Documentos de Académico
Documentos de Profesional
Documentos de Cultura
Copy 3
BEHAVIOR OF BOLTS IN
TEECONNECTIONS SUBJECT TO
PRYING ACTION
by
R. S. NAIR,
P. C. BIRKEMOE
and
W. H. MUNSE
SEPTEMBER, 1969
by
R. S. Nair,
P. C. Birkemoe
and
IN.
H. Munse
Illinois
September 1969
iii
ACKNOWLEDGMENT
This study was part of an investigation on riveted and bolted
structural joints conducted at the University of Illinois in cooperation
with the Research Council on Riveted and Bolted Structural Joints of the
Engineering Foundation under the general direction of W. H. Munse, Professor
of Civil Engineering, and the direct supervision of P. C. Birkemoe, Assistant
Professor of Civil Engineering.
The report presented herein was prepared as a doctoral thesis
by R. S. Nair, Research Assistant in Civil Engineering.
Professor W. H.
iv
TABLE OF CONTENTS
Page
CHAPTER
1.
2.
3.
INTRODUCTION
1.1
2.1
2.2
3
3
6.
7.
Experimental Investigations . . . .
Analytical Studies and Methods of Computing
Prying For ce . .
EXPERIMENTAL STUDY
4.1
4.2
4.3
4.4
5.
4.
Test Specimens .
. . . .
Materials and Fabrication
Test Procedures
Discussion of Results
8
8
9
12
12
14
15
22
ANALYTICAL STUDY
34
5.1
5.2
5.3
34
37
39
45
6.1
6.2
48
45
53
7.1
7.2
53
55
Design Procedure . . . . . . .
Discussion of the Suggested Procedure
CONCLUSIONS
59
A.
ANALYSIS OF TEE-CONNECTIONS . . . . . . . . . .
62
B.
86
8.
APPENDIX
Page
LIST OF REFERENCES
87
TABLES . .
89
FIGURES
102
vii
LIST OF FIGURES
Page
FIGURE
1.1
102
1.2
103
2.1
104
2.2
105
2.3
106
2.4
FORCES IN TEE-CONNECTION . . . . .
107
2.5
lOB
3.1
109
4.1
110
4.2
III
4.3
112
113
114
4.4
4.5
115
4.6
LOAD~NG
4.7
116
117
4.B
4.9
4.10
4.11
lIB
119
120
.....
4.12
121
viii
FIGURE
Page
4.13
122
4.14
123
4.15
124
4.16
125
4.17
127
129
130
131
4.21
132
4.22
133
4.23
134
4.24
135
5.1
136
5.2
137
5.3
EFFECT OF DISTANCE FROM BOLT LINE TO FACE OF WEB ON PRYING RATIO AT ULTIMATE IN TEE-CONNECTIONS WITH
A325 BOLTS . . . . . . . . . . . . . . . . . . .
138
139
140
4.18
4.19
4.20
5.4
5.5
ix
Page
FIGURE
EFFECT OF EDGE DISTANCE ON PRYING RATIO AT ULTIMATE
IN TEE-CONNECTIONS WITH A490 BOLTS . . . . . . . . .
141
142
143
144
145
146
147
148
149
150
151
A.1
152
A.2
153
A.3
154
A.4
155
A.5
156
5.6
5.7
5.8
5.9
5.10
5.11
6.1
6.2
6.3
6.4
6.5
CHAPTER 1
INTRODUCTION
In bolted structural connections in which the fasteners are subjected to direct tensile loading, tee-stubs are frequently used to transfer
load to the bolts.
type--a
(1,2) that prying action could substantially reduce both the ultimate load
and the fatigue strength of bolted tee-connections.
A general discussion
of ASTM A325 and A490 high strength bolts loaded in direct tension in
structural tee-connections and to formulate methods for the design of such
connections.
be considered.
The investigation included an experimental and an analytical study
of the factors which govern prying action in bolted tee-connections and the
effect of prying action on the behavior of the bolts in the connection.
In the experimental program, static and fatigue tests were performed
on connections made up of two identical tee-sections with their flanges
bolted together by means of four high strength bolts.
Such a "symmetrical"
The
~predicting
the two inner lines carry most of the external load applied on the connection;
the remaining fasteners are not very effective.
CHAPTER
In bolted tee-connections,
A descrip-
tion of the manner in which prying forces are developed in such connections
and a discussion of the effect of prying action on bolt behavior follow.
2.1
When
parallel to each other (Fig. 2.l(a)) and hence no prying forces are developed
between them.
In tee-connections in which the bolts are the critical elements
(i.e., where failure of the connection results from bolt fracture) the magnitude of the prying force usually decreases as the load on the connection
approaches the ultimate load.
The con-
4
loaded single-bolt connection shown in Fig. 2.2(a).
ject to prying action is shown in Fig. 2.2(b).
the bolt and the bolt pretension are the same in both connections.
The
These relation-
applied load is denoted as p(s), the bolt tension is denoted as T(s) and the
total compressive force between the connected parts is denoted as C(s).
there is no external load on the connection (i.e., when pes)
When
0) the bolt
(s)
(s)
=C
= To ).
(s)
compression, C(s), decreases until it becomes zero and the connected parts
separate.
is denoted as P, the bolt tension as T and the prying force per bolt as
Q.
The local compressive force between the tee-sections around each bolt hole
is denoted as C.
flange are ignored and it is assumed that the prying force and the local
compressive force around the bolt holes (Q and C, respectively, in Fig.
2.4(a
(Fig. 2.4(b
single~bolt
connection.
identical since the shear force (P+Q) on the portion of the tee-section flange
may not be distributed in the same manner as pes) in the single-bolt connection.
per bolt on the tee-connection, P, is less than the effective load per bolt
by an amount equal to the prying force, Q, as shown in Fig. 2.4(c).
The
shape of the T versus P relationship shown in Fig. 2.4(c) also reflects the
decrease in the prying force which usually occurs when the load on the connection is close to the ultimate load (as explained in Section 2.1, above).
Effect of prying action.
and applied load in the single-bolt connection and that in the tee-connection are compared in Fig. 2.5.
That
load PI applied to the tee-connection will result in the same bolt tension
Ultimate Load
The effect of prying action on the ultimate load of a connection
may also be illustrated with the help of Fig. 2.5. If the strength of the bolt
is denoted as T , the tee-connection and the single-bolt connection will both
u
in the tee-connection
is less than P
(s)
by an amount
P
u
P (s) _ Q .
u
Hence,
(2.1)
Equation (2.1) may, of course, be obtained directly from equilibrium considerations since, at ultimate, the local compression around the
bolt holes, C, is zero and the only forces on each tee-section, per bolt,
are P ,T
and Q
2.2.1, above, of the bolt tension versus applied load relationships of the
tee-connection and the single-bolt connection.)
bolt tension will have an adverse effect on the behavior of the tee-connection
under repeated loading.
CHAPTER 3
PREVIOUS INVESTIGATIONS AND EXISTING PRYING ACTION FORMULAE
3.1
Experimental Investigations
Experimental studies of the behavior of high strength bolts loaded
The
test
The
of bolts, most of-the applied load was carried by the bolts in the two inner
lines.
In the early stages of loading, the inner bolts carried the entire
load.
Douty and McGuire (3) conducted static tests on tee-connections
assembled with two lines of A325 bolts.
cluded the flange thickness, the edge distance (the distance from the bolt
line to the edge of the tee-section flange) and the bolt diameter.
The
were installed between the sloping surface of the tee-section flanges and
the bolt head and nut.
washers and changes in bolt pretension were found to have little effect on
the ultimate load of the connection.
nections with beveled washers withstood about ten times as many applications
of load before fatigue failure of the bolts as did connections without
beveled washers subjected to the same applied load.
structural tee-connections have been carried out by Chang (4) and Douty and
McGuire (3).
The
a~l
~P
8a
(3.1)
10
where
is the prying force per bolt, P is the external applied load per
bolt, g is the gage distance (the distance between the bolt lines) and a
is the edge distance.
as that which gave rise to a plastic hinge in the flange at the bolt line.
The model used by Douty and McGuire (3) in their analysis is shown
in Fig. 3.l(b).
tension, T; and the local compressive force on the bottom of the flange
around each bolt hole, C.
o.
By
eli~inating
I
2
3a ( a
4b
4b
1)
wt 4
2
30ab A
4
wt
2
30ab A
(3.2)
11
wt
1
2
Qu
a( a+ 1)
b
where
3b
2
30ab A
4
wt
+
2
6ab A
(3.3)
bolt,
A is
sectional area of each bolt, b is the distance from the bolt line to the
center of the web fillet and a is the edge distance except that if the edge
distance is greater than 1.25 b, a is equal to 1.25 b.
A method for the design of riveted tee-connections is presented
in reference 5.
eluded.
3b
4a
P.
(3.4)
(3.5)
12
CHAPTER 4
EXPERIMENTAL STUDY
The experimental program reported herein included static and
fatigue tests of symmetrical tee-connections, each assembled with four high
strength bolts.
selected for this purpose was a concentrically-loaded single-bolt connection (see section 2.2 and Fig. 2.2(a)).
Bolt calibration tests were performed on representative samples
from each lot of bolts used in the experimental program.
The mechanical
Test Specimens
The specimens tested included single-bolt connections and tee-
connections had the same geometry while some of the geometric parameters
of
Fig. 4.1.
1 1/16 inch thick cylinders with a 1 1/4 inch thick plate between them-gripped by a bolt of 3/4 inch nominal diameter running concentrically
through them.
13
were threaded on the outside so that they could be screwed into the pulling
yokes usually employed at the University of Illinois for testing bolts in
direct tension.
was 1 1/16 inches, the web thickness was 11/16 inch, the length of the
flange in a direction parallel to the web was 7 1/2 inches and the pitch
of the bolts in the flange was 4 1/2 inches.
from the bolt line to the edge of the flange, measured perpendicular to the
web), a, and the gage (the distance between bolt lines, measured perpendicular
to the web), g, were varied.
gage in the connections tested and the notation of the corresponding connection geometries are shown in the following tabulation:
Geometry
Designation
Edge Distance,
a
(inches)
Gage,
g
(inches)
TO
1 3/4
4 1/2
Tl
4 1/2
T2
2 1/2
4 1/2
T3
1 3/4
T4
1 3/4
The bolt holes in the tee-section flanges and the fill plate were 13/16
inch in diameter.
A listing of the specimens tested is presented in Table 4.1.
Four single-bolt connections and sixteen tee-connections were tested
14
under static loading.
letter denotes the type of test (F for fatigue tests, S or U for static
tests); the symbols in parentheses denote the connection geometry (S for
single-bolt connections, TO, Tl, T2, T3 or T4 for tee-connections); the
number following the geometry designation denotes the type of bolt used
in the connection (3 for A325 bolts, 4 for A490 bolts); and the number
after the hyphen denotes the sequential number of the test.
Thus,
F(TO)3-2 denotes the specimen used in the second fatigue tests of a teeconnection with geometry TO assembled with A325 bolts.
4.2
Materials
a) The bolts.
A325 and A490 were used in the tests of single-bolt connections and
tee-connections.
meter and 4 1/2 inch nominal length and had heavy hexagonal heads.
The A325 bolts were used with a single flat hardened washer
under the nut.
with 3 1/2 inch grip and three A490 bolts with 3 5/8 inch grip.
The
15
average tensile strength under direct tension was found to be 42.0 kips
for the A325 bolts and 51.8 kips for the A490 bolts.
Hardness tests on the bolts and nuts yielded readings of 25 to
27 Rockwell C for the A325 bolts; 32 to 35 Rockwell C for the A490 bolts;
87
to 93 Rockwell B for the nuts used with the A325 bolts; and 30 to 33
connections were fabricated from wide flange sections of ASTM A36 structural
steel.
thickness.
fabricated from the flanges of the wide flange beam from which the teesections were obtained.
