Upload
others
View
7
Download
0
Embed Size (px)
Citation preview
Lehigh UniversityLehigh Preserve
Fritz Laboratory Reports Civil and Environmental Engineering
1968
Static tests on unsymmetrical plate girders: maintest series, September 1968 (Schueller M.S. thesis)(70-21)W. Schueller
A. Ostapenko
Follow this and additional works at: http://preserve.lehigh.edu/engr-civil-environmental-fritz-lab-reports
This Technical Report is brought to you for free and open access by the Civil and Environmental Engineering at Lehigh Preserve. It has been acceptedfor inclusion in Fritz Laboratory Reports by an authorized administrator of Lehigh Preserve. For more information, please [email protected].
Recommended CitationSchueller, W. and Ostapenko, A., "Static tests on unsymmetrical plate girders: main test series, September 1968 (Schueller M.S. thesis)(70-21)" (1968). Fritz Laboratory Reports. Paper 263.http://preserve.lehigh.edu/engr-civil-environmental-fritz-lab-reports/263
Unsymmetrical Plate Girders
STATIC TESTSON UNSYMMETRICAL
PLATE GIRDERSMAIN TEST SERIES
byWolfgang Schueller
Alexis Ostapenko
" September 1968
Fritz Engineering Laboratory Report No. 328.6
STATIC TESTS ON
UNSYMMETRICAL PLATE GIRDERS
MAIN TEST SERIES
by
Wolfgang Schueller
Alexis Ostapenko
This work was conducted as part of theproject Unsymmetrical Plate Girders, sponsoredby the American Iron and Steel Institute, thePennsylvania Department of Highways, and theWelding Research Council. The opinions,findings and c9nclusions expressed in thisreport are those of the authors, and notnecessarily those of the sponsors.
Fritz Engineering LaboratoryDepartment of Civil Engineering
Lehigh UniversityBethlehem, Pennsylvania
September 1968
Fritz Engineering Laboratory Report No. 328.6
328 .. 6
TABLE OF CONTENTS
ABSTRACT
1 . INTRODUcrION
2 . TEST PROGRAM
2.1 Introduction
2.2 Test Specimens
2.-2.1 Introduction
2.2.2 Girder Specimens
2.2.3 Material Properties
2.3 Reference Loads
2.4 Instrumentation
2.4.1 Vertical Deflections
2.4.2 Diagonal Panel Deformations
2.4.3 Lateral Web DeflectionS
2.4.4 Strain Measurements
2.~ Test Set-Up
3. TEST PROCEDURE AND RESULTS
3.1 Introduction
3.2 General Girder Behavior--Load vs. VerticalDeflection
3.3 Strain Distribution
3.4 Web Deflections
3.5 Diagonal Panel Deformations
3.6 Behavior of Individual Panels and Modesof Failure
3.6.1 Test UG4.3
3.6.2 Test UG4.4
3.6.3 Test UG4.1
3.6.4 Test UG4.6
3.6.5 Test UG4.2
3.6.6 Test UG4.5
-i
iii
1
3
3
4
4
5
7
8
8
8
9
9
10
11
12
12
13
14
15
15
16
16
18
19
21
22
24
328 ..6
TABLE OF CONTENTS (ContTd)
3.6 Behavior of Individual Panels and Modesof Failure
3.6.7 Test UG5.3
3.6.8 Test UG5.4
3.6.9 Test UG5.1
3.6.10 Test UG5.2
3.6.11 Test UG5.6
3.6.12 Test UGS.5
4. DISCUSSION AND CONCLUSIONS
5 . NOMENCLATURE
6. TABLES AND FIGUREp
7 . REFERENCES
8. ACKNOWLEDGEMENTS
-ii
24
25
26
27
27
28
29
31
33
98
99
328.6
ABSTRACT
-iii
Two unsymmetrical plate girders made of A36 steel were tested
under statically applied loading. Each girder had six panels and
the geometry and the panel arrangement of both girder specimens were
identical except that one specimen had a longitudin~l stiffener.
The centroidal axis was at approximately one-third of the girder
depth with one flange having twice the cross-sectional area of the
other flange. Both girders had a length of 37 ft. - 10 in. and a
web depth of 48 inches. For one half of the girder length a 1/8 in.
web plate was used, for the other half a 3/16 in. web plate. The
resulting web slenderness ratios were 269 and 414. The panel aspect
ratios were: 0.833, 1.15, 1.46 and 1.77.
Girder specimens, the test set-up, the testing procedure and
the results are desqribed and discussed.
The primary objectives of the tests were: verification of a
new theory developed for the determination of the ultimate load of
unsymmetrical plate girders (not described), and investigation of
the behavior of the girder components under an increasing load. The
layout and size of the girder elements were selected in such a way
as to define the moment-shear interaction curve in .the region of high
moment and high shear. Only the end panels of the girder were
subjected primarily to shear.
328.6 -iv
Correlation between the test results and the theory was with
in 5 percent. A comparison of the ultimate loads for the trans
versely and longitudinally stiffened panels of the two girders
(which were otherwise identical in geometry) shows that the intro
duction of a longitudinal stiffener into panels subjected to
combined action of high moment and high shear resulted in an
increase in strength of approximately 44 percent. For panels
subjected primarily to shear the increase of strength was 13
and 20 percent.
328.6 -1
1. INTRODUCTION
The design of steel plate girders was first based on the
theore'tical web buckling strength. However , it has been shown by
a research team at Lehigh University that there is no consistent
relationship between the web buckling strength and the ultimate
strength of a girder.(l-4) The current American specification for
transversely stiffened girders for buildings is based on these
findings. (9) Research has been also conducted on the contribution
of longitudinal stiffeners to the static load-carrying capacity of
plate girders.(6,7)
However, present design specifications do not take into consi
deration the behavior or strength of plate girders which are un
symmetrical with respect to the horizontal axis. In bridge con
struction, such an unsymmetrical condition is encountered in the
orthotropic deck or the composite deck systems. To consider these
cases as symmetrical, as is the common practice now, may 'be quite
unrealistic. For simple span beams, this approach would be too
conservative since the web plate will be primarily in tension and
its buckling strength as well as the ultimate strength will be
significantly increased. On the other hand, for continuous beams
at the points of intermediate supports, a greater portion of the
web will be under compression and it will be subjected to a
combined action of shear and bending. For this case the symmetrical
de~ign approach may become unconservative.
328.6 -2
In 1966 a new research project was started at Lehigh University
with the general objective of determining the behavior of unsym
metrical plate girders and of investigating their static ultimate
strength. The experimental phase of this research consisted of two
series of tests: the pilot tests and the main tests. The purpose
of the pilot tests was to give initial information on the behavior
of unsymmetrical girders for the development of a theoretical
approach. (8) The main tests are described in this report. Prepar-
ation for the tests, the testing itself, and a discussion of the
test results are presented. A detailed evaluation of the test data
and the results of the theoretical study will be reported separately
later.
328.6 -3
2 • TEST PROGRAM
2.1 Introduction
The primary objective of the main tests was to verify the
correctness of the newly developed theory for the ultimate
strength of unsymmetrical plate girders.
Two 37 ft.-lO in. long and 48 inch deep girders were tested in
this series. The A36 steel was used. One of the girders had longi
tudinal stiffeners in addltion to the transverse stiffeners. A
variety of panel sizes was chosen for the girders in order to check
the theory for different panel aspect ratios, a. AlSO, two web
slenderness ratios were selected--a high value of S = 414 and an
average value of ~ = 269. The centroidal axis was shifted by about
1/6 of the girder depth due to the unequal areas of the flanges.
Figure 1 shows on a moment-shear interaction curve the girder
test results which have been already obtained in other projects at
Lehigh University. The shear, normalized by the ultimate panel
shear, is plotted on the ordinate and the moment, divided by the
ultlmate panel moment, is given on the abscissa. In the portion
Ql to Q2 the interaction is such that the web will fail, from Q2 to
Q3 the flange will fail. As can be seen from the ~igure, the regions
of pure shear, pure moment, and high shear and low moment are covered
quite well.* Only the region of high moment and high shear has not
been adequately investigated. For this region previous investi-
*Since symmetrical girders are considered to be a special case ofunsymmetrical girders, experimental results found for symmetricalgirders were also included in this plot.
328.6 -4
gators have assumed some relatively arbitrary functions to define
the continuous shape of the interaction curve.(1-3) The purpose of
the main test series was to investigate this region.
2.2 Test Specimens
2.2.1 Introduction
The original intention was to have homogeneous girders with
the same yield stress of about 36 ksi in all components. However,
for the girders tested in the main test series, it was reasonable
to exp~ct that the yield stress of the component parts would be
different since the webs were very thin and the flanges relatively
thick.* This fact was found to be true for girders previously
tested at Fritz Engineering Laboratory. The actual strength of a
supposedly A36 ksi steel was in the range of 27 ksi to 48 ksi. It
can be seen in Table 1, for example, that a 12 in. x 1.378 in.
flange(5) had a yield stress of 27.2 ksi; whereas, a 50 in. x .195
in. web(6) had a yield stress of 48.6 ksi.
To control this undesirable effect, the material was not
ordered from the mill, but selected in the stock yard. A portable
Rockwell Hardness Tester** was used to match yield stresses. Still,
the 1/8 inch web turned out to have a higher yield stress of 55.5
ksi than the anticipated 36 ksi.
* Girders used in practice are in general bigger than the testspecimens. Hence, because of the larger sizes of the girdercomponents the yield stress of the different girder parts isprobably more uniform.
**Manufactured by Riehle, Model PHT-2.
328.6 -5
Because it was undesirable to fail more than one panel per
test, the girders were designed so that the failure loads were
markedly different for each panel. Since the yield stress of the
girder components was not known initially, a most probably combin
ation of the yield stresses in the flanges and the web was selected
and ~he design was then checked for other possible yield stress
combinations.
2.2.2 Girder Specimens
Of the two girders te~ted, ~ne girder had transverse stiffeners,
the other had transverse as well as longitudinal stiffeners (Fig. 2).
The sizes of the girder elements, which will be given below, are
nominal sizes. The actual sizes are discussed in Article 2.2.3.
Both girders had the same lengths and panel arrangements. The length
was 37 ft.-10 in. and th~ web height 48 in. (Fig. 3). The aspect
ratios, defined as the ratio of the panel length to the panel depth,
were a = 1.77, 1 ..46, 1.15 and 0.833. One half of each girder had a
1/8 in. thick web and the other half had a 3/16 in. web. The result
ing slenderness ratios were S = 414 and 269. The small flange was
10 in. x 3/4 in. and the large flange 13 in. x 1 3/8 in. The longi
tudinally stiffened girder had the intermediate stiffeners only on
one side of the web, on the other side of the web, ~mall 12 in. x 4
in. x 3/4 in. plates were placed against top and bottom flanges (at
the points where diagonal braces were used) to prevent local defor
mations. The bearing stiffeners (4 in. x 3/4 in.) at the load and
at the two end supports were on both sides of the web. The size of
the horizontal stiffener was 3'in. x 3/4 in. placed on both sides of
the web at 12~ in. from the top flange.
