the wire size will
be determined by the voltage drop rather than just the ampacity.
We will first evaluate the voltage drop. The current is 2 X 40 Watts 24 volts = 3.3
amps. The total distance is 35 X 2 = 70 feet. Based on a 24-volt system the
acceptable voltage drop is 2% of 24 volts or .02 X 24 = .48 volts.
Voltage Loss Factor
.48 volts
3.3 amps X 70 feet
0.0021
Per Table 16-7, we see that the first conductor size with a voltage loss factor less than
this number is 12 AWG.
For the minimum ampacity required, we calculate it as follows (there are no
temperature derates):
Minimum Ampacity
4.1 amps
Referring to the values of THHN (90 C) in conduit in Table 16-4, we see that 14 AWG
has an ampacity of 25 amps, easily accommodating this current. Again, the voltage
drop in this case will determine the necessary conductor size.
16-3
Wiring
We will first evaluate the voltage drop for the array. Using 5 modules in parallel the Isc
for the array will be 5 X 3.45 = 17.25 amps. The total distance is 50 X 2 = 100 feet.
Based on a 24-volt system the acceptable voltage drop is 2% of 24 volts or .02 X 24 =
.48 volts.
Voltage Loss Factor
.48 volts
17.25 amps X 100 feet
0.00028
Per Table 16-7, we see that the first conductor size with a voltage loss factor less than
this number is 4 AWG.
We next look at the minimum ampacity required. Remember, for an array circuit, we
need to use both the NEC and UL safety factors:
Minimum Ampacity
Referring to the values of THHN (90 C) in conduit in Table 16-4, we see that 12 AWG
has an ampacity of 30 amps.
Point
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Current
Flowing
12 amps
3 amps
6 amps
3 amps
3 amps
3 amps
3 amps
3 amps
3 amps
9 amps
12 amps
16-4
Voltage
0 volts
0 volts
0 volts
0 volts
12 volts
24 volts
36 volts
48 volts
48 volts
48 volts
48 volts
Wiring
Referring to Figure 16-14, we first identify the curve for a 60A fuse. This is the third
curve from the top. We next find the time 0.1 seconds across the top and move down
to where this intersects the 60A fuse curve. Reading across horizontally, we find this is
about 900A of overcurrent.
An overcurrent of 400 amps for a 100-amp breaker represents a 400% load. Referring
to Figure 16-15, we find the 400 percent load on the bottom and then move up to
standard time delay range (top range shown). This indicates a time between 4-11
seconds to interrupt.
If we were to use a short time delay breaker (bottom range), the time is .03 and .2
seconds to interrupt.
Again looking at Figure 16-15, we find the 300% load at the bottom and move up until
we reach the 5 second mark. This occurs within the medium time delay range.
Refer to the appropriate diagram.
Refer to the appropriate diagram.
16-5
Wiring
Chapter Seventeen
Water Delivery Systems
Water pumping is a fast growing and wonderful application for photovoltaic power
systems. There is a natural match between the availability of sunlight and the need for
water. Most photovoltaic pumping systems do not use batteries thus avoiding a costly
and high maintenance component and increasing the reliability greatly. In rural
undeveloped areas there is a critical need for fresh underground water to help prevent
diseases spread by using surface waters. A low maintenance photovoltaic powered
water system can bring health and prosperity to remote villages, without the burdens of
paying for maintenance, spare parts or fuel. In this chapter we discuss issues
associated with photovoltaic powered water delivery systems including performance
characteristics of different types of pumps and sizing concerns.
The focus of this discussion is on direct coupled solar powered water pumping
systems. These systems output water in proportion to available solar insolation. They
pump more water during summer months, typically when more water is needed, and
avoid the need for batteries and regulators. Such direct coupled systems may not be
as efficient as battery coupled designs in the sense that they cannot pump during any
weather or at night. But the advantages of reliability, simplicity and low cost make
these types of systems very attractive for a wide range of applications and users around
the world.
17-1
Diesel
Engine
Generator
Disadvantages
No fuel required
Little maintenance
Environmentally Benign
Panel life is 20-30 years
Cost effective for small
power demands
Requires regular
maintenance and fuel
brought to site
Requires dependable
operator and service
support
Environmentally harmful
Expensive when
considering life cycle cost
17-2
17-3
17-4
Irrigation
Water can be pumped from nearby streams or canals or lakes to irrigate fields. Or it
can be draw up from even hundreds of feet. Photovoltaic systems can be permanently
mounted in the fields, be portable so they can serve multiple farmers, or be moved to
different fields as needed.
Drip irrigation is more efficient than flood irrigation and is especially suited for hot
climates. However, it involves more equipment and is more expensive. Photovoltaic
power systems can operate the pumps as well as the pressurizing systems to pass
water through filters that are often needed for drip systems.
In general solar water pumping for irrigation is not competitive when compared to diesel
pumping. An irrigation system usually requires a large amount of power for a short
seasonal period, whereas solar water pumping has its greatest advantages with small
scale year-round use.
However, there are niches for PV pumping with such applications as micro irrigation for
vegetable gardens and plots of less than 1 hectare. As a general rule solar pumping
systems for irrigation are most cost effective compared to diesel pumps where the peak
daily water requirements (volume-head product) are less than about 250 m4 and where
insolation is greater than 4 kWh/m2/day.
Consider solar water pumping if the following conditions are met:
Volume-Head
Product
(m4)
Average Daily
Solar Insolation
(kWh/m2/day)
Irrigation
17-5
Livestock Watering
Livestock watering using photovoltaics has been a high value application in developed
countries. The cost of maintaining utility lines to pumps in remote areas is often greater
than the total revenue for the power to run the pump. As a result there is a movement
in the US utility industry to offer photovoltaic pumps as a substitute for extending grid
power to such small and infrequently used loads.
Systems can be installed in remote ranch areas to provide a reliable water supply with
little maintenance and noise. Water can be pumped day and night with the use of
batteries. Or systems can be designed to work without batteries and accumulate the
water pumped during daylight in storage tanks, which then can deliver water day or
night.
Interesting applications such as pond aeration are also possible. Modules operate
pumps that bubble water into ponds to mix up temperature gradients and prevent or
postpone freezing.
Residential Needs
Homes have a variety of pumping needs including submersible pumps for wells,
pumping water to pressure lines, and water for irrigation and landscape maintenance.
A variety of specialized DC pumps for use with photovoltaics have been developed for
this purpose.
PV pumping for homes is typically a part of a larger residential solar electric power
system. In many cases standard AC pumps are used because they use the AC power
available from the inverter in the system.
17-6
Typical Human Water Requirements in Developing Areas (per person per day)
Region
Africa
Southeast Asia
Western Pacific
Eastern Mediterranean
Europe(Algeria, Morocco,
Turkey)
Latin America, Caribbean
World Average
Gallons
(Liters)
4 - 10
8 - 20
8 - 25
10 - 22
5 - 17
15 - 35
30 - 70
30 - 95
40 - 85
20 - 65
20 - 50
70 - 190
10 - 24
35 - 90
Gallons
Liters
12
15
40
12
4
2
13
4
8
45
57
150
45
15
10
50
15
30
17-7
Gallons
Liters
Developing Areas
Survival
Public Hydrant
Single Home Faucet
Multiple Connections
w/bath, toilet, sink
0.6
11
11-16
45-65
2
40
40-60
80-240
Industrialized Areas
Personal (including
cooking)
Shower
Bath
Sink
Washing Machine
Dish Washer
10
38
25-60
30-35
1-2
30-50
10-20
95-225
115-135
4-8
115-190
38-75
17-8
Battery Powered Or
Direct Couple To Array
There are basically two types of PV powered water delivery systems: (1) pump and
motor directly connected to an array (or using a maximum power tracking circuit); and
(2) motor connected to battery bank charged by a solar array.
The advantage of direct coupled water delivery systems is the underlying simplicity of
the design. There is more water pumped during summer periods of high insolation,
which is usually when more water is required, especially for livestock watering and
irrigation. During periods of low insolation there is less water pumped, but there is less
evaporation as well. The system has few components to fail and the initial cost can be
kept low. As photovoltaic and maximum power tracking technologies have become
more efficient and reliable over time the trend worldwide favors installing direct coupled
pumping systems over systems using batteries.
Battery Power or
Direct Coupled to Array?
Advantages
Disadvantages
Direct
Coupled
to Array
Simplicity
Reliability
Low maintenance
Low initial cost
Match to season
Low efficiency if no
MPT installed
Requires storage tanks
for 24-hour availability
Battery
Powered
Predictable supply
Higher efficiency
Supply starting surge
current
Maintenance
Complexity
High costs over time
Charge control failure
17-9
One of the disadvantages of direct coupled systems is the inefficiency of the match
between a DC motor and a PV array during periods of low irradiance. The array can be
designed to have the motor operating at the array Vmp during the peak hours of the
day, as discussed previously in the chapter on Output Curves. This means that each
module operates near 15 volts instead of being held back to around 12 for battery
loads. But during the morning and evening hours or days of low insolation the array will
operate the motor far below its optimum voltage and away from the array Vmp. The
array may be producing power, but it is not being efficiently passed to the load. The
usual solution for this problem is to include a maximum power tracking device between
the array and DC motor. It operates the array at its maximum power voltage and
converts the power into the most useful voltage and current for the motor. Of course it
adds somewhat to the initial cost, has some inherent inefficiency (usually 95%), and
adds some complexity back to the system. But batteries and charge controllers are still
avoided. And the overall efficiency of batteries (about 80% overall) is still lower than
the efficiency passed through a typical maximum power tracker.
Another disadvantage of direct coupled systems is the need for some form of water
storage if water is required at night or during low insolation days. This means
effectively higher initial costs, and perhaps some maintenance costs associated with
maintaining the tank and its support structure. But it can be argued that it is more
effective to store water than to store electricity in batteries.
The advantage of battery coupled systems is just the opposite. All the array energy is
passed to the batteries, and can be extracted during day or night. The motor can be
operated at its optimum voltage all the time, not just the peak hours of the day, and
even on overcast days, thus prolonging motor life and improving system efficiency. And
another key feature of battery coupled systems is the ability to supply a high surge
current to start difficult motors or counter a high torque portion of a pumping cycle.
The key disadvantage of battery coupled systems is that batteries are involved at all. If
they are treated well they can last for many trouble-free years. However they are
adversely affected by high temperatures, will require cleaning of the terminals and
water level maintenance, age with time, and can sulfate, corrode, and fail. Some form
of charge control is required to prevent overcharge, and a motor controller is necessary
to tell the motor when to operate and when to shut off. This all adds up to a more
complex system, with an associated higher probability of some sort of failure.
Which type of system is best for a particular situation depends on comparisons of initial
cost, costs over time, reliability, efficiency and availability.
17-10
Terminology
The most critical pieces of load information for doing water system sizing are the total
vertical lift, or head, and the volume of water per day required. Other details will help
to fine tune the design, but these two pieces of information are critical.
Flow: The rate at which water is delivered by the pump, usually measured in
gallons/minute or liters/second.
Volume: The total amount of water needed daily. Usually given in gallons/day or cubic
meters/day.
Suction Head: Vertical distance from surface of water to center of pump when pump is
located above water. There is no suction head for a submerged pump.
Discharge Head: Vertical distance from center of pump to surface of storage tank
water or point of free discharge.
Static Head: Vertical distance from surface of water to surface of storage tank water or
point of free discharge.
Static Head
Suction Head
Discharge Head
Pressure Head: If a final discharge pressure is desired, the pump must be able to
supply the flow with the needed energy. The Pressure Head is the equivalent feet of
head the pump needs to be able to pump to supply this final pressure.
Pressure (psi)
2.31
Friction Head: There is a loss of energy as water moves through a pipe. The smaller
the pipe diameter and the faster the flow, the greater the loss. The pump must be able
to supply enough energy to overcome these losses.
17-11
The Friction Head is the equivalent distance the pump must be able to push water to
have enough energy to overcome these losses. Tables of friction loss coefficients for
different size pipes, different types of pipe, and for fittings, are presented in the
Appendix. Factors are available for fittings, elbows, as well as for different pipe types
(steel, plastic, copper) and for different pipe diameters.
Draw Down: When the pump draws water from the well, the level of the surface may
drop depending on the ability of the surrounding earth to replenish the well. The draw
down is the distance from the surface of the water when it is being pumped to the
surface level when it is static. This can amount to many tens of feet, and depends on
rate of pumping. The higher the rate, the greater the draw down.
friction head
Hp
pressure head
(2.3 X psi.)
surface
mounted
pump/motor
Hd
discharge head
Hs
suction head
Hdr
drawdown
Total Dynamic Head: The final total head the pump must be able to deliver at the
desired rate of pumping, including all the previous heads.
17-12
17-13
1/2
3/4
1 1/4
1 1/2
90o Ell
1.5
2.0
2.7
3.5
4.3
5.5
40o Ell
0.8
1.0
1.3
1.7
2.0
2.5
1.0
1.4
1.7
2.3
2.7
3.5
3.6
5.0
6.0
8.3
10.0
13.0
Tee--Straight Run
3.3
4.5
5.7
7.6
9.0
12.0
17.0
22.0
27.0
36.0
43.0
55.0
8.4
12.0
15.0
18.0
22.0
28.0
Gate Valve--Open
0.4
0.5
0.6
0.8
1.0
1.2
11
13
12
14
19
GPM
3/8
1/2
3/4
1
4.25
1.38
.356
.11
2
15.13
4.83
1.21
.38
3
31.97
9.96
2.51
.77
4
54.97
17.07
4.21
1.30
5
84.41
25.76
6.33
1.92
6
36.34
8.83
2.69
8
63.71
15.18
4.58
10
97.52
25.98
6.88
15
49.68
14.63
20
86.94
25.07
25
38.41
30
35
40
45
50
(Equivalent vertical feet/100 feet of length)
&
17-14
1 1/4
.10
.21
.35
.51
.71
1.19
1.78
3.75
6.39
9.71
13.62
18.17
23.55
29.44
1 1/2
.10
.16
.24
.33
.55
.83
1.74
2.94
4.44
6.26
8.37
10.70
13.46
16.45
.10
.17
.25
.52
.86
1.29
1.81
2.42
3.11
3.84
4.67
GPM
3/8
1/2
3/4
1
4.30
1.86
.26
2
15.00
4.78
1.21
.38
3
31.80
10.00
2.50
.77
4
54.90
17.10
4.21
1.30
5
83.50
25.80
6.32
1.93
6
36.50
8.87
2.68
7
48.70
11.80
3.56
8
62.70
15.00
4.54
9
18.80
5.65
10
23.00
6.86
12
32.60
9.62
15
49.70
14.70
20
86.10
25.10
25
38.60
30
54.60
35
73.40
40
95.00
45
(Equivalent vertical feet/100 feet of length)
' &
GPM
3/8
1/2
3/4
1
6.2
1.8
.39
2
19.6
6.0
1.2
5
30
5.8
1.6
7
53
11.0
3.2
10
19.6
5.3
15
37.0
9.9
18
55.4
16.1
20
18.5
25
27.1
30
39.3
35
48.5
40
45
50
(Equivalent vertical feet/100 feet of length)
( &
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
17-15
1 1/4
1 1/2
.34
.51
.70
.93
1.18
1.46
1.77
2.48
3.74
6.34
9.65
13.60
18.20
23.50
30.70
.24
.33
.44
.56
.69
.83
1.16
1.75
2.94
4.48
6.26
8.37
10.79
13.45
.10
.13
.17
.21
.25
.34
.52
.87
1.30
1.82
2.42
3.10
3.85
1 1/4
1 1/2
2.1
3.2
3.9
5.3
7.6
10.2
13.2
16.2
19.4
1.5
2.1
2.8
3.5
4.2
5.1
2.2
3.9
6.2
6.9
10.4
14.3
18.7
25.4
30.0
39.3
Example:
A water pumping system is proposed to supply water to a storage tank
near a village. The piping used throughout is to be 1-inch diameter and the flow rate
will be about 10 gpm. It will draw from a nearby stream, so assume no drawdown.
The system uses a surface mounted centrifugal pump, so water is drawn up to the
pump and discharged above the pump. The total vertical lift will be 150 feet, from the
top of the stream to the top of the water in the tank (5 feet suction to pump, 145 feet
discharge from pump to water in tank).
There will be the following components:
370 feet of 1 inch plastic pipe; a check valve (spring type); three 90 deg. elbow joints;
two 45 deg. elbow joints/
Calculating the equivalent feet for the fittings (using Table 17-6):
90 deg. elbows:
3 X 2.7 feet
45 deg. elbows:
2 X 1.3 feet
spring check valve:
=
=
8.1 feet
2.6 feet
8 feet
18.7 feet
388.7 feet
26.7 feet
145
17-16
+ friction head
+
26.7
176.7 feet
17-17
Motor Type
Brushed DC
The traditional DC
motor in which
brushes conduct
current into the
spinning portion of
the motor.
Permanent magnets
produce a magnetic
field inside the motor
shell
Brushless DC
Advantages
High technology
motor using a
complex electronic
system to precisely
alternate the current
causing the rotor to
spin
Disadvantages
Simplest and
most efficient
motor for use
with PV systems
No complex
control circuitry.
Motors start
without a great
current surge and
will run slowly but
not overheat with
reduced voltage.
Efficient
No maintenance
required
AC Motors
Inductive motors
No brushes to
replace
Can use existing
AC motor/pump
technology, which
is cheaper and
available
worldwide. Can
handler larger
pumping
requirements.
Special Features
Brushes need
replacing
periodically
(Typical
replacement
interval, 2000 to
4000 hr. or 2
years)
Electronic
computation adds
extra expenses,
complexity,
increasing risk of
failure
In most cases, oil
cooled, cant be
submerged as
deep as water
cooled AC
Requires an
inverter to
change DC into
AC adding cost
and complexity
Less efficient
than DC units
Prone to
overheating if
current is not
sufficient to start
motor, or if
voltage is too low
Requires
maximum power
tracker for
optimum
performance
Only available in
small motor sizes
Increasing
current (by
paralleling
modules)
increases torque,
Increasing
voltage (by series
modules)
increases speed
Growing trend
among PV pump
manufacturers to
use brushless
DC, primarily for
centrifugal type
submersibles
Can be single or
three phase
Inverters are
designed to vary
frequency to
maximize power
to the motor in
response to
changing light
levels
17-18
Pump Types
Centrifugal
Surface
Centrifugal
Submersible
Straight or
Self-Priming
Submersible
Jet Pump
Positive
Displacement
Surface
Positive
Displacement
Submersible
Helical
Cavity
Diaphragm
Jack Pump
Vertical
Turbine
,
,
17-19
Straight Centrifugal
Installed above the water, it has a single impeller. Suction head is limited by
atmospheric pressure to approximately 20 feet, but can push above pump (discharge
head) hundreds of feet. Must be "primed" (filled with water) before each start-up.
Commonly used for surface water pumping, from streams, lakes, or to move water
along the land through pipelines.
17-20
Self-Priming Centrifugal
Has a chamber above impeller which holds water after shut-down so the pump is
"primed" before each start-up.
Jet Pump
Centrifugal with a portion of the flow returning to a venturi on the input side. This
increases suction head to 150 feet but with a reduction in net flow. The venturi can be
placed just in front of the impeller chamber, or all the way down at the input of the
suction pipe. Low cost solution for relatively low flow rates and heads. Low efficiency
however.
Vertical Turbine
Series of impellers in one long narrow cylindrical casing that is submerged below water
level, connected to the motor on the surface by a long drive shaft. Allows for deep
pumping at high rates by using huge motors above ground. Heads limited by shaft
length. Efficiency is reduced due to twisting, friction, vibration, and weight of shaft and
bearings. Typically used for large-scale irrigation with large AC or diesel motors.
17-21
17-22
Helical Cavity: Helical rod rotates inside a cavity of slightly different pitch, creating a
vertically moving cavity. Submerged under water and connected to motor above
ground by long drive shaft. Motor power is lost in friction, weight, vibration and twisting
of drive shaft and bearings. Requires high torque, may necessitate batteries or
maximum power tracker to supply surge to overcome tremendous friction. Can move
very gritty water, whereas high-speed centrifugals would be eroded quickly. Suitable for
low flow rates and moderate depths.
17-23
Jack Pump: Motor pulls on oscillating jack above ground which pulls on a long drive
shaft connected to a plunger with flapper valve below water level. Each cycle of the
jack moves a volume of water into the drop pipe, and displaces water previously
pushed into the pipe. Low flow rates achieved but great depths possible. Flapper valve
needs periodic replacement. Above ground mounting means can use wide variety of
either AC or DC motors.
A variation of the jack pump design has motor and small jack mechanism packaged in
narrow cylindrical casing like submersible. Pump and motor can be submersed below
water level.