The fill plate in the tee-connections and the block of steel at the
center of the single-bolt connections were fabricated from 1 1/4 inch thick
plates of A36 steel.
4.2.2
The various connected parts in the specimens were cut from the wide flange
shapes and plates described in Section 4.2.1, above.
parts were machined but no machining was carried out on their surfaces;
the mill scale on these surfaces remained intact.
Test Pr~cedures
4.3.1
Instrumentation of Bolts
A system of bolt instrumentation was adopted by means of which the
16
a)
with the help of a special extensometer which was developed at the University
of Illinois for use in tests of high strength structural bolts.
This
These
The sensitivity
computed from measurements of the average axial strain in the shank of the
bolts.
This strain was measured with the help of two resistance strain gages
from the strain gages were passed through small holes drilled through the
bolt head.
Metal lugs were glued to the shank alongside the gages to prevent
contact between the gages and the side of the bolt hole.
The bolt tension was assumed to be linearly related to the average
axial shank strain (the average of the strains indicated by the two gages on
the bolt shank).
accuracy of the method adopted for determining the bolt tension, three A325
bolts and three A490 bolts were instrumented with strain gages as described
above and loaded in direct tension in a universal testing machine.
One
instrumented bolt of each type (A325 and A490) was also loaded by torquing of
the nut with the bolt installed in a steel load transducer.
were from the same lots and had the same grip lengths as the bolts used in
the tests of single-bolt and tee-connections.
17
For bolts loaded in direct tension, the relationship between
bolt tension (as indicated on the testing machine) and average axial shank
strain (as indicated by the strain gages) was found to be linear for values
of bolt tension up to at least 95 per cent of the ultimate tensile strength.
In the bolts loaded by torquing of the nut (with bolt tension measured using
the load transducer), the relationship remained linear up to ultimate.
Typical bolt tension versus shank strain relationships of bolts loaded in
direct tension and torqued tension are shown in Figs. 4.3 and 4.4,
respectively.
The constant of proportionality between bolt tension and average
axial shank strain was found to be nearly equal for all the bolts, regardless of bolt type or method of loading.
per microinch per inch was used in the determination of the tensile force in
instrumented bolts installed in connections.
4.3.2
their tension
ver~us
tests were from the same lots as those used in tests of connections.
Two direct-tension tests, two torqued-tension tests and one test
under torqued tension followed by direct tension were each performed on A325
bolts with 3 1/2 inch grip (with 7/16 inch of thread in the grip} and A490
bolts with 3 5/8 inch grip (with 9/16 inch of thread in the grip).
In
addition, one bolt of each type was tested in direct tension with a grip
length of 3 3/16 inch (with 1/8 inch of thread in the grip).
A description
18
a)
Direct tension.
was~aded
in a
The elongation
of the bolt was continuously monitored with the U of I extensometer until the
ultimate load of the bolt was reached; readings were recorded at intervals of
bolt tension.
During this
latter stage of loading (i.e., after the ultimate load had been reached), an
approximate measure of
Torqued tension.
an electric load transducer was used for the measurement of bolt tension.
The load cell, a steel spacer and one or more hardened washers provided
the specified grip length for the bolt to be calibrated.
inserted through the load cell, spacer and washers, the nut was torqued with
a manual torque-indicating wrench until fracture of the bolt.
The bolt
tension (measured using the load cell) and bolt elongation (measured using
the U of I extensometer) were recorded at 30 degree increments of nut
rotation.
c)
Bolts calibrated
under this type of loading were instrumented with strain gages on the shank
for the measurement of bolt tension.
outside~
with a
of pre-selected magnitude (32 kips for A325 bolts, 40 kips for A490 bolts)
by turning the nut.
screwed into the pulling yokes used in tension tests of bolts (see Fig. 4.6)
and load was applied with a 120 kip universal testing machine until failure
19
of the bolt.
of applied load.
slipping a piece of foil between them) occurred, bolt tension was determined
from the shank strain.
from the load-indicating dial on the testing machine since the bolt tension
and the applied load were equal.
installed with a flat hardened washer under the nut while A490 bolts were
installed with washers under both the nut and the bolt head.
Thus, the
grip length of the A325 bolts was 3 1/2 inches (with 7/16 inch of thread in
the grip) and that of the A490 bolts was 3 5/8 inches (with 9/16 inch of
thread
in the grip).
The bolt in specimen S(S)3-2 was installed with a pretension of
28 kips, i.e., the nut was torqued until the bolt tension (as determined
from the shank strain) reached a value of 28 kips.
other specimens were tightened by the "snug plus half turn" method.
yokes normally used with the testing machine for loading bolts in direct
tension.
yokes.
connection (see Figs. 4.1 or 4.6) are pulled away from each other.
20
In each
cycle, the load was increased to a maximum value and then decreased to zero.
The
maximum load in each cycle was approximately 5 kips higher than in the preceding
one.
Bolt elongation
and bolt tension were recorded at small intervals of applied load during each
cycle of loading and unloading.
4.3.4
Tests of Tee-Connections
Tee-connections were tested under fatigue loading
as well as static
loading.
The static tests were of two different types denoted as type U and
type S.
Bolts
were installed with a snug load of approximately 5 kips followed by half a turn
of the nut.
nut.
A325 bolts were installed with a flat hardened washer under the
A490 bolts had washers under the bolt head as well as the nut.
The grip
length of A325 bolts installed in tee-connections was 3 1/2 inches (with 7/16 inch
of thread
in the grip) and that of A490 bolts was 3 5/8 inches (with 9/16 inch
21
The instrumentation of connections, the method of loading followed,
and the measurements made in each of the three types of tests (fatigue tests
and two types of static tests) are described below.
a)
Fatigue tests.
fatigue loading, either one or two of the four bolts in the flange of the connection were instrumented for measurement of bolt tension.
All four bolts had gage points drilled in them for the
0.0001 inch graduations were mounted at each edge of the tee-section webs
to obtain a measure of flange separation at the web.
Cyclic loading of constant amplitude was applied to each connection
in a 200
used.
machine.
applied was approximately 200 per minute (except when the fatigue machine
was stopped for the taking of measurements).
During the first cycle of loading on each connection, measurements
of bolt tension, bolt elongation and flange separation were made at intervals
of applied load of approximately 15 kips.
"statically."
22
b)
was installed in a 300 kip universal testing machine and tensile load was
applied until fracture of the bolts in the tee-section flanges occurred.
(In every tee-connection tested under static loading, all four bolts
appeared to fail simultaneously.)
c)
these tests was exactly the same as for the fatigue tests (see above) except
that in specimen S(T4)3-l all four bolts in the flange were instrumented
for measurement of bolt tension.
A 300 kip universal testing machine was used for loading the
connections.
the
In each cycle,
zero~
The
maximum load in each cycle was approximately 20 kips higher than in the
preceding one.
Discussion of Results
4.4.1
These idealized
relationships for a bolt of the same type with the same diameter, grip length
and length of thread. in the grip.
23
3/4 inch nominal diameter with a grip length of 3 5/8 inches and 9/16 inch of
thread in the grip are shown in Fig. 4.8.
Bolts in tee-connections are subjected to torqued tension during
the bolt-tightening (or pretensioning) process followed by direct tension
when external load is applied to the connection.
It is evident
24
calibration curve.
account the effect of the bolt variables (grip length, diameter, etc.) on
the direct-tension tension-elongation relationship.
Effect of grip length.
However, pre-
vious experimental studies (7,8,9) have shown that grip length has no appreciable effect on the tension-elongation characteristics of high-strength bolts
beyond the "proportional limit," especially in bolts loaded in direct
tension.
A325 and A490 bolts with only 1/8 inch of thread in the grip are shown in
Fig. 4.12.
of thread in the grip (and slightly larger grip lengths) are also shown in
the figure.
less thread in the grip, however, remained greater than that assumed in
the idealized relationships for the bolts with more thread in the grip.)
25
Similar results were obtained in the direct-tension calibration tests reported in references 7 and 9.
diameters other than 3/4 inch were neither tested nor analyzed.
However,
loading were obtained from static tests as well as from the first cycle of
loading in fatigue tests.
26
below.
The bolt
kips~
The
The magni-
tude of this drop in clamping force was larger for cycles of applied load of
larger amplitude.
two 'tee-connections
which behaved in markedly different ways, are shown in Figs. 4.14 and 4.15,
respectively.
The
load per bolt relationship of specimen 5(T3)3-l was very similar to that
of single-bolt connections (compare Figs. 4.14 and 4.13).
This indicates
In
27
The bolt tension versus applied load relationships of all teeconnections which were instrumented for measurement of bolt tension are shown
in Figs. 4.16 and 4.17.
assembled with A325 bolts; those of connections with A490 bolts are shown
in Fig. 4.17.
that of a single-bolt connection with the same type of bolt and the same
clamping force.
applied load relationships shown in Figs. 4.16 and 4.17 were obtained from
the first cycle of loading in fatigue tests and do not, therefore, cover
the entire range of load from zero to ultimate.
The number of bolts that were instrumented (with strain gages on
the shank) for measurement of bolt tension in each of the tee-connections for
which bolt tension versus applied load relationships were obtained is indicated in Figs. 4.16 and 4.17.
decre~sed
until,
at loads close to the ultimate load of the connection, the tensile forces
in all the bolts approached the strength of the bolts as determined from
direct-tension calibration tests.
b)
forces affect the bolt tension versus applied load relationship and the
28
where P is the applied load per bolt on the tee-connection and pes) is the
applied load on the single-bolt connection which gives rise to the same
bolt tension as load P per bolt applied on the tee-connection.
The prying
ratio, defined as the ratio of the prying force to the externally applied
load, may thus be expressed as follows:
(%)
pes)
- 1.
(4.2)
connection, Q per bolt, and the prying ratio at ultimate, (Q/P) , may,
u
T - P
u
u'
T
where P
(4.4)
is the strength of the bolts in the connection when they are loaded in-
29
is defined thus:
(4.5)
The prying ratios at ultimate and the bolt efficiencies of all the
tee-connections which were tested to failure under static loading were determined using Eqs.
The values of T
connections for which relationships between bolt tension and applied load
were obtained experimentally (see Figs. 4.16 and 4.17), the prying ratio
at a relatively low level of applied load, P , was determined using Eq.
(4.2).
The value of P
with A325 bolts and 95 kips for connections with A490 bolts.
These loads
fo~
were the same in all the tee-connections tested (see Fig. 4.2).
It may be observed in Table 4.2 that there were some differences
between the prying ratios at load P
connections, i.e., connections with the same nominal geometry assembled with
bolts of the same type.
30
shown in
Table 4.2 were determined from bolt tension versus applied load relationships which were based on measurements of the tension in only one of the four
bolts in a connection.
low levels of loading (see Eq. (4.2) and Fig. 2.5) was such that the resulting
values of prying ratio are highly sensitive to differences in the bolt
tension versus applied load relationship of connections.
The experimentally determined ultimate loads--and the corresponding
bolt efficiencies and prying ratios at ultimate--were almost identical for
tee-connections with the same nominal geometry and the same type of bolt.
No
two such connections had values of ultimate load that differed by more than
(where P
nections with an edge distance of 1 3/4 inches than in connections with edge
31
and
given values of the other geometric parameters) at which the prying ratios
at P
Increases in
in those assembled with A490 bolts than in those with A325 bolts (see Figs.
4.18, 4.19, 4.20 and 4.21).
than at ultimate
(compare Figs. 4.18 and 4.20 and Figs. 4.19 and 4.21).