328.6 -6
The load was applied to both girders at mid-span as shown in
Fig. 9. The shear and moment diagrams corresponding to this load
are also shown in the figure. It can be seen that the outer panels
were acted upon by high shear and low moment, and that the central
panels were acted upon by high shear and high moment. The point of
failure for each panel was plotted on an average interaction curve
shown in Fig. 10. The location of the failure points on the inter
action curve defines quite well the portion of high moment and high
shear. Actually, every panel has its own interaction curve with
its failure point located on the curve. However, for reasons of
simplification, the panel failure loads in Fig. 10 were shown all
together on one curve.
On purpose, three panels were designed with the same aspect
ratio, ~ = 1.77. The two outer shear panels were planned to give
an indication of the influence of the web thickness on the capacity
of the panels through direct comparison of the results. Comparison
of the outer shear panel with the central panel would show the
influence of the flexural stresses on the shear capacity. Another
interesting comparison was to be obtained by having the two central
panels designed for essentially the same load: one panel with a
thick web and a wide transverse stiffener spacing, the other panel
with a thin web and a close transverse stiffener spacing.
The reason for selecting the same geometry for both girders was
not only an economical consideration but, more importantly, the test
results from both girders could be directly compared and the impor
tance of the longitudinal stiffener could be evaluated.
328.6 -7
2.2.3 Material Properties
The actual dimensions of the component plates of the test
specimens (Table 2) were obtained from measurements on coupon
plates cut from the various plates before fabrication. Figure 4
shows the typical location of these coupon plates with respect to
the girders. The location of the coupon plates was a function of the
importance of the strength in the respective area. Thus, for the
web as shear carrier, the coupon plates were cut from the end,and
for the flanges, the COupoTI.plates were cut closer to the center of
the girder. The points on the coupon plates where the width and
thickness were measured are shown in Fig. 5. These measurements
were averaged and tabulated in Table 2.
Tensile coupons were cut from each of the coupon plates as
shown in Fig. 6. For each web coupon plate, three tensile specimens
were cut parallel to the direction of rolling and two perpendicular.
Figure 7 shows the shape and dimensions of the tensile specimens.
For the flanges an 8-inch gage length was used and for the webs a
4-inch gage length. Although ASTM specifies a gage length of 2
inches for sheet metal, a gage length of 4 in. was used in order to
get more conformity between specimens from different parts of the
girders.
Results of the tensile tests are shown in Tables 3 and 4. In
each case they represent average value~ from the tested tensile
specimens. The typical stress-strain diagrams of the flange and
web material are shown in Fig. 8. Although the flange shows three
328.6 -8
distinct portions of elastic, plastic)and strain-hardening ranges,
the web shows no separate flat yield and strain hardening portions.
The static yield stress was found by the intersection of two
straight lines. One line is drawn parallel to the elastic part of
the stress strain curve; the other line connects the three kink
points, which were found by stopping the machine during a tensile
test and allowing the load to settle to a static value.
The percentile elongation, the percentile reduction in cross
sectional area, and the ultimate strength were obtained from the
tensile tests and are shown in Tables 3 and 4. The chemical compos
ition of the plates as listed in the mill test reports is also shown.
2.3 Reference Loads
The theoretically predicted shear force, Vth , was used as ref
erence load. It is based on the theory for unsymmetrical plate
girders, which is being developed by Chingmiin Chern. The ratio of
the experimentally obtained shear force and the theoretical shear force
indicates the accuracy of the newly developed theory. The loads and
the accuracy for each test are given in Table 4. It is seen that the
error is in most cases within 5 per cent.
2.4 Instrumentation
2.4.1 Vertical Deflection
The locations for the readings of the vertical deflection of the
girder were as indicated in Fig. 23. Ames dial gages with the finest
division of one thousandth of an inch were used. The gages were
placed in such a way as to measure the displacement of the test panel
328.6 -9
relative to the displacements of the neighboring panels. Thus,
deflection of the panels adjacent to the test panel had to be
recorded. The deflection at the center line of the girder (under
neath the load) was used as an indication of the overall behavior of
the girder under load.
The same approach as described above was used for the deflec
tion readings of the longitudinally stiffened girder.
2.4.2 Diagonal Panel Deformations
Diagonal deformations of every tested panel wer~ measured in an
arrangement shown in Figs. 23 and 24. In Fig. 25 the details of the
attachment are indicated. A dial gage was fastened to a 1/8 inch
thick and l~ inch wide plate bent in the shape shown and clamped to
the vertical stiffener. A wire was fastened to the extension arm of
the dial gage at one end and to another plate at the other end.
For the longitudinally stiffened girder, the diagonal defor
mations were also measured for every test panel. However, in this
case,the diagonal deformations of the lower subpanel were also
recorded as shown in Fig. 24.
2.4.3 Lateral Web Deflections
The stations at which lateral deflection readiDgs were taken are
shown in Fig. 44. Every panel has its own coordinate system. The
x-coordinates for the different stations are given in the table of
Fig. 44. An aluminum dial rig was used to measure the lateral
deflection of the web. It is shown in Fig. 45. The dial rig consists
of a channel which is held on the top flange by a magnet and on the
328.6 -10
bottom flange by a bar. Fastened to the channel are angles which
support the dial gages at their ends. The advantage of this
aluminum rig against the one used by previous girder projects lies
in the small weight and in the flexibility of usage. The locations
of the dials or y-stations for the different tests. are given in
Fig. 45.
After each set of test readings reference readings were taken
to check against accidental movement of the dia~ gages.
2.4.4 Strain Measurements
Electrical resistance strain gages were used to measure strains.
The gages used were SR-4, A-l-S6 linear and SR-4, AR-l-S-6 rosettes
made by BLH Electronics, Inc.
For a typical shear panel, as shown in Fig. 52, the gages were
placed at points in the area of the tension field pattern which
develops after the buckling of the web. Strain gages were also
placed on the flange at points where hinges were expected to form.
(The concept of hinge development in the frame formed by the top
~no bottom flanges, and the vertical stiffeners was used in the
theoretical phase to find the ultimate shear capacity of the panel).
Since the strain gage results will be used in later fatigue studies,
the gages were also placed at points where fatigue cracks were
expected to occur, that is, along the web boundaries in regions
where the tension field is anchored, as shown in Fig. 52. The
flange stresses were of interest also. Similar reasoning was
applied to the layout of the gages for the panels under shear and
moment.
328.6
No strain gages were used at points of stress concentration
caused by either load application or support reactions.
The location and kind of gages are given for all panels of
both girders in Figs. 53 to 56.
-11
Strain-measurements were recorded by a B & F 96 channel strain
recorder* with a keypunch.**
2.5 Test Set-Up
An overall plan view of the test set-up is shown in Fig. 11.
The girders were tested in a hydraulic universal testing machine with
a capacity of 5,000,000 Ibs.*** The loads were applied at the mid-
span of the girder. The test specimens were simply supported on
rollers at both ends (Fig. 12). The load was transferred from the
machine crosshead to the girder through a spherical bearing block
(Fig. 12). The compression flange was laterally supported at the
vertical stiffeners by means of 2~ inch diameter steel pipes.
The lateral bracing against the machine column is shown in Fig. 13.
The bracing was designed to permit a sufficient vertical deflection
of the girder by the rotation of the pipes at the girder and at the
cross beam. As an additional precaution~... t:he pipes were slightly
inclined so that they would not influence the deflection of the girder.
A cross beam was used to support the lateral braces, as- can be seen in
Fig. 14. The cross beam itself is supported by two heavy columns
which were laterally braced to the fl~or (Fig. 14).
* B & F Instruments, Inc., Philadelphia, Pa.** IBM 526 PRINTING Summary Punch***Baldwin-Lima-Hamilton Corp., Philadelphia, Pa.
328.6 -12
3. TEST PROCEDURE AND RESULTS
3.1 Introduction
In this chapter the behavior of the girder and of the different
girder elements during testing is described. The ~esults of the
twelve tests are presented in the form of the following values
plotted against the applied load: mid-span deflection, web deflec
tion, diagonal panel deformation, and strain distribution. The
ultimate load was considered reached when a continuous increase of
the diagonal panel deformation in the compression direction was
observed with essentially no increase of the applied load.
Initial readings and readings at every load increment were taken
for all the measurements described in Art. 2.4. Smaller load incre
ments were taken in the inelastic range. Readings were made only
after the deformations had stabilized.
The coordinate system used for the following discussion is shown
in the nomenclature. Every panel has its own coordinate system with
the y-axis (vertical axis) located at the center line of the left
vertical stiffener,as shown in Fig. 44, and the x-axis is positive
to the right. The z-axis is perpendicular to the plane of the web.
The side of the specimen in the positive z-direction will be called
the near side, while the other side will be called the far side.
The transversely stiffened girder was tested first. The
sequence of testing is shown in Fig. 15. In order to determine
the behavior of the girder with the small flange on top and thus
328.6 -13
with the larger portion of the web in compression, panels 3, 4
and 1 were tested, in that sequence. After turning the girder
around, that is, putting the large flange on top, the case of the
larger portion of the web in tension was investigated. Panels 6,
2 and 5 were tested. The te~ting sequence of the longttudinal1y
stiffened girder is shownjn Fig. 21.
3.2 General Girder Behavior--Load vs. Vertical Deflection
The testing history and the overall behavior of each test
specimen is depicted in the load-deflection diagram. Load P is
plotted as the ordinate and the mid-span deflection as the abscissa.
The first deflection reading was taken at load No.1 (p=Ok).
Then the load was gradually increased up to load No.2. This
procedure was repeated up to the point where inelastic behavior or a
substantial increase in deflection per unit load was observed. At
this point the girder was unloaded. A seoond cycle was then started
and continued until the dynamic ultimate load was reacheq. Readings
were taken only after the load and the centerline deflection had
stabilized. The waiting period for the stabilization near the
ultimate load took approximately twenty-five minutes.
Residual stresses due to welding and initial deformations were
the reason for using two cycles. Thu~ the first cycle:was intended
to partially relieve any initial effects.
