Lehigh University

Lehigh Preserve
Fritz Laboratory Reports Civil and Environmental Engineering

1962

End-moment, end-rotation characteristics for beam

columns, May 1962
M. G. Lay

T. V. Galambos

Follow this and additional works at: http://preserve.lehigh.edu/engr-civil-environmental-fritz-lab-

reports

Recommended Citation
Lay, M. G. and Galambos, T. V., "End-moment, end-rotation characteristics for beam columns, May 1962" (1962). Fritz Laboratory
Reports. Paper 1346.
http://preserve.lehigh.edu/engr-civil-environmental-fritz-lab-reports/1346

This Technical Report is brought to you for free and open access by the Civil and Environmental Engineering at Lehigh Preserve. It has been accepted
for inclusion in Fritz Laboratory Reports by an authorized administrator of Lehigh Preserve. For more information, please contact
preserve@lehigh.edu.
 Welded Continuous Frames and Their Components

END MOMENT - END ROTATION CHARACTERISTICS FOR BEAM -COLUMNS

by

Theodore V. Galambos
and
Maxwell G. Lay

This work has been carried out as part of an investigation

sponsored jointly by the Welding Research Council and the
• Department of the Navy with funds furnished by the fo11pwing:

American Institute of Steel Construction.

American Iron and Steel Institute
Institute of Research, Lehigh University
Column Research Council (Advisory)
Office of Naval Research (Contract Nonr 610 (03»
Bureau of Ships
Bureau of Yards and Docks

Reproduction of this report in whole or in part is

permitted for any purpose of the United States
Government.

Fritz Engineering Laboratory

Department of Civil Engineering
Lehigh University
Bethlehem, .Pa.

May 1962

Fritz Engineering Laboratory Report No. 205A.35

 205A.35 i

S YN 0 PSI S

This report discusses the end moment - end rotation behavior of

wide flange steel beam-columns. It is shown that the theoretical moment-

rotation curves agree well with those obtained from tests on full size

columns~ provided that failure occurs by excessive deflection in the

plane of bending. As the rotation capacity of beam-columns is of con-

siderable interest, comparisons are made between the theoretical and

experimental values of this parameter.

Local and lateral-torsional buckling may also contribute to the

•
failure behavior of a beam-column and so approximate methods are pre-

sented.which allow prediction of both the occurrence and effect of these

phenomena. The amended rotation capacities are also compared with ex-

perimental results.

Curves are presented which enable rotation capacity to be pre-

dicted for any combination of slenderness ratio~ applied load and ratio

of end moments. Interaction curves for local buckling and lateral-tor-

sional buckling are also presented .

•
 205A.35 ii

TAB LEO F CON. TEN T S

1. INTRODUCTION 1

1.1 APPLICATION OF DEFORMATION CHARACTERISTICS 1

1.2 PREVIOUS WORK 2

1.3 PURPOSE OF THIS REPORT 2

LOAD-DEFORMATION BEHAVIOR OF BEAM-COLUMNS 4

3. EFFECT OF LOCAL AND LATERAL TORSIONAL BUCKLING 6

3.1 LOCAL BUCKLING 6

a. Flange Buckling 7

b. Web Buckling 8

c. Effect on Rotation Capacity 10

3.2 LATERAL TORSIONAL BUCKLING EFFECT 11

a. Unbraced Columns 11

b. Braced Columns 13

3.3 SUMMARY 14

4. EXPERIMENTAL SOURCES 16

5. COMPARISON BETWEEN EXPERIMENTAL AND THEORETICAL

. RESULTS 18

5.1 GENERAL 18

5.2 LOCAL. BUCKLING 18

5.3 LATERAL TORSIONAL BUCKLING 19

..
5.4 BRACED COLUMNS 20
 205A.35 iii

TAB L.E 0 F CON TEN T S (continued)

...
 205A.35
-1

• 1. I N T ROD U C T ION

Column research in the past has usually been directed towards the

strength properties of columns and beam-columns~l) However for many

plastic design and stability investigations it is also necessary to know

the deformation characteristics of the members, particularly in the ulti-

mate load. region.

For beam-columns these"characteristics may be defined by three

parameters at each joint: the end moment (M), the end rotation (8) and

the axial force (P). This report will consider the theoretical and ex-

perimental relationships between these parameters for beam-columns of

rolled steel wide-flange sections which are bent about their major axis.

1.1 APPLICATION OF DEFORMATION CHARACTERISTICS

Plastic design methods assume that a structure becomes a mechanism

at its ultimate load due to the formation of a sufficient number of plastic

hinges within the structure. All but the last hinge to form must undergo

a certain amount of inelastic rotation and it is necessary to ensure that

the member is capable of sustaining this rotation. Methods of calculating

the rotation requirement have been discussed elsewhere~2) However there

is little information on the rotation capacity which a beam-column can

provide to meet this requirement.

• Frames may also be analyzed by methods which check the stability

of the structure at each load increment. The stability check. requires a

 205A.35 -2

structure to return to its original configuration after a virtual dis-

•
turbance has been applied. As these procedures are based on deformation

analyses(3) (4) , they require a knowledge of the load deformation history

of the beam-columns for both the analysis and the stability check. Such

information is contained in the P-M-9 relationships. (5)

1.2 PREVIOUS WORK

For the elastic range it is possible to derive direct analytical

(6)
relationships between P, M and 9 . In the inelastic range the discon-

tinuities associated with yielding make such derivations impossible and

therefore inelastic P-M-9 curves are obtained by numerical and graphical

. procedures. The moment-rotation curves so derived include the effect of

residual stresses and are also applicable in the unloading range. A

..
collection of such M-P-9 curves is presented in Ref. 7.

1.3 PURPOSE OF THIS REPORT

There are two aspects of the present theory which require further

discussion:

(1) The moment-rotation curves only apply to in-plane bending

and do not consider the possibility of either lateral-

torsional or local buckling~7)

(2) The moment-rotation curves have not yet been experimentally

verified although they are-based on moment-curvature rela-

• tionships which are the outcome of .extensive experimental

work~8)
 205A.35 -3

, This report provides additional information on these two aspects.

Theoretical limits are given for .local and lateral-torsional buckling .

• Experimental results are abstracted from previous column tests and are

compared with these limits and with the moment-rotation curves .