4.4.3
Table 4.3.
on connections with A325 bolts, 100 kips on connections with A490 bolts)
was only slightly greater than the maximum allowable load on the connection,
computed neglecting prying, based on the allowable nominal stresses specified for bolts in buildings in reference 10.
as described above, was 71 kips for connections with A325 bolts and 95 kips
for connections with A490 bolts.)
~ycles
in
one case) while other connections withstood more than 2,000,000 load
applications without failure.
The range of average bolt stress in the first cycle of loading in
each tee-connection tested under fatigue loading is shown in Table 4.3.
This
32
range in bolt stress is the measured range in bolt tension divided by the
nominal cross sectional area of the bolt shank.
the tension in two bolts was measured, the range in average bolt stress
shown in Table 4.3 is the range in average stress in the bolt with the higher
range.
the fatigue tests that were performed on tee-connections are included in the
figure.
of loading, R, provided an accurate indication of the stress range throughout these fatigue tests.
the range of bolt stress after the first few applications of load was found
to be influenced by factors other than the magni tude o'f R, notably the
geometry of the connection (compare the behavior of F(TO)4-1 and F(T4)4-1
in Fig. 4.23).
33
Thus, as
shown in Fig. 4.24, if the bolt tension versus applied load relationship of
a connection is known (this relationship can, for example, be obtained from
a static test of the connection), the magnitude of the external load which
would cause failure of the connection after 2,000,000 applications can be
estimated.
Owing to the paucity of experimental data, no attempt was made to
establish a Rversus N relationship for values of N smaller than 2,000,000.
The available data indicate, however, that for each type of bolt investigated
(A325 and A490) a relationship between Rand N, independent of connection
geometry, does exist for values of N larger than about 300,000.
If such
34
CHAPTER 5
ANALYTICAL STUDY
The analytical investigation of bolted tee-connections consisted
of numerical analyses of a large number of such connections with different
geometries and bolt properties.
The parameters
which may be varied in connections analyzed using this computer program, the
assumed mechanical properties of the materials in connections, the method
of solution adopted and the output from the analysis are described below.
S.l.l
Variable Parameters
The parameters which may be varied in the analysis of tee-
connections using the computer program include the dimensions of the connection;
the number, diameter and grip length of the bolts; the bolt type (A32S or
A490); the bolt pretension or clamping force; and the yield stress of the
steel in the tee-section when it is under a uniaxial state of stress.
Further, the connection may be symmetrical, i.e., it may consist of two
identical tee-sections with their flanges bolted together, or it may consist
of a single tee-section bolted to a perfectly rigid base.
The inclusion of
a fill plate between the tees in a symmetrical connection can be taken into
account by increasing the bolt grip by an appropriate amount.
35
The figure also shows that the fillet between the web
The
(0
0
The idealized
are modified to reflect changes in grip length and diameter was described
in Section
4.4.1.
any specified value of pretension is shown graphically in Fig. 4.9 and was
discussed in Section 4.4.1.
5.1.3
A detailed
The magnitude of
36
each value of applied deformation, the following data are obtained as output
from the analysis:
1.
2.
3.
4.
bolt elongation;
5.
6.
7.
gation exceeds the value at which the bolts rupture or a mechanism is formed
by yield of the tee-section flange through its entire depth at locations
near the web and the bolt line.
definition, the maximum value of the applied load on the tee-section web
(item 1, above).
Example of results of analysis.
The important
37
plastic state at each level of loading are also indicated in the table.
The strength of each bolt in the connection, T , was 52.0 kips-u
this is the value assumed for A490 bolts of 3/4 inch diameter (see the
idealized bolt tension versus bolt elongation relationship in Fig. 4.8).
The ultimate load of the connection, P
per bolt (see Table 5.1).
( )
T - P
u
u
P
u
P
u
T
u
52.0 - 40.3
40.3
40.3
52.0
0.29
0.78
The analytical results shown in Table 5.1 could also be used to determine
the bolt tension versus applied load relationship of the connection.
this relationship, P
From
~ksi
It
5.2
Parameters Investigated
The influence of the following parameters on the behavior of tee-
connections assembled with A325 and A490 bolts was studied analytically:
1.
edge distance, a;
2.
This.
flange thickness, t;
4.
38
5..
bolt diameter, d.
The designation
connections of each geometry--one with A325 bolts and one with A490 bolts-were analyzed.
In the tee-to-
The
connections.
Connections of geometry TO, Tl, T2, T3 and T4 (see Table 5.2) were
assigned properties which approximated those of the tee-connections which
were tested in the experimental program; these connections were analyzed
with values of bolt pretension of 32 kips for A325 bolts and 40 kips for A490
bolts.
The bolt pretension in all the other connections analyzed was 28 kips
for A325 bolts and 35 kips for A490 bolts--these are the minimum permissible
values specified in reference 10.
39
5.3
Discussion of Results
The results of the analysis of tee-connections assembled with A325
and A490 bolts are summarized in Tables 5.3 and 5.4, respectively.
For each
connection, the ultimate load, bolt efficiency, prying ratio at ultimate and
mode of failure are shown in the tables.
in Tables 5.3 and 5.4; these are the values of applied load which might be
expected to cause fatigue failure of the bolts in the connections after
2,000,000 applications (see Section 4.4.3).
, and prying ratio at ultimate, (Q/P) u , presented in Tables 5.3 and 5.4
were obtained from the following equations:
- P
where P
expressed in kips per bolt, and T is the strength of each bolt in the
u
connection.
The value of T" selected was 42 kips for A325 bolts of 3/4 inch
u
is valid only when failure of the connection results from bolt fracture.
40
5.3.1
same geometry, bolt diameter and bolt type as connections which were tested
under static loading.
Tabl~
5.5.
in connections with A490 bolts), ultimate load, bolt efficiency and prying
ratio at ultimate closely approximate the values obtained experimentally.
It was concluded, therefore, that the analysis accurately determines the
behavior of bolts in tee-connections both at ultimate and at lower levels
of loading.
5.3.2
Changes in grip length have little effect on bolt tension and bolt
41
Separate
relationships are shown for connections assembled with A325 bolts and those
with A490 bolts.
The relationship
slope (see Figs. 5.3 and 5.4); increases in b result in an appreciable increase in the prying ratio.
Edge distance, a.
higher in connections with smaller values of edge distance (see Figs. 5.5
and 5.6).
which the prying ratio is maximum; further decreases in edge distance would
result in lower values of (Q/P)
there is a limit to the decrease in prying ratio that can be brought about
by increasing the edge distance.
The importance of edge distance as a parameter which influences
the prying ratio at
ultimat~
in the connection;
However,
42
may be regarded as a
single parameter.
and (Q/P)
values of wid .
5.3.3
up of two identical tee-sections bolted together was not appreciably different from that of a connection comprising one of the same tee-sections bolted
to a rigid abutment; it was further stated that inclusion of a fill plate
between the tee-sections in a symmetrical tee-connection had little effect
on the ultimate load of the connection.
Values of
43
always higher than for symmetrical tee-connections with the same tee-section
dimensions, without fill plates.
Inclusion of a fill plate between the tee-sections in a symmetrical
tee-connection is taken into account in the analysis of the connection by
increasing the bolt grip by an appropriate amount.
in grip length has little effect on bolt tension and bolt elongation at
ultimate, it does cause a decrease in bolt stiffness at relatively low
levels of load.
specified magnitude.
The value of P
in A325 bolts and 3 ksi in A490 bolts, is plotted against the ultimate
load of the connection, P , in Fig. 5.11.
u
without fill plates, the ratios of P
range:
to P
0.377 to 0.429 for connections with A325 bolts and 0.365 to 0.424
were
These differences in
44
further emphasize that although the differences in configuration do not
appreciably influence P ' they do affect P ; P is lowest for symmetrical
u
R
R
tee-connections without fill plates.
As noted above, the values of PR/P
connections without fill plates lay within a fairly narrow range (see Fig.
5.11); since the parameters a, b, t and wId
45
CHAPTER 6
FORMULATION OF SEMI-EMPIRICAL RELATIONS
The results of numerical analysis of tee-connections and data
from tests were used to formulate equations for the computation of the
prying force or prying ratio in bolted tee-connections.
Approximate for-
mulae relating prying force with the variable geometric parameters of teeconnections were first obtained by analyzing a simple structure which, in a
highly idealized manner, represented most of the important characteristics
of tee-connections.
(2a
The
The reaction,
were concentrated at the edges of the tee-section flange and that there
was no contact between the two flanges in the connection at any other point.
These assumptions are valid only when the load on the connection is
relatively high.
Tw~
were obtained:
flange was elastic through its entire length; a second solution was obtained
by assuming that there was a plastic hinge at the web-to-flange junction.
46
a)
Elastic flange.
bolt in the tee-connection, 0, was twice the deflection of the neutral axis
of the beam (Fig. 6.1) at the point of application of the load T on the
beam, i.e.,
(6.1)
T
yielded the following relationship
3ab
.9P
2a
2_ (Ed 20) wt 3
2+ 6a 2b +
d2
4T
(E)
2G at 2
(6.2)
+ (Ed
4T
0) wt
d2
where E and G are Young's modulus and the shear modulus, respectively, of
the tee-section m.-;.terial.
duced into the numerator and denominator of certain terms in Eq. (6.2).
It was assumed that the bolt elongation at ultimate,
u , was
(%)
3ab
u
wt
d2
(6.3)
47
where a is constant for connections with bolts of a given type and (Q/P)u
is the ratio of Q to P at ultimate.
2
connections with A490 bolts than for those with A325 bolts since d /T
is
In
follows:
M
o
where
CJ
CJ
Statical
(0 d)
(Ood2~
wt
4T )
o
4T
.9.
P
wt 2
d2
(6.4)
2
2
portional to d
ultimate:
(;)
b - S
wt
d
a + S
wt
d
2
2
2
2
(6.5)
48
and bolts of
bolts than for those with A325 bolts and would be higher for connections
with tee-sections constructed of higher strength steels.
6.2
analyzed, however, was a plastic hinge formed by yield of the flange through
Combinations of the important terms in Eqs. (6.3) and
The following formulae were found to agree closely with the results
of the finite element analyses and tests:
In connections with A325 bolts
(%)
2
I (t + 7wt )
b
u
"5
6a
7
\1
IOd 2
>
+ !:. (t + 7wt )
7\
IOd
(6.6)
b
>
3a
4"
49
(%)
IOObd
70ad
2
2
- 18wt
+ 2lwt
>
o
(6.7)
2
2
- l4wt
>
+ 2lwt 2
on (Q/P) u as determined
Equation (6.6)
analyses and tests of tee-connections are compared with the values computed
for the same connections using Eq. (6.6).
A similar comparison of
analytically and experimentally determined prying ratios with those computed using Eq. (6.7) is made in Fig. 6.3.
references
Figures.
50
In
were compared
(%)
wt
1
2
a ( a
b 3b
3b
Sa
20
30ab 2A
4
wt
+
2
6ab A
1)
>
>
0,
(6.8)
(6.9)
where
51
(6.7).
1
where T
(6.10)
(%),u
Equation (6.10)
is, of course, valid only if the bolt is the critical element in the
connection.
The externally applied load on a tee-connection, per bolt,
which would cause fatigue failure of the bolts in the connection after
that P
4 ksi for
3 ksi
aP
where a is .377 for connections with A325 bolts and .365 for connections
with A490
bolts~
(6.11)
(%) u
where
a
= 0.375
estimated using
52
Eq.
53
CHAPTER 7
DESIGN OF BOLTED TEE-CONNECTIONS
A method for the design of tee-connections assembled with A325
or A490 high strength bolts was developed.