328.6 -14
3.3 Strain Distribution
Linear strains were plotted to give an indication of the
behavior of the compression flange and of the longitudinal stif-
fener under load. The plot of the strain across the top flange
indicates the curvature and thus the lateral defleQtion cl the top
flange. For example, in test UG5.3 (Fig. 63), where panel failure
was initiated through lateral buckling of the flange, one clearly
notices that at the ultimate load (Load No. 14) buckling causes the
outer fibers to be in tension. However, the longitudinal stiffener
shown below does indicate lateral deflection but no buckling. at the
ultimate load.
Further, a plot of strain versus load, the strain gages being
located on opposite faces of the flange, gave an indication of the
vertical behavior of the flange. In test UG5.3, the strain was
equally distributed up to load No.3; thereafter, the strain distri-
bution was unequal because of vertical deflection of the flange.
Thus, the top surface was subjected to a larger compression strain
than the bottom surface. Figures 61 and 62 show that in test UG4.1
the premature failure of the panel in the first cycle was initiated
by vertical buckling of the top flange. Due to initial deformations
the initial strains on the top side of the flange are different from
those on the bottom side. Gage 71 on the bottom of the top flange
indicates that at load No.7 vertical buckling occurs and a
compression ·strain changes into a tension strain.
--lI
328.6 -15
The plots of the compression flange strains thus indicate
whether the vertical or lateral buckling of the flange occurred.
3.4 Web Deflections
For every test panel the x-station with the maximum lateral
deflections was plotted as shown in Figs. 46 to 51. The change of
web deformations due to an increase of loading for the respective
station is given. Furthermore, the variation of the lateral
deflection at the point of maximum deformation with the load is
plotted. One notices that in these plots the slope of the curve
decreases to a horizontal line at the ultimate load. The lateral
deflection at the ultimate load for larger panels was naturally
bigger than for smaller panels.
When comparing the lateral deformation pattern of the trans
versely stiffened girder with the one for the longitudinally stif
fened girder, it becomes obvious that the longitudinal stiffener
controls the deformations of the web. Due to its rigidity the
longitudinal stiffener . was able to force a nodal point in the
deflected shape of the web at its location.
3.5 Diagonal Panel Deformations
A good representation of the behavior of a test panel under load
is given by the load-vs-diagona1 panel shortening diagram. This
diagram also gives a good means of determining the ultimate load of
the panel. For test specimen UG4.3 (Fig. 31), for instance, the
ultimate load was reached at load No. 23. The horizontal curve
328.6 -16
starting at this point indicates that the rate of diagonal defor
mation is much higher than at the previous load points.
In the following discussions the diagonal shortening will
be used as an indication of the test panel behavior as well as
an indication of its ultimate capacity.
For the longitudinally stiffened girder the overall diagonal
panel deformations and the deformations of the lower subpanel were
measured and plotted. By comparing some of the plots, one notices
that the diagonal web deflections at the ultimate load are applroximately
equal when the panel failure is due to yielding of the lower sub-
panel (Fig. 39). However, when the panel failure is due to web
yielding of both subpanels, the overall diagonal panel deformations
are much greater than those of the lower subpanel (Fig. 40).
3.6 Behavior of Individual Panels, and Modes of Failure
Each test is described according to the sequence of testing.
Thus, the test numbers which were assigned according to the location
of the panel in the girder, may appear to be out of order. First,
the tests on the transversely stiffened girder UG4 are described
and then on the transversely and longitudinally stiffened girder
UG5. For a summary of the sequence of testing see Figs. 15 and 21 0
3.6.1 Test UG4.3
Test UG4.3 was the first test on girder UG4. Since the
computed moment capacity of panel 4 was approximately 10 percent
higher than the capacity of the test panel, the capacity of panel 4
328.6 -17
in bending had to be temporarily increased. A wooden diagonal
brace was put into the panel on each side to transfer some of the
compression force of the top flange into the bottom flange (Fig.
69). Also, the top flange was laterally supported at the center
of the panel. Thus, the lateral buckling capacity of the top
flange was increased.
The detail for the lateral bracing ·is shown in Fig. 16. An
angle was clamped and welded to the flange. The pipe bracing was
connected to the angle.
Development of the tension field was quite pronounced at load
No.8 (P=70k ). Yielding of the compression flange was first
observed at load No.9 (P=80k). The ultimate load was considered
kreached at load No. 23 (P=126.4). At this load, general yielding
of the compression flange throughout its thickness was observed,
and the rate of diagonal deformation was much higher than the rate
of loading,as can be seen from Fig. 31. Note that the load
deflection diagram (Fig. 26) was not completely flat, apparently
due to the moment gradient and strain hardening. At the ultimate
load the flange started deflecting vertically and horizontally over
the full width of the panel (Fig. 68).
The experimental ultimate load was 7 percent above the predicted
ultimate load.
328.6 -18
3.6.2 Test UG4.4
Before this test was conducted the moment capacity of the
already tested panel 3 was increased by welding a 2 in. x 8 in.
plate on the top flange. Since the moment capacity of panel 2
was only about 20 percent higher than of the panel .to be tested, a
provisional flange reinforcement was provided for it. The details
for the flange reinforcements are shown in Figs. 17 and 18. The
provisional flange plate was lying on four filler plates. Only
the end filler plate at panel 1 was welded to the flange and to the
plate reinforcement since at this point the compression force in
the flange had to be transferred to the plate. At the other end of
the provisional reinforcement plate, support was given by
bearing against the flange reinforcement of panel 3. The buckling
of the flange reinforcement of panel 2 was prevented by clamping it
at the filler plates to the top flange (Figs. 17 and 18).
The flange reinforcement was designed so that the panel capac-
ity was high enough to carry the failure load for the last panel to
be tested.
The shear capacity of panel 1 and the already tested panel 3 was
increased through tension diagonals applied at both sides of the web.
One inch diameter high strength steel bars with a yield stress of
156 ksi at 0.2 percent offset were used. The connection detail for
the tension diagonals is shown in Fig. 20.
328.6
The advantage of this method over welding a rectangular
steel bar diagonally to the web, as it was done in some previous
projects, lies in its flexible usage. This method is not depen
dent upon the size of the panel. The reinforcement can be used
for temporary panel strengthening, and it can be easily applied
and removed by student workers. Thus the dependence on the
-19
laboratory technicians was eliminated. Also the costs of fabri-
cation and application were reduced, since the same diagonals were
used for the testing of the longitudinally stiffened girder (UGS).
At load No.7 (P=llOk) yielding was first noticed in the
compression flange close to the point of load application. Complete
plastification of most of the top flange was reached at load No. 9
(P=139.Sk ). This load was considered the ultimate load because the
rate of diagonal deformation was much higher than the rate of load
ing (Fig. 32). Also, the local vertical buckling of the compression
flange started. This can be seen from the yield pattern of the web
in Fig. 67. The experimental ultimate load was 5 percent above
the predicted ultimate load.
3.6.3 Test UG4.1
Before this test was conducted, the moment capacities of panels
4 and 5 were increased as shown in Fig. 19. On the, top flange of
panel 5 a provisional reinforcement plate of 6 in. x ~ in. was
placed, the principle applied was the same as used for panel 2.
The flange reinforcement for panel 4 h'ad the same size as the one
for panel 3. The webs of panels 3, 4 and 5 were strengthened with
tension diagonals.
328.6
At load No.5 (P=llOk) the tension field was quite pro-
-20
nounced. Yielding in the tension field first appeared at load
No.6 (P=130k). At the same load, yielding started in the web of
the end post. The ultimate load was reached at load No.8
k(P=145.6). ,This value was considerably below the.theoretically
predicted value.
The failure was premature and 'it was considered to be due to the
yielding of the web of the end post. The end pos~ which can be
considered as a beam carrying the tension field, was not capable
of generating enough capacity for the development of the full
tension field. To strengthen the web of the end post a diagonal
steel bar was welded into the post thus creating arch action (Fig.
72). With this repair the girder was tested again. However, the
ultimate load obtained was only one kip higher than in the test
before.
This time, the failure was due to the vertical buckling of the
top flange. The flanges should be able to support the end post
carrying the tension field. Hence, the top flange acting as .a beam-
.column loaded with the vertical component of the tension field and
the compression force due to the end post column was not able to
resist this load and bent downwards. In order to test the panel
further, the top flange was reinforced with two 2 inu x 3 in. steel
bars which were clamped to the flange in order to increase its buck-
ling capacity and thus to induce failure due to tension field action.
The ultimate load was obtained at load No. 24 (P=163.2k ). Figure 73
shows the pronounced diagonal action of the tension field and its
328.6 -21
yielding along the field. The rate of diagonal shortening is quite
large as the horizontal portion of Fig. 33 between the load Nos.
23 and 24 indicates. The experimental ultimate load agreed exactly
with the predicted ultimate load ..
3.6.4. Test UG4.6
The girder was turned around so that the larger flange was in
compression and the larger portion of the web in tension.
Because panel 6 was to be tested (Fig. 15), diagonal tension
reinforcement was put. in panels 1 and 2. In addition, the
diagonals of panels 3 and 4 had to be relocated into the new
tension direction of the panels.
The girder tested in the first three tests with- the small flange
in compression was forced into a certain deformed shape. Now, with
the girder inverted, this deformation pattern had to be changed into
the opposite direction. The transformation of the initial web
deformations of panels 1, 3 and 4 into the opposite direction was
accompanied with sudden loud "bangs" of the webs. The process of
reshaping was finished at load No.5 (P=140k ). Yielding of the test
panels web in the direction of the tension field action was first
noticed at load No. S. The highest static load reached was at load
kNo. 10 (P=197.5). At this load the yielding of the web along its
tension field was quite pronounced (Fig. 74). It can be seen in
Fig. 35 that at this load the rate of ~iagonal shortening was much
higher than the rate of loading. The experimental ultimate load was
1 percent above the predicted load.
328.6 -22
3.6.5 Test UG4.2
Before this test was conducted, the provisional bottom flange
reinforcement in panel 2 was taken off, and the shear capacity
of panel 6 was increased by the placement of'tension diagonals.
The initial panel deformations underwent big changes 'as was
described in Art. 3.6.4. In order to give the web a chance to do
so, the tension diagonals were tightened only at the instant when
the panels showed development of a tension field action in the
new direction. If the bars had been tightened before testing, the
web panels would not have been able to develop much of the tension
field action; therefore, the tension diagonals would have had to
carry forces greater than their capacity.
At load No.3 with P=188k it was not possible to increase the
load because of the vertical buckling of the compression flange of
panel 1. This premature failure was due to the incapability of the
flange acting as a beam-column to carry the reaction of the end post
and to resist the vertical component of the tension field at the
same time. The buckling capacity of the flange was increased by
C-clamping a beam to the top flange.