•
 205A.35 -4

2. LOA D - D E FOR MAT ION B E H A V 10 R

• OF BEAM-COLUMNS

Column deflection curves form the basis for obtaining load-

deflection curves forbeam~columns~5) These column deflection curves

:. .
give the shape that an axially loaded column would take under a given

initial slope, 8. Figure 1 ~hows a set of such curves for constant

o
axial load. This particular family of curves is for strong axis bending

of as-rolled wide-flange members .and includes the effect of residual

stresses. However the curves do not include the effect of local or

lateral-torsional buckling.

Figure 2 presents a family of moment-rotation curves for constant

•
axial load. They are derived.from curves such as Fig. 1 by finding those

column deflection curve segments which satisfy the given conditions of

(5)
axial load, slenderness ratio and end-moment ratio. The slopes at the

ends of each segment are measured and this combination of moment and

rotation gives a point on the moment-rotation curve. The curves are for

various values of slenderness ratios. It is also possible to draw 8

o
contours which connect column segments derived from the same column de-

flection curve (dashed curves in Fig. 2.).

The resulting M-8 curves (Fig. 2) show how both moment capacity

and rotation capacity decrease with increasing slenderness ratio.

Figure 3 illustrates the effect of various axial load values. As the

"
axial load increases both the moment and the rotation capacity of the beam
 205A.35 -5

columns decrease. The influence of end-restraints can be seen from Fig. 4.

Forcing the member towards single curvature deformation reduces both the
. moment and rotation capacities.

It can be seen that at least some of the M-8 curves have an in-

elastic rotation range at ultimate moment, In order to measure rotation

capacity for plastic design purposes and also to give a measure of relative

curve sizes a simple definition will be proposed. The rotation capacity,

Rc ' of a beam column is defined as the ratio of two rotations:

( 1)

where 8 p and 8 are defined in Fig. 5. Both are measured at a moment

E
level 5% below the maximum moment. 8 is the rotation at this level during
E
loading and 8 p is the rotation at the same level during unloading. The

choice of 0.95 as the measuring level for R is to some extent arbitrary.

c
However, examination of the experimental curves showed that at this level

rapid unloading was either in process or was immdnent .

...

•
 205A.35 -6

3. E F F E C T 0 FLO CAL AND L ATE R A L

.
TOR S ION ALB U C K LIN G

It has been shown that a suitable collection of basic M-9 curves

are available(7) and it has been noted that these curves neglect the effect

of local and lateral-torsional buckling, These phenomena will cause devi-

ations from the available theoretical M-9 curves. If this deviation occurs

before the peak of the curve, both .moment and ro·tation capacity will be re-

duced, If it occurs after the peak, rotation capacity mayor may not be

affected. Lateral-torsional buckling may be prevented by a suitable bracing

system. However local buckling is a function of the geometry of the cross

section and will start when a certain critical strain has beenreached,(9)

3.1 LOCAL BUCKLING

Curves have been published(9) which give the critical strain for

" web or ,flange buckling as a function of the plate width-thickness ratio .

...
For a given axial load and a given cross section it is possible to compute,

from equilibrium conditions, the moment Ms corresponding to the relevant

critical strain. The column deflection curves are then searched and any

curves in which the maximum moment is less than Ms are considered safe

from local buckling. In this way a critical 90 , 9 0c ' may be found for
• each value of axial load and cross section (Fig.6). This critical 90 is

marked on M-9 curves such as Fig. 7 and represents the 9 0 contour at which

local buckling will occur. M-9 curves are not used beyond this contour.
 205A.35 -7

This simple searching technique applies only if the maximum column

moment does not occur at the end of the column, that is if the maximum

column moment is also the maximum column deflection curve moment. This is

the case for a beam-column such as AA' in Fig. 8. However many columns are

similar to BB I , and have their peak moment at a joint, that is at the end

of the member. The critical eo for such columns will be greater than

that found previously and it will be a function of slenderness ratio and

end restraint in addition to the normal variables. Thus an amended curve

for local buckling may be shown on Fig. 7 (dashed line) and this will be

more liberal than the previous cut-off contour. It is obtained by a some-

what more involved search of the column deflection curves. The boundary

between the two cases occurs when the node of the column deflection curve

is at the end of the column.

• a. Flange Buckling

Curves which show the critical strains at which local

buckling is assumed to occur for various bit ratios are re-

produced in Fig. 9 from Ref. 9. The solid line shows the curve

selected as governing these investigations. The maximum bit of

9.25 is slightly more liberal than that recommended for design

(l~
in the Commentary on Plastic Design in Steel (bit = 8.50)
but seems more in accord with later experimental results,

especially as it is now possible to specify the precise loading

conditions.

It can be seen that there are some members (bit> 9.25)

which cannot be assumed safe against local buckling and others

(b/t~ 7.25) which are not prone to local (flange) buckling.

 205A.35 -8

The following results apply to the large group of sections

...
falling in between, and, they give the loading at which local flange
.. buckling might be expected .

The values of critical 8 0 and axial load are determined

graphically from the column deflection curves, as previously

described. When 8 0 is plotted as a function of P it is found

that the section type causes little variation in the curves for

constant bit. In addition the variation between the curves for

the extreme cases of bit = 7.25 to bit = 9.25 form a small band

whose width is generally less than the presently available 8 0

increments. Fig. 10 shows the analytical results for the case of

equal end moments (solid lines) and also for the case where the

column is pinned at one joint and where the maximum column moment

occurs at the other joint (dashed lines).

From such curves it is now possible to predict the point

of occurrence of local buckling once the bit ratio is known.

b. Web Buckling

Calculation of the critical 8 0 v~lues when web buckling

I

governs follows a procedure similar to that used when flange

buckling is critical. Fig. 11 (taken from Ref. 9) gives values

of the critical'web b~ckling strains. The solid lines give th~

relationship between P/~ cry and df/w for various values of these

critical strains as measured py the maximum strain in the com-

pression flange. The dashed line shows the limit recommended in

the Plastic Design~ommentary(lO) for a section with d/df = 1.05

 205A,35 -9

., and A/Aw = 2.0. If the axial load and the section properties

are known the critical strain may be read directly from Fig. 11.