The
Hence, in the
Design Procedure
It is suggested that the procedure described below be used for
54
.Q
p
100bd
70ad
2
2
- l8wt
+ 2lwt
>
0
(7.1)
.Q
p
100bd
62ad
2
2
- l4wt
+ 2lwt
>
Check that
<
(J
(Ja'
(7.2)
is as shown in the
tabulation, below,
Allowable Stress,
Loading
Condition
(J
ksi
A325 Bolts
A490 Bolts
Static
40
54
Fatigue
30
36
55
and
where P is the total externally applied load on the
connection, A is the total nominal bolt area and Q/p
is the prying ratio obtained in step 2.
4)
are (P+Q)/2 at each bolt line, Q/2 at the edge of the flange
(except that if the edge distance is greater than 2t, force
Q/2 acts at distance 2t from the bolt line) and P on the
web.
where (Q/P)E
q.
is the prying ratio computed using Eq. (7.1) and the other
P , is as follows:
u
nT
P
u
1
(%)
(7.3)
u
56
nTu [
Aa
(%)
(%)
1 +
Eg.
or
where F
q.
It
Thus, the factor of safety which results from the use of the
F .
o
to range from 2.30 to 2.75 for connections assembled with A325 bolts and
2.10 to 2.55 for connections with A490 bolts.
was caused partly by differences in tensile strength between the bolts used
in the various studies.
57
7.2.2
It was found
that
>
aP
where a is 0.375 for connections with A325 bolts and 0.365 for connections
with A490 bolts.
be expressed as
anT
1
(%)
Acr
P
(%) Eq.
anT
u
Acr
a
Thus,
anT
::::
Acr
(7.4)
may
58
The values of
connections subjected to repeated loading are such that Eq. (7.4) yields
a value of F
of approximately 1.2.
and P
were
The values of
b~se
fill plates had somewhat higher factors of safety (see Tables 7.1 and 7.2).
59
CHAPTER 8
CONCLUSIONS
The maj or conclusions which \Vere arrived at in this study of the
behavior of high strength bolts in tee-connections with two lines of
fasteners are as follows:
1.
2.
The factors which govern the magnitude of the prying force-and the resulting decrease in ultimate load and fatigue
strength--in bolted tee-connections include the geometry
of the tee-sections, the size of the bolts and the bolt type
(A325 or A490).
3.
5.
60
()
100bd
=
u
70ad
2
2
- l8wt
+ 2lwt
>
>
(%)
- 14wt 2
2
2
62ad + 2lwt
100bd
u
2t.
u'
() u
61
62
APPENDIX A
ANALYSIS OF TEE-CONNECTIONS
A method was formulated for the
with the help of a digital computer.
~nalysis
of bolted tee-connections
i.e., it may consist of two identical tee-sections with their flanges bolted
together, or it may consist of a single tee-section bolted to a perfectly
rigid base.
the material of which the tee-sections are constructed and the mechanical
properties of the bolts are as described in section 5.1.
Presented below is a description of the structure analyzed (an
idealized representation of the tee-connection) and an explanation of the
analytical procedure followed.
A.I
The load on
the web of the tee-section is also assumed to be uniform over the length
of the connection.
structure.
The structure analyzed is shown in Fig. A.l(a).
one half of one tee-section resting on a rigid base.
It consists of
63
If
f(2~
64
where
tension in the bolt,
total compressive force in the set of springs,
and
~b
T-~
rigid base) the force-deformation relationship of the set of springs consists of a series of straight lines.
denoted as k
s,
s,
The value of
and {o }
s
= T .
0
in Fig. A.l(a
This force is so distributed over the thickness of the web (i.e., in the x
direction) that all points at the top of the web undergo equal vertical
displacement (0
in Fig. A.lea.
, reaches a value of
or 2) a mechanism is developed by
yield of the flange through its entire depth at two or more separate locations.
65
A.2
Analytical Procedure
The finite element method of stress analysis is used.
The com-
plexity of the structure to be analyzed and the availability of a highspeed digital computer were the primary factors which led to the choice of
the finite element method.
An additional source of
*The
66
it is analyzed assuming linear behavior of the springs and the tee-section
material.
2.
proc~ss
'The
If
67
Fig. A.l(a):
1.
3.
arrays of quantities.
{FA}
{UA}
{FB}
68
{FC}
{UC}
{ST}
FD
FD {I}.)
nc
follows:
{UA}
{FB}
{UC}
{FC}
{ST}
(A. 1)
The matrices [UAFA] , {UAFD} , [FBFA], {FBFD}, etc. are obtained by applying,
in turn, a unit yalue of each component of the force vector {FA} and a
unit value of FD on the structure and determining {UA} , {FB}, {UC}, etc.
for each loading, using the finite element method.
However, the
successive loading conditions after the first one requires little additional
computational time.
Presented below is a description of the various steps involved in
the finite element method of analysis as applied in the present instance.
69
such a way that no two adjacent squares are divided along diagonals in
the same direction.
The springs (representing the bolt) are attached to consecutive
nodes at the top of the flange.
A.2) represent all nodes at the bottom of the flange and the top of the
web, respectively.
Formation of element stiffness and stress matrices.
There are
matrices for each of the four types of triangular elements were formulated
in global coordinates.
used for formulating the matrices were those suggested by Clough (12) and
Zienkiewicz (15).
The assembly of the structure's stiffness matrix and the computation of stresses may be simplified if the structure is consider"ed to be
made up of square elements, each consisting of two triangles.
* The
There are
70
They are
This triangular
stress matrices of elements of type I, II and III (Fig. A.4) were computed
for the case of both plane stress and plane "Strain and are shown in Tables
A.I and A.2.
[k]e {a}
{a}
[S]e {a}
where
{a}
{a , a
X
,T
xy
and {F} and {8} are vectors of nodal forces and displacements, respectively,
defined as follows:
{F}
{Xl' YI , X2 , Y2 ,, Yn }
{a}
71
Assembly of the structure's stiffness matrix.
The structure is
Each partition
denoted as N and the partitions are numbered sequentially from the left.
The vectors of nodal forces and displacements in the ith partition are
denoted as {P.} and {o.}, respectively.
1
Kl
C
l
T
C
K2
C
2
T
C
2
K3
C .
3
C~l
2
3
1
(A.2)
forces and displacements in the ith partition while submatrices [C.] relate
1
top of the web are imposed by removing appropriate terms from the {a} and
{p} submatrices and corresponding rows and columns from the [K] and [C]
submatrices.
72
the ith partition (the other end of the spring being fixed) is taken into
account by adding k' to the appropriate location on the major diagonal of
s
[K. ] .
l
This
(K ]
l
[K ]
l
(K. ]
[K. ]
T-1
[C i-I] [K _ ] [C - l ]
i
i l
2,3, ... ,N
(A.3)
{PI}
{PI}
{po }
{P. }
2,3, ... ,N
-1
-1{p N}'
[K i ]
[KN]
({Pi}
- [Ci]{Oi+l})' i
(A.4)
(A.S)
The matrices (K.]-l and {P";} are computed sequentially using Eqs. (A.3).
l
The
73
obtained by multiplying the stiffness of the spring by the vertical displacement of the node to which it is attached.
(The structure
The symbols A, Band D denote the same points in structure I that they
denote in structure II and {FA}, {UA} , {FB} and {ST} denote the same
quantities in structure I that {FA}, {UA} , {FB} and {ST} denote in structure
II except that {FA} is an array of reactions rather than applied forces.
The downward displacement of the upper ends of the springs in structure I
is represented by
oc
prestressing operation; the upper ends of the springs are, thereafter, kept
74
fixed.)
,1 =
h b
hOI e AiCf )
Wlth tease
W.l
o
,1
0
{FA(r)}
{UA(f)}
{FA (f)}
nar
[R]nar, {UA} na
na
nar
[R]nar , {FA} na
na
naf
[F] na f , {UA} na
na
naf
[F]naf , {FA} na
na
(A.6)
are matrices which define which of points A are included in A(r) and which
ln A(f) .
o
transpose.
Since matrix
~iJ
[F]T{UA(f)}
[R]T{FA(r)} +
[F]T{FA(f)}
} CA.7)
75
" 1 acements an d f orces on A(r) an d A(f) In
" structure II are d enote d
Th e d lSP
as {UA(r)}, {FA(r)}, {UA(f)} and {FA (f)}.
As shown
(a)
oWlng
th
e lncrement,
respectively.
Let forces {FA} and FD be applied on structure II such that the
following conditions are satisfied:
{FA} (a)
{FA}
{FC}
FD{l}
nc
(A.8)
{t }(a).
(A.9)
and
Consequently,
{ST} (a)
{ST}
(A.lO)
{FE} (a)
{FB}
(A.ll)
76
identical.
also be
(A.12)
Comparison of the movement of the upper and lower ends of the springs in
structures I and II yields
{UC} + (0 (a) + o(a)) {I} .
c
w
{o }(a)
s
(A.13)
FD
+ o(a))
w
(A.14)
defined as the slope of the spring force versus spring deformation relationship, must be equal in structures I and II.)
The boundary conditions at the base in structure I may be
expressed as follows:
{UA(r)}
(i)
{O}
-(f) (i)
UA.
J
0, j
-(r) (i)
FA.
J
{FA (f) }
>
=
>
=
0, j
1,2, . .. , nar
(i)
{O}
77
(a)
{UA(r)}
- ( f ) (a)
-(r) (i-l)
-{UA
}
-(f) (i-l)
UA.
>
UA.
_( r ) (i-l) j
,
FA.
- ( r ) (a)
>
=
FA.
(A.16)
(a)
{FA(f)}
-(f)
= -{FA}
(i-l)
From Eqs. (A.8), (A.12) and (A.16) the following relationship are obtained:
- {UA(r)}
{UA(r)}
UA~f)
>
=
(i-l)
-(f) (i-l)
- UA.
o (a) j
w '
FA~r)
>
_( ) (i-l)
FA r
{FA (f) }
(A.17a)
- ,0 (a) {l}
-(f) (i-l)
- {FA }
(A.17b)
(A.17 c)
(A.17d)
[UAFA] {FA}
{UAFD} FD,
(A.18)
Using the matrices [R]
and [F] defined'in Eqs. (A.6) and (A.7), the following relationship may be
obtained from Eq. (A.18):
{FA(r)}
[R]{UAFD}FD).
(A.19)
[R][UAFA] [F]T[F]{FA}(i-l)
[R]{UAFD} k (o(a) + o(a)
s
CA.20a)
7B
{FA (f)}
and
{FA} (i-I)
[F]
(A.20b)
With assumed values of [R] and [F], {FA(r)} and {FA (f)} may be
obtained from Eqs. (A.20a) and (A.20b), respectively.
correctly assumed, the values of {FA(r)} and {FA (f)}
+ {UAFD}
-(f)
{VA
}
{FA(r)}
and
k (oCa)
s
+ oCa)
w
'
CA.2l)
(i-I)
(i-I)
[F] {UA}
(i-I)
(i-I)
[R] {FA}
If [R] and [F] are found to be incorrect, the grouping of nodes A into A(r)
and A(f) is changed as follows:
satisfied are included in A(r) and those nodes at which Eq. (A.17c) is
not satisfied are included in ACf ).
obtained and are then used in computing new values of {FA(r)} and {FA Cf )}.
This iterative process is repeated until the true values of [R], [F],
{FA Cr )} and {FA Cf )} are obtained.
{UA},
{FE},
CA. I) where
[R]T{FA Cr )}
{FA}
+ [F]T {FA Cf )}
} (A. 22)
and
FD
k CoCa)
s
+ o(a.
w
displacement,
o~a),
79
conditions of loading.
Application of spring prestress.
and
CA.23)
CA. 24)
where
nc
T
s
1:
i=l
s,i
nb
and
P
w
FB.
l
i=l
2.
{FA Cr )}, {FA (f)}, {FA} and FD are computed using Eqs.
and
and are denoted as {FA(r)}
FDCc) , respectively.