17-24
17-25
Pump
Self-Priming
Surface
Centrifugal
Use
Surface mounted
motor and pump for
high volume, low
head applications
commonly used for
irrigation from rivers
Advantage
Disadvantage
Ease of installation,
easily serviced, wide
range capabilities,
many pumps use
water as a bearing
lubricant
Vertical
Turbine
Jet
Medium-head,
medium -volume
pump
Submersible
Centrifugal
Ease of
maintenance with
surface mounted
pump
Self priming
Wide range of
capabilities
Good sand/tilt
tolerance
Low equipment and
maintenance costs
Can be used beyond
suction limit
Very reliable
Easy access to
motor and pump
Least expensive
intermediate-head
pump
Adaptable to very
small wells (50 mm)
Motor is directly
coupled to impellers
Easily cooled
because it is
submersed
Multistage impellers
accommodate a
wide range of heads
No noise, no
pumphouse
Use water as
bearing
Can pump from
great depth
Limited 5 meter
suction head
Relatively inefficient
compared to
centrifugal s which
have flooded inlets
Have a narrow
efficiency range with
respect to head
In abrasive
conditions, impeller
wear can lead to
reduced efficiency
Shaft losses reduce
efficiency compared
to submersibles
Shaft and borehole
alignment are critical
to successful
operation
Hard to install and
maintain
Relatively inefficient
compared to other
types of pumps
Sand/grit causes
impeller wear and
decreasing efficiency
Saline conditions will
corrode metal
housing
High capital cost,
highly expensive to
repair (must remove
pump from well)
Can be damaged by
running dry
.$$
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
17-26
Pump
Surface
Positive
Displacement
- Rotary or
Mono pumps
(Helical
Cavity)
Surface
PositiveDisplacement
Reciprocation
Piston
(Jack or
Nodding
Donkey
Use
High-head, low-flow,
down hole piston
and cylinder driven
by sucker rod from
surface
Advantage
Disadvantage
Very robust
Efficient over wide
range of head except
for under 20 meters
Simple construction
Self priming
No back-flow valve
required
Submersible
Positive
Displacement
(Diaphragm)
Adapted from Pump Selection, Water and Sanitation for Health Project, US AID p. 42
17-27
Positive Displacement
Submersible
10
Surface Suction
Centrifugal
Handpump
1
1
10
100
17-28
Positive Displacement
Volumetric movement
Large volumes
Lower volumes
Moderate head
High head
Unaffected by grit
17-29
Exercise
Model
Type
Head
17-30
Flow Rate
Power Needed
Primary Requirements
For Choosing A Pump
The first requirement that any pump must satisfy is the total head. If the head cannot
be reached then no water is pumped at all. If possible the total dynamic head should
be know including friction losses in the pipes and any drawdown that will occur. Often
all this information is not available. In that case certainly at least the suction and
discharge head must be known.
The second requirement is that the pump be able to deliver the required volume of
water in a typical day to the specified head. Direct coupled solar pumping systems will
operate only during daylight, and the output will vary with the daily insolation. A simple
approach to choosing a pump begins with estimating the amount of peak sun hours
available at the site, and dividing this into the total daily volume needed. This gives the
rate of pumping during peak sun hours that the pump must perform. Choose a pump
that can operate at the required head and that can output at this rate. The power
needed by the pump to operate at this rate and this head will then determine the size of
the solar array.
(1)
(2)
kwh / m2
1.0
Langley (cal/cm2)
0.01162
MJ / m
0.2777
Btu / ft2
0.003155
17-31
If one pump is not adequate to deliver the total volume perhaps the manufacturer offers
a way to combine pumps together. An example of a system by Grundfos is shown on
the next page that allows up to three pumps to be operated in the same borehole,
coupled either horizontally for large boreholes or coupled vertically for smaller borehole
diameters.
Exercise
!" #$
&""" #$
!"(""" #$
% ##$
&" '$
&& % ##$
17-32
Supplementary Considerations
For Choosing A Pump
Site, Well, and Water Considerations
The diameter of the borehole must be known because it can restrict your choices for
pump models. Typically submersible pumps are available in 4 inch or 6 inch diameters
or in larger sizes. A system designer who chooses a large pump on the basis of
meeting the volume demand must also confirm that the pump will indeed fit in the well
in the field!
The flow capacity of the well should be known so that the pump does not over pump
the well. If the rate of pumping far excess the recovery rate of the well from the
surrounding ground then the water level will continually drop during pumping, and the
pump could be exposed to air and pump in dry conditions. This might harm the pump.
If more volume is needed than can be delivered by a direct coupled pumping system
operating quickly during sunlight hours, then perhaps a battery coupled system should
be considered. In such a design a slower pump rate could be sustained 24 hours a day
perhaps staying under the natural recovery rate of the well.
Abrasive particles in the water are another consideration in choosing a pump type.
Very gritty water can cause high-speed impellers in centrifugal types to erode quickly.
Positive displacement type pumps should be considered because they are less affected
by grit. Or perhaps manufacturers of centrifugal pumps could supply an appropriate
filter to reduce the grit effect. Such filters will probably reduce output performance to
some degree.
17-33
Electrical Considerations
One electrical consideration is whether the system should be direct coupled to modules
or battery coupled. The usual trade-offs are simplicity, reliability, low maintenance, and
lower cost for a direct coupled system against 24-hour availability of water and higher
motor operating efficiency with batteries.
Another choice is whether a fixed array mount should be used or a sun position tracker.
Fixed structures are less expensive and can withstand extremely high winds. But
trackers can increase water output by 50-60% during summer months. The reduction in
the number of modules needed could equal or exceed the cost of the tracker.
If a pump model is chosen that uses an AC motor then an inverter will be required.
Most common commercial inverters require a battery bank for stable and narrow
voltage window. (Grundfos by way of contrast makes a special inverter designed to
operate directly connected to the solar array).
If a pump with a DC motor is chosen then the consideration is whether to use a brushtype commutated motor or a brushless electronically commutated design. Brushes will
have to be serviced periodically while electronic commutation circuitry could increase
the complexity and decrease the reliability of the overall system. Details of the motor
and control design must be considered.
17-34
17-35
Example:
500
A.Y. McDonald
211012DK
450
400
Power needed
at pump
Head (feet)
350
300
90 V, 1190 W
250
75 V, 880 W
200
66 V, 680 W
150
48 V, 350 W
100
34 V, 170 W
50
0
0
100
200
300 400
Flow (gph)
500
600
700
17-36
A.Y.McDonald
211009DP
350
300
Power needed
at pump
Head (feet)
250
200
90 V, 1190 W
75 V, 880 W
150
66 V, 680 W
100
48 V, 350 W
50
34 V, 160 W
0
0
200 400
The 211009DP model of pump has similar shaped curves but different values. This
model of pump cannot achieve the same total head, but can pump more volume.
Centrifugal pump curves always curve slowly downward, as seen in the A.Y. McDonald
curves. At lower head the same pump/motor will deliver much higher flow. Or
conversely, as head is increased the flow rate decreases drastically. This effect of
decreasing flow with increasing head is partially because at higher head. More of the
spinning impeller's energy is used to just hold up the column of water above it. Less
energy is available for moving the column. So flow rate goes down as head goes up.
17-37
Positive displacement type pumps show little decrease in flow rate with increasing head
compared to centrifugal pumps. This is because the column of water is held up by a
mechanical design of the pump (a flapper valve, diaphragm, or progressing screw, for
example) so little more energy is needed to hold up a large vertical column of water
than a small one. But positive displacement types tend to be able to pump less volume
than centrifugals types.
An interesting shape of curve occurs for a jack-type positive displacement pump. At
high head the nature of the curve is similar to positive displacement pumps. That is,
there is little effect of increasing or decreasing head on flow rate. But at low heads the
jack-type pump seems to be able to increase the flow more than the progressing cavity
types. The curves for the Solarjack SJA series of jack type pumps are shown as an
example. The power values given are power required at the pump.
900
800
700
Head (feet
600
500
400
300
200
100
0
0
2.5
7.5
10
12.5
15
Flow (gpm )
470 w atts
660 w atts
17-38
Pump Efficiency
Manufacturers may offer a pump model that can deliver to a required head at different
flow rates depending on the voltage or array power delivered. After a designer has
selected various pumps that can deliver to the specified head they must choose the
best pump among the choices. Perhaps the primary factor affecting this choice is the
efficiency of the pump at that particular head. Choosing a pump that operates at a
higher efficiency will mean that fewer modules will be required, so initial cost is lower.
The efficiency of a pump (including the motor) has a maximum at some value of head
and flow and is less anywhere away from that point. The efficiency is determined by
dividing the output power of the flow by the input electrical power to the motor.
Efficiency of Pump/Motor =
Ideal efficiencies of between 25% and 60% are possible although real world
performances will probably be limited to 40-50%. Added losses are due to heating in
the motor and friction and twisting losses in shaft type systems.
The formula for calculating efficiency divides the theoretical power needed to move
water by the actual power specified for the pump by the manufacturer.
Efficiency of Pump/Motor
Fc
Head x Flow x Fc
Electrical Power
if units are:
0.003
0.189
You may have determined that a number of manufacturers pumps will reach your
required head and also pump enough water. Use this formula to calculate the
efficiency at the required head for each of the pumps. Compare the efficiencies and
the flow rates or volumes, and choose the pump that has the highest efficiency, unless
other factors override this choice. You will then be delivering water to your head
requirement with the fewest modules per volume.
17-39
Example:
=
Efficiency @ 100 ft. =
=
Efficiency @ 50 ft. =
=
Efficiency @ 200 ft. =
=
Efficiency @ 150 ft. =
=
Efficiency @ 100 ft. =
17-40
=
Efficiency @ 100 ft. =
=
Efficiency @ 50 ft. =
So the DP choice is more efficient than the DK pump, and both pumps show the
characteristic peaking of efficiency in the midrange of the pump head range.
The are both about the same efficiency at 150 feet, showing that different pumps may
be similar in efficiency at particular operating points, but that over their range one model
is more efficient than another.
It is this type of analysis that needs to be done to decide which pump is the best choice
for a particular head and flow rate. Of course other must also be weighed, and pump
efficiency may be overridden if these other factors indicate a more cost-effective
solution.
17-41
Exercise
$
) *+ ,- !!"!-.
/
0% /1*
2""
)
3"4
-. "4
/1*
,- !!"!-.
5555555555
5555555555
5555555555 6
$
7 /
0% /1*
5555555555
5555555555
5555555555 6
$
8
$
9 $ : $9
!""
#
$%""
&'()*
)+,
+-!. /%"
&'(0 1
1
2222222222 3
4
17-42
17-43
Step
Action
Visit the site; determine the rough layout of the system and the
likely water source.
2
Obtain information on the well and water resources (i.e. well
diameters, depths, yields, drawdowns, seasonal river levels, and
monthly groundwater information). Consult with local authorities
and private welldriller who have knowledge of the area. (See
Supplementary Considerations for Choosing a Pump).
3
Estimate the likely daily demand for water for the applications
for each month. Be sure to consider water supply needs in the
future in for example a growing village population.
4
Gather monthly data on the solar insolation or peak hours
(convert from kWh/m2/day, MJ/m2, Langleys, etc.)
5
Calculate an estimate for peak flow rate, based on the average
daily insolation available.
6
Determine the total dynamic head, including drawdown and
friction losses if possible.
7
Choose a pump type (Positive Displacement, Submersible,
Floating or Surface Suction) that best fits the head, volume
requirements. (Recall the discussion previously on Optimum
Operation Ranges for Different Types of Pumps).
8
Choose motor/pump systems that will deliver to your required
head, and that can meet or exceed your flow requirement. (Use
manufacturers literature and help from your pump supplier).
9
Calculate the efficiency of the pump(s) at your required head.
Choose the pump with the highest efficiency that can deliver your
required flow rate.
10
Estimate the array size for each month, using the Array Size
formula presented next for this calculation, based on pump
efficiency and available peak hours of insolation.
11
Choose the largest array size as calculated above.
12
Compare manufacturers best case performance with your array
estimation. Adjust system size as necessary. Be conservative.
13
Reevaluate your decision. If you havent used the pump before,
get feedback from other users. Make sure that it fits other
considerations (i.e. computability with the water quality, ease of
maintenance, availability of support, etc.)
14
Complete the system design. (See Component Design
Considerations.
15
Procure and install system. (See System Installation).
,
17-44
H x V X Fc
P.H. x Fm x Ft x eff.
where:
H
V
Fc
=
=
=
P.H.
Fm
Ft
eff.
# Series Modules
# Parallel Modules =
17-45
Flow Rate
V
P.H.
kWh / m2
Langley (cal/cm2)
MJ / m2
Btu / ft2
1.0
0.01162
0.2777
0.003155
Pump/Motor Efficiency
Use the following formula for estimating the efficiency of the pump at the required head.
Efficiency
H x Flow Rate x Fc
Pump Electrical Power Requirement
17-46
Example:
30 m
7m
500 m
1.5 inch
PVC pipe
Step (3)
Step (4)
Local weather data is provided. The best month is June with 9.1
kwh/m2, and the lowest month is January with 5.7 kwh/m2. Use the
lowest month to be conservative. Convert to 5.7 peak hours.
Step (5)
Step (6)
drawdown + discharge
7 m + 30 m
37 m
=
=
500 m + 1.3 m
501.3 m
17-47
Friction Head
Step (7)
Step (8)
Step (9)
Efficiency
0.41 or 41%
1100 watts
75 nominal volts
15 volts/module
5 modules in series
# Parallel Modules =
Array design:
1100 watts
5 series modules X 75 Wp
2.9 modules
Round up to 3 modules.
17-48
Exercise
*
$ ;
""" #$ !!"
;
& %
#$
<
$
(
) =&4
$
) ,- -. -7
)
$
$( $
$
$
5555555555
) -.
5555555555
5555555555
5555555555
5555555555
5555555555 $
5555555555
) -7
5555555555
5555555555
5555555555
5555555555
5555555555 $
5555555555
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
17-49
Array
An important choice is whether a fixed array mount should be used or a sun position
tracker. Fixed structures are less expensive and can withstand extremely high winds.
But trackers can increase water output by 30 to 40% during summer months. The
amount gained depends upon your insolation levels, the time year, and your latitude.
Using a tracker could reduce in the number of modules needed, which could then equal
or exceed the cost of the tracker. Trackers, though generally very reliable, do add
complexity to the system and may need to be serviced overtime.
Inverters
There are a variety of PV pump manufacturers who have developed PV pumping
systems using inverters to drive single or three phase AC motors in direct coupled
systems. These inverters differ from those used for battery systems in that they not
only convert DC to AC but include circuitry that involves some form of impedance
matching as well. Typically these are variable frequency inverters that control the
frequency and hence voltage of an AC pump to best match the output of a PV system.
At this stage they have proved very reliable and efficient, though they add additional
cost to the overall systems as compared to a direct coupled DC motor/pump.
If a standard AC off-the-shelf pump is specified, then it is necessary to have a standard
commercial inverter with a battery bank. Though less efficient, because of their fixed
voltage operation, designer can optimize the motor/pump subsystem for maximum
efficiency.
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
17-50
17-51
17-52
17-53
Current
25
20
Direct
Couple
6836 gpd
15
10
5
0
0
10
20
30
40
50
60
70
Voltage
25
Current
Direct
Couple
9322 gpd
Noon
+ 673 watts
30
20
15
10
5
0
0
10
20
30
40
50
60
70
Voltage
17-54
17-55
MPT
10,055 gpd
noon
760 watts
20
15
10
5
0
0
10
20
30
40
50
60
70
Voltage
noon
760 watts
30
Current
25
MPT
10,055 gpd
20
15
10
5
0
0
10
20
30
40
50
60
70
Voltage
'
17-56
17-57
Current
25
20
15
10
Tracking
Direct
Couple
10,534 gpd
5
0
0
10
20
30
40
50
60
70
Voltage
(
+ Noon
+ 673 watts
+
25
Current
Tracking
Direct
Couple
13,745 gpd
20
15
10
5
0
0
10
20
30
40
50
60
70
Voltage
)
17-58
(End of Chapter)
17-59
CHAPTER SEVENTEEN
WATER DELIVERY SYSTEMS
17-1
17-2
17-2
17-3
17-4
17-4
17-5
17-6
17-6
17-7
17-9
Terminology
Calculating Total Head
17-11
17-13
17-17
17-19
17-20
17-22
17-23
17-25
17-31
17-33
17-33
17-34
17-35
17-35
17-60
Pump Efficiency
17-39
17-43
17-45
17-50
17-52
17-53
17-53
17-55
17-57
17-61
Chapter 17 Answers
Water Delivery Systems
Refer to manufacturer's literature.
a. Required volume = 10 m3/day. Insolation = 6 kwh/m2/day.
Peak sun hours = Insolation X Conversion Factor = 6 kwh/m2/day X 1 = 6 peak hours.
Required Flow Rate =
Convert to gallons:
Daily volume
Peak Sun Hours
10 m3/day
6 peak sun hours
1.67 m3/hour
441.2 gallons/hour
5000 gallons/day
5.23 peak sun hours
956 gallons/hour
956 gallons/hour
264.2 gallons/m3
3.62 m3/hour
Convert to m3/hour:
10,000 gallons/day
5.5 peak sun hours
1,818 gallons/hour
Convert to m3/hour:
1,818 gallons/hour
264.2 gallons/m3
6.88 m3/hour
17-1
Referring to the information for the A.Y. McDonald pump in Figure 17-12, we find the
curve for 680 Watts. At a head of 300 feet, the output is about 50 gph. This is equal to
50 gph 60 = .83 gallons per minute (gpm). We can calculate the efficiency as:
Efficiency of Pump Motor =
Head X Flow X Fc
Electrical Power
.069 = 6.9%
Looking at figure 17-14 for the Solarjack SJA pump, we look at the curve for 660 Watts
at 300 feet of head. The output is about 6.25 gpm. This is equal to 6.25 gpm X 60 =
375 gph. The efficiency is then:
Efficiency of Pump Motor =
78 Watts.
Head X Flow X Fc
Electrical Power
.417 = 41.7%
17-2
The estimated insolation is 6.5 peak sun hours. The farmer's required flow rate is then:
Required Flow Rate =
4000 gallons/day
6.5 peak sun hrs
615 gallons/hr
Array Wp
0.26 or 26%
75 nominal volts
15 volts/module
5 modules in series
# Parallel Modules =
=
Total array:
1028 watts
5 series modules X 75 Wp
2.74 modules
rounded to 3 modules
The actual flow rate will be 700 gph X 6.5 peak hours = 4550 gallons/day. The output
per module is:
Output per module =
4550 gallons
15 modules
17-3
303 gallons/module
Array Wp
0.41 or 41%
66 nominal volts
15 volts/module
# Parallel Modules =
=
Total array:
rounded to 5 modules
652 watts
5 series modules X 75 Wp
1.73 modules
rounded to 2 modules
The actual flow rate will be 850 gph X 6.5 peak hours = 5525 gallons/day. The output
per module is:
Output per module =
5525 gallons
10 modules
553 gallons/module
Clearly, the DP pump is much better suited for this application. At the required head /
flow combination, the efficiency of this pump is much higher than the DK pump. This
shows the importance of selecting the correct pump for the specific project
requirements.
17-4
Chapter Eighteen
Hybrid Systems
Multiple Energy Sources - Flexible
Approaches
Hybrid power systems include a combination of energy sources. The energy
sources might be solar arrays, wind turbines, diesel generators, biomass plants, or
others. Together, these sources provide the total energy required by the load.
These training materials consider mainly photovoltaic - diesel hybrids, although other
types of systems are briefly discussed.
The general purpose of the hybrid system is to get the best benefits of each
technology while reducing some of the disadvantages. A pure stand-alone PV
system, for example, requires very little maintenance but depends on the amount of
sun light for all its energy. On the other hand, a diesel generator can provide a
dependable amount of energy whenever it is required, but it requires fuel and
frequent maintenance. A hybrid system using both a PV array and diesel engine
can provide energy regardless of changing weather patterns with less maintenance
and fuel use than a dedicated prime-power engine. This illustrates how the features
and benefits of each energy source are blended in a hybrid system.
Hybrid systems can take many forms, and there is no single design that is best for
every situation. Instead, the hybrid designer must consider the different
technologies and components available and then choose a combination that best fit
the particular requirements. Good hybrid design consists of finding a balance
between many factors.
18-1
Hybrid Systems
18-2
Hybrid Systems
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Average
Horizontal
Insolation
Tilt to Optimize
Worst Case
2.65
3.55
4.82
5.92
6.73
7.41
7.16
6.44
5.44
4.10
2.78
2.31
4.94
5.30
5.65
5.55
5.18
4.85
4.85
4.85
5.07
5.47
5.51
4.87
4.77
5.16
Tilt to Optimize
Annual
Average
4.69
5.36
5.93
6.23
6.41
6.75
6.62
6.42
6.19
5.54
4.47
4.18
5.73
Since stand-alone systems are designed with the worst case in mind, they are
oversized for the rest of the time. This results in extra installed capacity. The
additional energy in the peak periods can not be used and is wasted. In effect, the
whole system performance suffers from the worst month. If the difference is severe
between the worst case and the other months, this can result in wasted energy
equal to 100% or more of the design load.