The column deflection curves are searched to find the

curves in which the critical conditions are first attained. The

resulting values of eoc are shown. in Fig. 12, although for clarity

the cases where the maximum moment occurs at the end of the column

have been omitted. It has also been assumed that d = 1.05 d f .

It was found that eoc is not sensitive to changes in section,

provided that d/w is constant. Furthermore the curves for the

extreme cases of d/w = 36 and 53 themselves form a narrow band.

When Fig. 10 (flange buckling) and Fig. 12 (web buckling)

•
are compared it is seen that the values of eoc are almost identical

• for the two cases. This occurrence greatly simplified local

buckling criteria as one set of curves may be used for both

cases. Fig. 13 illustrates this final result of the local buckling

analyses.

The bands enclosed by the curves A,B and C in, Fig. 13

represent the extremes of variation with section properties.

Curve A is for b/ t = 9.25 or d/w = 53, Curve B for b/t. = 8.25 or

d/w = 36, and Curve C for bit = 7.25. The dotted lines apply to

the p= 0 cases where the maximum moment occurs at the end of

the column, whereas A, Band C apply directly to the f = + 1.0 case.

•
For r = - LO the ~ = 0 curves may be used provided the slenderness

ratio of the column is taken as half its actual value.

 205A.35 -10

For most of Fig. 13 the width of the band is less than the

P presently available(7) (1962) 8 increments. The close coincidence

0

between the various critical 8 0 curves arises from the relatively

large increases in strain which occur between column deflection

curve increments in this region. If the curves are investigated

for a hinge formation criterion a similar critical 8 0 range is

indicated. In addition some of the results presented later in-

dicate that Fig. 13 may be a critical range, even for sections

whose bit and dlw values appear safe. This would arise from the

very large strain increases occurring in the critical region

rather than from' any limitations on the basic local buckling

• criteria.

•
From Fig. 13 cut~off lines may be drawn on the M-8 curves,

as has already been indicated in Fig. 7. When the column de-

flection curve node falls within the beam-column length these

cut-offs correspond to a 8 0 contour.

c. Effect on Rotation Capacity

The effect.~f local buckling on rotation capacity, as de-

fined in Fig~ 5, is shown in Fig. 14. The assumption that un-

loading occ4rs at local buckling may sometimes be conservative as

local buckling does not always cause immediate unloading. If the

local buckling contour falls beyond the defined rotation capacity

range there is no effect at all '(Fig. 14, lower curve). If it

•
falls within the range (upper curve in Fig. 14) it is assumed

that the load falls off with a characteristic identical to the

 205A.35 -11

original curve, AB.

Rotation capacity values obtained in this manner are

.. shown in Fig. 19. They will be discussed later in this report.

3.2 LATERAL TORSIONAL BUCKLING EFFECT

a. Unbraced Columns

At present no general solution is available for the

determination of M-8-P curves for lateral-torsional buckling.

A first movement solution has been given for the inelastic case

with a zero moment gradient(ll) (~ = + 1.00). The solution for

an 8 WF 31 section using this method is shown in Fig. 15 as the

family of solid curves. Results for three other sections. are

•
also shown.

The first movement solution only gives the moment at which

the equilibrium bifurcates. It cannot provide information on the

rotation capacity. However, consideration of experimental results

has shown that a good estimate of rotation capacity can be ex-

pressed by:

R' (2)
c
r
"

where R'c = rotation capacity as amended by lateral-

torsional buckling

ML = critical moment for lateral-torsional

buckling ,~

~ = in-plane maximum moment

..
R'
c = in-plane rotation capacity
 205A.35 -12

To apply the theoretical solution(ll) for ML to other loading

cases,Massonnet's equation(12) for equivalent moments can be

.
used. That is,
~

~ = ~j 0.3 + 0.4( + 0.313

I
2
(3)

where ME is the equivalent moment.

This allows unequal end moments to be converted~o an

equivalent pair of equal and opposite end moments. If ML ) ME

then lateral-torsional buckling is not critical for the particular

beam-column under consideration.

The analytical method(ll) predicts moments rather than

rotations. Hence the results can not be presented in the manner

adopted for modifying the M-8 curves for local buckling. The

rotation value results as read from M-8 curves are also ill-

conditioned for moment values near the maximum. It is an in-

herent advantage of column deflection methods that they predict

rotations and thus lead automatically to well-condi tioned, unique

solutions.

It has been shown(ll) that a significant section constant

in lateral-torsional buckling is the ratio DT where

(inch units) (4)

*Work on the same topic at Cambridge University produced

a similar cross sectional constant "T" w:here T = G x AK d 2 /I 2 ,
T
G = modulus of rigidity (Ref.l3).
 205A.35 -13

where KT St. Venant's torsional coefficient

A = cross sectional area

d = section depth
•

Results are available (see Fig. 15) for four sections (27 WF 94,

8 WF 31, 14 WF 142, 14 WF 246) whose D values are 219, 925, 1580

T
and 3712 respectively. If D
T
> 1000 there are loading conditions

for the section which preclude lateral-torsional buckling.

Figure 16 gives values of ML/~ to be used in the rotation

capacity equation. As ~ is a function of L/r y rather than L/r x

it is not possible to present the results more generally.

In Figure 19, later in this report, these ratios and

Eq. (2) are used to obtain the rotation capacity values for an

8 WF 31 section (DT = 925).

b. Braced Columns

The above principles for finding rotation capacity may

be applied to the segments between lateral braces for intermit-

tently braced columns (Fig. 17). The moments at A and E are

given and hence the moments at the brace points B, C and D can

be found directly from the revelant column deflection curve.

The curve used depends on the rotation capacity required and can

be selected by considering charts such as Fig. 2, with due regard

for local buckling (Fig. 13).

As the moment at the brace points depends on the par-

ticular column deflection curve used the results of analysis of

 205A,35 -14

a braced column can be presented more in the manner used for

local buckling. However bracing dimensions are too variable to

.. present general results and so the method will be illustrated by

an example. Fig, 17 shows a braced column - it is actually Test

A8 to be discussed later - and it will be checked to see whether

it is capable of delivering the full rotation capacity with the

bracing used. Local buckling occurs within the plastic plateau

(Fig. 2) and so the critical 8 0 for P/Py = 0.3 will be 0,06

(Fig. 13). From this column deflection curve the bending moment

diagram shown in Fig, 17 is constructed (the moments between

braces are represented by straight lines), Next the ratio of

moments givesf3 and substitution in Eq, (3) gives ME/MM for each

segment, ~ is known for each segment so ME is found. From Fig,

• 15 ML is obtained for each segment. As ME/ML is greater than 1,0

for segments AB and BC the column is unstable and thus the bracing

is not sufficient to allow attainment of the available rotation

capacity.

