Department of Electronic Engineering ELE2EMI Electronic Measurements & Instrumentation

Basic Op-Amp Circuit Analysis and Design

Chapter reference: Carr, chapter 12.

3.1 Outline
Differential op-amp symbol and properties Ideal op-amp characteristics Ideal op-amp circuit analysis Real op-amp characteristics Offset Currents and Voltages Compensating for Offsets Basic applications of the op-amp

3.2 Differential op-amp symbol and properties

I
+

Va

Io AV B Ng Vo

V+ I+
Other properties:

The differential op-amp has two inputs and one output. Its purpose is to produce a voltage Vo that is a multiple AV of the differential input voltage Va = V V+ . The voltage gain AV should be large. Although they are not always shown in schematics, every op-amp is connected to an upper power supply Vcc (which should be positive and greater than any normal output voltage) and a lower power supply VEE (which should be negative and lower than the op-amps output). 1

3.3 Ideal Op-Amp Characteristics

Innite open loop voltage gain, AV = Innite input impedance, Zin = Zero output impedance, Zout = 0 Innite bandwidth, B = Zero noise generation, Ng = 0 Power supply voltages are innite: VCC = + and VEE =

3.4 Ideal Op-Amp Circuit Analysis

What are the consequences of these characteristics? Since the voltage gain AV of the ideal operational amplier (op-amp) is innite and its output voltage Vo must be nite, therefore its differential input voltage Vi is zero; in other words V+ = V . Since its input impedance Zin is innite, its input currents I+ and I are both zero. Zero output impedance implies that the output voltage Vo is unaffected by the load circuit to which it is delivered. Innite bandwidth means that the op-amp works identically at all frequencies, from DC to daylight as the saying goes. The ideal op-amp is stable; this requires that it produce no noise internally; indeed, it must not permit any noise to enter it from any source, otherwise its innite voltage gain would amplify the noise innitely, producing extreme instability. Ideally the power supplies have innite voltage, since any nite output voltage Vo is assumed possible. 3.4.1 Inverting amplier circuit A typical inverting amplier circuit is shown below.

Rf Ri Vi
+

Vo

Since V+ is grounded, V = 0 also; and since I = 0, the currents in R1 and Rf are equal. So the resistors form a voltage divider:

Vi Ri 0V Rf Vo
from which we deduce the voltage gain, Vo Rf = Vi Ri which implies that the circuit does indeed invert the input signal. Note that the gain< 1 if Rf < Ri so we usually choose Rf > Ri . 3.4.2 Non-inverting amplifer circuit A non-inverting amplier circuit,

Ri

Rf

Vi

Vo

is similar, but the positions of the input Vi and ground 0V voltages have been reversed, leading to this voltage divider:

0V Ri Vi Rf Vo
and this voltage gain: Rf Vo =1+ Vi Ri which 1 no matter what the resistances are.

3.5 Real Op-Amp Characteristics

Finite open loop voltage gain, AV < Finite input impedance, Zin < Nonzero output impedance, Zout = 0 Finite bandwidth, B < Nonzero noise generation, Ng > 0 Finite supply voltages: 0 < VCC < + and < VEE < 0 Interestingly, the gain-bandwidth product GB (where G is another symbol for AV ) is nearly constant for each specic type of op-amp. Since in real-life the power supply voltages VCC and VEE are nite, any output voltage Vo had best keep between them, otherwise it will be clipped. (This is called hard limiting.) In fact, as Vo approaches one of the supply voltages, it is increasingly distorted because the formula Vo = AV Va becomes increasingly inaccurate. So, for linear performance (which is essential for good reproduction of the input signal) ampliers must be backed-off to less than maximum amplication. Comment: Oddly, some people (such as disc jockeys) set the volume on their hi- systems too high and listen to these nonlinear distortions for hours on end. Looking on the bright side, they are benetting the biomedical engineers who help to design bionic ears.

3.6 Offset Currents and Offset Voltages

Finite gain AV and input impedance Zin imply that the offset currents I+ and I and the offset voltage Va are nonzero. Lets investigate the consequences of each of these separately, for the inverting amplier circuit. 3.6.1 Effect of the Offset Current Supposing Va = 0 but I = 0, the currents Ii and If through the resistors are related according to Kirchoffs Current Law by: Ii = If + I Therefore, by Ohms law, Vi 0 0 Vo = + I Ri Rf from which we deduce that Vo = Rf Vi + I Rf Ri

Thus there is a zero offset in the output voltage, so the voltage characteristic (curve of Vo versus Vi ) is a straight line shifted vertically from the origin by the amount Vo = I Rf which depends on the offset current at the op-amps negative input. 4

3.6.2 Effect of the offset voltage If the offset currents I = I+ = 0 but the offset voltage Va is nonzero, then in the voltage divider we used to calculate the gain, the intermediate value which was zero is replaced by Va and the current equation becomes: Vi Va Va Vo = Ri Rf Rearranging terms, we obtain 1 1 Vi Vo = Va ( + ) Rf Ri Rf Ri Multiplying through by Rf and dening Vo as the amount by which Vo differs from its ideal value R Rf Vi yields: i Vo = Va (1 + Rf ) Ri