(c)'
{FA Cf )}
(c)'
{FA}Cc) and
80
3.
and
(w)'
{FA(f)}
(w)'
{FA}(w) and
FD(w) , respectively.
4.
FB( C ) , l. .
FB (W ) , l.
i=l
nb
i=l
where
and
81
6.
The assumed values of [R] and [F] are checked with the
help of Eqs. (A.17b), (A.17c) and (A.21).
An iterative
cycle of iteration.
7.
nc
L
i=l
(a)
.
S,1.
is denoted
as 0(0).
w
Matrices {FA} (0) , {UA} (0) ,{FB}(o) , {t s }(o) and {ST}(o) and
displacement 0 (0) represent the values of {FA}, {UA} , {FB}, {t }, {ST}
w
s
and 0
prestressing- operation,
{O} .
82
k s, 1.
(0)
follows:
1.
values first assigned to [R] and [F] are the values that
were found to be correct for the previous increment.)
2.
{FA
(r)
} and {FA
(i-I)
{UA ( f) }
3.
(f)
{UA
(f)
},
.
(i-I)
and {FA ( r) }
)
are obtained from Eqs. ( A.2l.
The values of [R] and [F] are checked with the help of Eqs.
(A.17b) and A.17c).
Thus, a
{FA (f)}.
4.
{FA} and FD are ob tained from Eqs. (A. 22) and {UA}, {FB},
{UC}, {FC} and {ST} are computed using Eqs. (A.l).
5.
{FA} (a), {UA} (a), {FB} (a), {t } (a), {os} (a) and {ST} (a) are
s
obtained from Eqs. (A.8) through (A.13).
83
and
xy
An
(0
)2 +
Y
(0
)2 + (0
)2 + 6T
>
xy
2a
2
o
(A.25)
where
and
vo
+ va
This criterion
has been found t<? accurately represent the behavior of steels with "flat-top"
yield plateaus (18).
The specified relationship between the total spring force, T ,
s
and the average spring deformation, 6 , consists of a series of straight
If it is found that
and 6
84
T(i
s
where 6(i) and T(i) are the spring deformation and force at the end
s
s
structure.
as described below.
2.
positive definite.)
85
use of statics.
at which the flange has yielded through its depth and the
reaction on the bottom of the flange is assumed to be concentrated at the outer edge of the tee.
Failure is assumed
to occur when (a) the spring force reaches its maximum value
or (b) a mechanism is developed through yield of the flange
through its entire depth at a second location.
4.
86
APPENDIX B
SUGGESTED DESIGN SPECIFICATION FOR ASTM A325 AND A490 BOLTS
SUBJECT TO APPLIED TENSION IN STRUCTURAL JOINTS
Bolts subject to applied tension shall be proportioned to support
the sum of the external load and any prying force produced by deformation
of the connected parts.
. . QP
. . QP
100bd
70ad
100bd
62ad
2
2
2
2
- l8wt
2~t
- l4wt
+ 2lwt
2
2
2
2
>
= 0
>
= 0
Allowable Stresses
Loading
Condition
Static loading
Repeated loading
ASTM A325
Bolts
40 ksi
30 ksi
ASTM A490
Bolts
54 ksi
36 ksi
2t.
87
LIST OF REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
Sterling, G. H., Troup. W. J., Chesson, E. and Fisher, J. W., "Calibration Tests of A490 High-Strength Bolts," Proceedings, A.S.C.E.,
Vol. 91, No. ST5, October 1965.
10.
11.
88
12.
13.
14.
15.
16.
Clough, R. W., Wilson, E. L. and King, I. P., "Large Capacity Multistory Frame Analysis Programs," Proceedings, A.S.C.E., Vol. 89,
No. ST4, 1963.
17.
18.
89
TABLE 4.1
LIST OF TEST SPECIMENS
Specimen
Designation
Connection
Geometry *
Type of
Bolt
Type of
Test
S(S)3-l
S(S)3-2
S
S
A325
S(S)4-l
S(S)4-2
S
S
A490
F(TO)3-l
F(TO)3-2
F(Tl)3-l
F(T2)3-l
F(T4)3-l
TO
TO
Tl
T2
T4
A325
F(TO)4-l
F(TO)4-2
F(Tl)4-l
F(T2)4-l
F(T3)4-l
F(T4)4-l
F(T4)4-2
TO
TO
Tl
T2
T3
T4
T4
. A490
"
"
U(TO)3-l
U(TO)3-2
U(TJ.,) 3-1
U(T2)3-l
U(T4)3-l
TO
TO
Tl
T2
T4
A325
Static
"
"
"
"
"
"
"
U(TO)4-l
U(TO)4-2
U(Tl)4-l
U(T2)4-l
U(T3)4-l
U(T4)4-l
U(T4)4-2
TO
TO
Tl
T2
T3
T4
T4
A490
"
"
"
"
S(TO)3-l
S(T3)3-l
S(T4)3-l
TO
T3
T4
A325
"
"
"
"
"
S(TO)4-l
TO
A490
"
*S
"
"
"
"
"
"
"
"
"
"
"
"
"
"
"
II
Static
II
II
"
Fatigue
"
"
"
II
"
"
II
"
"
"
"
"
"
"
"
90
TABLE 4.2
SUMMARY OF EXPERIMENTAL DATA ON BEHAVIOR
OF TEE-CONNECTIONS UNDER STATIC LOADING
Specimen
Designation
Edge
Distance,
inches
Gage,
inches
Prying
Ratio
at
Load P
Bolt
Efficiency,
per cent
Prying
Ratio
at
Ultimate
152
154
154
91
92
92
.10
.09
.09
157
93
.07
160
95
.05
165
98
.02
123
123
73
73
.37
.37
171
169
170
82
82
82
.21
.22
.22
182
88
.14
174
84
.19
202
98
.02
142
143
69
69
.46
.45
Ultimate
Load,
kips
1.75
1.75
1.75
1.75
1.75
4.50
4.50
4.50
4.50
4.50
.30
.33
.50
F(T1)3-1
U(T1)3-1
1.00
1.00
4.50
4.50
.00
F(T2)3-1
U(T2) 3-1
2.50
2.50
4.50
4.50
(a)
S(T3)3-1
1.75
3.00
.05
F(T4)3-1
S(T4)3-1
U(T4)3-1
1.75
1.75
1.75
6.00
6.00
6.00
1.08
1.08
1.75
1.75
1.75
1.75
1.75
4.50
4.50
4.50
4.50
4.50
.65
.50
.55
F(T1)4-1
U(T1)4-1
1.00
1.00
.4.50
4.50
.10
F(T2)4-1
U(T2)4-1
2.50
2.50
4.50
4.50
.30
F(T3)4-1
U(T3)4-1
1.75
1.75
3.00
3.00
.00
F(T4)4-1
U(T4)4-1
U(T4)4-2
1.75
1.75
1.75
6.00
6.00
6.00
.67
*Pa
a The clamping force in this connection was too high for computation of
the prying ratio (see Fig. 4.16).
91
TABLE 4.3
SUMMARY OF EXPERIMENTAL DATA ON BEHAVIOR
OF TEE-CONNECTIONS UNDER FATIGUE LOADING
Specimen
Designation
Maximum
Applied
Load,
kips
Range in Average
Bolt Stress*
in First Cycle
of Loading, ksi
Cycles
to
Failure,
thousands
72
75
72
86
72
2.0
5.7
1.1
4.1
7.4
a
3000
999
2300 a
2700 a
555
11.7
6.6
4.2
3.7
2.2
10.4
3.2
164
270
528
592
3000 a
32
3200 a
100
94
100
100
100
100
79
*The
92
TABLE 5.1
SELECTED RESULTS FROM ANALYSIS OF A TEE-CONNECTION
Applied
Load
per Bolt,
kips
Bolt
Tension,
kips
Reaction
on Bottom
of Flange
per Bolt,
kips
Distribution of Reaction
and
Regions of Yield*
Remarks
1.
0.0
35.0
35.0
11.9
35.8
23.9
18.6
37.9
19.3
22.7
40.4
17.7
28.2
- 44.3
16.1
*Portions
~
~
7sZ0----l
I
I
!
I
Number of bolts
= 4; length of
flange parallel
to web = 11.0.
(All dimensions
are in inches.)
93
TABLE 5.1 (Continued)
SELECTED RESULTS FROM ANALYSIS OFA TEE-CONNECTION
Bolt
Tension,
kips
Applied
Load
per Bolt,
kips
Reaction
on Bottom
of Flange
per Bolt,
kips
33.2
48.6
15.4
36.3
50.4
14.1
52.0
11.7
40.3
(Ultimate)
*Portions
Distribution of Reaction
and
Regions of Yield*
Remarks
The connection
became unstable;
failed by bolt
fracture.
94
TABLE 5.2
PROPERTIES OF TEE-CONNECTIONS ANALYZED
Designation
of
Geometry
t,
inches
a,
inches
b,
inches
w,
inches
Al
A2
A3
0.750
0.750
0.750
1.594
1.594
1.594
1.719
1.719
1.719
2.250
3.750
5.500
Yes
Yes
Yes
No
Yes
Yes
B1
B2
B3
B3R
1.062
1.062
1.062
1.062
1.793
1.793
1.793
1.793
1.863
1.863
2.660
2.660
2.250
5.500
5.500
5.500-
Yes
Yes
Yes
No
No
No
Yes
Yes
C1
C2
C3
C4
C5
C6
C6R
C7
C8
1.500
1.500
1.500
1.500
1.500
1.500
1.500
1.500
1.500
2.531
2.531
1.031
1.594
2.531
2.531
2.531
2.531
3.844
1.531
2.469
3.406
3.406
3.406
3.406
3.406
3.406
3.406
2.250
2.250
2.250
2.250
2.250
3.750
3.750
5.500
2.250
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
D1
D2
1.750
1.750
2.516
2.406
2.453
3.437
2.250
2.250
Yes
Yes
Yes
Yes
E1
E2
E3
E3R
E4
E5
2.000
2.000
2.000
2.000
2.000
2.000
. 2.875
1.625
2.875
2.875
2.875
4.625
2.562
3.812
3.812
3.812
3.812
3.812
2.250
2.250
2.250
2.250
5.500
2.250
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
TO
T1
T2
T3
T4
1.062
1.062
1.062
1.062
1.062
1.727
0.930
2.523
1.660
1.793
1.797
1.797
1.797
1.066
2.527
3.750
3.750
3.750
3.750
3.750
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
*Unless
Analyzed
with
A325
Bolts?
Remarks *
Analyzed
with
A490
Bolts?