Checking for eo = 0.05 shows that the bracing is adequate

for this degree of loading.

3~3 SUMMARY

Figure 18 shows the in-plane rotation capacities and also the

zones of influence of lateral-torsional and local buckling. The rotation

capacity values given are taken directly from the original M-e curves(7)

and do not include the effects of local or lateral-torsional buckling,

 205A.35 -15

Figure 19 gives the rotation capacity results as ammended by

• local and lateral-torsional buckling. The values were obtained by ap-

plying the modifications discussed beforehand to the in-plane rotation

•
capacities in Fig. 18. The modifications apply strictly to the 8 WF 31

section. For other shapes the appropriate modification factors may be

obtained from Figs. 13 and 16. However, for most cases the 8 WF 31 will

give conservative but useable results.

It is seen that local buckling will precipitate failure in the

f3 = 0 case, if the axial loads are low. Otherwise lateral-torsional

buckling is usually critical.

a. Limits at Zero Load

The limits of Fig. 19 as axial load approaches zero

can be found by considering available beam test results. For

the f3 = + 1.0 case at zero load the curvature distribution

will be uniform (Fig. 22a). Thus the critical local buckling

strain is reached simultaneously-along th: beam. A simple

analysis shows that the expected Rc value is 11.9, if 1atera1-

( 15)
torsional buckling does not occur. - Beam tests indicate this

to be .so if L/r y is less than about 40. If L/r y :> 40 1atera1-

torsional buckling will seriously curtail the rotation capacity.

As rx/r y can have values of the magnitude of three, the curves

in Fig. 19 «(3 = 1.0) will tend to qefinite

~
1imit~,
~.
less than

or equal to 12. Their precise determination requires further

inves tiga tion.

For thef3 =0 case the situation is somewhat different.

There the curvature distribution is not uniform but exhibits a

 205A,35 -16

pronounced peak which will be many times the averagecurva-

• ture (Fig 22b), Visually, or by conjugate beam theory, it

can be seen that the rotation capacity as affected by local

•
buckling will decrease asf' decreases from 1.0 to O. Analysis

gives a value of Rc = 2 for bit = 9.25 and Rc = 4 for bit = 7.25,

at zero axial load. Tests(16) on beams with moment gradients

indicate that this tendency exists. However, there is also

an indication that the local buckling theory is conservative

and that strains in excess of the critical may occur without

local buckling. Fig. 19 gives then a lower bound to these

results and tends to a limit of Rc = 2.0 at zero axial ~oad.

The appa~ent peak in the Rc curves for~= 0 arises

• ;
from the geometry of the Rc definition,~nd from the more favor-

able curvature distributions which res~lt from the addition of

the moments due to axial load (Fig. 22c).

 205A,35 ... 17

.
4. E X PER I MEN TAL SOU R C E S

This section presents experimental results in order to check the

theoretical processes discussed earlier in the report. Many tests have

been done on centrally loaded columns, but relatively few on beam-columns.

Of the reported beam-column tests only a limited number have been oriented

towards finding moment-rotation characteristics and of those tests the

rotations were often such that the recording instruments did not function

throughout the entire loading range,

"

This report is confined to a discussion of the "T" and "A" series

• .,.
tests performed at Lehigh University betweeh 1948 and 1959, These tests

are reported in Ref, 14, and Table I of this report gives the test de-

tails relevant to this investigation, The M-9 curves obtained are shown

in the Appendix,

Of the 42 tests in the two series ("Til Series - unbraced;

"A" Series - braced) only 20 were completely suitable for use in this

investigation, Some of the remaining tests were discontinued before un-

loading occurred (4), some were pin~ended (5), in some the rotation

readings were unreliable (5), and in two tests the moments were applied

before the axial load. There were six tests which were discontinued

before unloading but which gave some useful rotation information and these

are also recorded together. with the twenty fully applicable test~.
 205A,35 -18

Rotation capacity was measured by applying the definition of

Fig, 5 to the experimental M-8 curves, The results of this process

are shown in Table II for all tests, and Figs, 20 and 21 show fifteen

of the results graphically, Fig. 20 is for the unbraced tests with

r=s = + 1.0 and P/Py = 0,12. It can be seen that good agreement is ob-

tained between the test results and the predicted curve.

Fig. 21 is for the braced ~~'I series tests, In this case the

correlation between theory and experiment is not as consistent as it

was for thef3 =0 case, The variations between theory and experiment

will be discussed in the next section, however, Figs, 20 and 21 show

that a reasonable estimate of rotation capacity can be expected from

the preceding theory,

•
 205A.35 -19

•
5. C OM PAR I S O.N BET WEE N E X PER I M E NT A L
• AND THE 0 RET I C A LR E S U L T S

5.1 GENERAL

Table II records the predicted and experimental rotation capacity

results and also the difference between these two quantitites.The pre-

dicted rotation capacity includes the effects of lateral-torsional

buckling and local buckling where these are applicable, For the tests

carried to completion this difference (theory-experiment) had an average

value of - 0,06 and a range from - 1.10 to + 0.70. The majority of the

results were conservative and the most serious overestimate was + 0,70

for test A4 (See Fig. AS), This overestimate arose partly from the

experimental maximum moment being M/M

p
= 0.60 as compared with the

theoretical value of M/Mp = 0,55, Hence the rotation capacity was measured

at M/Mp = 0.57 which was still 3-1/2% above the theoretical moment. The

discrepancy can thus be traced to the definition of Rc which was chosen

in order to account for the more critical case of the predicted moment

being greater than the actual moment,

5,2 LOCAL BUCKLING

Table II indicates those tests for which an 8 WF 31 analysis

predicted a local buckling failure mode, Failure by local buckling,was

observed in six of the ten tests. The four remaining tests were all

4 WF 13 sections with high slenderness ratios, but it is noted that three

 205A.35 -20

4 WF 13 sections with lower slenderness ratios were amongst the six which
· failed by local buckling. The 4 WF 13 is a section for which bit = 6.04
• and from Fig. 9 local buckling is not expected to occur under any circum-

stances, however it is also a section which is relatively strong torsionally

and in short members local buckling is likely when the strains become

large and are not relieved by lateral-torsional buckling. In thi~; case

the limits of the Haaijer analysis(9) are exceeded and it is desirable to

base all calculations on the 8 WF 31 section although this will sometimes

'lead to conservative results.