Continuing the example from above, the system designed for the worst
case month has a worst case month (Dec.) of 4.77 sun-hours. The
best month is 5.65 sun-hours (February). So during February, the
system produces 5.65 / 4.77 = 118% of the energy necessary. If all
the energy over 4.77 sun-hours is excess, then an additional 18% of
the load energy is wasted.
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
18-3
Hybrid Systems
A properly designed hybrid system can make better use of the renewable energy.
This is because the hybrid system can be designed around the average case, not
the worst case period. In general, this results in a smaller renewable energy
component and less wasted power. It is important to remember that the renewable
energy source in a hybrid system is actively supported by another power source.
Exercise
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Average
Horizontal
Insolation
Tilt to Optimize
Worst Case
6.50
5.71
4.49
3.47
2.43
1.84
2.10
2.85
3.71
4.92
6.08
6.34
4.20
4.85
4.94
4.43
4.19
3.55
2.84
3.22
3.83
3.98
4.43
4.76
4.63
4.14
Tilt to Optimize
Annual
Average
6.20
5.91
4.83
4.13
3.21
2.50
2.84
3.59
4.14
5.07
5.91
5.99
4.53
!
"
#
$
b)
!
"
#
$
!
"
#
$
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
18-4
&
Hybrid Systems
Load Matching
Another advantage of hybrid power systems is a better match between power
production and the load. This can be as simple as being able to turn on a power
source for peak loads. A carpentry shop may use a small amount of power during
normal operation, but need more power to run a large lumber saw. It makes sense
to provide the baseline amount of smaller power for most of the day and then run an
on-demand source specifically to power the saw. Similarly, a remote village may
have a large peak demand for lighting and entertainment in the evening hours.
Step Load
12
16
20
24
Time
Sometimes the size of the load indicates that a hybrid might be useful. Large
applications require large current flows and higher voltages. These can be costly to
provide using only solar, for example.
For village or facility power systems, there is often a collection of different types of
load equipment. Different types of equipment may be suited to different sources of
power. Rather than try to supply all the loads with a single power source, a hybrid
may match different resources to each individual load. PV, for example, is very well
suited for water pumping for humans and agriculture. The source of power (the sun)
tends to be most available during hot, dry periods. Other loads may be better suited
for a diesel generator. Equipment that has high start-up surges for example may run
well with a generator. For example, an ice-making unit, which has a large
compressor, has high starting currents that can be supplied by a generator. 3-phase
loads are also typically run from a generator. While such loads can be run from
photovoltaics and appropriate power-conditioning units, it may be simpler to provide
an engine for a specific load.
18-5
Hybrid Systems
18-6
Hybrid Systems
18-7
Hybrid Systems
Better Flexibility for Loads Hybrid systems offer great flexibility in their ability to
meet changing loads. Since the generator can provide additional energy by simply
running longer, it is easier for a hybrid to accept an increase in the loads. Stand
alone PV systems can not handle a significant increase in the load above the design
load, and the only way to allow for growth is to over-size the system for the initial
load. Similarly, a prime power diesel can not easily accept load growth and must
either sized for the actual load or run poorly at low load levels initially.
18-8
Hybrid Systems
18-9
Hybrid Systems
Initial Cost
Recurring
Costs
Maintenance
Fuel Use
Load
Flexibility
System
Complexity
Use of
Renewables
Pollution
Noise
Hybrid
System
High
LowModerate
Moderate
LowModerate
High
Stand-Alone
PV
High
Low
Cycle
Charge
Moderate
Moderate
Prime Power
(Diesel only)
Utility
Low
Low - High
High
Low
Low
N/A
Moderate
Moderate
High
High
Low
N/A
Low
Moderate
Low-High
High
High
Low
Moderate
Low
Low
High
ModerateHigh
Low
Low
N/A
N/A
N/A
High
ModerateHigh
High
High
Low
Low
Moderate
LowModerate
18-10
Hybrid Systems
Photovoltaic Arrays
PV is well suited for use in remote areas, since it is extremely reliable and requires
very little maintenance. Many telecommunications loads use PV as the sole power
source, and hybrid systems are a logical extension of this application of PV. The
difficulties with photovoltaics are the initial cost and the requirement for unobstructed
land to place the array. Nevertheless, the advantages of PV make it one of the
more common renewable energy sources used for hybrid systems.
PV Arrays
Advantages
Very low
maintenance
High reliability
Long product life
Disadvantages
High initial cost
Requires land with
no shading
Seasonal resource
Engine Generators
Engine generators represent the single most common on-demand power source.
They are very common, compact in size and inexpensive to purchase. Their
disadvantages are that they require on-going maintenance and fuel, and produce
noise and pollution.
Engine Generators
Advantages
Disadvantages
Widespread use and Require on-going
support
maintenance and
fuel
Small size
Produce noise and
Low initial cost
pollution
18-11
Hybrid Systems
18-12
Hybrid Systems
energy available to a turbine on a 20-meter tower. The best advice is to monitor the
wind speed at a site for at least one year before making final decisions on WTG
installation. There are a number of excellent wind monitoring kits which can help
you perform this task.
Third, because of the extra energy farther off the ground, it pays to install as tall a
tower as possible. Care must be taken not to shade the PV array with this tower.
Summary of Energy Considerations with Wind Turbine Generators
1. Wind Turbine output depends very heavily on wind speed
2. Wind speeds can vary significantly over small areas
3. Height increases the amount of energy available
In summary, wind turbines can form an excellent generating source in a hybrid
system. However, since the output is so sensitive to the wind speed, and since the
wind speed is so sensitive to the particular climate, it pays to make a careful
resource assessment before committing large amounts of investment in wind
turbines.
Thermo-Electric Generators
Thermo-electric generators (TEGs) operate based on the principle that electricity
can be produced from the temperature difference between dissimilar metals. A fuel,
usually natural gas, is burned in the presence of a pile or stack of dissimilar metals,
producing a continuous DC current and voltage. Commercially available units have
been widely used in the gas pipeline industry, where there is an abundant supply of
cheap gas for fuel.
The basic principles of TEG's result in a number of specific advantages and
disadvantages. TEG's have no moving parts and there is no mechanical wear as
compared to an engine generator. TEG's can be started after a long shutdown,
which makes them suited for seasonal use. The thermoelectric affect works better
and is more efficient in colder weather. A TEG does, however, require fuel and fuel
storage. The efficiency is very low compared to other types of generators, typically
4% or less. The advantage of working well in cold climates becomes a
disadvantage in a hot climate. Finally, typical commercial units are only in the range
of 10 - 1,000 Watt output, limiting their practical use to smaller systems.
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
18-13
Hybrid Systems
Thermoelectric Generators
Advantages
Disadvantages
No moving parts
Requires fuel
Ability to start after
Low efficiency
long shutdowns
Do not work well in
Excellent
hot climates
performance in cold
Limited size range
weather
TEGs can be used in hybrid systems with small PV arrays. The best use for these
hybrids is in climates where the insolation varies dramatically over the course of the
yearsuch as far northern climates. Rather than oversizing the PV array by a factor
of three or more, a TEG can be used to supply some or all of the load during the low
insolation months. Since these months also tend to be the coldest, the TEGs are
most efficient. Finally, the waste heat from the TEGs can be used to supply space
heating for batteries and telecommunications equipment.
Micro-Hydro
Hydroelectric power is generated from the energy of falling water. A number of
small, commercial hydro plants are available in the 3 kW to 50 kW range, which can
be incorporated into a hybrid system. Since the output of a hydro plant is relatively
continuous, it lends itself to an AC bus type of system, with renewable or diesel
generators making up the extra energy required by fluctuating loads or seasonal
changes.
Power produced by a micro-hydro site depends directly on the height difference
(head) from the supply to the turbine, as well as the amount of flow. Since the
amount of water can vary seasonally, proper design of a micro-hydro system often
involves at least a year of data on the water flow.
The economics of this class of system are widely variable and depend mostly on the
annual availability of a stream, and on the ease of building and maintaining the
turbine. Clearing of debris and insuring maintenance are also concerns of
hydroelectric generators.
18-14
Hybrid Systems
Micro-Hydro
Advantages
Disadvantages
Requires no fuel
Depends on specific
sites
No pollution
Requires additional
May provide
civil works at site
economic power
Typically requires
some resource
monitoring
Maintenance
concerns
Utility / Grid
Although systems with a utility supply are sometimes not considered to be true
hybrids, there are times when the overall system behavior resembles a typical hybrid
system. These can be instances when there is utility power present, but it is not very
reliable. In such circumstances the utility is treated more like a variable energy
source rather than an on-demand power source.
Exercises
'
()*
!
*
()
*
18-15
Hybrid Systems
Bus Configurations
One of the basic distinctions in hybrid design is the bus configuration. Broadly
speaking, this is the general architecture of the system. Hybrids are usually
categorized as DC bus systems or AC bus systems.
DC Bus Hybrids
In a DC bus system, a generator / rectifier combination charges the batteries. The
center of the system is considered to be the common DC bus where the battery,
rectifier and PV array are all connected together. This is the most common hybrid
system configuration. This system was developed from the idea of cycle charged
diesel generators, and it is ideal for telecom hybrid systems, since the loads are
typically DC. Note that a DC bus system can supply AC loads by adding a separate
inverter to provide AC power.
The PV array charges the battery based on the available solar energy. The DEG is
normally only started when the batteries reach a low voltage state and need to be
recharged. The DEG is connected to a dedicated battery charger that supplies DC
current. Sometimes there may be an emergency transfer switch so that the
generator can power an AC load directly. Under normal conditions, however, the
engine only runs to charge the battery and to provide the additional energy needed
for restore the battery to full charge.
DEG
Rectifier
AC
PV
Array
DC
(+)
(-)
Battery
Bank
DC
Loads
18-16
Hybrid Systems
DEG
Rectifier
AC
PV
Array
Transfer
Switch
DC
DC
AC
(+)
(-)
Battery
Bank
Inverter
AC
Loads
DC
Loads
!
"
AC Bus Hybrids
AC bus hybrids utilize the generator to power some or all of the loads. The batteries
are charged using a bi-directional power conditioning unit that serves as both
inverter and rectifier. The focus of the system is considered to be the common AC
bus where the generator, inverter and AC Loads are all connected together. Since
the generator takes a much more active role in the system, these are also
sometimes called interactive hybrids. An AC bus hybrid may have some DC loads,
but it is more common to have only AC loads.
The PV array still supplies energy to the battery depending on the solar insolation.
The generator however, can now charge the batteries and supply power to the load.
The bi-directional inverter can take power from the battery bank to supply AC power,
or can draw AC power from the generator to charge the batteries. Some bidirectional units are capable of operating in synchronized or parallel mode where
both the inverter and generator can supply the load together. This allows for great
flexibility in handling peak loads.
18-17
Hybrid Systems
DEG
AC
Loads
Inverter/
Rectifier
AC
PV
Array
DC
(+)
(-)
Battery
Bank
DC
Loads
"
Exercises
#
1
2
"
$
0%
-
-
'
18-18
Hybrid Systems
Hybrid Operation
Daily Operation
A hybrid system will go through shorter and more intense cycles than a stand-alone
system. While a stand-alone system may remain at almost a full state of charge for
long periods of time, most hybrid systems will cycle through the battery every couple
of days. Rapid changes in the battery result caused by a sequence of array and
rectifier charging.
The figure below illustrates a sample hybrid system's performance over 4 days. The
PV array and rectifier both contribute about half the total energy in the system. The
battery bank provides about 1.5 days of storage. This results in a system with rapid
cycles and heavy utilization of the battery. Three lines are shown, one depicting the
average battery state of charge, one line showing the energy provided by the PV
array, and one line showing the rectifier contribution. For this example, the load is
assumed to be constant.
D ay 1
D ay 2
D ay 3
D ay 4
B a tte ry
P V A rra y
R e c tife r
%&'()
The figure shows the battery initially at nearly full charge. During the first day, the
battery state of charge falls as the load draws power. The PV array replaces some,
but not all of the energy used by the load. On the Day 2, the load again draws down
the battery, with some of the loss being replaced by the PV. At the end of the
second day the battery is discharged sufficiently to bring the rectifier on line. The
rectifier provides a large amount of current in a short interval of time, bringing the
battery back nearly to full charge. As the rectifier shuts down additional PV energy
18-19
Hybrid Systems
finishes the charging. At the end of the third day, the load keeps drawing power
from the battery. On the fourth day, the load continues to draw down the battery
with some replacement from the PV array. This is very similar to Day 1 and we can
see that the system will continue to run in cycles with the battery eventually reaching
a low point, causing the generator to start.
Performance Factors
The pattern of charge and discharge cycles shown above is typical of hybrid
systems. Obviously, the cycles may not be identical each day because of several
changing factors:
1. The load may change from day to day.
2. The battery may be at a different state of charge at the start of the day.
3. The weather may change, resulting in more input from the PV array.
4. The generator may be programmed to run differently, such as providing an
equalize cycle.
The annual performance of a hybrid system is typically measured by the following:
Total Energy Production
Contribution (%) by each source
Number of battery cycles
Generator run hours
Number of generator starts
Generator fuel consumption
Each of these factors is determined by the size of the various system components.
Some of the relationships may not be obvious at first. However, if you stop to think
about what is actually happening within the system, then the explanation should be
apparent. Below are some general guidelines for most hybrid systems. Note that
computer programs must be used for detailed performance results.
Battery Size The size of the battery affects how many cycles the battery goes
through. The smaller the battery the more frequently the battery will be charged
and discharged.
Generator / Rectifier Size The sizes of the generator and rectifier mainly
determine how long the generator will run for each charge cycle. They do not affect
how much of the total energy comes from the generator. So, to keep the annual run
hours low choose a relatively large engine.
Array Size The size of the array does affect how much of the annual energy is
supplied by PV. Obviously the bigger the array the more energy it will produce.
Inverter Size In a DC bus system the size of the inverter does not affect the
annual performance directly. In an AC bus system, the size of a bi-directional
inverter affects the generator run-time similar to the rectifier size.
18-20
Hybrid Systems
Cycle
Charge
Prime Power
Emergency
Back-Up
*+
18-21
Hybrid Systems
Exercises
,
0%
-
18-22
Hybrid Systems
Component Selection
PV Array
The photovoltaic array for a hybrid system must meet the same strict requirements
as a stand-alone application:
Component
Selection
36 cells for
adequate
voltage
Rugged
design
High energy
output
15+ year
warranty
Test
certification
5. Test Certification Most modules are now tested to a variety of national and
international standards. In the US, the Underwriters Laboratory has a test for PV
modules. In Europe, the Commission of European Communities (CEC) and
International European Community (IEC) have standards for module testing.
By choosing a reputable supplier you will ensure the high quality of the product as
well as getting strong support.
18-23
Hybrid Systems
Battery
The battery bank is a key element in a hybrid system. The battery is literally at the
center of the system, and the quality of the battery can influence the entire system
performance. A well-designed system makes good use of the battery storage. A
poorly designed system suffers when the battery does not perform. The battery
bank also represents a significant cost in the system and is worthy of extra attention.
Component
Selection
Choose a
deep cycle
battery
Batteries designed for float applications and UPS operation are definitely not
appropriate, and neither are SLI (auto) batteries. Marine type cycling batteries are
a marginal choice, while wheel chair and golf-cart batteries may be used in a careful
design. Care should be taken with sealed (gelled or absorbed electrolyte) VRLA
type batteries, since most of these were developed for low cycling applications.
They are also much less forgiving of the variable rates of charge and discharge
experienced in a hybrid system.
18-24
Hybrid Systems
Engine / Generator
The engine generator is a critical component of a true hybrid system, since it
supplies a significant part of the energyit is not just a backup. There are a wide
range of choices for construction, fuel and accessories that must be considered.
Component
Selection
Use an engine
rated for
prime power
Be sure the
engine is
designed for
remote starts
Use air-cooled
engines for
small systems
Choose fuel
type based on
application
Add proper
accessories
for the system
Choice of Fuel
The choice of fuel for the generator is of primary importance. Generators using
diesel fuel are known as diesel electric generators (DEGs) and are very common
around the world. Diesel fuel is commonly available (although the quality can vary
significantly). Service and trained diesel technicians are also comparatively easy to
find, making diesel generators a good choice for many types of systems. Diesel
gensets are often rated for 20,000 to 50,000 hours of operation. In cold weather
(approaching 32 F or 0 C) diesels can have starting problems although these can
be addressed through the addition of glowplugs and appropriate grades of fuel.
18-25
Hybrid Systems
Diesel Fuel
Advantages
Disadvantages
Readily Available
More expensive than
gasoline engines in
Difficulty with cold
small sizes
weather starting
Noise and pollution
Fuel can be stored
for several months
Gasoline/petrol generators are very lightweight and compact for the amount of
power they produce, making them ideal for automotive applications and light duty
contractor generators. They also tend to be the least expensive generators
available. Their two major limitations are fuel storage and design life.
Gasoline/petrol is much more difficult to store in large quantities because it is more
volatile (meaning it burns easier) than other fuel options. This makes it prone to
explosions if not stored properly, or if there is a leak in the system. In addition these
engines are typically rated to provide only about 1,000-2,000 hours of engine life
(compared to 20,000-50,000 hours for diesel gensets) making them suited only for
small backup systems such as cabin power systems.
Gasoline (Petrol)
Advantages
Disadvantages
Readily Available
Hazardous to store
large quantities
Usually technicians
are available
Fuel can not be
stored for long
Engines are small
periods
and inexpensive
Engines typically not
rated for long life
Another common fuel choice for engines in commercial hybrid systems is liquefied
petroleum gas (LPG). These are sometimes modified DEGs. While LPG is less
efficient in terms of fuel consumption compared to gasoline, it is much cleaner and
requires less engine maintenance interval, thus justifying its use in many situations.
Certain environmentally sensitive locations such as National Parks will only allow
LPG generators because of concerns regarding pollution. LPG is volatile and
requires specialized fuel feed systems.
18-26
Hybrid Systems
Natural Gas
Advantages
Disadvantages
Very clean burning
Typically available
only in specific
Less engine
applications
maintenance
Specialized
Low pollution
technicians
Remote Starting
It is very important to ensure the engine selected has remote start capability (unless
an operator will be on the site at all times). The engine start circuit must be
specifically designed to allow remote starting, otherwise it will not operate properly in
a hybrid system. An engine for a hybrid system must also include an automatic
choke that will automatically adjust the air mixture between starting and normal
operation.
Even if the engine can be remotely started, it is also important to determine the type
of signal(s) needed to start and stop the engine. The simplest system is a two-wire
start system in which there is a single pair of wires. When the wires are connected
the engine will start and continue to run. When the wires are disconnected, the
engine will stop. Other types of start systems take 3 or more wires and have
separate signals to start the ignition and to operate the engine. These type of
systems are more complicated, since the control system must now determine how
long to close the starter circuit, whether to allow repeated crank attempts, etc.
18-27
Hybrid Systems
Whether or not the engine has a two-wire start system, the important factors are that
the engine is designed for remote starting and that the type of start system is closely
coordinated with the system controls.
Generator Accessories
Engine-generators in hybrid systems have special duties, compared to a traditional
prime diesel or standby diesel engine. Hybrid engines work in a demanding
environment, with frequent starts, extended hours of operation and typically
infrequent maintenance. For these reasons, a number of the following accessories
may be justified.
18-28
Hybrid Systems
investment. The system should be checked, however, to ensure that it does not
overcharge the batteries.
18-29
Hybrid Systems
Control Systems
Component
Selection
Identify all
necessary
control
functions
Design the
control
systems as
part of the
total system
Starting Engine
A hybrid control system needs to be able to start the generator under the
appropriate circumstances. The type of relays and control points required
depend on the start circuit of the engine. In addition, the controller needs the
proper data to determine when the engine should be started, which may
include battery voltage, current and amp-hour readings, time of day or other
parameters.
Stopping Engine
The engine must be stopped at the conclusion of a cycle.
Regulating Array
The power from the array into the battery must be regulated. The relays used
to disconnect the array must be sized for the maximum array current
(including edge of cloud effects).
High Voltage Disconnect
A high voltage disconnect protects the battery from over charging by
disconnecting the battery from all charging sources. The relays must be sized
according the maximum output for each source.
Low Voltage Load Disconnect
A low voltage load disconnect protects the battery from over discharge. The
relays used to disconnect the load must be sized for the maximum load
current.
Exercise Engine
Engines that sit unused for a long period of time may experience problems
starting. To avoid this, hybrid systems sometimes include an engine exercise
function, which periodically starts and runs the engine for a short time.
18-30
Hybrid Systems
Equalize Battery
In addition to normal battery charge cycles, the system may include an
equalize cycle to reduce any voltage differences between battery cells.
System Alarms
Engine Alarms
One of the most important functions of a hybrid system controller is to monitor
the engine. The control system must be able to recognize potential
dangerous conditions and shut the engine down before physical damage
occurs. There are certain specific indications that should be monitored as a
minimum. While the control system does not need to have a separate alarm
signal for each of the following situations, it should be able to respond to any
of them.