Rigid base
Rigid' base
Rigid base
1.25
1.25
1.25
1.25
1.25
in.
in.
in.
in.
in.
fill
fill
fill
fill
fill
plate
plate
plate
plate
plate
95
TABLE 5.3
SUMMARY OF RESULTS OF ANALYSIS OF TEE-CONNECTIONS
ASSEMBLED WITH A325 BOLTS
Connection
Geometry
P4ksi' *
Mode
of
Failure
Ultimate
Load,
kips
per bolt
Bolt
Efficiency,
per cent
A3
10.4
12.1
13.6
16.7
28.8
34.2
40
69
81
.46
.23
BI
B2
B3
14.0
17.0
14.3
28.8
42.0
37.1
69
100
89
.00
.13
CI
C2
C3
C4
C5
C6
C7
C8
17.9
14.2
10.6
12.1
12.7
14.4
15.8
12.7
42.0
34.7
26.7
27.6
29.6
36.2
42.0
31.3
100
83
64
66
70
86
100
75
.00
.21
.57
.52
.42
.16
.00
.34
DI
D2
16.3
14.2
38.5
33.0
92
89
.09
.27
'11
El
E2
E3
E4
E5
17.7
13.7
14.2
18.0
14.3
42.0
35.0
35.3
42.0
35.3
100
83
84
100
84
.00
.20
.19
.00
.19
II
TO
Tl
T2
T3
T4
19.3
18.1
19.4
24.3
15.9
37.6
37.5
37.8
42.0
31.4
89
89
90
100
75
.12
.12
.11
.00
.34
kips
per bolt
Al
A2
* Externally
Prying
Ratio
at
Ultimate
Flange
Bolt fracture
II
Flange
Bolt fracture
II
II
"
II
"
II
II
II
II
II
"
"
"
II
"
"
"
"
"
96
TABLE 5.4
SUMMARY OF RESULTS OF ANALYSIS OF TEE-CONNECTIONS
ASSEMBLED WITH A490 BOLTS
Connection
Geometry
kips
per bolt
Ultimate
Load,
kips
per bolt
Bolt
Efficiency,
per cent
Prying
Ratio
at
Ultimate
A2
A3
12.2
14.0
32.7
37.4
63
72
.59
.39
Bolt fracture
B3
B3R
15.0
18.2
40.3
40.3
78
78
.29
.29
"
"
C1
C2
C3
C4
C5
C6
C6R
C7
C8
18.7
16.5
11.4
12.6
13.0
15.4
17.3
17.3
13.0
46.8
38.8
28.2
30.0
33.3
39.7
39.8
47.3
35.4
90
75
54
58
64
76
76
91
68
.11
.34
.84
.73
.56
.31
.31
.10
.47
"
"
D1
D2
17.7
15.3
44.5
38.0
86
73
.17
.37
"
E1
E2
E3
E3R
E5
18.5
15.3
16.6
17.5
16.5
47.7
38.8
39.1
. 39.1
41.3
92
75
75
75
79
.09
.34
.33
.33
.26
"
"
TO
T1
T2
T3
T4
20.6
19.9
20.6
25.9
17.6
43.3
43.3
44.5
52.0
35.6
83
83
86
100
68
.20
.20
.17
.00
.46
*Externally
P 3ksi '
Mode
of
Failure
"
"
"
"
."
"
"
"
"
"
"
"
"
"
"
"
"
97
TABLE 5.5
CO:MPARISON OF RESULTS OF ANALYSIS OF TEE-CONNECTIONS
WITH EXPERIMENTAL DATA
Connection
Geometry
P ,*
R
kiEs Eer bolt
from
from
Analysis Tests
Ultimate Load,
kiEs Eer bolt
from
from
Analysis Tests
Bolt Efficiency,
Eer cent
from
from
Analysis Tests
Prying Ratio
at Ultimate
from
from
Analysis Tests
19.3
18.1
19.4
24.3
15.9
18.5
l8.o+a
20.8
23.5
15.1
37.6
37.5
37.8
42.0
31.4
38.2
39.2
40.0
41.3
30.8
89
89
90
100
75
91
93
94
98
73
.12
.12
.11
.00
.34
.10
.08
.07
.02
.36
43.3
43.3
44.5
52.0
35.6
42.5
45.5
43.5
50.8
35.5
83
83
86
100
68
82
87
84
98
68
.20
.20
.17
.00
.46
.22
.16
.19
.02
.47
20.6
19.9
20.6
25.9
17.6
19.0
21.0
22.9
25.0+
18.0
*PR
98
TABLE 7.1
ALLOWABLE LOADS AND FACTORS OF SAFETY FOR TEE-CONNECTIONS
WITH A325 BOLTS SUBJECT TO REPEATED LOADING
Connection
Geometry
P ,*
f
Source of
Data on
P
**
Maximum
Allowable
Load,
kips per bolt
Factor of Safety
Against
Fatigue Failure
of Bolts
12.1
13.6
Analysis
Analysis
8.6
10.1
1.40
1.35
B2
B3
17.0
14.3
Analysis
Analysis
13.3
11.1
1.28
1.29
Cl
C2
C3
C4
C5
C6
C7
C8
17.9
14.2
10.6
12.1
12.7
14.4
15.8
12.7
Analysis
Analysis
Analysis
Analysis
Analysis
Analysis
Analysis
Analysis
13.3
10.8
7.9
8.4
9.0
11.5
13.3
9.4
1.35
1.31
1.34
1.43
1.40
1.25
1.19
1.35
Dl
D2
16.3
14.2
Analysis
Analysis
12.5
10.3
1.30
1.37
El
E2
E3
17.7
13.7
14.2
18.0
14.3
Analysis
Analysis
Analysis
Analysis
Analysis
13.3
10.9
11.3
13.3
11. 5
1.33
1.25
1.25
1.35
1.24
13.la
14.6a
Test (1)
Test (11)
10.9
12.5
1.20
1.18
A2
E4
E5
Tl
T2
T3
T4
TO
Tl
T2
T3
T4
19.3
18.1
19.4
24.3
15.9
Analysis
Analysi-s
Analysis
Analysis
Analysis
11.5
11.1
11. 7
13.3
9.4
1.68
1.63
1. 66
1.82
1.69
18.5
18.0
20.8
23.5
15.1
Test
Test
Test
Test
Test
11.5
11.1
11.7
13.3
9.4
1.61
1.62
1.78
-I. 76
1.60
*Pf
** Figures
99
TABLE 7.2
ALLOWABLE LOADS AND FACTORS OF SAFETY FOR TEE-CONNECTIONS
WITH A490 BOLTS SUBJECT TO REPEATED LOADING
Connection
Geometry
P ,*
f
kips per bolt
Source of
Data on
P
**
Maximum
Allowable
Load,
kips per bolt
Factor of Safety
Against
Fatigue Failure
of Bolts
12.2
14.0
Analysis
Analysis
9.6
11. 2
1.27
1.25
B3
15.0
Analysis
12.1
1.24
C1
C2
C3
C4
C5
C6
C7
C8
18.7
16.5
11.4
12.6
13.0
15.4
17.3
13.0
Analysis
Analysis
Analysis
Analysis
Analysis
Analysis
Analysis
Analysis
14.8
11.9
8.8
9.3
10.0
12.5
14.8
10.3
1.26
1.38
1.29
1.35
1.30
1.23
1.18
1.26
Dl
D2
17.7
15.3
Analysis
Analysis
13.5
11.3
1.31
1.35
E1
18.5
15.3
16.6
16.5
Analysis
Analysis
Analysis
Analysis
15.0
11.8
12.3
12.6
1.23
1.35
1.31
Test (11)
16.5
1.20
E2
E3
E5
1~29
Tl
T2
T3
T4
TO
Tl
T2
T3
T4
20.6
19.9
20.6
25.9
17.6
Analysis
Analysis
Analysis
Analysis
Analysis
12.5
12.0
12.9
15.8
10.4
1.65
1.66
1.60
1.64
1.69
19.0
21.0
22.9
25.0
18.0
Test
Test
Test
Test
Test
12.5
12.0
12.9
15.8
10.4
1.52
1. 75
1. 78
1.59
1. 73
12.1
12.5
12.3
1.50
1.39
1.42
18.2
17.3
17.5
Analysis
Analysis
Analysis
*Pf
** Figures
aPf was obtained from an applied load versus cycles to failure diagram.
100
TABLE A.1
ELEMENT STIFFNESS MATRICES
l+B
Element type I:
A
0 -A-B
0
-B -A-B
0
0
B
-1
0
A -B
0
l+B -A-B -B
A
l+B
B -1
l+B
0
l+B
B
-B
A -1
l+B -A-B
l+B
l+B
-1
B
C
symmetric
l+B
A+B
l+B
-B
-B
l+B
-A
-1
-1
-A
-B
-B
A+B
l+B
A+B
l+B
l+B
symmetric
l+B
A+B
l+B
-B
-B
B
-1
-A
-B
-B
B
A
1
0
0
-A
-1
C
symmetric
B
Plane stress:
Plane strain:
A = I-v'
v,
B=
I-v
2
, C=
1-2v
2(1-v)
~ [1_~2 ]
, C
t [
2
E(l-v)
]
(l+v) (1-2v)
0
0
0
0
-1
-B
-B
-A
l+B
-A
-B
-B
-1
A+B
l+B
101
TABLE A.2
ELEMENT STRESS MATRICES
C
2
[
[
Plane stress:
Plane strain:
A=-
= v,
-1
-A
-B
-1
-A
-1
-A
-B . B
-1
-A
-B
-1
-B
~A
I-v
=--
, C
0
0
1
A
-B
-A
-1
B
A
1
1
A
0
0
1
a
A
1
-B
1
A
B
[1~v2 ]
1-2v
v
B = 2(1-v) , C
I-v'
1
a
E(l-v)
]
(l+v) (1-2v)
!J
102
- l+-
-H
Cpo
--.l
4.,..
ql
.cpo
~
~
a)
Beam-to-Column Connection
b)
FIG. 1.1
Hanger Connection
103
t
Bolt
lines
a)
Symmetrical Tee-Connection
b)
Te~-Section
FIG. 1.2
104
a)
b)
FIG. 2.1
105
t
1...---+--+----1I
..t
....
'I
... ~
'_ ... )
4:;=-
a)
Single-Bolt
Connection
FIG. 2.2
b)
Tee-Connection
106
t pes) ,
-...,..-+--+-~---,
a)
(s)
Separation of
connected parts
occurs here
r--.
CJ)
E-<
C
0
-M
CJ)
C
Q)
E-<
.u
..-i
co
FIG. 2.3
107
2P
~II::
:t
(P~),
I H : total shear
~-f--~~--~-f--- ~~~~:!ference
C, total compressive
force
a)
Forces on Tee-Section
b)
Forces on Portion of
Flange Around Bolt
H
.,
c
0
-r-l
(J)
c
Q.)
H
.w
r-1
r:o
o ~--------------------------~
o
Applied Load per Bolt, P, and
Effective Load per Bolt, (P+Q)
c)
FIG. 2.4
FORCES IN TEE-CONNECTION
108
T
u
--------------Tee-connec t ion
I
I
Single-bolt
connection
H
..
I
I
I
Tl
OM
CI)
(l)
r--- Qu
---I
I
I
+J
M
0
P=1
I
I
r---QI
~I
I
I
I
PI
I
I
I
I
I
I
PI
I
I
I
I
(s)
Pu
P
u
(s)
FIG. 2.5
109
2P
assumed points
of inflection
Q, resultant of
prying force
g
a
a)
gage distance
edge distance
2P
Q~ a+- b~
b)
FIG. 3.1
b-+a~
110
I
I
I
J
-+
~I~
rl
FIG. 4.1
III
~ +
~-+-
11"------1
2~1I+
2
3"
3"
2~"~
21
t
I
-t--t--+
+
+
-4--+---$+
l'
I
ia--r
a1
I
~
~-----~~~-----------------------------------~+-~-----~~~
g
~1-.::t"~lco
'I
.--i
I
1"
72
1
1
_%-+"_-+-1_"--+_}--t"
L....
42
~r
~I
13/16 in. diam. holes for bolts
of 3/4 in. nominal diam.
~-+-+
I
-4--4--4I
-t--t.--+
I
FIG. 4.2
Geometry
Designation
a
(inches)
(inches)
TO
1 3/4
4 1/2
T1
4 1/2
T2
2 1/2
4 1/2
T3
1 3/4
T4
1 3/4
C"')
112
50
,......
?'----
40
Q)
".-1
...c
CJ
cU
CfJ
E:1
o..bO
".-1
~ ".-1
30
.w
'"
CI)
Q)
o .w
".-1
CfJ
~
Q)
E-4'"'d
Q)
+.I
.:.J
...-1
0
cU
20
CJ
~ ".-1
'"'d
".-1
CI)
cU
'-"
10
/
V
A325 Bolt
3/4 in. diameter
3 1/2 in. grip
......