In tests A9 and Tl3 local puckling was observed when theoretically

predicted but unloading did not occur until late~ in the test. In test

• A9 the end rotation at unloading was twice itsva~ue when local buckling

was first observed and this would account for the large underestimate of

rotation capacity for this test. The inability of local buckling to always

cause immediate unloading will lead to conservative results, on the other

hand it is not considered desirable at this stage to utilize members in

which some local buckling has occurred.

From the limited samples available it would appear that the theory

predicts both the point of occurrence and the effect of local buckling,
, ..
but is likely to be conservative,in both estimate~o

5.3 LATERAL-TORSIONAL BUCKLING

Ten completed tests failed by lateral-torsional buckling (Table II)

a~di~ nine of these this failure mode had been predicted. The method

also appears to give a good estimate of rotation capacity in all(~ ranges.

 205A.35 -21

5.4 BRACED COLUMNS

The out-of-plane slenderness ratio L between brace points in

ry
the "A" Series tests was about 40 (Table I). Lateral movements were ob-

s~rved in a number of tests, but lateral-torsional buckling was only con-

sidered as the failure mode in test A8. ~he three tests on the torsionally

strong 4 WF 13 (AS, A6 and A7) had no predicted out-of-plane buckling and

their results agreed well with those predicted by the basic M-8 curves

(Table II). Of the six remaining tests three (A2, A3 and A4) were 8 WF 31

tests and some reduction in rotation capacity was expected and~:observed..

In the three tests on the torsionally weak 8 B 13 (A8, A9 and A10) , large

rotation capacity reductions occurred as predicted. Local buckling

occurred in test A9, this phenomenon being independent of the bracing

•
spacing, It would appear that for torsionally weak sections the

bracing requirements for attainment of full rotation capacity could be-

."
come excessive,

5.5 MOMENT-ROTATION CURVES

The accuracy of prediction of rotation capacity is also a measure

of the accuracy of the prediction of the general M-8 curve shape. In

the "A" Series braced tests (Table II) AS, A6 and A7 had no predicted or

observed lateral or local buckling effects, hence they should best fit

the basic M-8 curves derived by OjalvoP). Fig, A6 in the Appendix~

shows this to be the case and these tests are strong confirmation of the
"

Ojalvo theory. Tests AS and A6 show particularly good correlation. In

addition these are 4 WF 13 tests whereas the column deflection curves

were derived for the 8 WF 31. For tests A2, A3 and A4 the correlation
 205A.35 . . 22

is still good although the O)alvo curves appear conservative. As ex-

pected the results for the tests (A8, A9 and A10) on the torsionally

weak 8 B 13 do not show good agreement, due to out-of-plane action.

However for the "T" Series tests there are some marked. deviations

between the test results and portions of the in-plane M-8curves (Figs.

A.l - A.4). These arise mainly from (1) differences in maximum moment
,
'.
and (2) differences in rotation capacity and indicate that the basic
,i:l '

curves should only be used where in-plane, local buckling free, behavior
',,'

is assured.

5.6 ROTATION CAPACITY VALUES

It has .been shown that for local buckling the results based on

the 8 WF 31 may be used for all sections. For lateral-torsional buckling

the same applie,s if the section has a similar DT value (925) or if it is

stronger and greater refinements are not required. In these cases

Fig. 19 gives Rc values directly. Otherwise Eq. 2 is evaluated and used

in conjunction with Fig. 18. (Note that L/r y may also vary.)

Attention is drawn to the magnitude of both the theoretical and

experimental Rc values. For the ~w¢pty.completed tests in Table II the

average experimental,value was only 1.62. Theoretically, it can be seen

from Fig. 19 that only columns with low slenderness ratio and axial loads
.. can be expected to provide Rc :> 3.0.
 205A.35 ~23

There are many columns which do not fall into these categories .
•
Highly loaded columns in tall buildings or slender columns in low

buildings must be analyzed carefully if they are required to provide

some rotation capacity during the course of their structural behavior.

Finally it is noted that rotation capacities would not normally

be required to an accuracy of greater than 0.25.

 205A.35 -24

•
6. CON C L U S ION S

This report has presented an analysis of the presently available

moment-rotation curves for beam-columns. Certain modifications have been

proposed and the basic theory and the modifications have been checked

against experimental results. The following conclusions may be stated:

a. The methods formulated for the prediction of local and

lateral-torsional buckling and their effect on rotation capacity

give results which appear to be reasonably consistent when ap-

plied to test results.

b. As the work in this report is largely based on the

( 5)
theory presented by Ojalvo ,possibly the most interesting

outcome of the experimental analyses is that they offer con-

firmation of Ojalvo's theory and do not disclose any incon-

sistencies. When the assumptions of in-plane behavior are

fulfilled the agreement is excellent. When local and lateral-

torsional buckling occur there appears justification for using

the approximate modifications proposed in this report to predict

rotation capacities.

c. A method has been presented which gives an interaction

'0 curve between axial load and column deflection curve for local

buckling. It has been shown that one curve provides a suffiently

accurate prediction of flange and web buckling and that the be-
 205A.35 ...·25

havior of all sections may be reduced to a few simple cases •

.
In this form the criteria may be applied to any column loading

case and presents a limit on the usable length of the moment-

rotation curve. The estimates of rotation capacity so obtained

are sometimes conservative as local buckling is frequently not

catastrophic.

d. A less complete solution is presented for lateral-

torsional buckling, and more refined methods of analysis are

required for the final solution of this problem.

e. Most columns appear to have a.definite rotation capacity

although in many cases it is not very great. Because of this it

..
seems necessary .to check the rotation requirements of hinges

forming in the columns of plastically designed frames. If the

requirement exceeds the capacity predicted by this report it

would seem advisable to either use a stronger member or to carry

out the design by·one of the stability methods now available.

f. For the spacing of lateral bracing of beam-columns it

has been shown that the column deflection curve may be used to

obtain the actual in-plane moments at a brace. Hence design

checks may proceed more logically. The bracing requirements

presently used(.lO) appeared adequate although care must be taken

• if the member is torsionally weak, as' it is unlikely that the

full rotation requirement will be attained.