- Low oil pressure
- Engine over-temperature
- Engine over-speed
- Engine over-crank (engine has failed to start)
Battery Voltage
An essential feature of a controller is to protect the battery bank from overcharge and over-discharge. The actual control functions for high voltage
disconnect and load voltage disconnect are explained above. In addition, it is
often useful to sound an alarm so that the user is aware that the voltage is out
of normal limits.
Data Monitoring
System Voltages
System voltages can provide a quick check on the state of the system. A
number of readings through the course of a day also depict array
performance, load use and approximate battery state of charge.
- Array Voltage
- Load Voltage
- Battery Voltage
Current
The current in each of the main circuits is also an important variable. From
the current, one can calculate array output, load demand and charging
efficiency.
- Array Charging Current
- Load Current
- Battery Current (Charge and Discharge)
Temperature
Temperature information is used to adjust the battery charging algorithm and
may also be important to monitor for the load equipment.
- Battery Cell Temperature
- Load Enclosure Temperature
- Ambient Temperature
18-31
Hybrid Systems
Miscellaneous Parameters
Depending on the system, there may be additional items that should be
recorded. For example:
- Fuel level
- Wind speed
- Site insolation
Manual Controls
In this control scheme an operator is always available. When the battery voltage
falls to a certain point the operator turns the diesel generator on and supervises
charging of the battery. When the battery is fully charged the operator shuts the
engine off manually.
While this is the simplest technique in terms of control hardware it is also very prone
to errors in judgment or attention. It can also be very costly both in terms of operator
labor and in consequences of a simple lapse in attention. This type of labor is often
classified as long hours, low effort and poses a special challenge to maintaining
skilled labor.
18-32
Hybrid Systems
Types of Chargers
Component
Selection
Choose the
right
combination of
factors:
- Cost
- Efficiency
- Temperature
rating
- Power
Quality
Diode Bridges (half bridge, full bridge and three phase bridge)
These units are the most basic type of battery chargers and tend to be very cheap.
They offer very little control of the end of charge. Single-phase half bridge chargers
have very significant ripple even when filtered heavily. Single phase, full bridge
chargers have less than 5% ripple and are very efficient; however, they have no
inherent control.
SCR Chargers
This class of chargers is based on silicon-controlled-rectifiers (SCRs). In essence
these are diodes that can be turned off thus offering a method of regulation. SCR
rectifiers tend to be inexpensive and moderately efficient but also have a reputation
for being extremely noisy in electrical terms. This affects both the AC input circuit
and the DC output. The high frequency spikes caused by switching the SCRs off
tend to cause problems with the alternators and voltage control circuits used with
engine generators. Most manufacturers require 20-40% oversizing to ensure proper
operation.
18-33
Hybrid Systems
Ferroresonant chargers
These chargers produce very clean power using ferroresonant transformers and
diode bridges. Their drawback is size, weight and low efficiency, but they are
available in a wide range of voltages (up to 260 V) and currents (up to 800 amps).
18-34
Hybrid Systems
85-92% efficient
18-35
Hybrid Systems
Inverters
Inverter technology can be critical to hybrids that are specified to supply AC power.
There are three main classes of DC-AC converterssquare wave, modified square
wave, and sine wave. The technology follows the same basic classes as battery
chargers discussed above. The best choice is a single or 3-phase, transistorized
PWM, sine wave inverter provided the cost is appropriate.
Power Quality
Component
Selection
Choose the
right combination of
factors:
- Cost
- Efficiency
- Temperature
rating
- Power
Quality
(Worst Quality)
Sine Wave
(Best Quality)
DC Charge Current
If the inverter is bi-directional, then the DC charging current power quality should
also be examined. Refer to the section above on battery chargers for this
discussion.
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
18-36
Hybrid Systems
Choose
structures /
enclosures
based on:
- Weather
- Battery
location
- Corrosion
- Vandalism
protection
18-37
Hybrid Systems
Some system-level design is necessary to make the proper choices during the basic
sizing. It is very important to keep in mind the overall system operation during this
process. There are many interactions between components in a hybrid system and
the designer must ensure that all the overlapping requirements are met.
This section addresses some of the fundamentals of system sizing and design
specifically for PV / Diesel hybrids. Although this particular combination of power
sources is given emphasis many of the design principles are valid for other types of
systems.
Trade Offs in Hybrid System Design
Initial Cost
Customization
Flexibility
Design Limits
Simplicity
Standardization
Operating Cost
)
18-38
Hybrid Systems
Get complete,
accurate load
information.
For DC and
AC loads:
- voltage
- power
- duty cycle
- peak power
For AC loads:
- frequency
- 1 ph or 3 ph
- power factor
Not only the size of the load is important, but also the hourly load profile. The
system must be able to accommodate any peaks. The load profile might also
suggest certain control strategies. For example, in an AC bus design load peaks
should be coordinated with the generator since this improves system efficiency.
Example:
DC Load Calculation
For a certain telecommunications application, the load is given as:
225 Watts at 24 Volts for 14 hours per day
100 Watts at 24 Volts for 10 hours per day
So
225 W x 14 hours = 3,150 watt-hrs/day
100 W x 10 hours = 1,000 watt-hrs/day
Total load = 4,150 watt-hrs.
4150 / 24 hrs = 173 W continuous. 225 W peak
18-39
Hybrid Systems
Example:
AC Load Calculation
In a certain rural community, the community leaders would like to allow
the following equipment per household. There are 40 households total
in the community:
2 x 50 watt fluorescent lights (12 hours per day), power factor = 0.6
1 x 25 watt radio (6 hrs per day), power factor = 1.0
For one household
2 x 50 Watts x 12 hours = 1200 Watt hrs/day
1 x 25 Watts x 6 hours = 150 Watt hrs/day
Total load = 1,350 Watt hrs /day. 1350 / 24 hrs = 56.3 Watts
continuous
Peak power = 2 x 50 Watts + 25 Watts = 125 Watts
The apparent power is:
2 x 50 Watts / 0.6 power factor = 167 Volt-Amperes (VA)
1 x 25 Watts / 1.0 power factor = 25 VA
Total apparent power = 192 VA
For 40 households
Total load = 40 x 1,350 Watts = 54 kWhrs/day.
Total real power = 40 x125 Watts = 5000 Watts
Total apparent power = 40 x 192 VA = 7680 VA
Total power factor = 5000 W / 7680 VA = 0.65
18-40
Hybrid Systems
Bus Configurations
Determining the Correct Configuration
The choice of whether to use a DC bus or AC bus is determined by the loads and
the overall system operation. If all the loads are DC power, then a DC bus makes
the most sense. If the majority of the loads are AC then an AC bus system might
make sense. AC bus hybrids are particular useful when the generator is required to
support part of the load.
Design
Guidelines
For a DC-only
system, use a
DC bus.
For an AConly system,
use an AC
bus.
For a mixed
system,
evaluate both
options.
18-41
Hybrid Systems
Current
2.5 A
5A
10 A
Loss
1.25 W
2.5 W
5.0W
% of Total
1%
2%
4%
.'
Design
Guidelines
Choose a
higher voltage
to reduce the
current.
DC systems:
Match the
voltage of the
largest load
18-42
Hybrid Systems
Example:
A 100A disconnect for the 120V system may be less than 10% of the
price of the disconnect for the 12V system. Other components will
have cost savings as well.
18-43
Hybrid Systems
Battery Sizing
Stand-alone photovoltaic systems will often provide 5-7 days or more of battery-back
up for the load. For critical applications or in locations where the solar insolation is
poor this may increase to 10-15 days or more.
The autonomy of the battery bank is the number of days that the battery can supply
the load under no charging conditions. Thus a battery bank that can support the
load for 5 days without charging is said to have 5 days of autonomy.
Design
Guidelines
A larger battery reduces
the number of
engine starts
and battery
cycles
Start with a 23 battery
Check the
number of
cycles for the
battery
For a hybrid system the battery bank is usually much smaller, often
2-3 days of autonomy. The reason that the battery can be smaller in
a hybrid system is because there is an on-demand power source.
When the battery gets low the system can put a generator in action
and replenish the battery as needed. In a stand-alone system the
battery functions as a reserve of energy. This reserve should be
kept as full as possible to guard against periods of poor weather.
The battery in a hybrid system is serving a slightly different purpose.
The battery allows the system to control the use of each power
source. By providing storage the battery allows the generator to run
at optimum times. It also provides a means of utilizing the energy
gathered by the array. In advanced hybrid control strategies there
are even times when the system deliberately seeks to leave the
battery at a low state of charge.
The smaller the battery capacity, the more frequent the cycles. This
results in a shorter battery life. A good hybrid design strikes a
balance between a smaller economic battery and reasonable life. A
common error about hybrid performance is that a small battery
increases the amount of diesel energy contributed to the system.
On a first order this in not true. A smaller battery results in more frequent diesel
operation and more generator starts, but the total amount of energy produced is
about the same. By using high-cycling batteries and a reliable start system a hybrid
can use a smaller battery and potentially save significant money.
Verify the
maximum
charge rate is
less than C/5.
Since the size of the battery does determine the maximum charge current the
battery capacity does have an effect on the size of the battery charger. This in turn
affects the maximum size of the generator because the generator should not be too
much larger than the battery charger. Forcing a very small engine size will increase
the total hours of operation, which might not be desirable.
18-44
Hybrid Systems
Example:
18-45
Hybrid Systems
Since engine generators use internal combustion of fuel-air mixtures to produce their
energy they are subject to a derating factor for altitude. An engine will simply not
produce as much power at higher altitudes. The typical derating is -3.5% for every
300 meters above its rated altitude. The rated altitude could be sea level, 150
meters above sea level, or some other starting altitude depending on manufacturer.
Temperature is also a factor due to the change in the volume of air and in the
capability of cooling and exhaust air to carry heat away from the combustion
chambers. Typical derating is -0.36% for every oC above a rated temperature (often
30 oC). Humidity can also cause derating of up to 6%.
18-46
Hybrid Systems
Example:
=
=
=
-3.5%
300 m
-0.035
300m
-0.27
2300m
or
-27 % derating
=
=
=
So the 10,000 watt generator could really only produce about 7,300
watts at that altitude. If 10 kW were really required, then a larger
generator would have to be installed.
The example above shows that the rated power must be larger than then required
power to allow for the derating. You would divide the required power by [1 - the
derating factor] to get the sea level rating of the inverter you would need.
18-47
Hybrid Systems
Example:
18-48
Hybrid Systems
Exercises
/
62
4666
1
,666
"
2
62
%
/:-
1
:-
4666
"
2
0
966
62
,;66"
2
18-49
Hybrid Systems
A larger array
results in
more energy
from PV and
reduces the
amount of
energy from
other sources
Tilt the array
to get the best
annual output.
Size the PV
array
between 25%
and 75%
contribution.
Example:
Array Tilt
Angle
20
30
40
50
60
Annual
Wh / year
93,849
97,031
97,506
95,454
90,959
Lowest Month
Wh /day (Month)
158 (Dec)
186 (Dec)
207 (Dec)
220 (Dec)
228 (Dec)
18-50
Highest Month
Wh /day (Month)
329 (Jun)
314 (Jun)
293 (Sep)
278 (Mar)
272 (Oct)
Hybrid Systems
18-51
Hybrid Systems
Site:
Average Temperature:
DC Loads
Description
Quantity
Watts
Hours/day
Daily Demand
(DC Whrs)
Total
AC Loads
Description
Quantity
Watts
P.F.
VoltAmperes
Hours/day
Daily Demand
(AC Whrs)
Total
Total AC Whrs
=
Inverter Effic.
18-52
+
AC Input (Whrs)
=
DC Whrs
Hybrid Systems
DC
Amps
AC
Volts
kW
kVA
Hz
Phase
1 or 3ph
Other Characteristics
Rectifier
Fuel
Type
Total
Derating
Continuous
Rating
Surge
Rating
Estimated
Peak
Estimated
Surge
Capacity
(kWh)
Type
Number in
Parallel
Number in
Series
Quantity
Tilt
Angle
Number in
Parallel
Number in
Series
Generator
Inverter
Total
Load
Battery
PV Array
Other
Renewable
18-53
Hybrid Systems
18-54
2000 Ah
2500 Ah
3000 Ah
4000 Ah
The 3000 Ah capacity cells are the most appropriate, since they are the smallest
size that provides at least 3 days of autonomy. The maximum charge rate of C/5 for
this size battery is approximately:
3000 / 5 = 600 Amps.
Battery Charger and Generator
Suppose we have the following rectifiers to choose from at 48V (Efficiency = 0.80):
Amps
Input (kW)
Phase
100
6
1 ph
150
9
1 ph
200
12
3 ph
400
24
3 ph
800
48
3 ph
The 400A rectifier is a little smaller than the maximum allowed charge rate, but will
still charge the batteries at an appropriate rate (about C/8).
We next determine the size of the generator. Suppose that the following generators
are available and the ratings are given for 900 m and 25 C:
Capacity
Fuel
Phase
25 kW
Diesel
3 ph
27.5 kW
Diesel
3 ph
30 kW
Diesel
3 ph
40 kW
Diesel
3 ph
24 / 0.75 = 32 kW
24 / 0.90 = 26.7 kW
We need somewhere between 26.7 and 32 kW installed output from the generator.
We must now allow for any deratings due to the elevation and temperature.
Temperature
Elevation
Total derating
18-55
Generator Contribution
The customer has stated a preference to reduce the amount of generator run time.
However, given the size of the load it will not be economical to power the loads
exclusively from PV. We might select 500 hrs as the maximum desired generator
run time in one year (typically 1 maintenance interval per year).
The rectifier output is 400A x 48 V = 19.2 kW. Assuming an additional 10% system
losses (0.90 multiplier) and overall battery efficiency of 80% we can calculate the
energy output of the rectifier / generator combination as:
500 x 19.2 x 0.90 x 0.80 = 6,912 kWh/year
PV Array
The PV array needs to make up the remaining portion of the annual load = 12,629 6,912= 5,717kWh /year. The average insolation is 5.5 peak sun hours and we
assume the following multipliers for array output:
High temperature
Dirt
Battery Eff
=
=
=
0.85
0.9
0.8
5717
5.5 x 365 x 0.85 x 0.9 x 0.8
18-56
AC Bus Hybrid
=Z
Battery Capacity
kWh
= 4.3 Y
=5Z
Rectifier Rating
kW
= 0.48 Y
= 0.56 Z
Generator Rating
kW
= 0.67 Y
= 0.78 Z
Annual Generator
Energy
= 0.3 Z
= 365 Z
PV Array Pmax
kW = Annual PV Energy
Rating
Peak Hrs X 365 X 0.6
= Annual PV Energy
Peak Hrs X 365 X 0.5
18-57
Example:
0.3 x 24
7.2 kW
5 x 24
120 kWh
=
=
0.56 x 24
0.78 x 24
=
=
13.4 kW
18.7 kW
=
=
24 kWh/day x 365
days/year
8760 kWh
=
Annual Energy From PV
=
8760 - 4800
=
3960 kWh
Assume that the region receives about 5 peak hours of insolation
daily.
Array Power Required
=
=
3960
5 X 365 X 0.5
4.3 kW
So we find that a 4300 watt array would provide the energy that the
generator did not supply during its 500 hours of annual operation.
18-58
All Systems
Battery bank autonomy
=
3 days
Max battery depth of discharge =
70%
Max battery rate of charge =
C/10
Rectifier efficiency
=
80%
Generator derating
=
90%
General system efficiency =
90% (Due to wiring losses,
etc.)
Battery total energy efficiency
=
80%
PV Array temperature factor
=
90%
PV array dirt factor
=
90%
Inverter efficiency
=
85% (AC Bus systems only)
DC Bus Systems
AC Bus Systems
Y = daily DC load (kWh)
Z = daily AC load from inverter
(kWh)
=Y
= Z / inverter efficiency
= 1.2 * Z
=3*Y
= 3 * Y / Max. DOD
= 4.3 * Y
6. The rectifier
rating is based on
the size needed
= C/10 * Battery Capacity
to deliver the
.9
maximum charge
rate to the battery: = 4.3 * Y / 10
.43 * Y
.9
.9
= .48 * Y
.5 * Z
.9
= .56 * Z
18-59
6. The generator
rating is based on
the size needed
to power the
rectifier:
7. The annual
amount of energy
produced by the
generator in
kWh/year is given
by:
8. The inverter
rating is assumed
to be:
9. The annual
amount of energy
produced by the
PV array in
kWh/year is given
by:
10. The PV energy
is related to the
array rating as
shown by:
= .48 * Y
.9 * .8
= .56 * Y
.9 * .8
= 0.67 * Y
= Gen Run Hrs * final DC load
output
= 0.78 * Y
= Gen Run Hrs * final AC load
output
(Not applicable)
Array Energy =
PV Array
kW * peak hours * 365
days/year * deratings
where deratings = heat factor
* dirt * battery eff
= .85 * .9 * .8
So the PV array
rating is given by::
= 0.6
= Annual PV Energy
peak hours * 365 * 0.6
Array Energy =
PV Array
kW * peak hours * 365
days/year * deratings
where deratings = heat factor
* dirt * inverter eff * battery
eff
= .85 * .9 * .85 * .8
0.5
= Annual PV Energy
peak hours * 365 * 0.5
18-60
PV System Maintenance
The maintenance of a PV array in a hybrid system is no different than standard
maintenance in a regular PV system. A good check of the system will include:
1.
2.
3.
4.
This is a very important point in a hybrid system. Even though the energy
contribution of the diesel segment may be 50% or more the duty cycle is typically
much less than this often 10% or less. Because maintenance is related to the
number of hours rather than the total energy delivered this offers a tremendous
advantage to a hybrid system.
Consider the following maintenance schedule which is typical of a small, air-cooled,
prime diesel engine:
Maintenance Task
Interval
Oil Change
250 hrs
Decoke
1,500 hrs
Overhaul
6,000 hrs
!
18-61
Cost Estimate
$25 to $100
plus travel if
required.
$250-500 plus
travel for a
typical two
cylinder air
cooled engine.
$1,500-3,000,
plus
transportation
and loaner of
replacement
engine.
Note that many standby engines have only a 100-hour oil change interval that, if
running in a prime power application, would necessitate nearly 90 oil changes per
year -- almost twice per week! Engines made for prime power often have options
that can extend the primary maintenance interval to 500, 1000 or even 1500 hours
dramatically reducing the maintenance requirements.
Using the hybrid sizing example in the previous section the generator runs
approximately 500 hours per year vs. 8760 hours for a continuous prime diesel. The
following table combines the maintenance schedule with the operation of the
engines to calculate the total number of maintenance visits required:
18-62
Maintenance
Interval
Hybrid
Prime Power
Task
(Hours)
(500 hours)
(8760 hours)
Oil Change
250
2 per yr
35 per yr
Decoke
1500
1 every 3 yrs
6 per yr
Overhaul
6000
1 every 12 yrs
3 every 2 yrs
$%&!'
(
)*
In addition the higher loading of the hybrid engine will keep the engine running more
efficiently reducing some of the common problems with diesel generators -- clogged
fuel injectors, heavily carbonized valves, slobbering and glazing. This means that
the intervals for extended maintenance items such as top-end overhauls and full
overhauls might be extended even further than for a variable load prime diesel.
18-63
Battery Maintenance
The second most maintenance intensive item in a hybrid system is the battery.
Unlike batteries in PV systems that are treated quite gently hybrid system batteries
are exposed to frequent deep discharges (typically two per week or more) and
variable rate charges (often at high rates). This last point is very important -charging must be coordinated between the various generators.
For example a system may include a diesel engine, a couple of wind turbines and a
substantial size PV array. The engine/battery charger is designed for a high rate
recharge (C/6 or so). If the battery charger is on and the wind suddenly picks up or
the sun comes out the charge rate may exceed the battery manufacturers
recommendations causing damage to the battery and a potentially dangerous
situation with excessive gassing and hydrogen evolution. If VRLA batteries are used
this point is even more critical.
In general the batteries should be checked at regular intervals. Unless there are
serious access constraints to the site the batteries should be examined at least
every 3 months. They should be filled with water, equalized (if the controller does
not have an automatic equalization function) and checked for tight connections,
evidence of corrosion, etc. Once a year a complete record should be made of
individual cell voltages and any suspect cells tested and replaced if necessary.
Since most hybrid systems use flooded cycling batteries they often use increasing
amounts of water as they age. This should be taken into consideration when
planning maintenance visits. In a village type system a local villager can be trained
to check the water level at weekly intervals, keep records and report to the
maintenance crew if there are any thirsty cells. In telecom systems this is not
possible; however, careful monitoring of separate battery strings might show if one is
drawing excessive amounts of current during charge -- a sure sign of a problem cell.
In general the use of a high quality controller and battery charger will help ensure
lower maintenance requirements and longer life of any hybrid system battery.
18-64
Controls Maintenance
The amount of maintenance on the system controls is usually very small and
consists primarily of checking that all the wiring connections are tight and that there
are no problems with the fuses or surge protection.
Some systems with on-site data storage may use this time to download the data or
to replace storage modules for transportation back to a central site.