3.7 Compensating for Offsets

A solution to the problem of an offset current or voltage is to use a pre-offset to cancel the error V that it caused. Interestingly, a bias voltage is used to compensate for an offset current, and a bias current to compensate for an offset voltage. We will now consider the specics of offset compensation for the inverting amplier circuit. 3.7.1 Bias voltage compensates for offset current

Rf Ri Vi I+ RC
+

Vo

The compensation resistor RC introduces a bias voltage I+ RC which acts as an offset voltage Va = I+ RC so it will produce a zero offset Vo cancelling that due to offset current I provided that: I+ RC (1 + Rf ) = I Rf Ri

If we use a variable resistor for RC we can adjust it until the output voltages zero offset is nil (or as near as we can achieve); this will occur when: RC = I (Ri ||Rf ) I+ 5

3.7.2 Bias current compensates for offset voltage

IS Va Ri Vi
+

Rf

Vo

By Kirchoffs Current Law and Ohms Law, the bias current IS contributes a term IS Rf to the output voltage. The effect of the offset voltage Va will be cancelled when IS Rf = Va (1 + thus we should choose IS = Rf ) Ri

Va Ri ||Rf

But how do we adjust the current? One way is to use a potentiometer circuit as the current source:

VEE () IS

Rpot Vtap R1 Va

VCC (+)

The voltage Vtap tapped by the potentiometer can be varied anywhere in the range from the lower supply VEE to the upper supply VCC . The value of the bias current IS equals IS = Vtap Va R1

Example: if the power supply rails are10V and Va 0 and R1 is 100 k then the magnitude of IS can be varied up to 0.1A which seems ample. (The diagram in Carr has Rpot = 20 k and R1 somewhere between 10 k and 100 k, but the particulars are sure to differ between circuits.) Equating the two formulas for IS we nd that the required tap voltage Vtap is proportional to the input offset voltage Va = V that it compensates for: Vtap = Va (1 + R1 ) Ri ||Rf

3.8 Basic Applications of the Op-Amp

3.8.1 Voltage Follower

Vi

Vo

Also known as the Unity Gain Buffer, this is a special case (Rf = 0 and Ri = therefore gain = 1) of the non-inverting amplier. Notice that Vo = Vi even though there is no current connecting them. This is a very useful property as it permits non-ideal voltage sources (with large resistances) to drive non-ideal loads (small resistances) without being loaded down (reduced in voltage).

3.8.2 Comparator

VCC V V+
+

Vo VCC Vo VEE
slope =

AV VEE

Va

Va = V V +

A V

When used directly with inputs that differ signicantly so that linearity is completely broken, an opamp acts as a comparator, which is a circuit with analog inputs and an output that is practically digital as it spends most of its time on one or the other of the power supply rails depending on which input is higher:

V+ > V Vo = VCC V+ < V Vo = VEE 3.8.3 Summer On the other hand, linear operation of the op-amp allows many useful analog computer circuits, such as the summer (summation circuit, not the season). This design exploits Ohms Law and Kirchoffs Current Law.

Rf
R1 V1 V2 R2 +

Vo

Note that this summer is essentially an inverting amplier with two input signals joining at the negative input terminal of the op-amp. 3.8.4 Integrator The characteristic equations of basic circuit elements allow a variety of analog computation circuits. For example, since current is the time derivative of charge, we can use a resistor (V = IR) and a capacitor (Q = CV ) to integrate current and thereby voltage.

If Cf Ii Ri Vi
+

Vo

Here, Vi = Ii R and QC = CVC while Ii = If = Vo = 0 VC = 3.8.5 Differentiator

QC dt so Ii dt = 1 RC Vi dt

QC 1 = C C

By reversing the positions of capacitor and resistor, the integrator is converted into a differentiator:

Rf Ci Vi
+

Vo

3.8.6 Logarithmic Amplier Bipolar Junction Transistors (BJTs) have a logarithmic relationship between collector current Ic and base-emitter voltage Vbe : Vbe = VT ln( where Is = its reverse saturation current. Recall that: VT = kT qe Ic ) Is

where k = Boltzmanns constant, qe = the magnitude of the electron charge, T = the absolute temperature of the transistor. We can use this property of BJTs to construct a simple analog circuit that computes logarithms. This is known as a logarithmic amplier.

Ic Ii Ri Vi
+

Q f

+ Vbe

Vo

In the circuit above, Vo = Vbe and Vi = Ic Ri so (if the offset current is zero) the input-output relation is: Vo = VT ln( 3.8.7 Antilog Amplier By exchanging the transistor and the resistor, we obtain a simple antilog amplier, which calculates the exponential function. Vi ) Is Ri

If Rf Qi

Vi

Vo

In the forward active region of the transistor, the antilog ampliers input-output equation is: Vo = Is Ri exp( 9 Vi ) VT