 205A.35 -26

'7. ACKNOWLEDGMENTS

This study is part of a general investigation

''Welded Continuous Frames and Their Components" currently

being carried out at Fritz Engineering Laboratory of the

Civil Engineering Department of Lehigh University under

the general direction of Lynn S. Beedle. The investigation

is sponsored jointly by the Welding Re~earch Council, and

the Department of the Navy, with funds furnished by the

American Institute of Steel Construction, the American Iron

and Steel Institute, Lehigh University Institute of Research,

the Bureau of Ships, and the Bureau of Yards and Docks. The

Column Research Council acts in an advisory capacity.

The authors express their thanks to all those whose

experimental and theoretical work has been drawn upon and used

in this report, and to Mrs. Dorothy Fielding who did the typing

and Mr. Richard Sopko for his assistance with the drawings.

-0

.'
 205A035 -27

80 NOM E N C L A T U R-E

2b Flange width of section

d Depth of Section

df Distance between flange centroids

rx Strong acis radius of gyration

ry Weak axis radius of gyration

t Flange thickness

w Web thickness

A Cross-sectional area

Aw Web area
•
DT Lateral-torsional buckling coefficient, Eq. (4)

L/r Slenderness ratio

M End moment

ME Equivalent moment, Eqo (3)

ML Lateral-torsional buckling moment

MM Maximum moment

M Local buckling critical moment

S
~ Moment at first yield

P Axial Load

P =A (J
Y Y
-~
Rotation capacity

R' Rotation capacity modified for lateral-torsional buckling

c
Ratio of enp moments on a column, the larger moment in
the denominator and the moments measured in opposite
directions.
 205A.35 ... 28

• Em Maximum strain in compression flange

cry Yield strain

9 End rotation

90 Column deflection curve slope at zero deflection

9 0c Critical column deflection curve for local buckling

:J
arc
Abscissae of M-9 curve defined in Fig. 5

Residual stress parameter

0y Yield stress

•
 ·' .
TABLE I

TEST DATA (FROM REF .14 )

Force Ratio Bracing Spacing

Test L/ rx Section
!l- Value P/Py . (Measured From Maximum Moment End)

T-7 111 4 WF 13 - 0.56 0.26

T-12 55 8 WF 31 + 1,00 0.12
T-16 CIl
41 8 WF 31 + 1.00 0.12
T-17 .j.J 56 4 WF 13 - 0.50 0.12
CIl
T-19 Q)
E-l
28 8 WF 31 + 1.00 0012
T-20 "C
56 4 WF 13 + 1 00 0 0,12 No
T-21 Q)
(J
56 4 ToJF 13 - 0,55 0,47 Bracing
T-23 til
1-1
83 4tIF 13 0 0.11
T-26 .0 84 4WF 13 + 1.00 0.12
T 31
M § 112 4 WF 13 0 0.12
T-32 112 4 WF 13 + 1.00 0.12
A-2 55 8 WF 31 0 0.65 71" .·..·12.1" (one brace)
A-3 55 8 WF 31 0 0.32 67.5" - 124.5" (one brace)
A-4 CIl
.j.J 55 8 WF 31 0 0.49 67" - 125" (one brace)
A-5 CIl
Q) 110 4 WF 13 0 0.33 36'1 37" - 119 " (two braces)
A-6 E-l 112 4WF 13 0 0.50 36" 37" - 119" (two braces)
A-7 "C
Q) 112 4 WF 13 0 0,16 36" 37" - 119" (two braces)
A-8 (J
til 52 8 B 13 0 0,30 30" . 34" 36" - 67.5" (three braces)
A-9 1-1
l:l:l 52 8 B 13 0 0.12 30 " 30" 38" - 70" (three braces)
A-10 52 8 B 13 0 0.60 30" 30" 38" - 70" (three braces)
T-4 "C
Q) 55 ti tJF 31 - 0.50 0.12
::l 111 4 WF 13 - 0.56 0.10
T-9 t:: CIl
T-13 .~
.j.J
.j.J
CIl 55 8 WF 31 0 0.12 No
T-24 t:: Q)
83 4 WF 13 - 0.52 0.12 Bracing
oE-l
T-29 (J
CIl 84 4 WF 13 - 1.00 0.13
T-30 ~
0 112 4 WF 13 - 1.00 0,12

Section COilstants:
·Dor (in) rx/r y
c
4WF 13 2360 1. 73 N
\0
8WF 31 925 1. 73
8 B 13 364 2.75
 -31

.IP is constant I
z
o
...
(,)
IJJ
...J
~
IJJ
o
•

p p

COLUMN LENGTH

Fig. 1 COLUMN DEFLECTION CURVES

"
 .. . •

0.7

I
/
I
I /
tr:l
I
Z I
? 0.5 I I
~ M I I
i~M
~ . y 0.4
~
,
I
I
I
I

t:::l I I
·1
, MI p .'
~ ~~/.
~
Py =0.3.
~~jZ1:;:::::::Ie==k~
t-3
H
o
·z , I O'"RC =O. 3 O"y
oy =33 ksi
,,
C":l
I I
,~
<: "
tr:l
tJ)
02
•

I
I . I
! -I- Strong Axis Bending

0.1 I I

,,
I
I
I ,
I
I I
o O. 0.02' 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 0.11 0.12 0.13
8. END ROTATION I
W
N
 -33

P =0.12 P.
•

1.0 P=0.20P.

13= 0
L/r=60
M
My O.5 -I-
P=0.3Py '
P=0.4Py
P ,p
M
_,0-0 ----0-+
• 0.04 0.08 0.12
8
Fig. 3 INFLUENCE OF AXIAL LOAD ON M-~ CURVES

1.0

13 = -1.0
P= 0.3ACY
Strong Axis
L/r= 60

{3=0 '-:I-
13 =+ 1.0 , ,13M M
P +-0--0---o-+p

0.04 0.08 0.12

8
Fig. 4 INFLUENCE OF END RESTRAINT, ON M-e CURVES
 -34

MAXIMUM MOMENT MM
MM t--.L-------------===-=---__

END
MOMENT
M

Fig. 5

DEFLECTION

C.RITICAL 80 CONTOUR

p COLUMN LENGTH p,
Fig. 6 CRITICAL COLUMN .DEFLECTION CURVE·
 -35

, L/r=40

0.8
---I-, I
~AMENDED
CUT-OFF·
/
I ,
I
I
M I
My OA- ~ ,
M-8
CURVES

F
[
f, o 0.04
8, END ROTATION
Fig. 7 CRITICAL 9 o CONTOUR
.