It is also a good idea to use the maintenance interval as a chance to review the
system operation. Does the system operate as intended? Is there any unusual
behavior that requires attention. Reviewing the performance data can be a way to
identify potential problems before they become serious. The system set points and
control strategies can sometimes be adjusted to accommodate new changes in the
load demand or different conditions than were originally expected.
18-65
Microwave repeaters
Fiber optics relays
Rural telephony
Cellular phone sites
Television transposers
Telecommunications power sources demand very high reliability. Some systems will
include a second engine generator as a backup. This is based on the idea that a
system should have duplicates any essential component or system. Sometimes this
is referred to as "1 + 1 redundancy." Another approach used to assure performance
is to over-size some of the components. A telecommunications system may include
a much larger battery or have a very high PV array contribution. These features
allow the system to operate for a longer period of time after a diesel failure, hopefully
giving a repair crew sufficient time to reach the site and make repairs.
A system designed for telecommunications must also give thought to protecting the
load equipment. Unlike a village power system where all the loads are physically
separate from the power system a telecomm hybrid usually needs to include some
space for the load equipment. This should be removed from the generators and
batteries. In addition there may be heating and cooling requirements for the load
equipment. The system designer must include the power necessary to heat or cool
the enclosure in addition to the load equipment draw.
18-66
Village Power
Many rural communities in the developing world do not have electricity. A small
amount of electricity can make a dramatic change in the lifestyle of these people by
providing lighting, refrigeration and access to television and radio programs. Hybrid
systems can be used for rural electrification projects. Of those communities that are
already electrified many are powered by a diesel generator and already have
distribution lines. In communities without power a hybrid offers the ability to supply
power while reducing fuel costs, pollution and noise.
Village load profiles are often well suited for a hybrid system particularly if there is a
significant lighting component. Since the lights are primarily used during the evening
hours the load level may be several times the daytime value. This results in very
inefficient prime power diesel systems, but the same load variation can be handled
by a properly designed hybrid power system. The presence of a generator to supply
power on-demand is also a cost-effective means to power large commercial loads
(e.g. an ice-making facility) that would be very costly for a stand-alone system.
Hypothetical Village Load
4.8 kW Peak / 50 kWHr/Day
5.00
4.50
4.00
Load, kW
3.50
3.00
2.50
2.00
1.50
1.00
0.50
0.00
0:00
2:00
4:00
6:00
0:00
Time of Day
+,-
There are many social aspects to bringing power into a village. How will the
lifestyles of the villagers change? What new uses for the power will they find? The
size of the load can often change according to special occasions or holidays.
Sometimes the total load in the village grows dramatically after the system has been
installed.
18-67
Facility Power
Another application of hybrid systems is for facility power. This could be any
isolated site with a collection of diverse loads. Some example of facility applications
include the following:
Hotels
Government offices
Medical clinics
Military facilities
Recreation Sites (Marinas, Campgrounds, etc.)
While the characteristics of each system can vary significantly these applications
require fairly large amounts of power and typically include AC loads. In addition the
load demand is rarely constant and may be composed of a number of variable
loads. It is important to carefully define the load requirements of the system and
ensure that the hybrid will have sufficient capacity to meet peak loads.
Since many facility power applications are existing sites the hybrid power system is
often intended to supplement an existing diesel-based power system. In such
circumstances the primary customer concern is to reduce the amount of diesel
operation. This is in effort to reduce the amount spent on fuel and maintenance.
Pollution and noise may also be significant secondary considerations especially for
ecologically sensitive sites.
One feature of a facility power system is that someone is usually designated to
provide maintenance for the system. This can be an advantage over village power
projects where on-going maintenance can be difficult to support.
18-68
Design Examples
Pulau Pemanggil, Three Islands Project, Malaysia
The Pulau Pemanggil system is located in Malaysia and is designed to provide
power for a small community located on an island near the coast. The original load
requirements were calculated as follows:
ITEM
Avg.
No. of
Wh/Day Units
per unit
276
20
380
3
834
1
464
1
1048
1
3970
1
740
1
Daily
Energy
Wh/Day
5520
1140
834
464
1048
3970
740
Annual
Energy
(kWh)
2014.80
416.10
304.41
169.36
382.52
1449.05
270.10
Domestic House
Domestic House + Shop
Mosque
Community Hall
Police Post
School
Clinic (excluding vaccine storage
system)
Power Plant House
904
1
904
329.96
Total Average Wh/Day
14620
5336.30
Add 10% Safety Factor (to account
1462
533.63
for cable losses, load variations,
etc).
Total Average Energy required.
16082
5869.93
.**,'
Array
Battery
Generator
Inverter
Design Load
7.2 kW peak
124 kWh @ 120 Volts
26 kW diesel generator
10 kW, 240 VAC, 50 Hz, 1 phase
16.1 kWh/day @ 4.5 Sun hours
The system was originally designed as a PV system with the generator primarily
used for backup. The PV array was to supply most of the loads. After the system
was installed the loads grew significantly to the point where the generator supplies a
substantial portion of the loads.
PV Array The photovoltaic array uses high-efficiency, single crystal modules.
There are three ground mounted arrays.
Controls The system uses one set of controls for the PV array and another set of
controls for the generator. In addition there is a data acquisition system that records
performance data for the system.
Battery The system uses a single string of sealed batteries.
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
18-69
11.3 kW
Rectifier
7.1 kW
PV Array
10 kW
Inverter
124 kWh
Battery
240 VAC
Loads
/#
0+*1
For additional information on this system contact Siemens Showa Solar, Singapore.
18-70
1500 W peak
24 kWh @ 24 Volts
4.8 kW
4 kW, 120 VAC, 60 Hz, 1 phase / 24 VDC
6.5 kWh/day AC loads @ 4.5 Sun hours
18-71
4.8 kW
Generator
4 kW
Inverter/
Rectifier
1.5 kW
PV Array
24 kWh
Battery
/#
0*#
For additional details refer to the brochure located in the Appendix. The SunWize
Power Station is manufactured by SunWize Technologies, Inc. located in Kingston,
NY, USA.
18-72
11 kW peak
200 kWh @ 24 Volts
4.8 kW 24 VDC Output, Dual Units.
1.5 kW, 120 VAC, 60 Hz, 1 phase / 24 VDC
70 kWh/day @ 4 Sun hours
PV Array The photovoltaic array uses single crystal modules. About 75% of the
array is mounted on the shelter the rest is ground mounted.
Generators The engine generators are propane fueled. There are two DC
generators for the site. One generator is designated as the primary engine and the
other serves as backup (1 + 1 redundancy).
Note that the array size is very large while the generator is relatively small for the
given load. This system was designed to operate at 80-90% PV contribution but
also to allow for significant load expansion. This way the load may increase without
any need to add additional hardware.
System Controller The system controller at this site allows for remote monitoring
of key system values including fuel levels. This allows the customer to access the
site via a telephone modem and determine the status of the system.
Inverter There is a small inverter to provide power to 120 VAC loads within the
shelter. Most of the equipment loads operate directly off the 24-volt power bus.
Batteries The battery bank is composed of sealed batteries. Because the
generator provides a relatively small amount of charge current the risks of
overcharging are well managed.
Shelter/Enclosure The load equipment is contained in a 24' concrete shelter.
The shelter also provides a mounting location for some of the PV array. Cooling
fans ensure that the shelter temperature does not exceed the design limit.
* Note: There are actually two hybrid systems located on McPherson peak both of
serving similar loads. The system in this example is slightly larger than the other
system.
18-73
70 kW
PV Array
200 kWh
Battery
24 VDC
Loads
1.5 kW
Inverter
120 VAC
Loads
/#
0
For additional information on this system contact Solar Electric Specialties in Willits,
California, USA.
18-74
Exercises
*
2.324
!
"
18-75
(End of Chapter)
18-76
CHAPTER EIGHTEEN
HYBRID SYSTEMS
Multiple Energy Sources - Flexible Approaches
18-1
18-1
18-2
18-2
18-3
18-5
18-6
18-7
18-9
18-10
18-11
18-11
18-16
18-19
Component Selection
PV Array
Battery
Engine / Generator
Control Systems
Battery Chargers / Rectifiers
Inverters
Structures and Enclosures
18-23
18-23
18-24
18-25
18-30
18-33
18-36
18-37
18-38
18-39
18-41
18-59
18-61
18-61
18-61
18-64
18-65
18-66
18-66
18-67
18-68
18-69
18-77
Chapter 18 Answers
Hybrid Systems
Looking at the table of insolation data, the worst case for the system tilted at 55 is 2.84
peak sun hours and occurs in June. In comparison, the average number of sun hours
for the hybrid system tilted at 25 is 4.53. Therefore, the first advantage to using the
hybrid is that we can design around an insolation value of 4.53 sun hours, rather than
2.84.
The highest insolation for the system tilted at 55 is 4.94 sun hours, occurring in
February. A stand-alone system designed for the worst case of 2.84 will lose much of
this energy. The amount of excess energy is
Maximum Insolation
Worst Case
4.94
2.84
1.74
18-1
Hybrid Systems
1. DC Bus (DC Loads Only) This system configuration makes the most sense for
large DC loads. The generator provides power to the rectifier, which then charges the
battery. It is the simplest to design and control, and it does not require a sophisticated
inverter. Good applications for this type of system include telecommunications
systems.
2. DC Bus (DC and AC loads) A DC bus hybrid with mixed loads has some of the
same control simplicity as a basic DC bus system. AC loads are connected to a
dedicated inverter that draws from the battery. This works best if the AC loads are
generally small and do not have much variation. This system again works well for
telecommunications, perhaps for a system including an equipment shelter with lights or
other small AC loads.
3. AC Bus (AC Loads) This type of system has the most flexibility in supplying
variable AC loads. The generator is connected in parallel with the output of the inverter.
This allows generator power to flow directly to the loads and is very efficient. The
inverter must be more sophisticated and the control system needs to include more
features. This type of system works well for a village with many AC loads or a facility
power system, e.g. a small eco-tourist hotel.
An AC bus hybrid can still supply DC loads directly from the battery bank. The bidirectional inverter keeps the battery charged, supplying any DC loads that may be
connected. Note that for larger hybrid system, the battery bank voltage may be too high
for most normal DC loads.
Choose any three of the following factors:
Battery Size
The larger the size of the battery bank, the fewer cycles the generator will make.
A large battery bank will also reduce the number of cycles on the battery, thereby
extending its life.
Generator/Rectifier Size
The larger the generator / rectifier combination, the shorter the time necessary to
recharge the battery bank. This does not change the total amount of energy
provided by the diesel; it just reduces the hours of generator operation.
18-2
Hybrid Systems
Array Size
The larger the PV array, the more renewable energy is produced and the less
energy will be required from the generator.
Inverter Size
The size of the inverter affects the size of the AC loads that the system can
handle. In an AC bus system, it also has the same effect as the rectifier.
Other items, such a load use, weather and control strategies are also possible.
a. System B is the cycle-charged system. System A is a prime-power diesel system.
b. Although the engine sizes can vary by the application, System B probably would
have a larger engine. The larger the engine size in a cycle-charged system, the fewer
the overall hours of operation (same as for a hybrid system). A large engine will require
a large rectifier to properly load the engine. Increasing the size of the generator in
System A does not reduce the run time (the engine is always on). A large engine in a
prime-power system will actually reduce the efficiency of the system.
c. For System A, the issues will be those associated with a high number of run hours.
This includes frequent oil changes and overhauls. The system will also require frequent
re-fueling. System B will have lower maintenance requirements in terms of oil changes
and maintenance. However, the start system will have to be reliable, so the start
system and starting battery will need to be properly maintained.
We use the general engine derating figure given in the text. The derating is calculated
as:
Derating
-0.035 X (3000)
300
0.35
or 35% derating
18-3
Hybrid Systems
We now need to take into account deratings for both temperature and altitude. The
altitude derating was calculated in the previous problem to be 35%. The temperature
derating is:
Temperature Derating
.0036 X 10
= .036
= 3.6% derating
The final output will be the rated power multiplied by both derating factors:
Output Power
We first calculate the derating factor for the elevation.
Derating
-0.035 X (1200)
300
.14 or 14%
Power REQUIRED
[ 1 - Derating Factor]
20,000 Watts
[ 1 - .14 ]
Therefore, in order to have 20 kW available at 1900m, the engine must have a rating of
at least 23.3 kW at the rated altitude of 700m.
18-4
Hybrid Systems
The sizes of the various installed components (array, battery, rectifier and generator)
may be different than the sizes resulting from the "rules of thumb." There are a number
of reasons for these kinds of differences, including:
It is important to remember that the rules of thumb are a rough guide. The hybrid
system designer still needs to evaluate the sizing and selection of components to meet
his/her specific needs.
18-5
Hybrid Systems
Chapter Nineteen
Installation and
Troubleshooting
As you read this chapter, consider the benefits of planning and the costs of a mistake
once you have begun the installation process. As we have seen throughout this
course, the calculations needed for a typical photovoltaic installation are simple. It is
the proper consideration of all the small details that distinguishes a professional
installation from an amateur one.
Also consider the cost of repair or maintenance visits to a remote site compared to the
cost of initially planning the system for extra reliability. There is never a free lunch,
even with a photovoltaic system. But you can pay a little at the beginning of the life of a
system, or a lot more over time, especially when it is difficult or costly to access the site.
The information presented in this section is intended as a survey of technical topics,
and does not constitute a complete course covering all aspects of installing systems. It
is expected that the student will continue his or her own education, to insure that proper
safety and design concerns are included in any system design, installation,
maintenance and repair.
The technical information and suggestions for installation, operation, use and
maintenance made herein are based on Siemens Solar Industries' knowledge and
experience. They are believed to be reliable, but such information and suggestions do
not constitute a warranty, expressed or implied.
Careful planning, knowledge of electricity, and safe practices will reward the
system installer with a dependable, low maintenance system.
19-1
Safety Issues
The safety record for photovoltaic power systems has been excellent, and we in the
industry want to keep it that way. There is the danger of physical harm from working
with equipment, as well as the danger of electrical shock from working with solar
modules, batteries, inverters, and engine generators.
Inadequate grounding.
Poor workmanship.
19-2
Know the test equipment and tools you are working with.
The best safety system is an alert mind, a skeptical mind, and a slow hand.
19-3
Personal Safety
Remove any jewelry from around neck or from hands or wrists before working
around batteries.
Wear protective clothing while working with batteries. This includes acid-resistant
gloves, apron, and eye protection.
Have fresh water easily accessible in case acid splashes on skin or eyes.
19-4
Disconnect battery bank from any sources of charging or discharging before working
on batteries.
Design battery area with adequate ventilation, and adequate protection from the
environment.
Lift batteries only in a manner approved by their manufacturer, and never by the
terminal posts.
Use lifting aides, dollies and other moving aides to transport batteries. Do not
simply pick up batteries and carry over terrain. Plan the transportation process.
Tools should be shorter than the length between terminal posts, to reduce the
possibility of dropping and causing a short circuit.
Have non-working or grip end of metal tools covered with a strong non-conductive
plastic coating or tape.
19-5
Exercise:
19-6
19-7
Plan to locate batteries, controls, inverters, and electronic components such that
they are protected and access is controlled (locked room).
Plan to protect PV panels from animals, people, and falling objects. Consider
fencing surrounding the array and structure.
Consider the shortest possible wire runs, to minimize voltage drop/wire sizes, and
wire protection, in the placement of equipment.
19-8
a solar siting device to determine shading for various times of the year
19-9
The device can even be used to estimate quantitatively the amount of shading that
might occur due to objects in the suns path. Values are assigned to each hourly
segment that indicate the percentage of total solar path contained in each hourly
segment. Since the sun is up for fewer hours during the winter, each hour of the suns
path during those months will constitute a greater percentage of the whole than an hour
segment during the summer.
The installer can draw the horizon onto the card, including any objects that intercept the
suns path. Each hourly segment that is obscured during a particular month can be
added up to give the total percentage of the suns path that is blocked. This can then
be subtracted from the expected total solar insolation for a typical day during that month
to give the real insolation that would be available. But all effort should be made at the
outset to site the array away from objects that block the suns path.
19-10
h e ig h t
a rra y
c ritic a l
a ng le
d is ta n c e
An array should be placed far enough away from an object so that there is no shading
between the hours of best insolation, usually from 9:00 a.m. to 3:00 p.m., on the day
with the longest shadows, December 21 in the Northern Hemisphere and June 21 in the
Southern Hemisphere. It is very helpful to have a solar site locator to determine
this (as discussed earlier). If you do not have access to a siting device, use the
following calculations to insure that the array will be located away from potential
shading.
19-11
A graph is presented to aid in deciding where to place an array for no winter shading.
Read up from the latitude of the site to the curve for the hour when no shadows are to
reach the array (again usually 9:00 a.m.). Then read across to find the Spacing Factor.
Multiply this factor times the height of the object to calculate the distance the array must
be placed away from the object.
24
20
16
12
8
4
0
0
20
40
60
Latitude (degrees)
7 a.m.
8 a.m.
9 a.m.
Example:
19-12
Example:
The northern pan handle of Texas falls between the 10o E and 12 oE
lines of the map. This means that the north point of a compass points
about 11o EAST of true north. So true north would actually be about
11o WEST of where the needle points.
19-13
Array Considerations
When laying out the array, the system designer should consider the following:
Mechanical Considerations
Electrical Considerations
Aesthetics
19-14
Equipment Location
The site visit is also an opportunity to plan the location of the control equipment, battery
bank, and other wiring concerns. Ease of access to the equipment and safety should
be considered. Building new cabinets or structures, or modifications of existing ones,
may be planned at this time.
The location of a ground rod or locating of a suitable grounding pipe should be planned.
The quality of the soil may impact the type of grounding scheme used. A great distance
between the array and the control and battery systems may cause the designer to
consider two ground rods, one at the array and another near the battery and controls.
This would impact the grounding wiring plan.
Equipment location is very important. One should consider the following:
Plan to post a clearly written method for disconnecting power to equipment before
servicing and for reconnecting
Consider that the batteries will need to be vented and protected from sparks. Do
not locate the controller or inverter over the batteries
Do not locate an engine generator in the same room as the batteries if they are
vented lead-acid type
Plan for properly sized electrical disconnect switch or circuit breaker for all source
circuits and to isolate major components such as the controller and the array
19-15
Foundation Design
Purpose of Foundations
The purpose of a foundation is (a) to prevent the array from sinking; (b) to maintain
proper array orientation with respect to the sun; and (c) to prevent the array from
"uprooting" during severe winds. Many factors are involved when choosing an array
foundation: site accessibility, local topography, soil properties, local building codes, and
availability of labor.
Types of Foundations
The slab type foundation requires a large mass of concrete and relatively flat terrain.
The slab can be poured at the site or prefabricated slabs can be transported to the site.
It is thus not well suited for remote applications where the cost of transporting cement
to the site is prohibitive. Nor is it desirable in very rugged terrain because of the site
excavation that would be required before placing the foundation.
The block type foundation is most ideally suited for rugged terrain and remote locations
because it is relatively light and transportable, and can be prefabricated wherever
cement and equipment are available. Little excavation is needed and the blocks can
usually be positioned fairly easily, minimizing alignment problems. Blocks must be
assembled with reinforcing steel, and all cavities fully filled with concrete or mortar.
A variation of the block type foundation is to use sauna tubes (with concrete) or wooden
pressure treated posts to make a beam foundation. In this type the tubes or posts are
set in below the frost line. Beams of wood, steel or aluminum create a frame across
the tops of the posts and tubes. Bolted to this frame is the module mounting hardware,
typically a standoff type. The beam foundation is a compromise between the slab and
block types. It is well suited for "rolling hill" terrain, and provides easy alignment
between adjacent arrays.
There is no foundation that is applicable to all situations all the time, for any such
foundation would have to be grossly over designed and would be uneconomical in most
situations. The foundations discussed represent the most common designs used for
PV power systems. Although variations in these foundations exist, one of them will
generally be suitable for a particular application.
If a pedestal or tracker structure is to be used, consult the manufacturer for the
recommended foundation. Often simply setting the vertical pole in a hole filled with
concrete will be enough.
19-16
19-17
Loading Values
The foundation weight must be determined as a part of the foundation design process.
The weight depends upon expected array loading and soil type.
Array loading includes wind force pushing against or lifting the array, snow pack that
might accumulate on the array, and the dead weight of the modules. The Siemens
Solar Standard Support Structure has been designed to withstand 50 lb./ ft2 loading
acting directly on the solar panels. It would take a 140-mph wind (225 km/h) to produce
this force or one meter of snow. A foundation should be designed to support an array
during these "worst load" conditions.
R4
R1
R2
R3
The reaction forces to this loading are shown below, both for horizontal and vertical
forces. The values shown are not the required foundation weights, but are the forces
that will be transmitted to the foundation during worst load conditions. For example, if
an 8-module structure at 45-degree tilt was subject to worst load conditions, the vertical
force transmitted to one front foot would be 75 lbs. and a rear foot would experience
700 lbs. vertical force. The front foot would be pushed horizontally with 325 lbs. and
the rear foot would feel 300 lbs. push horizontally. To remain stationary, the foundation
must react with at least this much force in the opposite direction.