0
0
1000
2000
3000
4000
FIG. 4.3
5000
113
50
40
7-
,.-..
,..-i
,..-i
QJ
cJ
Cf.l
o..""d
-..4
,.!G
...
~
m
0
30
,..-i
:>-.
0,..0
-..4
Cf.l""d
~ QJ
QJ +J
cJ
+J -..4
,..-i""d
0 ~
20
i=O -..4
Cf.l
'-'"
10
Ultimate load,
[3..5.2 kips
A325 Bolt
3/4 in. diameter
3 1/2 in. grip
7/16 in. thread in
grip
1000
2000
3000
4000
FIG. 4.4
5000
114
f
Pulling yoke
Heat treated
steel inserts
SECTION A-A
FIG. 4.5
115
L
A
SECTION A-A
FIG. 4.6
60
50
_---+- __ -
40
_____
-'I
CJ)
p...
'r-!
t::
o
.r-!
CJ)
30
D i r : t Tension
~ --'"j-
--,----~~
~-=r-::1'
::\
t::
QJ
Torqued Tension
H
+J
r-4
pC:!
20
idealized relationship
A325 Bolt
10
10
20
**
3
3
in.
in.
30
40
50
60
70
80
90
100
110
FIG. 4.7
.......
(j"I
60
I
Direct Tension
50
---
40
(/)
P.
Torqued Tension
,,-1
p'"
,,-1
30
(/)
Jj
Q)
H
.lJ
H
P4
20
A490 Bolt
idealized relationship
1---+--
____
______
10
20
______
____
30
~~
40
____
______
50
60
______
~~
80
70
____
____
______-L____
90
100
--
110
I-'
I-'
'-.I
118
F'
Direct-tension
relationship (given)
IE
~_-----r-----"'" D
Torqued-tension
relationship (given)
/
A' ~
/
/
/
/
/
/
/
J 0'
Bolt Elongation
119
45
40
A325 Bolt
3/4 in. diameter
3 1/2 in. grip
7/16 in. thread in grip
32 kip pretension
35
CI)
--------.....
30
o s::
o
OM
CI)
Q.)
25
OM
20
+-l
.--i
0
o:l
.w
CI)
s::
...
b!l
s::
.--i OM
o
s::
p:::) Q.)
.w
Q.)
CI)
p..
OM
S::...:x!
p..
OM
...:x!
..
s::
10
15
s::...c:
OM b!l
OM
Q.)
CI)
.w
C'Ij
l-I
Q.)
Q.)
l-I .w
U4-<
s::
10
- - - - Experimentally determined relationship
5
0
0
10
20
30
FIG. 4.10
50
40
3
C'Ij
120
55 __--------__--------~--------~~--------~--------~ 15
-- --- - - - -
----~
10
A490 Bolt
3/4 in. diameter
3 5/8 ino grip
9/16 in. thread in grip
40 kip pretension
40
CJJ
..!G
~
c:
o c:
35
+J
....-l 0.-1
~
30
+J
rl
0
p:j
25
c:.c
.-1 bO
.-1
QJ
co
$-I
QJ
QJ
$-I
4-l
c: co
20
- - - - Experimentally determined relationship
15
10
10
20
30
40
FIG. 4.11
CJJ
QJ
CJJ
c:QJ
..
bO
0
-.-1
E--!
CJJ
P.
CJJ -.-1
c:~
QJ
P.
-.-1
c:
c:
o
0.-1
50
3
121
60
50
40
30
20
(/)
0...
01"'1
~
..
01"'1
(/)
10
A325 Bolts
0
60
Q)
H
+-J
..--l
0
50
I=Q
40
30
grip
20
A490 Bolts
10
0
0
20
40
60
80
100
120
140
160
FIG. 4.12
122
50
40
CJ)
0.-ri
,.!s:!
..
s::
30
0
-ri
CJ)
s::
OJ
H
+J
M
0
i=Cl
20
10
Failure of connection
Increasing load
Decreasing load
--0
10
20
30
40
FIG. 4.13
50
123
50r---------~--------~~--------~--------~~--------~
40~--------_+----------+_--------_4--------~~--------~
CIJ
oM
-Ie
--
oM
CIJ
~
Q)
H
+.J
.-i
20~--------_+--------~~--------~----------~--------__4
*Average
of measurements
"from two bolts
10r---------~----------~--------~--------~~--------~
Failure of connection
Increasing load
Decreasing load
O~--------~----------~----------~----------~--------~
40
o
20
30
10
50
FIG. 4.14
124
50
40
en
0..
-r-l
~
30
-)I:
"'
0
0
-r-l
en
<lJ
H
+J
.-l
20
0
1=0
*Average
of measurements
from four bolts
10
Failure of connection
Increasing load
----
Decreasing load
0
0
10
20
30
40
FIG. 4.15
50
125
60
S(TO)3-l
50 ... One bolt
instrumented
40
30
10
..!G
'"
C
0
or-!
Cf.l
Q)
H
.j..J
50
t-
....-i
0
1=0
--V ~
-- ~
20
Cf.l
0or-!
F(TO)3-2
Two bolts
instrumented
-_
r-
40
F(TO)3-l
One bolt
instrumented
FCTl)3-l
One bolt
instrumented
v/
30
20
10
0
/
10
-----
.. ~---
...
/
A
Failure of connection .
Tee-connection
---20
30
40
50
Single-bolt connection
10
20
30
1
40
FIG. 4.16
I
50
60
126
60
50
F(T2)3-l
One bolt
instrumented
40
---
30
20
CJ)
P.
OM
,.!:::!
10
c:'"
0
OM
CJ)
c:
a>
+J
.-I
50
Relationship for
single-bolt connection with sam'e
clamping force
not available.
0
o:l
----V ---V
..".
30
20
10
0
V
o
10
F(T4)3-l
One bolt
,instrumented
,.......--
~---V
-~
-- ...
.....".",.
....--
S(T4)3-l
Four bolts
instrumented
40
S(T3)3-l
Two bolts
instrumented
/
Failure of connection
Tee-connection
----
20
30
40
50
Single-bolt
Connection
10
20
30
40
50
60
127
60
50
!-
S(TO)4-l
Two bolts
instrumented
40
CJ)
0.-
-1""1
~
10
...
~
-1""1
CJ)
Q)
H
-+-J
...-1
50
r-
0
p::)
F(TO)4-2
Two bolts
instrumented
20
V
lL
~~--
30
40
10
-:~ ~---
30
20
10
--~~---
----~
V
---- ---- -L
V
F(Tl)4-l
One bolt
-instrumented
---
6-
Failure of connection
Tee-connection
- - - - Single-bolt connection
Vi
F(TO)4-l
Two bolts
instrumented
20
30
40
50
10
20
30
40
50
FIG_ 4.17
60
128
60
50
t-
-- - p..----;/
40
30
20
U)
0..
.r-l
~
..
10
.r-l
U)
Q)
E-t
~
,.....,
50
f-
~~-- --
30
20
V
10
F(T3)4-l
One bolt
instrumented
--- ---'"
V
IV
F(T4)4-l
Two bolts
instrumented
40
10
k(
F(T2)4-l
One bolt
instrumented
V
IV
F(T4)4-2
Two bolts
. instrumented
...
V
l7
---/
;'
Failure of connection
Tee-connection
- - - - Single-bolt connection
20
30
40
50
10
20
30
40
50
60
129
1.2
1.1 .....
('j
-..
P-4
1.0
0.9
*Pa
0.8
-......
0'
'-'"
-+: .,
0.7
('j
p..;
+J
0.6
0
ri
+J
p:::
0.5
bO
-r1
:>,
H
P-;
V --n-
0.4
"
...
"'-
"\
,
0.3
............. to-
/ /
1//
,/'
o
~
~---
0.2
0.1
Gage
~/
0.0
1
2
"
1
4 1/2 in. -
11.
2
FIG. 4.18
130
1.2
0
1. 1-1. 01-
O. 9-
--
C1j
p..
f5
*Pa
/
/
O. 8
.........
CY
'-'
~
..
O. 7
C1j
+J
C1j
O. 6
-M
+J
C1j
/Q
O. 5
co
O. 4
I~
O. 2
O. 1
b3
O. 0 2
O. 3
...-
~-C ]
01"1
H
p..
~t
I
p..
V
Edge distance
/
4
1 3/4 in. -
Gage, g, inches
FIG. 4.19
131
0.6
A490
bolts
0.5
I
Gage
= 4 1/2 in.
-...=='
~
..........
0-
0.4
..
Q)
+J
(lj
oM
+J
1'"""i
:::>
0.3
+J
(lj
OM
+J
~
bO
oM
0.2
."",
/'
>-.
---
--
--- ---.
.........
) ...........
/[
,....
0.1
)~
/"
...---- ~-8---
t----..,,-
"-
),
0.0
1
21.2
FIG. 4.20
132
O.6r---------~I----------~I--------~I~--------~--------~
O.5~-------~I-------+I--------~------~--------~
IV
/
~~
SO. 4 t------+----+-----+-------:/;L---t-/~-___t
;
/ /~
~
]
/
/
~
/
/
~
O.3~--------~--------~--------~~----,+---+---------~
.~
O.2~--------+---------~---/~----~I~--------+---------~
I
/
/
O.l~------+-----/~~--~_r---~------~------~
/
/
O. 0 ' - - _...
-..:&.-L--.....:/_=--=-=-:..a-./_ _ _~_ _
- - 4 -_ _- - - . d
Gage, g, inches
FIG. 4.21
oM
Ul
~
pq
bO
~
ri
30
20
'1j
m
o
r-
TO*
4-4
M
(J
>.
T4*
Q}
10
c..:>
.j..J
Ul
H
or-\
,"
or-\
s:1
Ul
Ul
Q)
'.
T~
~*
""-
"
a""
T2
A325 bolts
pC!
~
m
Q}
~
4-4
Q}
bO
r::
~
.................
.j..J
20
.TO
"- "'---
T1*
""- .....
$..I
.j..J
Cf)
,~
---
---
t-4T3
---
A490 bolts
100
200
500
T1
1000
2000
5000
FIG. 4.22
25
., I
"rl
(/)
...
20
Q)
r-1
(j
>-
F(TO)4-1
(R = 11.7 ksi; N
164,000)
(R = 10.4 ksi; N
32,000)
'51
~
Q
",.1
15
(/)
(/)
Q)
~
II /
C/)
~
r-1
0
I=Q
10
Q)
bO
m
~
Q)
~
lH
0
Q)
F(T4)4-2
= 3.2 ksi; no failure)
~ (R
bO
Q
o
1
2-
10
100
1000
10,000
I--'
LV
+:--
135
= nominal area of
bolt shank
R
f
OL-______________________
o
Applied Load
FIG. 4.24 METHOD OF ESTIMATING THE LOAD WHICH WOULD CAUSE FAILURE
AFTER 2,000,000 CYCLES WHEN APPLIED TO A PARTICULAR TEE-CONNECTION
136
Bolt row
Bolt line
~
C
l
I.
I~
C
3
.. I
C
2
.ITI .
C
6
1O
: grip length
FIG. 5.1
...
137
~r
Bolt
1"
,a
line \
16\
b
--i
f-'.--
"I
-L-
w:
d:
bolt diameter
FIG. 5.2
0.6------------~--~--------r-------~
a = 2 1/2 in.
wid 2 = 4.01 in. -1
1 3/4 in.
1 1/16 in.
2
wid = 6.69 in. -1I
t
0.5
Analysis
Eq. (6.6)
;j
,,-...
Eq. (6.7)
P-l
-......
0'
'--'
0.4
It
OJ
~
B
*, ,
0.3-
cd
'M
~
cd
0.21
iVI
?,-r
bO
'M
:>..
P-l
0.11
0.0
.,
~2
2!