DEFLECTION = ~
I I
BS . AA
,!1
\

•
p
- -
p

~L/rx MAXIMUM MOMENTS

t
Fig. 8 LOCATION OF. COLUMN SEGMENTS
 -36

40
"P=o
P=O.OI

30

~= Coefficient of Restraint
20
EQ.. _

10

Ef ..
0---------·'
o 5 10 15 20 25
bIt.

Fig. 9 FLAN.GE LOCAL BUCKLING

0.6
~=9.25--~ ~
, ---- .
~=o

P/~ 0.4
~ =7.~5.- ----~~. . . ~.

Solid Lines:
........ -- -- --
--.,
-- -- -- ..

~ =·+1.0
0.2
•

. 0.02 0.04. 0.06 0.08 0.10

CRITICAL 80
Fig. 10 CRITICAL 90 ,VALUES FOR FLANGE LOCAL
BUCKLING
 . -,37

0.75

AA =2.0
w

1;-=1.05
~.
0.50

P
Away

0.25

o
30 40 50 60

ALLOWABLE WEB DIMENSIONS

•

0.6
0..>-
Q:
0
~
0
«
0
...J.
0.4
Ii
~~w
~ 0.2
X
• «
.- ',.'

'. 0.02 0.04 0.06 0.08 0.10

CRITICAL 80
Fig. 12 CRITICAL 80 VALUESFQR WEB LOCAL BUCKLING
 -38

ABC
0.60

Values For
050 ~=O

t
~~
0.40
"
.........

-.. . . . ;
.......... "" " " " L/r =30
~ ......................
~~ .......................
~........... ............
030
......... " L/r=40 "
~= + 1.0
~
~"" "" ........... ,
........... ~/r=50 .......................
~ =9.25 .......... ""
............
0.20 curveA<d ...... L/r=60
......... ..................... .
w= 53 ..................
...............
~ =8.25
curveB<d .
0.10 -=36
w
CurveC-~=7.25

0.04 0.05 0.06 0.07 0.08 0.09

• 80

Fig. 13 CRITICAL COLUMN DEFLECTION. CURVE FOR

FLANGE AND WEB LOCAL BUCKLING
 -39

ROTATION CAPICITY AMEN ED

FOR LOCAL BUCKbJ.NG
•. '~~:',;. . .:~, ~_.~.;,; :..;.~ ", 0;. ".
.:.".•

-...:"?:...... \:~: .. "'-"•..,.':,-' '.,

-MyM IN-PLANE ROTATI N

CAPACITIES
\
i80e
LOCAL
BUCKLING
CONTOUR
END ROTATION, 9

. .' . .

..

. Fig. 14:: EF_F~CT OF LOCAL BUCKLING ON ROTATION CAPAC I TY

I.
 -40

•••••• DT:: 3712

-. -e_- DT:: 1580

- - - - DT:: 925
'. ----DT = 219
\.
\.
\" ..
0.8 ..
"
..
"
..
\
\
\ ,,
".
, ...
..
0.7

", , " "- ...

.... -1-
\.
\
' ,, '., ..
.....
{3=+ID

0.6 .\.
, ".,
.... ,
'. " .... ..
\
.. ' .
\

,,
"
'. \. ...

M
Mp
0.5
" " ,,
\
.

' ..~
\ •• P-
\ . .•'P."-
.y
\., ..
. ' : e.
\
\
\.
...
..
...
." \. .. ··•
\ .... "" ,
\ \
0.4 ~< ..
\\ \.
, .. \ .
\.
",
"
., :. -=0.3
P
\ . Py
, \
\
:•
:.
:
\ \
0.3 "" ... \ \· :'.
\. ." .~. \ : :
\.: ,'~.
,: , .
\
\
..
: . \
\l
0.2 \ :£'=0.5',
..p
\. y "
,\
\
\ . \l
.... \ :
. \i.
\
,, \

.0.1
,
,
t

o 50 100 150· 200 250 300

.. SLENDERNESS RATIO,L/ry

Fig. 15 INTERACTION CURVES FOR LATERAL-TORSIONAL

BUCKLING
 -41

L= 50ry
1.0 •.••.•••••.••••.••.••••• e

... .. .

.. ...
. l -..
.. ~ :~~?ry
L=50ry
LIFIOOr
•••
y
-
.'lo.. ." ••••
09
.'. ..- ......,--.--....
., _ . _._._._....... e.
.'., ...
I .. ' • •• •

...... ............... -._.-._. , .,....

.... _ . _ : e. • •

.'... ~

0.8 •~ : L 200
: = r y ',.\ .'.•
,

L= 50r y
.'., --".---
0.7
' ....' ; __

0.6
" ""
.. ....
\:..
"'t-~

\; ---
..... ' .... ....':..
..... ....i ........ ....._-
•
0.5
.....
r:,~~'~.I ' - .....

,, '"'.i
00 ',<"
,\ 0 ~
,., ~ I
•I \11"J-
0.4 ,,(, \ ,,
.
, .~
\ 0
·I
·1 r ,
,q • II
\
'J- I . U'
\
0.3 \
\ ·io
I ... 0<
\
\
\ · \
\
\ \
0.2 L - -_ _.....I..._ _....L......&.-......._ _L - - _ - - - L - . L L . ...L- I.-

o 0.1 0.2 0.3 0.4 0.5 0.6

f.
Py
•

Fig. 16 REDUCTION FACTOR (MdMM) TO ALLOW FOR

LATERAL-TORSIONAL BUCKLING
 -42

PIA CTy=0.30
(1~ DI~NS I ,
A SM=0.866M y
3d' 9.25 35.3
BIoo--'---.--..L......f-+-+----L..-+-+-+-----1--+-t-+--
, 64.25" 19.8 75.6
TestA8
C ~... -___Ir___-----II"-+-_+_----t-----+__+_---~t--+---
100.5" 31.0 118.1

D~04-_____jr___-----L.._+_-------I.__+_----&_._+--

8813
L=168" .b...=197
r
BRACING y
POINTS

Et----------L....--------------L--
P/Auy=0.30

M=0.866My
C
80 =0.06 0.918 0.9600.8660.832 0.622,<PR
B .....- - 1 0 .795M y 1
T
0.7800.891 0.795 0.710 0.610' <P b
A
L
0.710 0.857- 0.62 0.53 .605

8BI3
BENDING
MOMENT' 0.0 0.548 0.44 0.24 0.54
DIAGRA

E o
' ..
•

Fig 0 17 , CALCULATIONS, FOR A BRACED COLUMN

 -43

,,/
".
,-..- -- _

'LOCAL BUCKLING
- 0

ZONE.