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
19-18
The reaction forces have both positive and negative values because the forces act in
both directions. The forces will be in one direction when the array is loaded from the
front, but will act in the opposite direction when the loading is on the back of the array.
Vertical forces are resisted by the sheer weight of the foundation. Resistance to
horizontal forces is related to soil density, cohesiveness, and soil aggregate and to
foundation weight and design. In general, an engineering soil analysis may need to be
conducted on site, and the results analyzed to determine required foundation weight
and design.
The values presented below are not weights, but the force that the foundation must
resist. The final weight and shape of the foundation to resist these forces are derived
from knowing the soil characteristics as well as the bulk weight of the foundation
members. All values are for both directions with respect to the R vectors.
R1
R2
R3
R4
525
425
350
325
300
300
275
275
300
275
325
250
250
250
250
250
250
250
250
250
275
325
525
125
150
175
225
250
300
325
375
425
400
575
675
700
700
700
700
700
700
725
725
850
4-Module Structure
15
20
25
30
35
40
45
50
55
60
65
175
150
150
125
100
75
50
50
50
50
125
425
300
200
125
50
50
50
50
75
100
125
8-Module Structure
15
20
25
30
35
40
45
50
55
60
65
250
125
100
50
50
50
75
150
200
275
475
300
175
200
250
275
300
325
350
350
350
425
!"
#$%
19-19
!"
&$
19-20
19-21
Electrical Connections
All wire connections, switches, receptacles and breakers should be mounted in
electrical boxes. Following are the common ways of connecting wire.
Wire or Conduit to a Box
Wire to Wire
Wire to Terminal
!"
'(
')!*$
Control Center
A primary candidate for pre-wiring is all the control and main safety equipment. This
typically includes the main disconnect switches for the array and battery, as well as the
charge controller(s) and any alarms and metering equipment, and AC and DC
distribution boxes or load centers. One of the great advantages of pre-wired and codeapproved power centers such as the products made by Ananda Power Technologies is
that you do not have to pre-wire and test all the main control and safety equipment.
It is often desirable to pre-wire and actually pre-assemble these components onto a
plywood board for transportation to the site as an assembled unit. Careful
measurements must have been taken at the site to insure the board will fit properly with
the required safe spacing from walls or other equipment.
Even if transporting to the site fully assembled is not possible, pre-assembly is a good
way to test the overall layout and design, and insures that all details are checked. Ends
of wires can be wrapped with a number or letter code, and a legend made, so that in
the field the proper connections can be made quickly.
Locate the control center to avoid temperature extremes. Use insulated enclosures for
cold weather, shaded enclosure for hot weather. Many small system controllers are
mounted in rainproof boxes such as a NEMA 3r. This should be separate from any
battery compartment unless you are using maintenance free batteries that do not emit
gasses.
Array
The array should be pre-assembled and wired, to again insure that all hardware is
included. Module interconnect wires can be pre-cut and have soldered crimp
connections or wire stripped ready for installation into the module junction box. The
path for the main wires from an array can be planned, and means for securing these
wires against wind can be planned. The method for grounding the array and mounting
structure should be worked out.
19-23
Battery
Planning for batteries involve how to safely handle and move them. Due to their large
weight, it may be necessary to use a mechanical lift to move them. In addition, it is
important that there is sufficient width allowed to move batteries in and out without
damaging them.
It is best to have batteries delivered a few days before installation. The battery supplier
should charge them before delivery. Before taking them to the site, the installer should
verify their state of charge and if necessary bring all batteries to an equal full charge.
This is also the best time to screen batteries. Quality control can be a problem with
batteries in many Third World countries. By checking voltages and observing how they
take a charge, batteries can be screened for weak cells and poor performance before
placing them in the field.
Pre-wiring of the battery bank insures that proper lugs or clamps are obtained and presoldered to heavy gauge wire. Also a method can be practiced for attaching any
temperature sensor to the battery cases, for charge controller temperature
compensation.
Other Wiring
The full system of array, battery, controls, loads, and inverters should be planned. Wire
and conduit can be obtained and cut, and interconnects and fittings acquired. The
mounting and wiring of any inverter can be planned. An AC and/or DC load center or
distribution box can be obtained and pre-wired with appropriate circuit breakers or
fuses. All switches for loads can be examined for the best way to interconnect the
wires. Using closed lugs on the switches will assure reliable operation even if the
connecting screws loosen slightly.
19-24
Transportation Plan
Appropriate transportation equipment must be obtained. The vehicles must be large
enough to carry all the equipment in a safe and secure manner. All possibilities for
weather should be considered to be prepared for snow, wind storms, rain, and rough
terrain or other problems.
There may be a need for small lifting and transporting equipment such as two-wheel
hand trucks or dollies to move the batteries or inverters.
19-25
Tool Selection
Some tools for installing and maintaining photovoltaic power systems are listed below.
Of course other equipment for building and construction will be useful. Careful planning
of the installation process will help prevent loss of time due to missing tools and
accessories.
Optional
Clamp-on DC ammeter.
Knife
Hammer
Light meter
Chisel
AC Amp meter
!"
19-26
Installation Processes
As opposed to the pre-installation processes at some comfortable location, the actual
installation process involves working quickly and safely at the installation site to install
and commission the PV power system and loads.
Foundation Alignment
The most critical element of foundation installation is alignment. The points where the
mounting structure is to meet the foundation must be level and any mounting bolts must
be spaced correctly. It is critical that careful measurements be made both for spacing
and for flatness.
The orientation of the foundation must face true (solar) south (unless local climate or
load factors require facing away). Care should be taken to adjust for magnetic
corrections to compass readings.
19-27
Exercise:
/
01
+
&
,
2/
. '
20
34
5
.
.
6
'
'
"
21
2
2
/
"#
8$
3
.
"
0
00
19-28
Step
Check that all tools and mounting hardware are available and in
working order. Check that foundation is properly installed and
locations for mounting feet (if applicable) are spaced correctly.
Remove modules from shipping boxes and inspect. Check electrical
output of each module at Voc, Isc and using a 50 watt 6 ohm resistor to
approximate maximum power.
Place modules face down, side-by-side, on a clean flat non-abrasive
surface. The shipping boxes will work as a "table" if desired. Arrange
the modules with the junction boxes located so that series and parallel
wiring will be efficient.
Place the two support structure "C" channels on top of the modules,
parallel, with the "C" facing inward and the mounting holes facing down
aligning with the mounting holes in the modules.
Attach the modules to the structure channels using a 1/4" stainless
steel bolt, a flat washer, a lock washer and a nut in each of four places.
Fasten only finger tight at this point. Continue for all modules on the
structure.
NOTE: Up to 8 modules can be mounted on a 9-foot long structure.
There should be about 1 foot of empty space left between the lowest
module and the ground, to allow for debris to blow free and not build up
against a module surface. Modules should be mounted flush to the top
of the structure. Empty spaces can be filled with a cross-brace,
available from Siemens Solar.
Attach all module interconnect wires. Use sunlight resistant type UF
wire or equivalent. Remove about 5/8" of insulation (3/4" for AWG #8
wire). Do not use wire lugs. Insert stripped wire through foam seal and
under terminal screw slider. Tighten screw securely using a
screwdriver. Attach large sized strain relief clip enclosed with modules.
For small wires (AWG #14 and #12), use the second smaller strain
relief clip also.
Assemble the two rear support legs to the correct length to give the
proper final tilt angle. Use four 3/8" bolts, washers and nuts to fully
assemble each support leg.
Attach four-foot angles to foundation.
Attach each rear support leg to a rear foot angle using 3/8" bolt, flat
washer, locking washer and nut.
Move module assembly into place with modules face up, and attach
bottom of side channels to front foot angles.
Lift up assembly and bring ends of rear support legs up to meet the
assembly. Attach each rear support leg to a side channel using a 3/8"
bolt, washer, locking washer and nut.
Tighten all 1/4" nuts and bolts to 80 in-lbs., and all 3/8" nuts and bolts
to 230 in-lbs.
8
9
10
11
12
!"
19-29
Control Center
Assemble all controls, disconnect switches, alarms and meters and load centers as preplanned.
Make sure there is the required safe space between any boxes and walls or pipes or
other equipment.
Check that all connections are secure and clean, that all wiring is color coded or
marked for correct polarity and that all wire lugs are fastened tight to their wires.
Before hooking up the inverter, make sure the polarity is correct. Most inverters can
be damaged if hooked up with the wrong polarity.
Leave a schematic of the system wiring (secured to the control center if there is
appropriate room) for later maintenance and troubleshooting. Label all main switches
so that an uninformed person with the operation guide can safely and easily disconnect
the system in an emergency.
19-30
Other Wiring
Connect all equipment with proper wire and conduit if necessary.
Make sure all wiring is secured by gently but firmly pulling on all connections. Check
that insulation is not cut during installation work, or when wrapped around corners or
near edges of conduit.
Make sure that the color coding of the wire used is consistent and conforms to local
codes.
Commissioning
The full system should be tested and put into commission before the installers leave the
site.
All safety equipment and switches and circuit breakers should be cycled and tested.
Some controllers and inverters have a very specific hook-up sequence to bring them
on-line. Be prepared to follow that sequence.
All hard-wired loads should be operated and all outlets should be tested.
Restricted equipment should be locked and the keys secured.
A system description with circuit diagrams and an operation/maintenance plan should
be presented to the client. The designer or installer should read through the
operation/maintenance plan with the client. Perhaps have them sign a copy asserting
that they have read the instructions. A copy of the operation/ maintenance plan should
be left with the client. All major components and switch gear should be labeled and be
easily accessible for service.
19-31
19-32
Weekly Maintenance
Monthly Maintenance
Annual Maintenance
19-33
Battery Maintenance
Physical Considerations
Observe water level in every battery cell and fill to lever indicated with distilled water.
Inspect all terminals for corrosion and loosened cable connections. Clean and tighten
as necessary. After cleaning, add anti-oxidant to exposed wire and terminals.
Examine outer surface of batteries for film of liquid. This may indicate excessive
gassing, bubbling electrolyte out of battery, it also may indicate a controller problem.
Confirm that batteries are not sitting directly on floor. Some elevation allows for
cleaning and refilling water to flow completely away from battery cases, and prevents
bottom of battery from operating at lower temperature that top of battery.
Make sure the battery enclosure is well ventilated. Inspect any ventilation pathway or
system for blockage.
19-34
To determine voltage, measure the open circuit voltage of each battery cell, or of each
block of cells if internally connected. The best time of day is late afternoon, after the
array has had the most time to charge the battery. Ideally, this should be done after
disconnecting any array and load connections and waiting at least 15 minutes after any
charging or discharging has occurred. Sometimes it is helpful to first discharge the
batteries for a few minutes to remove any surface charge from the battery plates.
Remember to compensate for temperature.
Use the manufacturer's chart of state of charge vs. Voltage (or the figure presented in
the chapter on Battery Technology) to determine the state of charge from the voltage
reading. If the batteries must remain connected, then measure the charging or
discharging current. Use the manufacturer's charts for voltage at different currents to
estimate the battery state of charge.
Presented below are typical values for state-of-charge, specific gravity and voltage, for
lead acid batteries in temperate climates (25 deg C):
Battery Voltage
SOC
Specific Gravity
12 volt
24 volt
100%
1.265
12.68
25.35
90%
1.250
12.60
25.20
80%
1.235
12.52
25.05
70%
1.225
12.44
24.88
60%
1.210
12.36
24.72
50%
1.190
12.28
24.56
40%
1.175
12.20
24.40
30%
1.160
12.10
24.20
20%
1.145
12.00
24.00
10%
1.130
11.85
23.70
0%
1.120
11.70
23.40
!"
!-
.-/
$$&'
19-35
Array Maintenance
Physical Considerations
Replace any broken modules. If full replacement is not possible at time of visit check if
the electrical requirements allow for wiring around the broken module, to allow
continued array output without the broken module.
Wash module surface as needed, using soft cloth and water. Perform cleaning in early
morning or wait until evening, to avoid thermal shock to glass. Perform washing only
when modules are not in direct sunlight, when the sun is positioned below the horizon.
Check module tilt angle using tilt indicator or level and protractor. Tilt angle should be
within five degrees of that specified by sizing calculations.
Check module azimuth angle using compass. Compensate for local deviations of true
or solar North from magnetic North. Normally azimuth is true south in the Northern
Hemisphere and true north in the Southern Hemisphere. However different angles may
be chosen depending on local weather conditions and load profile matching. Azimuth
angle should be within 15 degrees of that specified by sizing calculations.
Confirm that no objects cast shadows on the array surface between 9 a.m. and 3 p.m.
Verify that all bolts are secure, and that the mounting structure is well attached to any
foundation.
Examine all wiring connections for corrosion or looseness. Clean and tighten as
necessary.
Check that all junction boxes are covered and that any seals are in place.
Inspect modules for broken cells, de-lamination, and discoloration. These alone may
not result in any noticeable deterioration in electrical output, but may be useful if other
problems are observed.
19-36
Electrical Considerations
It is very helpful to periodically verify the electrical output of the array.
Measure Short Circuit Current (Isc) using clamp-on meter or passing current directly
through meter. Isc is directly proportional to sunlight intensity. Use a reference cell or
light meter to independently check light level. The ratio of light level to 1000 w/m2
should be the ratio of Isc to literature value.
Measure Open Circuit Voltage (Voc) using meter. Voc decreases with increasing
temperature. Use a reference cell or light meter to independently check light level and
a temperature gauge to measure ambient (air) temperature. Use the graph of cell
temperature above ambient temperature to estimate actual cell temperature. Calculate
drop in Voc for operation above 25 deg.C by using voltage factors presented in Chapter
8. Compare calculated value to measured value.
Actual module current will probably not be at literature values, due to low irradiance and
high temperatures. A rough guideline for module current into a battery is given below.
If a module is connected to a 12-volt battery and operating under common outdoor
conditions (irradiance about 800 w/m2 and ambient air temperature about 30 deg.C),
then the following currents should be observed:
19-37
3.5 amps
2.5 amps
2.4 amps
1.6 amps
6 ohms
# Modules in Series
# Modules in Parallel
R = Vmp/Imp = 6 ohms
(Vmp, Imp)
2
1
0 0
8
12
16
Voltage (volts)
20
0
R e s is tiv e L o a d S iz e d T o
C h e c k A rra y
R = 6 ohm s
# S e r ie s
# P a r a lle l
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
19-38
Load Maintenance
Lubricate all moving parts of motors and loads as recommended by the manufacturer.
Check refrigerator door seals. Clean refrigerator coils.
Clean or replace any air filters.
19-39
19-40
Troubleshooting
The following guide is presented for troubleshooting of generic photovoltaic power
systems. A more detailed and relevant guide can be created for a specific system with
input from the various component manufacturers.
Symptom Chart
Symptom
Probable Cause
Array
Wiring
Controller
Load
A,B,C,D,E
F,G,H
I,J,K,M
Load inoperative
A,B,C,D,E
I,L
F,G,H
I,J,N
F,G,H
I,M
!"
0
Exercise:
(
"#
!
7
.
19-41
Corrective Action
Array
A
B
C
D
E
Loose wires
Improper wiring
Broken module
Shading
Improper
orientation
Tighten
Reconfigure
Replace
Remove obstruction, or relocate array
Reorient
Excessive voltage
drop
High resistance
connections
Short circuit
Wiring
F
G
H
Controller
I
J
K
L
M
N
Loose
connections at
terminal block
High resistance at
relay contacts
Inoperative
Relay fused shut
Electrical
interference
Load shedding
occurred
Tighten
Replace controller
Replace controller
Replace controller
Install filter, or remove source
Wait for battery to recharge, check actual
load usage against sizing calculations.
Load
O
P
Q
Failed
Excessive current
demand
Generating
electrical noise
!"
19-42
Field Evaluation
An example of useful field evaluation procedures is presented next. Field technicians
or engineers investigating the status or performance of an installed system can use
such worksheets. All parameters may not apply in all systems, but these provide you
with the basis for a thorough assessment of how a system was installed and is working.
The first field evaluation worksheet is for complete system evaluation. This can be
used for commissioning a system or for checking a system after some time of
operation. It is especially useful in the context of checking a system installed by some
other organization than your own, either by government agencies checking the quality
of subsidized systems, or by a company called in to service a system installed by some
other organization.
The second field worksheet is for checking the operation set points of a charge
regulator. This simple procedure is needed in the field because the system may not
cycle through its extremes of operational values during a typical field visit within a short
period of time. The field technician or engineer must force the regulator to operate to
its set point values to check for proper operation.
19-43
Array
(V)
(A)
(A)
(W)
__________
__________
__________ (V)
__________
__________
(V)
(A)
(A)
(W)
BATTERY SUBSYSTEM
Battery Manufacturer:
Battery Chemistry/Type
Battery Bank Configuration
Total Number of Batteries
_______________
Model #: _______________
_______________
(antimony, calcium, VRLA)
_______________ series X _______________ parallel
_______________ (#)
o
Battery Specifications at 25 C
Cell/Battery
Capacity
__________ (Ah)
Nominal Voltage
__________ (V)
Bank
__________ (Ah) Rate __________ (C/#)
__________ (V)
Open-Circuit Voltage
at 100% SOC __________ (V)
at 50% SOC __________ (V)
Specific Gravity at 100% SOC __________ (#)
at 50% SOC __________ (#)
Cycle Life
19-44
ELECTRICAL LOADS
1.
2.
3.
4.
5.
6.
7.
8.
Device
______________
______________
______________
______________
______________
______________
______________
______________
DC/AC
__________
__________
__________
__________
__________
__________
__________
__________
Amps/Watts
__________
__________
__________
__________
__________
__________
__________
__________
Hrs/Day
__________
__________
__________
__________
__________
__________
__________
__________
Ah/Day
__________
__________
__________
__________
__________
__________
__________
__________
____________________
Model # _______________
____________________ (series, shunt, PWM, constant-voltage)
Controller Specifications:
Nominal Voltage
Maximum Rated Currents: PV Array
________ (V)
________ (A)
Load
________ (A)
________ (V)
________ (SS/EM)
VRH:
Leg:
________ (V)
________ (pos/neg)
________ (V)
________ (SS, EM)
19-45
INVERTER/POWER CONDITIONER
Manufacturer: ____________________
Model #: _______________
Waveform
____________________ (sine, quasi-sine, square)
Output Voltage Regulation
__________ (%)
Harmonic Distortion
__________ (%)
Minimum Input/Operating Voltage
__________ (V)
LVD Function On Inverter DC Input
__________ yes
________ no
________ (V)
19-46
Sub Array
Array
________ (V)
________ (V)
________ (A)
________ (A)
2
________ (W/m )
O
O
________ ( CR) Ambient
________ ( CR)
2
________ (A)
________ (V)
________ (A)
Battery Subsystem
o
3. ______ (V)
6. ______ (V)
9. ______ (V)
12.______ (V)
Maximum ________ (V)
Specific Gravities:
3. ______ (#)
6. ______ (#)
9. ______ (#)
12.______ (#)
Maximum ________ (V)
________ (A)
________ (Vend)
________ (min)
________ (Vrest, min)
Electrical Loads
DC Load Currents 1. ________ (A) 2. ________ (A) 3. ________ (A) Voltage ______ (V)
4. ________ (A) 5. ________ (A) 6. ________ (A)
AC Load Power
Total DC Load Current ________ (A) Total AC Load Power ________ (W)
All Operational? ________ yes ________ no
Special Load Operational Performance (water volume/head, illumination levels, etc.)
_________________________________________________________________________________
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
19-47
________ (VR)
________ (LVD)
________ yes
________ (A)
________ (VRH)
________ (LVDH)
________ no
Inverter Operation
DC Voltage (V)
__________
__________
__________
DC Current (A)
__________
__________
__________
DC Power (W)
__________
__________
__________
AC Power (W)
__________
__________
__________
Efficiency (%)
__________
__________
__________
__________ (W)
System Operation
Steady-State Operation (no controller regulation, no changes in load or irradiance)
PV Array Current
________ (A)
DC Load Current
________ (A)
Inverter DC Current
________ (A)
Controller Quiescent Current
________ (A)
Sum of all currents should add to zero _________ (check)
19-48
19-49
For these reasons, a procedure is needed to force the regulator to operate to its set point values, to allow
the field personnel to witness first-hand whether the regulator is operating within acceptable limits.
Regulation Voltage (VR): the maximum voltage the controller allows the battery to reach before the
array is disconnected.
Regulation Voltage Hysteresis (VRH): the voltage span between the VR and the voltage at which
the array is reconnected to the battery.
Low Voltage Disconnect (LVD): the battery voltage at which the controller disconnects the load,
defining the maximum DOD.
Low Voltage Disconnect Hysteresis (LVDH): the voltage between the LVD and the voltage at which
the load is reconnected.
Battery Voltage
Charging
Discharging
Time
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
19-50
Disconnect the PV array, battery and DC loads via switches or circuit breakers provided in the system.