Yielding in the test panel first appeared in the web in the
direction of the tension field action at load No.8. At load No. 12
a sudden weld crack developed at the bottom angle connection of the
diagonal reinforcement for panel l (Fig. 70). The diagonal rein- ,
forcement had rotated the bottom flange locally about the vertical
stiffener. This rotation caused cracking of the welds connecting the
328.6
plate (used for the provisional flange reinforcement) to the
bottom flange. By clamping the plate to the flange, the propa-
gation of the cracks was sufficiently arrested.
-23
The ultimate load of test panel 2 was reached at load No. 20
k(P=238.5). Yielding was observed in the web along its tension
diagonal and on the top of the compression flange. Als~ vertical
buckling of the flange seemed to start. The rate of diagonal
deformations was much" higher than the rate of loading. The experi-
mental ultimate load was 12 percent above the predicted value.
328.6 -24
3.6.6 Test UG4.5
Before panel 5 was tested, its provisional bottom flange
reinforcement was removed. Panels 1 and 2 had -to be reinforced
to increase their moment capacity. A plate was welded to the
bottom flange of panel 2 and beams were clamped to.the top flange
of panels 1 and 2. Reinforcement diagonals were provided for
panel 2.
Yielding first appeared in the web of the test panel at load
No.8 (P=220k ). The ultimate load was reached at load No. 14
(P=272k ). At this load the web showed the typical yield pattern
of· a plate under shear which had not buckled (Fig. 75). The
diagonal shortening was quite pronounced as is indicated by the
approximately horizontal portion of the curve in Fig. 36. The
experimental ultimate load was 4 percent above the predicted load.
This was the last test on the transversely stiffened girder.
3.6.7 Test UG5.3
Panels 1 and 4 of the longitudinally stiffened girder had to be
reinforced since their capacities were only about 10 percent higher
than that of panel 3. The shear carrying capacity of panel 1 was
increased by diagonal tension reinforcement,and the moment capacity
of panel 4 was increased by a wooden compression diagonal and by a
steel beam (lOWF54) clamped to the top flange.
At~load No. 3 (P~80k) yielding first appeared on the top of the
compression flange. At load No.5 (P=137k ) the flange yielded
throughout its thickness and over the full length of the panel.
328.6
However, this yielding did not affect the carrying capacity of
the panel. Yielding of the longitudinal stiffeners and yielding
-25
in the web of the smaller subpanel and along the tension diagonal
kof the larger subpanel appeared at load No. 11 (P=167). The
ultimate load was reached at load No. 14 (P=18Sk ). The failure
was due to lateral buckling of the compression flange (Fig. 79).
The web of the larger subpanel was yielded along its diagonal
tension field (Fig. 78). The ultimate load was 4 percent below
the predicted load.
Interesting was the deformation and yield pattern of the
smaller subpanelts web (Fig. 79). The web was deformed in a'
seemingly sinusoidal wave pattern inclined under approximately 45
degrees. Yielding appeared on the compression side (valleys) of
the waves.
3.6.8 Test UG5.4
Before this test was conducted a 8 in. x l~ in. plate was
welded on the top flange of panel 3 to increase the moment capacity
of that panel. Also, tension diagonals were added to help to
carry the shear. The lateral stability of the top flange of panel
2 was somewhat increased by clamping a wide-flange section to it.
Yielding of the top flange in panel 4 started at load ,No. 7
(P=179k ). Tension field yielding was first noticed at P=190k . The
ultimate load was reached at load No.8 (P=193.Sk ). The failure of
the panel seemed to be due to extensive yielding of the web of the
lower panel in the direction of the tension field action (Fig. 80).
328.6
Though. the compression flange was yielding, it did not become
-26
unstable and thus seemed not to initiate the failure of the panel.
The web of the smaller, upper subpanel was also yielding. Inter-
esting was the fact that two different modes of web yielding
appeared. One half of the panel showed the same yield pattern as
UGS.3, that means, tension field yielding at about 45 degrees.
However, the other half of the panel showed the yield pattern of
a plate before buckling (Fig. 80). At the ultimate load the rates
of both diagonal and vertical deformations were much higher than
the rate of loading (Figs. 39 and 37). The experimental ultimate
load was 8 percent below the predicted value.
3.6.9 Test UG5.1
Before this test was conducted the top flange of panel 4 was
reinforced by welding a 8 in. x l~ in. plate to the top surface
and providing diagonal reinforcement (Fig. 22).
At load No.6 (P=160k ) yielding first appeared in the web of
the larger subpanel in the direction of the tension field action.
At the same load yielding was noticed in the web of the smaller
8ubpanel. The ultimate dynamic load was reached at load No. 11
(P=200.Sk ). It was not possible to increase this load. Since
kthis load dropped to a static value of P=194.5 and load No. 10
kdropped only to P=195. 5 , ,load No. 10 was considered as the static
ultimate load. The failure of the panel was caused by yielding of
the upper panelTs web along an inclination of approximately 45
degrees, and plastification of the lower panelTs web in the diagonal
328.6 -27
tension field direction (Fig. 82). At the ultimate load the
diagonal panel deformations were quite pronounced (Fig. 82).
The predicted ultimate load agreed exactly with the experimental
load.
3.6.10 Test UG5.2
Before this test was conducted the shear capacity of panel
No.6 was increased by the addition of tension diagonals.
Yielding was first noticed at load No.9 (P=208.Sk ) on top of
the top flange; web yielding first appeared at load No. 11 (P=221.Sk).
The ultimate load of the panel was reached at load No. 14 (P=236k ).
The failure was du~ ,to plastification of the lower panel's web along
its tension field action. The web of the smaller upper panel showed
only one yield line inclined under 45 degrees and thus the upper
panel apparently did not reach its capacity (Fig. 83). From the
diagonal deformation plots it can be seen that at load No. 14 the
slope of the curve is about· horizontal. The experimental ultimate
load was 1 percent below the predicted ultimate load.
3.6.11 Test UG5.6
Before this test was conducted panel 2 was reinforced with
tension diagonals and on the top of the top flange a 8 in. x 3/4 in.
plate was welded (Fig. 22).
The highest static load reached was load No.8 (P=223.Sk ). The
failure of the panel was due to shear failure of the web. The web
of the lower (larger) panel was yielding along its diagonal tension
field and the web of the upper (smaller) panel was yielding in shear
328.6
without having developed a tension field (fig. 81).
One notices in Fig. 42 that at load No.8 the load-vs-
-28
diagonal deformation curve is about horizontal. For this test the
experimental maximum load was 12 percent below the predicted value.
3.6.12 Test UG5.5
At first it was intended to weld a 8 in. x 3/4 in. plate on
the top flange of panel 2. But in order to eliminate the welding,
the plate was clamped to the top flange by fitting it exactly
between the block holding the angle (for the tension diagonals) and
the flange reinforcing of panel 3. The shear capacity of panel 6
was increased by tension diagonals in panel 6.'
kAt load No.8 (P=233.5 ) the top flange of the test panel
started yielding. Yielding in the web of the lower panel first
appeared at load No. 9 (P~244k). The ultimate static load was
reached at load No. l2A (P~272.Sk). Due to plastic flow of portions•of the web it was not possible to increase the load further. The
lower web plate was yielding along the diagonal tension field, the
web of the upper panel was plastified along one tension field
inclined at approximately 45 degrees (Fig. 84). Diagonal defor-
mations were quite extensive at the ultimate load. The experimental
ultimate load was 7 percent below the predicted value.
328.6 -29
4. DISCUSSION AND CONCLUSIONS
The primary objective of the main test series was to obtain
information about the behavior of every panel of the girder up to
its ultimate capacity. The ultimate load capacities of every panel
are given in Table 5. The overall behavior of the test panels as
the load increased is shown in the load vs. diagonal deformation
plots, the web and flange behavior is represented by the lateral
web deflection and linear strain plots. The overall girder
behavior is expressed by· the mid-span deflection plots.
The average accuracy of the test results for the transversely
and longitudinally stiffened girder is 5 percent, with the largest
deviation being 12 percent. Thus, the newly developed theory seems
to predict the ultimate capacity of unsymmetrical plate girders
quite well.
Since the longitudinal. stiffeners control the web deflections,
two separate tension fields were developed in the subpanels. How
ever, tension fields did not run diagonally through the whole upper
subpanel as was assumed by other investigators;(7) they seem to be
parallel to each other and inclined at approxi~ately 45 degrees.
Comparing the results of the two girders one notices the
increase of strength due to the longitudinal stiffe~er. The increase
in strength of shear panels 1 and 6 i~ approximately 13 percent and
20 percent. The increase in strength of the two central panels, 3
and 4, (high moment and high shear) is approximately 44 percent.
328.6
However, the results of panels 2 and 5 of the two girders
cannot be compared because for the transversely stiffened girder
the large flange was on top)and for the longitudinally stiffened
girder the large flange was on the bottom.
-30
It is interesting to note that the increase in panel strength
derived from the introduction of a longitudinal stiffener is
considerably greater in the panels subjected to a combination of
high shear and high moment (44 percent) than in the panels under
shear alone (13 and 20 percent).
328.6 -31
5 . NOMENCLATURE
"z~y
Test
/- Panel _I ~ x
b
~_a------...--1
a panel length (in.)
b web depth (in.)
b l distance from bottom of top flange to center of
longitudinal stiffener (in.)
t web thickness (in.)
v deflection in y-direction (in.)
w peflection in z-direction (in.)
x,y,z cartesian coordinate axes
distance from neutral axis to junction of compression
flange and web (in.)
p load applied at midspan of girder (kips)
I
experimentally obtained ultimate load (kipsL
modulus of elasticity (29,600 ksi)
moment of inertia of girder section (in. 4)
Vex experimental ultimate shear (kips)
328.6
s
theoretical ultimate shear (kips)
aspect ratio, alb
web slenderness ratio, bit
strain, alE (in./in.)