2
~2
21
2
3l
2
Distance From Bolt Line to Face of Web Minus 1/16 inch, b, inches
~
FIG. 5.3
l0
00
0.6-.------~-----r------r_----I
a
= 1 3/4 in.
= 1 1/16 in.
a = 2 1/2 in.
wid 2 = 4.01 in. -1
Analysis
::J
1/
.' j
- - - - Eq. (6.6)
- - - Eq. (6.7)
.........
0'
'-'"
0.41
II
Q)
~
cU
"M
~
r--t
0.31
II I I
cU
"M
~
cU
bO
Q
0.21
//
"M
:>...
:/.
0.1,
r.
II
J
~2
2!
2
~2
2!
2
3!
2
Distance From Bolt Line to Face of Web Minus 1/16 inch, b, inches
FIG. 5.4
LU
~
0.6
1 3/4 in.
t = 1 1/16 in.
-1
w/d 2 = 6.69 in.
'\,
\
0.5
Analysis
- - ' - - Eq. (6.6)
- - - Eq. (6.7)
;:::3
,--.
-P-l
0'
0.4
...
Cd
wid
'r-!
M
:::>
3 1/2 in.
<lJ
~
f\,
4.01 in.
-1
0.3
Cd
0
'r-!
~
Cd
0.2
~~.
r-. ........
bll
......~
..........
'
'r-!
:>..
H
P-l
- .....
0.1
--
--.-"'r---...
----- - t-=:::-- --- --
~--==
~
2 in.
~.-
0.0
I-'
+:--
0.8
1 3/4 in.
1 1/16 in.
wid
0.7
= 6.69 in.
---,,-...
;:1
---
...........
CY
wid
-1
3 1/2 in.
= 4.01 in.
-1
Analysis
Eq. (6.6)
Eq. (6. 7)
0.6
......."
...
(l)
.w
rd
or-!
.w
r--l
0.5
P
.w
rd
or-!
.w
rd
p:::
0.3
7'
<If!:.
~ "-
L.
L.
,',
,~
bJ)
s::
or-!
>.
t'--- _
0.2
--....;;;
0.0
i
4
I--'
+:--
I--'
0.6
0.5
,.......::l
P-l
...........
0'
'-"
...
Q)
.w
m
S
or-f
.w
r-I
0.4
=1
wid
2 1/2 in.
4.01 in.
-1
Analysis
Eq. (6.6)
Eq.
(6.7)
0.3
"" ,
.w
"
or-f
.w
1 3/4 1n.
.
w/d 2 = 6.69 in.- 1
5/8 in.
0.2
bO
r.:l
or-f
~
P-l
0.1
0.0 5
8
'"
3
7;
"8
I!
8
II
2
1~
8
124
"
2
21:
8
FIG. 5.7
+:-N
0.6
~I
\
\ \\
P-I
CY
1\
f\
",--.
'-..
0.4
'-"
0
~
0
',...j
~
Ctl
0.2
rY:
-1
Analysis
Eq.
Eq.
(6.6)
(6. 7)
'~~
'" ~
''''''~~
<
"" , -1/<
""
"
Ctl
= 4,01 in.
'~~...J
,
-1/
,,~t0.
~,
,,"
rl
" ",,~\
0.3
"
',\
ClJ
Ctl
",,
',...j
"
w/d
2 1/2 in.
~" "'-,
= 1 3/4 in.
-1
= 6.69 in.
\
;j
lYl/d
1 5/8 in.
b
\
..
0.5
~,
,,'-
"
"
~""l
-('
l)
bO
r::
',...j
H
P-I
0.1
~~,
'~
3
7
8
Il 11:
12
8
Flange Thickness, t, inches
8
12
4
12
8
2!8
f-J
FIG. 5.8
0.6
\,
0.5
,\
~
;j
r--..
--..
CY
.,
(l)
oW
ttl
~\
',,'\
0.3
\\
'M
l'
,,~,
"
\'
\' f\
',',,
\~
\,
oW
ttl
oW
ttl
'\
"\
" .,..
'M
t~ \
'-',,\
oW
.-;
0.2
'M
\.
f\,' \
\e\
"
bO
P-l
0.1
a
0.0
= 1 5/8 in.
= 1 3/4 in.
a
b
3/4 in.
t
L
6"
- - - Eq. (6.7)
",
0.4
'-'
Eq. (6.6)
~\
r\
P-l
---
,\
Analysis
10
..
= 1 3/4 in.
= 2 1/2 in.
2 5/8 in.
3 1/2 in.
1 1/16 in.
1 1/2 in.
\,
'\
"-
\
\
_ _ _ _ -------------'-_
10
"W
2 ' inches-
""
""
10
!-I
+:-+:--
FIG. 5.9
0.6
~ \,
'\
'\
"
...
oW
til
'M
oW
0.3
\\~
"
,..;
\\,\
II
oW
til
\\
0
. 'M
oW
til
'~
\..,
(J)
Eq. ( 6 . 6 )
I\'\.
",\
\'\
, ,
- - -
\\
,~
0.4
.'\
",\
;j
'\~\
Analyses
II
~\ \,
" ",\
,-.....
0'
........,
0.5
.............
~\
\\,
0.2
p..
bO
\\\
'r-!
~
'\.\~
1-l
0.1
a
b
= 1 5/8 in.
= 1 3/4 in.
= 3/4 in.
0.0 6,
2 1/2 in.
= 1 3/4 in.
= 2 5/8 in.
3 1/2 in.
= 1 1/16 in.
1 1/2 in.
10
10
"'
'"
"
10
w
1
d 2 ' inchesI--'
-l>VI
FIG. 5.10
146
30F-----~,----~,------~,----~,----~,----~,~----~,----~,----~,~----r-,----,
25
20 CI.l
CI.l
15
t-
10
t-
Cf)
+J
rl+J
Orl
r:Q
:::
-ri
0
..0
----;::~~;4
"",,,...,
L
H
P.
CI.l
cU
5 >-
CI.l
Cl) P.
H 'ri
CJ~
:::
'"
..
;:jp::;
cU
25 -
U Cl)
"'0
,..c: ;:j
CJ +J
ri ri
,..c: :::
20 -
cU
04-4
Cl)
-ri
rl
P.
P.
<
Sl
..,.,~
4 ksi
R = 4 ksi
.377 P
......
Connections with
A325 Bolts
.429 P
I
I
o Symmetrical tee-connections
with fill plate
o
to rigid abutment
15
.. .
./
"",
x
"",~.
~
~",.. ......... ~
/.
",..
~~
,..,.,
,,/"
".
(All the data shown was
~
...AI"
obtained analytically)
. / ...... ,.- ,...
10 ~
./~.,.,,~
Connections with
",..
A490 Bolts
,....-:~"...,
5r".,...~
P 3 k Sl.- .424 PU
/",R = 3 ksi
~~
p . = .365 P
~, 1
I
I
~ kSl I
IU
I
I
I
1
00
5
10
15
20
25
30
40
35
45
50
,..,.,
\2:
FIG. 5.11
".,...
x Connections of tee-section
bO
cU
"d
't
Symmetrical tee-connections
without fill plate
"d::E
O ......
",../
",..
p::;
CI.l~
~..;;;-'
Cl)
CI.l
c9
~"..., .......-'
,.., ~.~
",.. ~"
Cl)
Cl)
..
o
Cl)
+J
55
147
y
Beam with
depth t;
width w
T
~
____________________
1 ____________________
~~
IQ
I
----- x
t
I
I
(a)
(b)
FIG. 6.1
x=b
148
1.0
0
I
0.8
~----------~----------~--------~~--------~~--------~
0.6
0.4~----------~--------~~--------~----------~----------~
~------~--~~~------~--------~~--------~----------~
0.2
o
o
O.O . .~______~__________~________~~________~________~
0.2
0.0
0.4
FIG. 6.2
0.6
u
0.8
1.0
149
1.0
I
0
0.8
0.6~--------~----------~--------~-'r-------r---------,
'"
,j..J
0.4~----------~--------~~r-------~----------~r---------~
cO
"ri
,j..J
,.....,
;:J
,j..J
cO
"ri
0.2~------~~~~~~-+----------r---------~---------;
,j..J
cO
o
O.O~~
______
________
0.2
0.0
__________
________
0.6
0.4
__________
0.8
1.0
FIG. 6.3
150
1.0
A325 bolts
Analyses,
Tests, A325 bolts
0
0.8
0.6
I
0.4
0.2
~
0
.0
I
0.0
0.0
0.2
0.4
0.6
0.8
1.0
FIG. 6.4
151
1.0
C/)
Cl.l
C/)
r-i
cU
<:t:
0.8
'"0
cU
C/)
+.J
C/)
Cl.l
0
H
0.6
....
..
::l
--P-l
...........
0'-"
....
Cl.l
0.4
I
+.J
cU
'r!
.j..J
r-i
0
cU
0.2
.j..J
cU
p::;
bD
'r!
0
0
H
P-l
.j..J
0
'r!
CD
0.0
0.0
0.2
0.4
0.6
0.8
1.0
FIG. 6.S
152
y
(a)
center line
of tee-section
(b)
CI)
co
'r-!
~
0..
CI)
r-1
C\l
.w
/::,
FIG.
A.I
153
D. ,
1
i=l, 2, .. , nc
FIG. A.2
154
(/)
v ~ [7
/
(l)
/ r'\..
r'\.
p
H
+oJ
oW
['.
~V
A/ ~
'Y
"..."-V
..
['\..
.'\.. V
L.L
/~
v~r7
"- V["'\.
.Ie
v.
f'...
V
f'-.
[7
. . . . . . A ...
. . . . VYyV'VV
r::t:l
l'..
r::t:l
:z;
Pt-l
/ f'\...
-=\~
/'vo
"
LL~
V
/ f'\...
'\..
r::t:l
/ ~
YVY'VYYV'IY
Cf.l
:z;
/ ['\..
V
/ I'-..
'\.
V
[7
VY
[7
A A ..........
"V
[7
~
.... A
V'
lY
J i'-.
....
.A'"
'\..
~
/
. A
(1-N)
0...
. A
'\
l--4
AA
[7
/ r'\..
[\"
,y 'yYYI
"'"
V
S V "" "\ V ~ V "
7 !"\::
I'-.
'
"
V f7f'... ~
V ~V
f'...
"
V "- V ~ V " "-7 V
.", / "' "- V V ~
V
"" '" "'V "- V ~ V '\V
H
H
:z;
H
,,[/--S
"V
:>
H
/ ['\..
r::t:l
[7
:z;
/ ~
['\..
U
r::t:l
Cf.l
I
r::t:l
r::t:l
H
17
V '""- ~
"
./
V" ~
--/""'- ~ " V '" "~
"" Vf". /~
V ""'" '"~ '" ""'-'"t/
V '" '" ~
'" '" ~
~
V
"
:"
"
"
-/
V
"" V
~
" ~ " "
/
V
/ f'\.
~V
/ ['\..
/ ['\..
''\
"'1/
17
1/
['\..
'\
[7
;;
1/
/ f'\...
r7
<
/ ~
'\
17
uo
Z uo 1=:n:=l.lB d
1 uo 1==l1==l.lBd
155
I....
-I
Ci:l
Type II
Type I
Y, v
Ci:l
X, u
Type III
Thickness of element
FIG. A.4
156
(a)
:5
Reactions:
{FB}
Spring forces:
{t }
s
Spring deformations: {c5 }
s
Stresses:
{ST}
Reactions:
(b)
{FA};
displacements:
{UA}
Reactions: {FB}
Spring forces:
Forces:
Stresses:
A.5
Spring deformations:
{ST}
Forces:
FIG.
FD{l}
nc
{FA};
{FC}
displacements:
{UA}
{DC}