Remove all conductors from the terminals of the charge controller.
Connect a small battery or electrolytic capacitor to the charge controller battery terminals.
The battery voltage must be the same as the nominal charge controller voltage.
If a battery is used, the capacity should be low, approximately 5-10 amp-hours (ensures that the battery can be cycled in a
short period of time)
If a capacitor is used, the working voltage must be above the array open-circuit voltage. The capacitance should be between
5,000 and 50,000 micro-farads.
The array nominal voltage must be the same as the nominal charge controller voltage.
The module (or series string) should have a low output current, but large enough to quickly charge the small battery if used.
Approximately one to three amp modules should be sufficient.
The resistive load should be large enough to discharge the battery at a high rate, at least C/1 but not greater than C/0.25.
Furthermore, the load must consume a lower current than the test PV array provides.
The charge controller should now be cycling between the voltage regulation and array reconnect set
points.
Connect a voltmeter to the charge controller battery terminals.
The voltmeter must have the capability to record minimum and maximum voltages.
Wait for the controller to complete a full regulation cycle, and note the minimum and maximum voltages
recorded on the meter.
These voltages will be the array regulation (VR) and array reconnect set points.
Next, reset the minimum and maximum voltage recording for the voltmeter.
Disconnect the test array from the charge controller array terminals (or shade the array).
Leave the load connected and adjust the load for a current higher than the PV array current.
Reconnect (or uncover) the array and wait until a maximum voltage is recorded.
19-51
Charge Controller
Under Test
Test PV Module
or Array
Variable
Resistance
Load
Voltmeter
Small Battery or
Electrolytic Capacitor
19-52
Special Considerations
The relative sizes of the test battery (or capacitor), the PV array and test load must be correct to
ensure the operation of the controller between all set points.
Disconnects and overcurrent devices should be used on the battery connection to the charge
controller during the tests.
Special controller algorithms and timing circuits may require special considerations for these tests.
Repeat the test a few times to ensure that accurate data is measured. Average the final results from
all of the measurements.
19-53
[End of Chapter]
19-54
CHAPTER NINETEEN
INSTALLATION AND TROUBLESHOOTING
19-1
Safety Issues
National Electric Code
General Safety Practices
Safe Practices With Batteries
19-2
19-2
19-3
19-4
19-7
19-9
19-16
19-21
19-22
19-25
Installation Processes
Foundation Alignment
Array Assembly Procedure
Battery Bank Assembly
Control Center
Other Wiring
Commissioning
19-27
19-27
19-27
19-30
19-30
19-31
19-31
19-32
19-32
19-33
19-34
19-36
19-39
19-39
19-40
19-41
Field Evaluation
19-43
19-55
Chapter Twenty
Economic Analysis
It is not uncommon to have a request to compare renewable generators (wind, PV
hydro) with non-renewable generators (diesel engines, grid-connected systems). It
is very difficult to compare these technologies fairly since their cost structure is
entirely different. For example, a diesel generator has a low initial cost while a PV
system to supply the same energy requirement is significantly more expensive. On
the other hand, the PV system, once installed, uses no fuel and has very low
maintenance costs, while the diesel generator requires constant purchases of fuel
and maintenance on a regular basis. An approach to economic analysis that looks
at the total life costs of these different systems is required. This is discussed next.
20-1
System Costs
To properly account for all the expenses in a power system that operates over many
years, three different types of costs should be analyzed. These are: (1) initial capital
costs of purchasing equipment and installation; (2) recurring costs that occur every
year of operation such as fuel and maintenance; and (3) nonrecurring costs that may
occur on an irregular basis, such as equipment replacement or repairs. Each of
these costs is discussed next.
Capital Costs
Generators
The cost of a generator which uses fossil fuel (gasoline / petrol, diesel, or LPG) as
its fuel source varies depending on the construction of the engine. Typically,
gasoline is used for portable generators because of the high weight / power ratio of
these engines. However, these engines are typically only designed for a lifetime of
2,000 hours or less, and require near continuous maintenance. In addition, the
storage problems (flammability and "shelf life") associated with gasoline make
diesel or LPG the fuel of choice for prime power generation.
Diesel generators are typically specified either as "standby" or "prime" duty cycle
engines. The prime rating is usually less than 90% of the standby rating. In
addition, prime diesels usually have longer maintenance intervals than "standby"
engines, savings costs later in the life cycle process. Typical life of a diesel
generator is 20,000 to 40,000 hours. Single generators are usually sized to meet
annual peak loads, which means that there are typically sized at 6-8 times the
average load.
The cost of generators depends on the fuel, "duty cycle" and size. Gasoline
generators are typically less expensive than diesel or LPG generators. Units
designed for "prime" duty are usually significantly more expensive than those rated
for "contractor" or "standby" duty. Finally, larger generators typically cost much less
than small generators. AA typical example is that a generator which is four times as
large will only cost two as much as a base case generator.
For specific examples, a recent "Grainger" catalog lists a gasoline powered 4 kVA
"contractor" generator at $1,089 (approx. $270 per kVA), while a "standby" diesel
generator rated for 8 kW continuous sells for $7,500 (approx. $940 per kW of
generator capacity) and a 35 kW standby generator sells for approximately $15,000
(approx. $430 per kW).
20-2
BOS costs such as control systems, fuel tanks, fuel containment structures,
generator buildings, redundancy (dual engine systems) and storage / power
electronics (batteries and rectifiers for telecom systems) add additional costs to a
generator system. Translated into terms of $/kW of load, the capital cost of diesel
systems is usually $10K-$30K per kW of load.
PV Modules
PV modules are typically sold in terms of $/W(p) (or "dollars per peak watt") for a
specific module size. Although there are some discounts for bulk purchases, the
cost of a PV system over a wide range of power outputs is much more linear than
the cost of a diesel generator.
A typical PV array will cost $6,000 to $8,000 per rated kilowatt ($36K-$48K per kW
of load), plus array structure and BOS. The array that is required to meet a specific
load is typically sized for worst case solar insolation, and is thus 6-8 times the rating
of the load. Thus a 6 kW(p) PV array might be required to meet a 1 kW load in a
typical environment.
Batteries
PV systems usually have batteries equivalent to 5-10 days of storage, with the
batteries rated to 80% of design capacity. Thus, a 1 kW PV system would have
120-240 kWh of useful capacity, equivalent to 150 to 300 kWh of rated capacity.
Batteries usually cost $125-$250 per kWh of capacity, so the net cost of a battery for
a 1 kW load PV system is $19K-75K.
Diesel based telecom systems usually have batteries rated to 3-8 hours of load, and
so are significantly less expensive than PV batteries.
BOS
A PV system must include a support structure for the PV array, charge controller to
regulate battery charging, and housing for the system battery. A diesel based
telecom system usually includes housing for the telecom equipment, batteries, and
battery chargers and often for the diesel generators as well.
Line Extensions
Capital costs for line extensions are usually based on the distance that the load is
away from an established distribution transformer. Typical line extension costs for a
small system range from $1K to $5K per km, plus the actual connection charges ($12K). Stringent environmental requirements for remote sites can increase the cost of
extension by up to a factor of 10 times.
20-3
Recurring Costs
Recurring costs are those costs that are incurred every year of operation. For a
fossil-fuel generator, this includes fuel and engine maintenance costs, while for a PV
system, it is typically limited to checking the batteries one to four times per year.
Fuel costs are usually segregated from maintenance costs because of a difference
in inflation rates between the two categories. If the system is financed, the recurring
costs include loan payments.
Fuel
Annual fuel costs are dependent on the fuel consumption and the cost of fuel at the
site. Fuel consumption in a gas or diesel generator is a function of how highly the
engine is loaded. For example, a diesel engine that is loaded at 25% of its rated
capacity will still use 40% of its full load fuel consumption. Another way to express
these numbers is as specific fuel consumption or kWh per liter of fuel. A fully loaded
diesel has a specific fuel consumption of approximately 3.1 kWh/L, meaning that it
will deliver 3.1 kWh for every liter of fuel consumed. At 25% load, specific fuel
consumption goes down to approximately 1.9 kWh per liter, while at 12% load the
engine will only deliver approximately 1.2 kWh for every liter of fuel used. This is a
very important factor in comparing renewable and nonrenewable technologies, since
most prime diesel plants operate at an average load of between 12% and 25% of
rated capacity. Well-designed hybrid power systems, however, operate the engine
at 75-95% of its rated load, making them much more fuel-efficient.
The cost of fuel at the site is the sum of the purchase price of fuel plus the delivery
costs. While delivery costs for some sites are negligible, the delivery costs at many
remote sites (mountaintop telecom installations, remote villages, and island
communities) can easily double or even triple the purchase price of fuel.
Maintenance
System maintenance is the other recurring (annual) cost. This category includes
equipment maintenance, site maintenance, system supervision, etc.
For PV and wind systems, the generators typically require very little maintenance.
However, if there are batteries in the systems, these will require inspection and
topping up approximately every 3-6 month.
20-4
Qualifications
Can be
performed by a
Also inspect air and fuel filters, trained technician
fuel system, starter battery, and at the site.
system electrical connections.
1500 hours Decarbonization (top end
Can only be
overhaul). In addition to oil
performed by a
change tasks, replace air and
trained engine
fuel filters. Remove the
mechanic, site
cylinder head and clean valves, work is possible.
valve seats, injector nozzles,
etc. Replace all top end
gaskets.
6000 hours Full overhaul. In addition to
Requires a
tasks listed above, perform full competent
engine overall. Clean all
mechanic.
combustion and inlet/outlet
Typically
chambers. Replace all engine
performed in a
gaskets. Replace camshaft /
repair shop which
crankshaft bearings, valves,
means the engine
valve springs, injectors, fuel
must be
pumps, pistons, piston rings,
transported to the
starter battery, start solenoid
shop, and a
and engine starter, fuel
replacement
solenoid, fuel pipes, and other
engine provided
parts as necessary.
to the site.
Cost Estimate
$25 to $100 plus
travel if required.
$250-500 plus
travel for a typical
two-cylinder aircooled engine.
$1,500-3,000,
plus
transportation and
loaner of
replacement
engine.
For a prime power engine which is running continuously (8,760 hours per year), the
sample maintenance schedule listed above would require 35 oil changes, 6
decarbonizations, and 1.5 overhauls per year. Engines made for prime power often
have options that can extend the primary maintenance interval to 500, 1,000 or even
1,500 hours, dramatically reducing the maintenance requirements. Note that many
standby engines have only a 100-hour oil change interval, which would necessitate
nearly 90 oil changes per year almost twice per week!
Loan Payments
If the system is financed through a bank or loan agency, the amortized loan
payments can be treated as a recurring cost rather than a capital cost.
20-5
Non-recurring Costs
The last major category of costs are those which only happen every once in a while.
These include replacement of major components, major system maintenance tasks,
etc. For remote power systems, typically non-recurring costs include replacements
of engines/generators and replacement of system batteries.
Engine Replacement
Generator lifetimes are typically dependent on the construction and fuel type. A
portable gasoline / petrol generator will probably last only 500-2,000 hours. A diesel
generator has an effective lifetime of 20,000 to 50,000 hours, and LPG engines that
are based on diesel construction can last 100,000 hours or more.
Battery Replacement
While many telecom-style batteries are rated for 20 year lifetimes in float
applications, the harsh environments and cycling duty of remote PV or hybrid power
system typically limit useful battery life to 7-10 years at best. Life cycle cost
techniques can be used to help evaluate battery tradeoffs -- e.g. should you use a
high quality battery which will last 14 years, or use a lower cost battery and simply
plan on replacing it every 5-7 years.
Exercise:
20-6
Economic Factors
Life Cycle Term (term)
In performing a life cycle cost analysis, the primary economic factor is the "life cycle" of the
system. The length of "term" of the analysis is chosen to be the service life of the longestlived component. For example, a system which compared the costs of a diesel generator
and PV system over the course of a only a single year would ignore the fact that the PV
system will continue to operate for many years without additional expenses.
In the case of PV system comparisons, the useful life of a PV module is on the order of 2030 years, so this should be the "term" chosen for the analysis. Similarly, a comparison
between standard "prime" diesel generators and "cycle charged" systems would probably be
based on the life of the battery in the cycle charged system, or about 10 years.
20-7
A full tank of fuel for a generator costs $100 today. It will need to be replaced
next year. If the discount rate is 10%, then the amount of money needed
today to pay for $100 of fuel next year will be given by
PV
1
1+ 0.10
$100
$90.91
So $90.91 is the present value of the future $100 needed to pay for the fuel in
one year. If the $90.91 was put aside this year and invested at the discount
rate of 5%, then in one year we would have $100.
If there is price escalation happening as well, then the future cost of the tank
of fuel will be slightly higher than the price today. If the price escalation of
diesel fuel is 5%, then the price in one year will be given by
FV
(1 + 0.05)
$100
$105.00
So you would need to have $105.00 in one year to pay for the fuel.
The NPV of next years fuel will be the escalated price discounted back to the
present. This involves combining the two formulas above.
NPV
(1 + 0.05)
(1+ 0.10)
$100
$95.45
So the net present value of the tank of fuel to be purchased next year is
$95.45 in todays money. Putting this much aside today will increase in value
at the discount rate of 10% and be enough to cover the escalated price of
$105.00 that will be needed one year from now.
20-8
Exercise:
!
!
"
#
$
%
&
'
(
!
)
*
'
!
+
,
*
-.
)
&
/01+!222
/032!222
)
-
&
/0+2!222
/04!+22
* /
)
03222
5$
%$% !
3+6
76
76
16
20-9
Capital Costs
Capital costs are input into the formula with a factor of 1.0 since they are assumed
to occur in "year 0," that is, before the start of system operation.
20-10
DR FE 1 + DR
(Eqn 20.1)
where
AnnFuelCost
FE
DR
Term
(Note: If the discount rate is equal to the fuel escalation rate, the first term has a
"division by zero" which will cause it to go to infinity. This can be remedied by using
a number with a very small difference, i.e. 0.06999 instead of 0.07 for a 7% discount
rate. In any event, if the escalation factor is the same as the discount rate, then the
Fuel Factor (the term in the brackets) is simply equal to the life cycle term. So the
total life cycle fuel cost would simply be the annual fuel expenditure times the
number of years in the term. The same concern applies to the Maintenance Factor
described in the next section)
Recurring Maintenance
The formula for life cycle maintenance costs is:
1 + GE 1 + GE Term
LCCMaintCost = AnnMaintCost *
1
DR GE 1 + DR
(Eqn 20.2)
where
AnnFuelCost
GE
DR
Term
20-11
Non-Recurring Costs
The formula for nonrecurring costs is:
LCCReplCost =
1 + GE RY
ItemCost
*
1 + DR
(Eqn 20.3)
where
ItemCost
GE
DR
RY
Note that the costs for each non-recurring cost need to be calculated separately,
then summed together to get the total nonrecurring costs.
(Eqn 20.4)
where
LCC$/kWh
LCCFuelCost
is calculated above
LCCMainCost
is calculated above
LCCReplCost
is calculated above
Term
kWh/d
20-12
Exercise:
8
!
,
5595
.
0
,
55:
5
.
0
,
55"5
.
0
,
*,
550;*,.
0;*,
5
'
<
9<
"
*,
&
5<@
9
<@
:
5
$
A
5
"
B
-
'<
9<
"
?,;
55
95
42
=6
>6
346
42
0+2!222
04!222
:
5
0+22
&5
3
032!222
"B
B32
20-13
Sample Analyses
This section presents analysis of four different power system options for a sample
application, a small village power supply. The costs associated with these
calculations are estimated only -- they are not intended to be a formal analysis of the
options presented. Rather, they are intended to show the method used and how a
comparison of various power solution options might be presented. A thorough
evaluation would require more detailed designs, actual system and operating costs,
and a careful selection of economic factors.
The analyses will compare four typical energy solutions:
20-14
Site Factors
Location
Madras,
India
13 deg.N
Load Factors
Daily
Economic Factors
Life Cycle
20 years
Term
Latitude
Annual
15,330
Discount
12%
kWh
Rate
Longitude
80 deg.E Total in
306 MWh Fuel
10%
Life Cycle
Escalation
Distance
15 km
Surge
20-40 kW General
7%
from grid
Estimate
Escalation
Fuel Cost
$0.40 /liter
(delivered)
42 kWh
20-15
PV System
A PV array to meet this load in this location would be approximately 16.5 kWp (4
series X 55 parallel 75 Wp modules). A battery bank with 5 days of autonomy is
recommended with a total rated capacity of 196 kWh DC @ 10 hour rate (4080 Ah
@ 10 hr @ 48 volts).
The major cost during the life of the system would be the replacement of the battery
bank. It is estimated that the battery will need replacement after year 7 and again
after year 14. A small cost for battery maintenance trips during each year is
included.
The resulting LCC analysis shows that the total life cycle (20 year) system cost,
including initial capital and maintenance and battery replacement, comes to
$168,486 NPV (discounted to constant dollars). The system produces 306,000 kWh
in 20 years, for a LCC Cost/Energy ratio of $0.55/kWh.
20-16
L o c atio n
M ad ra s, Ind ia
R ec u rrin g C o sts
Fuel
--------------
A nn ua l H ou rs
--------------
M ainte na nc e
G en era l
Q ua ntity
2 20
1 96
1
1 0.0
0
0
C os t
$ 42 0.0 0
$ 0.1 4
$ 1,0 00
$ 0.7 0
$0
$0
E xte nd ed C os t
$ 92 ,40 0
$ 27 ,44 0
$ 1,0 00
$ 7,0 00
$0
$0
$0
$0
$0
$0
T o ta l Initia l C ap ita l
$ 12 7,8 40
L ite rs/h ou r
--------------
C os t/u nit
--------------
F re qu en cy/yea r
1 .0
C os t/E ve nt
$ 35 0
A nn ua l C os t
--------------
E xte nd ed C os t
$ 35 0
$0
$0
$0
$0
T o ta l R e currin g C o st
$ 35 0
N o n -R e c u rrin g Item s
Item
C os t (tod ay)
B attery R ep l + 10 % la bo r
$ 30 ,18 4
B attery R ep l + 10 % la bo r
$ 30 ,18 4
Economic Factors
Item
LCC Period (years)
Discount Rate
Fuel Escalation (FE)
General Escalation (FE)
A ve ra g e E n e rg y R eq u ire m e n t
4 2 kW h/da y
Value
20
12%
10%
7%
Y ea r
N P V F ac tor
1
0 .95 5
2
0 .91 3
3
0 .87 2
4
0 .83 3
5
0 .79 6
6
0 .76 0
7
0 .72 6
8
0 .69 4
9
0 .66 3
10
0 .63 3
11
0 .60 5
12
0 .57 8
13
0 .55 2
14
0 .52 8
15
0 .50 4
16
0 .48 2
17
0 .46 0
18
0 .44 0
19
0 .42 0
20
0 .40 1
T o ta l N o n-R ec urrin g C o st
N P V C os t
$0
$0
$0
$0
$0
$0
$0
$ 20 ,94 6
$0
$0
$0
$0
$0
$0
$ 15 ,21 5
$0
$0
$0
$0
$0
$ 36 ,16 1
NPV Factor
16.64
12.82
Cost Summary
Initial Capital
All Non-Recurring Costs
Annual Fuel
Annual Maintenance
$127,840
$36,161
Annual Cost
-------------$350
NPV Factor
16.64
12.82
20-17
The graph below illustrates that the array is sized large enough to operate all the
loads even during the lowest insolation month (July), and that there is excess array
energy produced in every other month of the year. The array is deliberately
oversized so that it can adequately operate the load in every month. The oversizing
is evident by the fact that the array potential is larger than the load demand
throughout the year (except for July). The largest oversizing is evident in March,
where the array could have produced almost 40% more energy than was needed by
the load.
No generator is included in the standalone system, so there are no generator starts
in any year.
S ys t e m P e rf o rm a n c e
1
2000
1800
1600
1400
1200
1000
800
600
400
200
0
JAN
0
FEB
MAR
APR
G e n S ta rts
MAY
JUN
JUL
M o nthly Lo a d
AU G
SEP
O CT
A rra y (us e d )
N OV
DEC
A rra y (p o te ntia l)
!
20-18
A cash flow analysis shows that there is a large initial outlay for all the equipment,
and a battery replacement in year eleven. All other costs are small. The total
accumulated cash flow (not discounted) comes to $278,333 over the life of the
system.
$140,000
A c c u m u la te d C o s ts
$ 2 7 8 ,3 3 3
$120,000
$250,000
$100,000
An n u a l C o s t
$200,000
$80,000
$150,000
$60,000
$100,000
$40,000
$50,000
$20,000
$0
$0
1
10 11 12 13 14 15 16 17 18 19 20 21
Ye a r
"#$% &
20-19
Prime Generator
A prime diesel generator power system would include two (redundant) generators,
to extend life and insure reliable power availability. They will be sized at 20 kW
each, but could range from 20-40 kW, to handle the possible surge requirements of
a variable village load profile (recall the discussion of surge requirements for
complex large systems in the chapter on Loads). The lowest of the possible sizes is
chosen here, to be most conservative in the cost comparison.