longitudinal stiffener position bl/b
yield stress (ksi)
ultimate tensile stress (ksi)
-32
328.6 -33
6. TABLES AND FIGURES
328.6 -34
I.. ,
Test Flanges Webs
Number 12 Tlx3 /4 Tl 1 14 Tlx l}zTl 112 Tlxll/16 Tl 118 Tl xl Tl 50 Ttx3/16 tT 150 TlX~1l
See Reference 5
F8 27.2 29.5 31.4 43.6
F9 27.2 29.5 31.4 43.6
See Reference 6
LB1 37.6
LB2 37.0
LB3 36.0
LB4 34.9
LB5 35.3
LB6 33.1
LS1-T1 30.5 46.8
LS2-Tl 29.4 39.4
LS3-Tl 29.8 38.2
LS4-Tl 30.5 48.6
Table 1 Actual Yield Stress (ksi) Obtained inPrevious Girder Projects For A36 Steel
Panel Panel Larger Flange .Smaller Flange Web IGirder No. Length Thickn. Width Thickn. Width Thickn. Depth
(in4)(inches) (inches) (inches) (inches) (inches) (inches) (inches)
UG4 1 85.0 1,388 13.0 .756 10.0 .119 47.9 14339
2 55.0 1.388 13.0 .756 10.0 .119 47.9 14339
3 70.0 1.388 13.0 .756 10.0 .119 47.9 14339
4 85.0 1.388 13.0 .756 10.0 .182 47.9 15166
5 40.0 1.388 13.0 .756 10.0 .182 ·47.9 15166
6 85.0 1.388 13.0 .756 10.0 .182 47.9 15166
.UG5 1 85.0 1:388 13.0 .754 10.0 .119 47.9 15852
2 55.0 1.388 13.0 .754 10.0 .119 47.9 15852
3 70.0 1.388 13.0 .754 10.0 .119 47.9 15852
4 85.0 1.388 13.0 .754 10.0 .183 47.9 16519
5 40.0 1.388 13.0 .754 10.0 .183 4749 16519
6 85.0 1.388 13.0 .754 10.0 .183 47.9 16519
Table 2 Plate Dimensiorts
LNr'0co
m
Il.NV1
Tensile Tests Mill Report
UG-4cry CJu % E·long. %- Area Chemical Composition cry
Component (ksi) (ksi) Reduct. C Mn P S (ksi)
1 3/8 in. flange 34.1 65.5 31.5 52.2 .17 .54 .011 .025 39.7
1/8 in. web Hor. 55~5 72.1 23.9 50.9 .18 1.17 .010 .028 53.7(4.1.1, 4.1.2)
Panel 1 Vert. 69.2 78.9 18.8 39.7 .18 1.17 .010 .028 53.7-'
1/8 in. web Hor. 56.1 72.0 15.1 45.3 .18 1.17 .010 .028 53.7( 5.1)
Panels 2, 3 Vert. 62.1 75.1 21.8 42.4 .18 1.17 .010 .028 53.7
Hor. 36.5 61.3 29.7 57.3 .07 .42 .009 .0333/16 in. webPanels 4,5,6 Vert. 34.7 57.0 26.6 52.7 .07 .42 .009 .033
3/4 in. flange 34.1 67.1 34.9 58.2 .23 .93 .010 .021 42.9
Table 3 Material Properties UG~
VJrvco
m
ILNen
Tensile Tests Mill Report
UG-4 cry CYu % Elong. %Area Chemical Compsoition crReduct.
yComponent (ksi) (ksi) C Mn P S (ksi)
1 3/8 in. flange 32.1 6'4.9 32.4 54.3 .17 .54 .011 .025 39.7
1/8 in. web Har. 56.2 72.2 23.4 47.5 .18 1.17 .010 .028 53.7
( 5 6' 1. 1, 5. 1. 2 )Panel 1 Vert. 60.4 75.8 24.7 42.7 .18 1.17 .010 .. 028 53.7
1/8 in. web Hor. 59.3 75.8 23.3 45.3 .18 1.17 .010 .028 53.7
(5.1)Panels 2, 3 Vert. 55.3 72.6 22.6 45.6 .18 1.17 .010 .028 53.7
3/16 in. web HoI'. 36.2 60.6 31.6 55 .. 7 .07 .42 .009 .033
Panels 4,5,6 Vert. 34.5 58.0 28.4 50.8 .07 .42 .009 .033
3/4 in. flange 34.4 67.2 30.7 55.0 .23 .93 .010 .021 42.9
3/4 in. stiffener 34.4 64.3 31.2 54.3 .23 .93 ~ 010 .021 42.9
Table 4 Material Properties UGS
LNtvco(J)
ILN-....J
j
328.6
Specimen (1 B Y1 Ye V Vth Vex/Vthex(inches) (kip) (kip)
UG4.1 1. -/7 414 32.461 81.6 81.5 1.00
UG4.2 1.14 414 15.607 119.25 106.0 1.12
UG4.3 1.46 414 32.461 63.2 59.0 1.07
UG4.4 1.77 269 32.669 69.9 66.5 1.05
UG4.5 0.83 269 15.339 136.0 131.0 1.04
UG4.6 1.77 269 16.399 98.75 98.0 1,01
UG5.1 1.77 414 .263 29.804 97.75 98.0 1.00
UG5.2 1.14 414 .263 29 .804 118.0 119.0 0.99
UG5.3 1.46 414 .263 29 > 804 92.5 96.5 0.96
UG5.4 1.77 269 .263 29.354 96.75 105.0 0.92
UG5.5 0.83 269 .263 29.354 136.25 147.0 0.93
UG5.6 1.77 269 .263 29.354 111.75 126.0 0.88
Table 5 Test Results
-38
HIGH SHEAR3 Symm. (Basler)4 Symm. (Cooper)I Symm. (Cooper)
I Unsymm. (Dimitri)7 Symm. Long. (Cooper)
16 Tests
LNt'0(X).m
HIGH MOMENT
2 Un-s. (Dimitri)
PURE BEN DING
9 Symm. (Basler)2 Unsymm. (Dimitri)§. Symm. Long. (Cooper)16 Tests
Q3
1.0M/Mu
~Vu
o
PURE SHEAR
5 Symm. (Basler)
3 Uns. (Dimitri)8 Tests
Fig. 1 Summary of Past Girder TestsI
LNLD
11-
~8 II Web 3~611 Web
18'-11" ct 18'-11" II. • I • •
I ~- - I ~u- II l I
\ ""'" I(1/8 x1/4) v l~ I
I~ v (3/16 x 114)
Ia=J..46 ....
a =1.77 r a=1.55 I a =1.77 a =.833 a=L77I I,....." I 600 \!/ "-
(l/8 x5/16) v \I
~J/ v (3116 x 5/16)I
I 60°/1\I 8
i. 71-1 11 I::b 4 1-711 .I.T 51-lO ll .110"1. 7 1 _1 11 i 3'-4" : 7 1-1 11 .1.. ... .. .371-10"
GIRDER UG4
0"
LNtVcoOJ
1/8 x5/16)
II
IIIIIIIIII
B
!)8~600I 60°8
GIRDER UG5
Panel Layout As UG4
II
!3/16 x 5/16)
Fig. 2 Elevations of Test Girders UG4 and UG5 I~o
LNf'.)
enmlOll X 3/4
11 T. fl
--13/411
I0 L,--5 11 X 3/4 II V. S1:01 I
121/211 f
3 11X 3/4
11L. S1:
48 11 4811X !Ja ll
( 3/161I )W. fl
N~t+t-I 351/2
11 N.A.--
4 11X 3/4
11 S1: f!.:I n II'~ I I
11'1 I ..
13 11X J 3/8
11 B. fl.,3/8
11 ,3/8",- fSECTION A-A SECTION B-B
Fig. 3 Vertical Sections Through UG4 and UG5 I..p::.l--J
4 1 -0"
±7-d~1~! ~ Of Girder
LNr'0OJ
en
oI
-v
~811 Web 12.
'Is II Web ~
101 -11"
37'-1011
3/1611 Web fl
181-11 11
I'-~ l-
2'-0"
.T *B Flange Plates H~I I I I
1_ As Requ. -' _ Max. Length Available _1_ As Requ. -I21-0"ri
I--------~cg]
Long Stiffener ~ I
Fig. 4 Location of Coupon PlatesI~I'V
328.6 -43
=-~
I A B C 0 E I
-
2 ....---- ---- ~-----2~ ;,.C\I
-3' AI BI CI 01 E' :3
2" I. 11 11 •1. 11" .1. 11" .1. 11" .1 211t~=- r-.:=-
48"
!~6" WEB COUPON PLATE
Thickness At I A,AI, B, B',C, CI, 0, 01, E, El
Width At l 1-1,2-2,3-3
j:, I:, :[nI 101'z" 101'2" I -,dO... • I • .. 172-
24"
--.A C E
~=Q £:!
AI CI El
2" 1.. IO~211 IO~2"I I
~t
.::- ·1- __Iyzll
24"
I~
FLANGE COUPON PLATES
Typ.Thickness At: A,A"C,C'.E,EI
Width At l A,AI,C,C"E,E '
~: I:, I:· I:, :.~]~2" 1.1 14 1 .1 r .1. II" .1 2:
~e II WEB COUPON PLATE
=-.A C E j=Q ~
Al CI E'
I"~ =-t~
~1211
I!/e II FL ANGE COU PON PLATE
LONGIT. STIFFENER COUPON PLATE
Fig. 5 Typical Location of Measurement Pointson Coupon Plates
328.6 -44
Coupon JtUG 4.2UG 5.2
2 Coupon ft'sUG 4.7UG 5.7
Coupon JlUG 4.3
) ~UG5.3
lOll x 314" Top Flange I UG 4.3.1I-----------.......J UG 5.3.1
Coupon ItUG5.4UG4.4
r2 Coupon Jl1s
~~--::-II--=-'~--------.---~------.,~ UG 5.5P :3 x ~4u Long. Stiffn. ~ UG 5.6
Coupon It UG 4.1.2UG5.i.2
I/SII Web
Tensile Specimen
UG4.l.iUG5.1.1
r-~~~.....--Coupon Jt UG 4.1.0UG 5.1.0
Va II Web
UG 4.1UG5.1
Fig. 6 Location of Tension Coupons
328.6
•
R = 2 ~2"
I~211 •
-45
G=all.. -- 5 11
- -gil
- 5 11
- - =
- ±24 11
--TENSILE SPECIMEN FOR
FLANGES E LONG. STIFFENERS
III •
R =211--
•
G=4 11.. -- 4 11 5"
- - 4"~ - --
±1511
- -WEB TENSILE SPECIMEN
Fig. 7 Tensile Specimens
328.6 -46
3/41FLANGE TENSILE SPECIMEN
,007.005
Initiation of StrainHardening
.003.001
17.8
35.6
8.9
26.7STRESS
KSI
36.5
29.2
STRESSKSI 21.9
14.6
7.3
3/16" WEB TENSILE SPECIMEN
Fig. 8 Stress-Strain Curves for Tensile Specimens
328.6
p
, I
I~~ .,-,.