The generators would be alternated, so they would both age approximately the
same, each operating for 4380 hours per year. The manufacturer would
recommend replacement after 30,000 hours of this relatively continuous operation,
which would occur after 7 years of system operation (30,000 hours / 4380 hours
each generator = 6.8 years), and then again after year 14 of the system life.
Fuel and maintenance costs would add up to a considerable expense during the
system life. The fuel consumption rate is estimated to be 3.0 liters/hour as the
generator is expected to operate on the average at only about 25% load. Operating
for 8760 hours/year at this rate consumes 26,400 liters/year at a cost of $10,775.
Maintenance is expected for oil changes every 250 hours, decoking every 1,500
hours and major overhaul every 6,000 hours (with full replacement needed after
30,000 hours).
The resulting LCC analysis shows that the total life cycle (20 year) system cost,
including initial capital and fuel, maintenance and generator replacement, comes to
$354,816 (discounted to constant dollars). The system produces 175,200 kWh in 20
years, for a LCC Cost/Energy ratio of $1.16/kWh.
20-20
Location
Madras, India
Recurring Costs
Fuel
Diesel fuel (liters)
Maintenance
Oil Change
Decoke
Overhaul
General
Non-Recurring Items
Item
Quantity
2
1
0
1
0
Cost
$12,000
$1,000
$10,000
$1,000
$10,000
Extended Cost
$24,000
$1,000
$0
$1,000
$0
$0
$0
$0
$0
$0
Total Initial Capital
$26,000
Liters/hour
Annual Hours @ 25% Loa d
8760
3.00
3.0
Period (hours) Frequency/year
250
35.0
1500
5.8
6000
1.5
0.0
Cost/unit
$0.41
Annual Cost
$10,775
Extended Cost
$5,256
$1,752
$2,190
$0
$0
Total Recurring Cost
$9,198
Cost (today)
$26,400
$26,400
Economic Factors
Item
LCC Period (years)
Discount Rate
Fuel Escalation (FE)
General Escalation (FE)
Value
20
12%
10%
7%
Cost/Event
$150
$300
$1,500
$300
Year
NPV Factor
1
0.955
2
0.913
3
0.872
4
0.833
5
0.796
6
0.760
7
0.726
8
0.694
9
0.663
10
0.633
11
0.605
12
0.578
13
0.552
14
0.528
15
0.504
16
0.482
17
0.460
18
0.440
19
0.420
20
0.401
Total Non-Recurring Cost
NPV Cost
$0
$0
$0
$0
$0
$0
$0
$18,320
$0
$0
$0
$0
$0
$0
$13,307
$0
$0
$0
$0
$0
$31,627
NPV Factor
16.64
12.82
Cost Summary
Initial Capital
All Non-Recurring Costs
Annual Fuel
Annual Maintenance
$26,000
$31,627
Annual Cost
$10,775
$9,198
NPV Factor
16.64
12.82
20-21
A cash flow analysis shows that there is an initial outlay for all the equipment, and
then generator replacement in year 8 and 15 of system life. The escalated cost of
generators after 7 and 14 years is shown. The fuel and maintenance costs
continually add up during system life, and also increase in cost as their values
escalate over time. The total accumulated cash flow (not discounted) comes to
$420,312 over the life of the system.
This analysis shows the sensitivity of long-term prime diesel system operation cost
to the discount rate value used. There are substantial costs occurring in far future
years. The higher the discount rate, the more discounted the far future costs are
reduced in the LCC type analysis (with all future costs being brought back to initial
constant dollars). So different discount rates can end up drastically affecting the life
cycle cost of such a system with much of its cost occurring in the far future.
C a s h F lo w A n a ly s is - P rim e D ie s e l
$1,400,000
$160,000
A c c u m u la te d C o s ts
$1,226,510
$140,000
$1,200,000
$120,000
$100,000
$800,000
$80,000
$600,000
$60,000
Accumulate d C ost
An nu al C ost
$1,000,000
$400,000
$40,000
$200,000
$20,000
$0
$0
1
10 11 12 13 14 15 16 17 18 19 20 21
Y e ar
20-22
PV-Diesel Hybrid
The hybrid system designed for this example included enough solar array to supply
40% of the annual energy load, with the generator operating to provide the
remaining energy throughout the year. The battery needed for such a system can
be quite small, as there is no need for many days of autonomy because the
generator is available on demand. A battery bank of 2 days of autonomy was
used, with the generator turning on and recharging the battery whenever it dropped
to 75% depth of discharge (deep cycling batteries). This resulted in a total battery
capacity of only 101 kWh (C/10 rate), a reduction of 50% of the capacity needed for
the Pure PV system.
One generator is installed with a capacity of 13 kW, sized based on charging the
battery bank at its maximum allowable charge rate. The fuel consumption rate is
estimated to be 4.8 liters/hour, as the generator is expected to operate at full load
when it is needed to recharge the battery. This is of course larger than the rate used
for the prime diesel example, because the generator was only operating on the
average at about 1/4 of full load. The hours of operation at this full rate are now
estimated to be only about 1296 hours/year (instead of 8760 hours), only about 15%
of the amount of hours for the prime diesel system example.
Operating for 1296 hours/year at this full power rate consumes 6220 liters/year at a
cost of $2,550. This is less than 25% of the cost for the prime diesel system.
Maintenance is estimated at the same periods as for the prime diesel system, with
oil changes every 250 hours, decoking every 1,500 hours and major overhaul every
6,000 hours. Maintenance costs are estimated to come to $1253/year, less than
15% of the prime diesel costs.
At the end of the full system life term of 20 years, the diesel is expected to have
operated on 25,920 hours total (20 years X 1296 hours/year), less than the
estimated life of 30,000 hours. So the generator will not have to be replaced during
the system life, compared to two replacements during the prime diesel systems life.
The battery life is expected to be similar to the pure standalone system, due to the
more intensive deep cycling of the batteries but less sulfation resulting from less
time at partial state of charge. The generator is expected to start a total of about
108 times per year, so the batteries will cycle down to 75% depth of discharge at
least that many times. This would mean about 756 such deep cycles in 7 years.
Although some battery manufacturer literature states longer expected cycle life to
this depth of discharge, the accumulated loss of life over real time was estimated to
result in the need to replace the batteries after year 7 and again after year 14 of the
system life.
The resulting LCC analysis shows that the total life cycle (20 year) system cost,
including initial capital and fuel, maintenance and battery replacement, comes to
$141,300 (discounted to constant dollars). The system produces 175,200 kWh in 20
years, for a LCC Cost/Energy ratio of $0.46/kWh.
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
20-23
76
SP75
5.7 kW
Ge ne ra tor Info:
Pow e r:
Fue l Co nsum p tion
Outp ut De ra ting Fa c to r
13 kW
4.80 lite rs/ ho ur
100 %
2
88
90
132
101
2745
48
75
85 %
80 %
275 a m p s
Da ily
Mo nthly Arra y kWh Ge n kWh
Loa d kWh Loa d kWh Sup p ly
Sup p ly
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NO V
DEC
ANNUAL To ta ls:
PV/ Diesel Ra tio
42.0
42.0
42.0
42.0
42.0
42.0
42.0
42.0
42.0
42.0
42.0
42.0
d a ys
%
%
kWh @ C/ 108
kWh @ C/ 10
Ah
Vo lts
%
# G en
Sta rts
Ho urs of
Fue l
Op era tionUse d (liters)
1302.0
1176.0
1302.0
1260.0
1302.0
1260.0
1302.0
1302.0
1260.0
1302.0
1260.0
1302.0
562.3
553.5
627.2
579.1
545.5
453.5
422.9
461.4
486.1
509.5
481.4
509.5
672.0
672.0
672.0
672.0
756.0
756.0
924.0
840.0
756.0
840.0
756.0
756.0
8
8
8
8
9
9
11
10
9
10
9
9
96.0
96.0
96.0
96.0
108.0
108.0
132.0
120.0
108.0
120.0
108.0
108.0
461
461
461
461
518
518
633
576
518
576
518
518
15330
6192
9072
108
1296
6220
PV 40%
Ge ne ra tor 59%
20-24
Location
Madras, India
Quantity
1
1
0
1
76
101
1
10.0
0
Recurring Costs
Fuel
Diesel fuel
Annual Hours
1296
Liters/hr
4.80
Maintenance
Oil Change
Decoke
Overhaul
General
Non-Recurring Items
Item
Cost
$6,000
$1,000
$10,000
$0
$420.00
$0.14
$1,000
$1.00
$10,000
Extended Cost
$6,000
$1,000
$0
$0
$31,920
$14,190
$1,000
$10,000
$0
$0
Total Initial Capital
$64,110
Cost/liter
$0.41
Annual Cost
$2,550
Extended Cost
$777
$259
$216
$0
$0
Total Recurring Cost
$1,253
Cost (today)
$15,609
$15,609
Economic Factors
Item
LCC Period (years)
Discount Rate
Fuel Escalation (FE)
General Escalation (FE)
Value
20
12%
10%
7%
Cost/Event
$150
$300
$1,000
$300
Year
NPV Factor
1
0.955
2
0.913
3
0.872
4
0.833
5
0.796
6
0.760
7
0.726
8
0.694
9
0.663
10
0.633
11
0.605
12
0.578
13
0.552
14
0.528
15
0.504
16
0.482
17
0.460
18
0.440
19
0.420
20
0.401
Total Non-Recurring Cost
NPV Cost
$0
$0
$0
$0
$0
$0
$0
$10,832
$0
$0
$0
$0
$0
$0
$7,868
$0
$0
$0
$0
$0
$18,700
NPV Factor
16.64
12.82
Cost Summary
Initial Capital
All Non-Recurring Costs
Annual Fuel
Annual Maintenance
$64,110
$18,700
Annual Cost
$2,550
$1,253
NPV Factor
16.64
12.82
20-25
The graph below shows the array and generator performance over a typical year.
The array is sized to meet only 40% of the load, so the generator is operated every
month, varying from 8 starts in January-April (about once every 3-4 days) to 11
starts in July (about once every 2-3 days). All of the array potential output is utilized,
in contrast to the pure standalone system where the array met the load in July and
was oversized all the rest of the year, peaking at nearly 40% oversize in March.
System Performance
1400
1200
1000
800
600
400
200
0
JAN
FEB
MAR
Gen Starts
APR
MAY
JUN
JUL
AUG
Monthly Load
SEP
OCT
Array (used)
NOV
12
11
10
9
8
7
6
5
4
3
2
1
0
DEC
Array (potential)
20-26
A cash flow analysis shows that there is an initial outlay for all the equipment, and
then battery replacement in year 8 and 15 of system life. The escalated cost of the
battery bank in those future years is shown. The fuel and maintenance costs are
quite small compared to the prime diesel system. The total accumulated cash flow
(not discounted) comes to $349,600 over the life of the system.
C a s h F lo w A n a ly s is - H y b r id S y s t e m
$ 7 0 ,0 0 0
$ 4 0 0 ,0 0 0
A c c u m u la te d C o s ts
$ 3 4 9 ,6 0 0
$ 3 5 0 ,0 0 0
$ 6 0 ,0 0 0
Annual C ost
$ 2 5 0 ,0 0 0
$ 4 0 ,0 0 0
$ 2 0 0 ,0 0 0
$ 3 0 ,0 0 0
$ 1 5 0 ,0 0 0
A c c u m u la te d C o s
$ 3 0 0 ,0 0 0
$ 5 0 ,0 0 0
$ 2 0 ,0 0 0
$ 1 0 0 ,0 0 0
$ 1 0 ,0 0 0
$ 5 0 ,0 0 0
$0
$0
1
10 11 12 13 14 15 16 17 18 19 20 21
Year
20-27
Grid Connect
A grid connection would entail the costs of extending the line, estimated for this
example at $6,000 / km, and the cost of a step-down transformer (estimated at
about $7,000). There is no diesel fuel, but there is an electricity charge estimated to
be $0.10/kWh, and an annual maintenance cost for the line of about $30/km.
The resulting LCC analysis shows that the total life cycle (20 year) system cost,
including initial capital and electricity and maintenance, comes to $128,279
(discounted to constant dollars). The system produces 175,200 kWh in 20 years, for
a LCC Cost/Energy ratio of $0.42/kWh.
Of course, this analysis does not account for any of the issues associated with grid
power in remote areas, such as power quality or outages. There can be real
economic costs attached to low voltages and poor power factor, in that customers
might have to pay for power conditioning equipment at their end of the wire, or may
have unacceptable operation from motors or electronic equipment. And power
brown outs or outages will have very measurable impact on productivity. This
aspect of the grid extension alternative do not show up from this simple type of
analysis, and should also be considered when comparing true costs of various
power solutions.
The cash flow analysis shows the large initial cost of extending the grid and then the
small-accumulated cost of the electricity and line maintenance over the life of the
system.
C as h F lo w An a lys is - U tility E x te n sio n
$120,000
$250,000
A c c u m u la te d C o s ts
$213,322
$100,000
An n u a l C o st
$80,000
$150,000
$60,000
$100,000
$40,000
Acc u m u la te d C o s t
$200,000
$50,000
$20,000
$0
$0
1
9 10 11 12 13 14 15 16 17 18 19 20 21
Ye a r
20-28
Location
Madras, India
Recurring Costs
Fuel
Electricity
Maintenance
Line Maintenance
Non-Recurring Items
Item
Economic Factors
Item
LCC Period (years)
Discount Rate
Fuel Escalation (FE)
General Escalation (FE)
Quantity
15
1
Cost
$6,000
$7,000
Extended Cost
$90,000
$7,000
$0
$0
$0
$0
$0
$0
$0
$0
Total Initial Capital
$97,000
Annual kWh
15,330
Cost / kWh
$0.10
Annual Cost
$1,533
Extended Cost
$450
$0
$0
$0
$0
Total Recurring Cost
$450
Cost (today)
Value
20
12%
10%
7%
Cost / km
$30
Year
NPV Factor
1
0.955
2
0.913
3
0.872
4
0.833
5
0.796
6
0.760
7
0.726
8
0.694
9
0.663
10
0.633
11
0.605
12
0.578
13
0.552
14
0.528
15
0.504
16
0.482
17
0.460
18
0.440
19
0.420
20
0.401
Total Non-Recurring Cost
NPV Cost
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
NPV Factor
16.64
12.82
Cost Summary
Initial Capital
All Non-Recurring Costs
Annual Fuel
Annual Maintenance
$97,000
$0
Annual Cost
$1,533
$450
NPV Factor
16.64
12.82
20-29
Comparison of Results
The real value of a LCC analysis is not in the absolute number but in a comparison
of values from various alternative choices. Now we can compare the four energy
solutions side-by-side, to see how their values and components relate.
Economic Factors
Hybrid
System
$64,110
$42,438
$16,052
$18,700
Prime Diesel
Pure PV
Initial Capital
$26,000
$127,840
Annual Fuel (NPV)
$179,314
$0
Annual Maintenance (NPV)
$117,874
$4,485
All Non-Recurring Costs
$31,627
$36,161
(NPV)
Total LCC Cost
$141,300
$354,816
$168,486
Cost/kWh (NPV)
$0.46
$1.16
$0.55
Accumulated Cash Flow
$349,600 $1,226,510 $278,333
/"" !
"#$"
Utility
Extension
$97,000
$25,512
$5,767
$0
$128,279
$0.42
$213,322
One useful method of comparison is to graph the component costs. Each solution is
graphed below, with the initial capital, annual fuel and maintenance costs, and any
major non-recurring expenses shown. This makes it easier to see which alternative
has the largest initial capital costs, and which has greater costs over time. For
example, the diesel has the lowest initial costs of all, but accumulates the greatest
operating costs of all.
$350,000
$300,000
Annual
Maintenance
(NPV)
$250,000
$200,000
$150,000
Annual Fuel
(NPV)
$100,000
$50,000
Initial Capital
$0
Hybrid
System
Prime Diesel
Pure PV
Utility
Extension
Siemens Solar Basic PV Technology Course
Copyright 1998 Siemens Solar Industries
20-30
Notice that the benefits of the hybrid solution over the pure PV solution are not just
in a lower LCC, but also in a dramatically lower initial cost.
Another interesting method of comparison is to look at the cash flows over time.
Each alternative is graphed below against the same scale, to more easily see when
their accumulated (non-discounted) expenditures cross one another. Each system
is assumed to be producing the same amount of total energy annually, so this
comparison shows when each system becomes cost effective compared to the
others.
For example, in the first years of the systems, the prime diesel system is producing
energy for the lowest accumulated cost. At about 5 years, the accumulated costs of
fuel and maintenance push the prime diesel system above the total accumulated
cost of the hybrid system. Therefore after about 5 years, the hybrid system is the
most cost effective (total energy / total expenditure). And after about 7 years, even
the pure standalone PV system, with its very high initial costs, is more cost effective
than the prime diesel system.
$1,400,000
$1,200,000
$1,000,000
$800,000
$600,000
$400,000
$200,000
$0
0
10
12
14
16
18
20
Year
Hybrid System
Prime Diesel
Pure PV
Utility Extension
20-31
Parametric Analysis
In this analysis, various factors, or parameters, are varied while holding all other
factors constant to determine the sensitivity of the final result to the initial
assumptions. Typical factors to compare through parametric analysis include:
Fuel Cost
Discount Rate
Inflation Rate
Maintenance Rate
Grid Extension Distance
Grid Electricity Cost
Battery Cost
Diesel Cost
PV / Diesel Ratio
Alternative
Sensitivity
0.20
0.40
1.00
Standalone
PV
Prime Diesel
0.55
0.55
0.55
none
0.86
1.14
2.00
high
Hybrid
0.39
0.46
0.66
moderate
Grid
0.42
0.42
Extension
'/"""
0.42
none
This analysis shows that PV and grid options are not at all sensitive to fuel cost
(since they use no fuel directly), while hybrids are moderately sensitive and diesel
systems are highly sensitive. Therefore any assumptions about diesel fuel cost
have a tremendous effect on the life cycle costing of the Prime Diesel alternative,
and should be validated as much as possible for this type of analysis to have real
meaning.
20-32
Alternative
Discount Rate
Sensitivity
6%
12%
18%
Standalone
PV
Prime Diesel
0.66
0.55
0.49
low
2.00
1.16
0.76
high
Hybrid
0.66
0.46
0.36
moderate
0.38
low
Grid
0.50
0.42
Extension
)/""(
This analysis shows that the capital intensive projects (pure standalone PV and grid
extension), where most of the expenses occur at the beginning of the project, are
not very sensitive to discount rate, while the project with the highest proportion of
recurring expenses in the future (prime diesel) is highly sensitive. Once again, this
points out that assumptions about discount rate must be carefully exposed and
validated for the conclusions of this type of analysis to have meaning. The
assumptions of risk and real financial conditions must be carefully considered,
especially since we are considering alternatives that have very different future costs.
A high discount rate for example will make a Prime Diesel alternative look
inexpensive compared to a PV solution, whereas a low discount rate will show a
Prime Diesel system to be very expensive over time.
20-33
LCC Sensitivity
Another interesting analysis is to evaluate the total Life Cycle Cost of the hybrid
system for different PV/Diesel ratios. It is interesting to find the ratio of PV energy to
generator energy that gives the lowest life cycle cost. Other factors such as initial
cost may take precedence, but this type of sensitivity analysis helps to show how
much of a hybrid the system should be to have the lowest cost over time.
Below is shown the analysis for the system discussed earlier, with PV/Diesel energy
ratio changing from 0/100 (no PV) to 100/0 (no diesel). It is interesting to note that
the total cost over time actually goes down as the size of the solar array increases.
This indicates that at least in this particular example that the costs of fuel and
maintenance in the future outweighs the higher initial cost of solar modules. It is
only at very high PV/Diesel ratios (95% and higher) that the cost actually goes up
again.
LCC ($/kWh)
0.49
0.48
0.47
0.46
0.45
0.44
0.43
0.42
0
20
40
60
80
100
Percent PV
In summary, life cycle cost analysis is a powerful tool that can be used to compare
technology options for remote power requirements. The techniques described can
easily be used to compare water pumping alternatives, grid connection options and
other choices.
20-34
(End of Chapter)
20-35
CHAPTER TWENTY
ECONOMIC ANALYSIS
20-1
20-1
System Costs
Capital Costs
Recurring Costs
Non-recurring Costs
20-2
20-2
20-4
20-6
Economic Factors
Life Cycle Term (term)
Discount Rate (DR)
Fuel & General Escalation (FE, GE)
Net Present Value
20-7
20-7
20-7
20-7
20-8
20-10
20-10
20-11
20-11
20-12
20-12
Sample Analyses
Sample Application Parameters
PV System
Prime Generator
PV-Diesel Hybrid
Grid Connect
Comparison of Results
20-14
20-15
20-16
20-20
20-23
20-28
20-30
Parametric Analysis
Sensitivity to Fuel Cost
Sensitivity to Discount Rate
LCC Sensitivity
20-32
20-32
20-33
20-34
20-36