-47
I.. L..I
): i ______1):
SHEAR DIAGRAM
MOMENT DIAGRAM
Fig. 9 Loaded Girder
328.6 -48
UG 4.1 UG4.3 UG4.4
jL~:~:~~~~~~~~c::IJ]]~:~:~:~:~:~:~:~:D 1
""'----......&.....---a. -.-.-_..A--_ M/ Muo 0.2 0.6 1.0
VERTICALLY STIFF. GIRDERSmall Flange in Compression
0.2
UG4.11.0 ...,.---.r\
V/Vu
0.6
UG 4.2 GU 4.5 UG 4.6
I~i
0.2
1.0..--.U__G4•6
V/Vu
0.6
100...-......&.....----10_...-..--...._..-....--._ M/Muo 0.2 0.6 1.0VERTICALLY STIFF. GIRDERLarge Flange in Compression
UG 5.1 UG5.2
1.0 UG5.6
V/Vu
0.6
UG5.4UG 5.1 UG5.3 UG5.5
j;:i:«i.:.:.:.:._.::.:.:..: :.:.:.:.:.: ~':'~'::':':I08::............-:-:.. :-:-:- :.:..:.: :.:.:-: :":"~: III •••••••••••••••....:.-.:...:.
UGS.2 UG5.4 UG5.6
0.2
-......------------ M/Muo 0.2 , 0.6 1.0
LONGITUDINALLY STIFF. GIRDER
Fig. 10 Location of Panel Failure Pointson Interaction Curve
LNI'Vco
01
LaterialBraces
Wlterial BeamSupport
Column of 5 MUlion lb.Machine
!!!!!iii!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!~~~~~~~~~~~ ~~~~~~ ~~~~ ~~~~ ~~ ~~~~~~~~ ~~~~~~~ ~ ~~~ ~~~~~~~~~~~~~~~·L····,··'···',········'········,·····················...;...
K- Frame
Girder
End Support
N
~Fig. 11 Plan View of Test Set-Up
1~t.O
328.6 -50II!. 1811
X 23/4Bolted to MovableCross Head
Lg. Bear. ItItg. Bear. I!.
or(2'·011 LgJ
From) Stock Yard.
!.IOllx101lx211
(or Equival)der ( ~
nge)
TATION
"'-I'
II i 1
3 11 r I !...itSll l 1([1 LI
14" x 3!1 XII-S"14" X3"x'2LS'
3 11 1---l':--Sll<fl Round 8
/4"
IS3~1 BI4F 320(I4VF320
SECTION ALOAD APPLIC
Fig. 12 End Support
~ of Machine andGirder
I
6'-2"8 11
I"¢> Pins b.e.
Edge of Slots on5 Million lb. Machine
I~~~IIE!!:=::===========tiH----t--r <tof Beamand Pipe
Platform of TestingMachine
Fig. 13 Lateral Support of Pipe Brace at Machine
328.6 -5012. lallx 2 3/4Bolted to MovableCross Head
Lg. Bear. 12.lLg, Bear. f2.
ar(21"0" LgJ
From) Stock Yard.
I!.IOllxIO"x21l
(or Equlval)der (lJnge)
TATION
"'1'
II I I
3" r I !.....h611 L TI LI
1411 X 3'1 XI!..6"14" X3"x'2L61
3"1--- I 1'- 6"~ Round 814"
16 341
814F 320(I4YF320
-I ;;0 " ~SPherlcal.
l':"ri:;~ Brg. Block
(Fla
SECTION ALOAD APPLIC
Fig. 12 End Support
1"1Jl Pins b.e.
Edge of Slots on5 Million lb. Machine
<t of Machine andGirder
I
i 21tz""I ~
2~211¢ Pipe Lg,41-1"
3~2 x 5~2" X 3/all L
Wooden Shim
Itt~~18~=:==========~rtJ-t-+lr tof 8eamand Pipe
Platform of TestingMachine
Fig. 13 Lateral Support of Pipe Brace at Machine
SECTION AT SU PPORT
LNJ\.)
co
en
211
a~211 Lg.
4 1-011
11-6 11
Hydraulic JacksEach 100 kips
2Y2 11 x Pipe Lg.41-411
SI-2 Yall
aYF 64 Lg~ 151-0 11
3~1I X 5 Y211
X 3/SIi L
3 1-7!t2 11
IIL33~
Parts Have tobe ClampedTogether
6 1-a ll
Fig .... l4 End Support Bracing IU1\--I
328.6
Q) ® ® ® ® ®...... to- -.J_
GIRDER UG4
-52
TEST UG4.3 ~aterial Bracing
LWooden Diag.
TEST UG44 rPrOViSlonal (ReinfOrCing!. SlI X 2"-----.- Reif. !.S"xVa"
TEST UG4.1 Reinforcing Ie. SIIX 2" ProvisionalRelnf. It 6" x ~2"
TEST UG4.6
/'/'
......../'
,/
,//"
......../'
/'
TEST UG4.2
lTake Reinforcing off
TEST UG4.5
........ '".""
,//'
,/.",.'/
................
.................
................
"""-"
Take Reinforcing off
Fig. 15 Summary of Sequ~nce of Testing Girder UG4
328.6 -53
Clamp
SECTION
Note:Clamp Angle to Flange
ELEVATION•
Fig. 16 Bracing of Panel 4 at Centerof Panel (UG4. 3)
3'811 For all x ~81J I!.
4~2" Intermediate Clamp
1'411 For SIiX Y2 1J e.
3'8"
Fig. 17 Section nA ft of Figures 18 and 19
A
A
Clamp
4~211
3/all
all x 2 11 x 7 1-311 Lg.LNrvco
OJ
DETAIL "BII
all ~I II 4 1 7 11 Lx· '/8 x-g.
9" x 3 If X 5/S II
CD
IIAII
®
II BII
@ @
l4~2il ~ 2 Y2"
SECTION A-A OVERALL VIEW
all 71 IINote: Clamp X'8 plate to flange
at third point at plate.
Fig. 18 Flange Reinforcement of Panels 2 and 3(UG4.4) I
Ul~
3/8VJ(\.)
OJ
OJ
DETAIL II ell
B
B
Clamp
4~211
SECTION 8-8
all x 2 11 x 7 1-3 11 Lg.
6 11 'I. II 8 1 4 11 LX 1'2 X _. g.
9 11 X 3 11 X 5/8 II
II ell IIAII
o I@II ®
2~2~L 4~~ I ~ _II A~2"OVERALL VIEW
Note: Use 4 clamps for flangereinforcement
Fig. 19 Flange Reinforcement of Panels 4 and 5(UG4.l)
IVlen
L SIiX 6 11 X }e"
LNtoOJ
Ol
Shim
ELEVATION
III High StrengthSteel Bars
SECTION
Fj.g. 20 Connection Detail for Diagonal Braces IlJ1OJ
328.6
I
! ! I
CD I ® I @ ® 1®1 ®I I I I----011 a...- ,....;1-
GIRDER UG5
TEST UG5.3
-57
r"-, I
TEST UG5.4
LWooden
Tension Dieg.
i
I
I
TEST UG5.1 R . f . m ell x 11~211~ eln orclng rL. "
II I I II "
TEST UG5.2
TEST UG5.6II I I II
TEST UG5.5
t:;:;:;:;:;:;:;:;:;:;:;:;:;:;:;:;:;:;:;:;
Fig. 21 Summary of Sequence of Testing Girder UG5
328.6 -58
CD ® @ @ ® ®
41~ II l4~~L 2~J 4lJ2 ::.1~ ~ --LOCATION OF FLANGE REINFORCEMENT
51 II'16
TYPICAL FLANGE REINFORCEMENT DETAIL
Fig. 22 Flange Reinforcement'for Girder UG5
328.6
TEST UG4.3
-59
V3.1 V3.2
TEST UG 4.4
V3.3 V3.4
V4.2
TEST UG4.1
V4.3 V4.4 V4.5
VI.-I VI.O Viol V 1.2 V 1.3
TEST UG 4.6
j-D-f WiV 6.3 V6.4 V6.5
TEST UG 4.2
J f1 !-D-lV2.1 V2.2 V2.3
TEST UG4.5
V5.3 V5.4 V5.5
Fig. 23 Dial Gage Instrumentation For GirdersUG4 and UG5
328.6 -60
Fig. 24 Instrumentation for Diagonp.l Panel ·Deformation (UG5)
'-C-Clamp
C-Clamp~---
TOP ELEVATIONC-Clamp
BOTTOM ELEVATION
Fig. 25 Conne'ct_ion Detail of Diagonal Instrumenta_tion
328.6
200
160
120
p(KIPS)
80
40
o 0.4 0.8 1.2
(INCHES)
1.6 2.0
-61
Fig. 26 Load-vs-Centerline Deflection (UG4.3)
200
PUlt =163.2k
160Pth =163k
11,"'-'\,..1 8/°7
,0
120 /65/
P 4/°(KIPS)
80 3/0
40 2/0
III0 0.4 0.8 1.2 1.6 2.0
(INCHES)
Fig. 27 Load-vs-Centerline Deflection (UG4.l)
328.6
240
200
160
P(KIPS)
120
o
PUlt =238.Sk
0.4 0.8 1.2(INCHES)
1.6 2.0 2.4
-62
Fig. 28 Load-vs-Centerline Deflection (UG4 .. 2)
320
240
p(KIPS)
160
80
o
PUlt =272k
Pth
-=262k -
0.4 0.8 1.2 1.6 2.0 2.4 2.8 3.2 3.6
(INCHES)
Fig. 29 ~Load-vs~Centerline Deflection (UG4.5)
r
LNtvcoen
o15
r- ..... -- '11:::::::~ill~2~D- - -ij I---I- -- _1_ -.4 ::::::::::::::::::::\ "-------'-.................. - t A
" . --- tSTU
/0--/9o 0
7
t:::". - -- -- ---. __
0 5o
40r //03/°2
120
160
p(KIPS)
80
o 0.4 0.8 1.2 1.6 2.0
(8/8y)'(~~y)
Fig. 30 Load-vs-Centerline Deflection (4.4) ImLN
1
-- -----------------------1
328.6 -64
200
0.4
25
0.3/
02627
Compression DiagonalDeflection
0.2
(INCHES)
24. __---------;°25_+--_ --::.::~r&_----==:=:o:....--
0 20
0.1
o/
160
o
p(KIPS)
80
Fig. 31 Load-vs-Diagonal Deformation (UG4.3)
200
160
120
p(KIPS)
o 0.04 0.08 0.12
(INCHES)
Fig. 32 Load-vs-Diagonal Deformation (UG4.4)
•
328.6 -65
200
0.90.7
Compression Diagonal Deflection 23 24- 0 - c -0r==:;;;;::=~--!t'4-
O~S0.3II0.1
160
120
o
·40
p(KIPS)
80
(INCHES)
Fig. 33 Load-vs-Diagonal Deformation (UG4~1)
240
200
160
p(KIPS)
120
80
Itl. =238.Sk--- 19
Compression DiagonalDeflection
Tension DiagonalDeflection
o 0.1 0.2 0.3
(INCHES)
0.4 0.5
Fig. 34 Load-vs-Diagonal Deformation (UG4.2)
328.6 -66
0.40.2 0.3
(INCHES)
0.1o
\
enSion Diagonal DeflectionP =197.5k
200 ult _~~....--~~, _........:-----~Compression DiagonalI~ 10 ~OOU //'" .......... \ Deflection
!tf:9
,::/ /'0 1170
160 117 ~t6'l
120 JIS 16 0 12 0 0 16
P 4
UOO
(KIPS)
80 0
40 ~
Fig. 35 Load-vs-Diagonal Deformation (UG4.6)'
280
240
200
160
P(KIPS)
120
Tension O-iagonalDeflection·
o 0.02 0.04 0.06
(INCHES)
0.08 0.1
Fig. 36 Load-vs-Diagonal Deformation (UG4.5)
o o o o o
UG 5.5
k k~-= 294 _~_ [PUlt =272.5
"'-...., .....- .....,"'''' ~13
12A
014
015
LNI\.)
coOJ
SCALE I 1 I I -I 1 INCHESo 0.4 0.8 1.2 1.6 2.0
Fig. 37 Load-vs~Vertical Deflection eUGS)
Im......J
328.6 -68
Compression DiagonalDeflection
115 0
r Tension Diagonal=18Sk , Deflection
Pult /--. 13 _-----."4/0 - '0 12----------=-==-8-- - ----...--
/0 1I~0
t /000 6 10160
120
200
o 0.1 0.2 0.3 0.4
(INCHES)
0.5 0.6
Fig. 38 Load-vs-Diagonal Deformation (UG5.3)
0.3
Compression DiagonalDeflection
0.2o
120
iTenSion Diagonal Deflection
_ k Lower Subpanel200 ~It - 193.5 8 9-:c *=<* 0
o~I I ! 7
!i /6ll/O~
(KI~:~ !1/04
40 ff3
f.(INCHES)
Fig. 39 Load-vs-Diagonal Deformation (UG5.4)
328.6 -69
0.3
•
0.2
•
0.1
~enSion Diagonal Deflection
(Lower Subpanel)• La 0.,..-- o~-::--; 10
./0~/8
I //7iil
d6
/lBIJ- 0
4
/32
Pult =195.5k----
40
o
120
160
200
p(KIPS)
80
(INCHES)
Fig. 40 Load-vs-Diagonal Deformation (UG5.1)
0.36
CompressionDiagonalDeflection
0.30
--::..........i!ilrI~--1 -I-~
0.12 0.18' 0.24
(INCHES)
0.06
,/Q ...__0"",,,0 ~"""'12 13
/0 /0 IIo 0' 10
(/~9! 7
! /06 TensionDiagonal//05 Deflection
114
/32
80
o
40
200
160
P(KIPS)
120
Fig. 41 Load-vs-Diagonal Deformation (UGS.2)
328.6 -70
240Pult = 223.5k
Tension DiagonalDeflection
0.3
Compression DiagonalDeflection
0.240.12 0.18
(INCHES)
0.06o
0 8
11.7
200
160f
0/5
P(KIPS)
120
I80 [340 oe2
Fig. 42 Load-vs-Diagqnal Deformation (UG5.6)
Tension DiagonalDeflection
Diagonal
0.9
/
0 Ir50 CompressionDeflection
fl71i ~----""[\~ii~~I
12A
0.6
12
(INCHES)0.3
2
o
60
120
180
2801-----
240
Fig. 43 Load-vs-Diagonal Deformation (UGS.5)
328.6 -71
STATIONS FOR PANEL 2
®x
3~611 Web
Stations In Other PanelsAre Analogous
..
r~--~-IIW_eb ~,+
x - coordinates for every panel in inchesPanel
x Xl X 2 x3 X 4 Xs x6
x 7 x8 X9 xIO·0
I 7 16 26 37 48 59 69 78
2 -3~ 2 7 15 23 32 40 48 52 3/8 56 3/8
-=:::t 3 7 17 29 41 53 63c.98
4 7 16 26 37 48 59 69 78
5 3 6 15 25 34 37 43
6 -3 3 7 16 26 37 48 59 69 78 82
1 3 7 16 26 37 48 59 69 78 88
2 3 7 15 23 32 40 48 52
Lfl 3 -3 3 7 17 29 41 53 63 67c.9;=J
4 3 7 16 26 37 48 59 69 78 88
5 3 6 15 25 34 37
6 -3 3 7 16 . 26 37 48 59 69 78 82
Fig. 44 Stations for Late·ral· Web Deflections
INrv
-- y= 0.0 - y=o.o I ~»>>>>>)1-- ~;~.o en1.75 -- 1.69 en
4.75 4.69 I I ( , n 4.75...
7.75 7.69 I I ( ). U 7.75
10.75 10.69
15.7515.75 I I ( ). U 15.69
k: M 22.75 k M 22~9 L II 22.75
b H 2~75
\ B 38.25
I R 43.25
b M 3435
I t, <: <: <: , , ,,<'t <:,«<, <<<:<:<<<i 4 7.9
38.69
\: H 43.19'
k II 31.19
1 , l~~"'-\."~~~"~~-- 48.0
L R 38.25
L II 31.25
It!: H 43.25
I ,,{ <: <: <:«,«s<{<,«<:,(,«(,(1 4 7.9
UG4.1, UG4.3, UG4.4 UG4.2,UG4.5,UG4.6 UG5
Fig. 45 Vertical Location:-: of Lateral WebDeflection Readings
J-....Jtv
INI\.)
coen/1
lilo Q)
/ 1/1 y= 15.75"
, IiII,l04
o 0 0
160
0.4 0.8 1.2 W(INC HE'S)
23/Pu1t
20t ·,0 25P I leo,ol7
II 0'
(KIPS) 80 '9' 13 UG 4.3
I x =2911
40 1?-!5 y = 15.75
200
160L 10 15 p.0-0---0---- ult
91
P 120t t7
(KIPS) 80 I: UG 4.4I;4" X =37 11
401-;3 y = 15.75 11
I I
0 0.4 0.8 1.2 W(INCHES)
UG 4.3x = 2911
UG 4.4x ~ 37"
Fig. 46 Web Deflections for UG4.3 and UG4.4I
.....jLN
200
IN1'0OJ
OJ
y=22.6S"
Load No. illo
UG 4.6
x =3711
y = 22.6811
0.4 0.8 1.2 W (INCHES)
160
200
24160 \- "l 914 1~r22PUlf
.p 120t /0---rr/f2"(KIPS) ~
80 3/ f. f UG 4.1°401- 2/ X =5911
° II~/ I ; II~ 18. Y = 31.25
0.4 0.8 1.2 W (INCHES)
II9----0 --F!ult,0
1860
scip I~Ot 41
(KIPS,) 3/80
2/
40"-
Lpad No.
y =31.25"
UG 4.1x ~ -59"
UG 4.6x =37
11
Fig~ 47 Web Deflections for UG4.1 and UG4.6 I......,J
-P:>
LNf'Vco.en
UG 5.4x =37"
Load No.
UG 5.4
x = 37 11
y = 34.75 11
UG 5.3
X =2911
y = 23.75" $"
I,
o 0.4 0.8 1.2 W (INCHES)
16.0
200
0.4 0.8 1.2 W(JNCHES)
8 9200~ 7/0~D
i rult.160t-°6JS
P 1201-/4·(KIPS) 80~3
I40 1j,-2 .
12 13 146 II O.....O-o-P0_0' ult
P 5/(KIPS) 120t 40
/
·80 3140·J
UG 5.3x = 29 11
100
'\"-o
y=23.7511
'Fig. 48 Web Deflections for U85.3 and U85.4 I......,JUl
lJ'1tvco01
0-°\~ ;0CD9
W(lNCHES)-0.4 -0.8
UG 5.6X = 37'1
200 ~ 8o.!2-~--PUlt7c1
160~ J
p 120~05(KIPS)
UG 5.1 l y=2"l75"80 .3X =59"
40 If y =34.75"I
o -0.4-0.8 W(INCHES)
240~ 9Load
t...___o-.-R2001-1 ult
06I
P160H5
.(KIPS) 120II
UG 5.6
.401-x =37
11
y =27.75 11
I
UG 5.10 0.4 0.8 W (INCHES)
x = 59 11
Fig. 49 Web Deflections for UGS.l and UGS.6 I-.......Jm
o -0.4-0.8 -1.2 W(INCHES)
14240~ 12cP......o,,-Pult
o 13
2001-8110
07
1601-/I ~5
P 120'(KIPS)
80h(a UG 5.2
I x =2311
40 1t2 Y=27.751•
I
Load No.
240~ 20LN
1~?--PUlt ~TI'\.)
en
A 200 80(f)
7&/11 160 51/0 1r P 120
3/0_ ~.. q (KIPS)80 UG 4.2
40r:/ x = 23" -1\\ '1=-27.7511
y= 31.19 11 U7i y = 31.1911
ift I , I
0 0.4 0.8 1.2 W(JNCHES)
UG 4.2x =23"
UG 5.2x =23 11
Fig. 50 Web Deflections for UG4.2 and UGS.2 I....j-.J
j
2801- 13 14""'-RIN
~,J\.J
12r ult enI .
2401-1o~9 OJ
'082001- 71
60
P . lOOt /(KIPS) 120 jO
0~2J UG 4.58. ..
x· = 25
y=38.69'1... 40~ y = 38.69 1
'
I ". , , ,
0 -0.4 -0.8 W (I-NCHES)
280~ 10 12_b3_p' ItY .u$/0
240~o
2001-/-0
P 160t15 , \(KIPS) 120 0 0
UG 5.5 V801!-3X = 25 11
~
40 tt- Y = 27.7511
I I
0 0.4 0.8 1.2 W(INCHES)lJG 4.5 UG 5.5X :; 2...5 : x = 25 11
IFig. 51 Web Deflections for UG4.5 and UG5.5 -......Jco
328.6
Fatigue Crack
TYPICAL SHEAR PANEL
Fatigue Crack
TYPICAL PANEL UNDER S 4M
Fig. 52 Location of Fatigue Cracks
-79
328.6 -80
CD ® @ @ ® ®
UG-4
-24.0
- y=O.O-2.0
-13.0-16.67
-31.33-35.0
-46.048.0
787470- - -~42-72 -76 -8~
1941271 0'1
38 40 10 8 52 54
-96 14 16 2471
61£18
~~
~ 44"- 467148
20L!.2256
5824
"-60 98-
26 28 62 64168 34 32
~30 L!S6 36~_ 82 _86 90_
-84 -86 -92
TYPICAL FOR PANELS 4 $ 2
See Table Below For Actual Gage Layout
PanelLinear Rosettes a b
Gages Used Used Inch Inch Note:
68-98 32-66 10 15 YaOdd numbers are
2 applied to gages on
468,74-80 2-30 10 30Ye
the opposite face.
86-88,94-98
Fig. 53 Strain Gage Location (UG4)
328.6 -81
- 24.0
y =0.0- 2.0
- 31.33
- 35.0
-13.0
-16.67
787470
-46.0-. 48.0
J 1.0_1. b 8:'1. b _1_0_1 ~II
- - --72 -76 80- 32,,:g ~12 164 ?I42 a lO 34 36
,,: 6652~54 -
6 50
16 14
"~ 18~94 '=f' 98
96
58~5660 28 26
6830~-
44 4662 22 20
1L48 1 24~90_82 86_ --84 - -92
TYPICAL FOR PANELS 1,3,5 ¢6
See Table Below For Actual Gage Layout
PanelLinear Rosettes a b
Gages Used Used Inch Inch Note:
I 62-80 2-36 10 30~8Odd numbers are
86-88 applied to gages on
5 62-92 20-486 11 5/8
the opposite face.94-98
662-76 2-12,20-30
10 30Ye82-92 44-48,94-98
3 62-76 2-12,20-30 10 22 5/886-88 44-60
Fig. 54 Strain Gage Location (UG4)
328.6 -82
CD ® ® @ ® ®UG - 5
y=o.O
-25.25
-30.438
-35.625
-6.063
_12.125-12.875-14.875
726864
-46.0-48.0
2~ I..O~.. b _1_ b _1 ..0 ~ ~"TYPICAL FOR PANELS I, 2 _ :3
- ... .-- .- .-66 70 74
54
"8052 50
!! 84- -78~12 82- 86-36
I '?I8 10 56 34 32 ,~ ~~ 2
1'4 62-16 14
618~
2224
-60 20~
26 28 58 46 44 40 38
1£30 I 48~ 42~_88
-gO
See Table 'Below For Actual Gage Layout
Panel Linear Rosettes a bGages Used Used Inch Inc'h
I 56-58,60-90 2-24,36-42 10 30~8Note'
50-54 Odd numbers are
2 56 -58,64-902-18,32-42
10 15 ~8applied to gages on
26-30 the opposite face.
:3 56-58,62-70 2-30,44-48 10 22 5/876-82,88-90
Fig. 55 Strain Gage Location (UG5)
328.6 -83
-y=o.o
- 25.25
- 30.438
- 35.625
_'2.'25=12.87514.875
-46.0-48.0
___ •••,. b ..,. b ~~a j ~IITYPICAL FOR PANELS 4, 5 $ 6
56 60 64- -- --- - -58 62 66
68 72 76t--- - - - ~
t-----
~30-70 -74 - ~
1522
7178
26 28 4 6
~~
", 12
14 16 1071~18
8
24 22
1£20 50-
3840 3234154 46 44
1£42 1£36 48~_80
-82
See Table Below For Actual Gage Layout
PanelLinear Rosettes a b
Gages Used Used Inch Inch
4 50-82 2-36 10 30'18 Note:Odd numbers are
5 52-82 2"18,26-306 11 5/8 applied to gages on
30-48 the opposite face.
6 50-82 2-36 10 30~8
Fig. 56 Strain Gage Location (UG5)
Load Nos.
-Boao -4000
14
-3000 -2000STRAIN
9 7 5 3
-1000
LNf\.)
CD
Gage No. 73 (75) ~ en
Gage No. 74 (76)-.
o (JL in./in.)
Fig. 57 Strain Distribution Across Top Flange (UG4.4)
-p(KI PS)
50
150
100
o (p. in. lin.)-1000-3000 2000STRAIN
~\-4000
ItM:e===o-e...
\ 10 9~O~~~
~-16\\~d' 5oe~
~.4\3\
2°
-5000
Fig. 58 Load-vs-Strain of Strain Gages 73 and 75(UG4.4)
IOJ~
l.Ntvco(j)
Gage No. 74(76)
Gage No. 73 (75)
9 51317
o~"" 0\ i 00_ 0_ 0 0 0
2223Load No.
STRAIN -5000 -4000 3000 -2000 -1000 o (JL in./in.)
Fig. 59 Strain Distribution Across Top Flange(UG4.3)
Q "e oo-~~ 1\10~ - -e 23 22 18 ~o~......... _. ;>~
24 17 "'~~.\I, 15 0",
o. ;>~ 13 o~II ~o.9'\
7°\5°
\3°
100p
(KIPS)
50
-5000 -4000 -3000 -2000
STRAIN
-1000 o (JL in./in.)
Fig. 60 Load-vs-Strain of Strain Gages 73 and 75(UG4.3)
IcoV1
LNr-vOJ
OJ
Fig. 61 Strain Distribution Across FlangeThickness (UG4.1)
150·
~~~9
2
·50
II
010
+1000 (fL in./in.)
8
p(KIPS)
0-
100
o
7,°6
\5~~4~\
\-1000-5000 -4000 -3000 -2000
STRAIN
Fig. 62 Load-vs-Strain of Strain Gages 69 and 71(UG4.1)
IcoQ")
Load No.LN
6 5 3 I\.)
ro
Gage No. 67 (69) OJ
--------- -.............. '", I__ 14
--------Gage No. 68 (70). ,
STRAIN -5000 -4000 -3000 -2000 -1000 0 1000 (fL in./in.)
Fig. 63 Strain Distribution Across the Top Flange(UG5.3)
Load-vs-Strain of Strain Gages 68 and 70(UG5.3)
Strain Distribution Across the,L9ngitudinalStiffener (UG5.3)
Fig. 65
Gage No. 80 (82)
Gage No. 79 (81)
o (fL in./in.)
:36 5
-1000
12o
Load No.14
Fig. 64
STRAJN - 2000
100
150
13 0 A30-, 14 \1214 12~ II~
II' 6l6
0
I\ 50~&O~+o '\~ 0f!l. o~
e 0
4°~0 I P
3 \ (KIPS)
-150
2
0
\"
-2000 -1000
STRAIN
o (fL in./in.)
I0)
-.....J
328.6 -88
Fig. 66 Test Set-Up
Fig. 67 Flange Failure and Web Yielding (UG4.4)
328.6
Fig. 68 Yielding of Top Flange (UG4.3)
Fig. 69 Deformed State of Test Panel (UG4.3)
-89
328.6 -90
Fig. 70 Weld Failure (UG4.2)
Fig. 71 Web Panel Failure (UG4.2)
328.6 -9l
Fig. 72 Top Flange Failure and Reinforcingof Compression Flange (UG4.1)
Fig. 73 Web Deformations at Failure
328.6 -92
I Fig. 74 Web Failure (UG4.6)
Fig. 75 Web :railure (UG4.5)
328.6 -93
I
'Fig. 76 Test Set-Up After Last Test--LookingTowards West (UG4.5)
Fig. 77 Test Set-Up After Last Test--Looking,Towards East (UG4.S)
328.6 -94
Fig. 78 Panel Deformations (UG5.3)
Fig. 79 Yielding in ~Flange and Upper Web Panel(UG5.3)
328.6 -95
Fig. 80 Failure of Lower Web Panel (UG5.4)
Fig. 81 Failure of Upper and Lower Web Panel (UG5.6)
- - ---------------.
328.6 -96
Fig. 82 Web Failure (UG5.l)
Fig. 83 Web Failure (UG5.2)
328.6 -97
Fig. 84 Web Failure Due to YieldingAlong Tension Field (UG5.5)
Fig. 85 Last Test on Girder UG5 (UG5.5)
328.6 -98
7 . REFERENCES
'1. Basler, K., and Thlirlimann, B.STRENGTH OF PLATE GIRDERS IN BENDING, Proc. ASCEVol. 87, No. ST6-, August 1961
2. Basler, K.STRENGTH OF PLATE GIRDERS IN SHEAR, Proc., ASCEVol. 87, No. ST7, October 1961
3. Basler, K.STRENGTH OF PLATE GIRDERS UNDER COMBINED BENDING ANDSHEAR, Prac., ASCE, Vol." 87, No. ST7, October 1961
4. Basler, K., Yen, B. T., Mueller, J. A., and Thllrlimann, B.WEB BUCKLING TESTS ON WELDED PLATE GIRDERS, BulletinNo. 64, Welding Research Council, New York,September 1960
5. Yen, B. T., and Mueller, J. A.FATIGUE TESTS OF LARGE-SIZE WELDED PLATE GIRDERS,Bulletin No. 118, Welding Research Council, New York,November 1966
6. D'Apice, M. A., Fielding, D. J., and Cooper, P. B.STATIC TESTS ON LONGITUDINALLY STIFFENED PLATE GIRDERS,Bulletin No. 117, Welding Research Council, October 1966
7. Cooper, P. B.STRENGTH OF LONGITUDINALLY STIFFENED PLATE GIRDERS,Proceedings of ASCE, Vol. 93, No. ST2, April 1967
8. Dimitri, J. R., and Ostapenko, A.PILOT TESTS ON THE ULTIMATE STATIC STRENGTH OFUNSYMMETRICAL PLATE GIRDERS, Fritz EngineeringLaboratory Report No. 328.5, June 1968
9. American Institute of Steel ConstructionSPECIFICATION FOR THE DESIGN, FABRICATION, ANDERECTION OF STR~CTURAL STEEL FOR BUILDINGS,AISC, New York, 1963
328.6 -99
8. ACKNOWLEDGEMENTS
This report was prepared as part of a research project on
unsymmetrical plate girders conducted in the Department of Civil
Engineering, Fritz Engineering Laboratory, Lehigh University,
Bethlehem, Pennsylvania. Dr. David A. VanHorn is Chairman of the
Department and Dr. Lynn S. Beedle is Director of the Laboratory.
The sponsors of the research project are the American Iron and
Steel Institute, the Pennsylvania Department of Highways (the Bureau
of Public Roads) and the Welding Research Council. The interest and
financial support of the sponsors and the advice of the Task Group
on Unsymmetrical Plate Girders of the Welding Research Council
Committee on Plate Girders are gratefully appreciated.
The authors wish to thank the Fritz Engineering Laboratory staff
for their.cooperation and assistance in conducting the testing
program. Greatly appreciated is the help offered by Chingmiin Chern
in the preparation and actual testing of the girder specimens.
Appreciation is also expressed to Charles E. Anderson, Jr. for his
assistance during testing and preparation of the final report. The
assistance of Irving J. Oppenheim and Jorgen G. Ollgaard during the
testing period is gratefully acknowledged. Thanks are also due to
Miss Marilyn L. Courtright for typing this report, to John M. Gera
for preparing the drawings and to Richard N. Sopko for taking the
photographs.