L/r=4b
L/r=50
0.2

/ L/r=80 L/r=60
L/r=120 L/r= 100

•
2.0 4.0 6.0 . 8.0 ·10.0
Rc ,ROTATION CAPACITY

0.6 L/r=80
L/r=60
113 =+ 1.0 I
t--+---l~~r-' L/r =40
L/r=30
~-~../ L/r=20

LOCAL.
/4-BUCKLING

2.0 4.0 6.0 8.0 10.0

ROTATION CAPACITY

Fig. 18 IN-PLANE ROTATION CAPACITIES ( = 0, = + 1.0)

 -44

02
,
, I
I
I
•
1.0 2.0 3.0 ~ 4.0 5.0
., ROTATION CAPACITY, Rc

0.6
I~=+1.01

~,

•
1.0 2.0 3.0 4.0 5.0
ROTATION CAPACITY, Rc
Fig. 19 AMENDED ROTATION CAPACITIES <(S= 0, ~ = + LO)
 -45

{3=+1.0
u P
a:: 3.0 THEORETICAL P =0.12
y
CURVE
....>- Completed
C3 Tests
~
~. o Test Results
02.0
z
o
~
6a:: 1.0

20 . 40 60 80 '100 120
SLENDERNESS RATIO
Fig. 20 RES ULTS FOR UNBRACED TES TS (~= + 10 0)
•

•
4.0
~VF31
I-BRACE
u
a:: 3.0 8BI3
3 BRACES
....>-'
a
~
~
02.0
z
o
'~

~
oa:: 1.0
4YFt3 '
2 BRACE;S

'.
0.1 0.2 0.3 0.4 0.5 0.6
P/Py
Fig. 21 RESULTS FOR BRACED TESTS (~= 0)
 -46

¢CRIT ,---------------., ¢ C~IT

CURVATURE DIAGRAM

CURVATURE DIAGRAM

( b)
¢CRIT

CURVATURE DIAGRAM

M
P ~~=========~---p
(C)
•

. Fig. 22 CURVATURE DISTRIBUTIONS FOR VARIOUS

LO~ING CASES
 205A.35
-47

••

APPENDIX

. EXPERIMENTAL AND THEORETICAL

.• MOMENT-ROTATION CURVES

(Also see Table I)

•
 -48

" -- .........
.........
"'"
0.8 /
/
/
""
I "
I ,
I "-
/ "T12 THEORY
0.7
I
I
I
I
O. I ~ T32 THEORY
I ~
I :-..
I T32 TEST
0.5 I
I
1!l
M .. I
p I
•
.I Theoretical Curves Are .
·0.4 I For In- Plane Behavior
I
I T 7 ;L/rx=1I1 ;~=-0.56 P/~=0.26
I TI2 ; L/rx=55;~=+ 1.00 P/~=O.l2
0.3
I T32;L/rx=1I2 ;~=+ 1.00 P/Py =O.l2
I
I
I
0.2 I
I

0.1

•
..
0.02 0.04 0.06 0.08 0.10
8, END ROTATION

Fig. Al TEST RESULTS (T SERIES)

..
'
 -49

--- ------------------
0.9
I
I
/
".....

/
_.::..:---
TI7,TEST
~....."
T17, THEORY
.......
.... ,
/ ,
I
I "
I I "
I "
I
0.8 I II " '}16, THEORY

I I
I I
I I
0.7 I I
I I
I
~
,,
I , ,
0.6 I
I "
\ \ T26,THEORY
T26,TEST \
\
\
0.5 \
M
• Mp
0.4
'.

Theoretical Curves Are

For 1n - Plane Behavior
T16; L/rx=41;~=+1.0 P/Py =O.l2
T17; Llrx=56;~=-0.50 P/Py =O.l2
T26; L/rx=84;~= 11.00 P/Py =0.12

0.1

, 0.02 0.04 0.06 '0;08 0.10

• 8, END ROTATION

Fig. A2TESTRESULTS (TSERIES)

 -50

0.9

. /
/ "
,..
"",.... .----.. ........
- --/
""'-Y.'/
/ .........
..........
-------
--- ----
T31
THEORY
/ ....................
I I .......
I I
0.8 I I
I T31
I
, I
TEST

0.7
,
I
.l Tl9
0.6 I TEST
I
M I
Mp I
I

•
0.5 t
I
I /
/'
--- -- --
T21 ---.
THEORY
l /
II
0.4 I
I
I
1
Theoretical Curves Are
0.3
For In- Plan.e Behavior

T19; L/rx=28; ~=+ 1.0 P/Py=O.l2

T21; L/rx=56 ;~= -0.55P/Py=0.47
0.2
T3t·, L/rx , ~ == 0
=112' PIP.y =012
•

• "0.02 0.04 0.06 0.08 0.10

• 8, END ROTATION

.Fig. A3 TEST RESULTS (T SERIES)

 -51

•
;., .
0.02 0.04 0.06 0.08 0.10
8 I END ROTATION .
•
Fig. A4 TEST RESULTS (T SER~ES)
• • .. .

0.8
Local
--- --------- -- -- --- Buckling

0.7
. TEST A-3
P/Py =0.326
....... -....
-- .....
' .... ,
......
"
"'\ \
\
\
\
\
\
\
\
M \
Mp \
0.4

-,
........

0.3 "- \
TEST A-2
P/Py =0.647..
\
\
\
-I- Theory ------
Experiment 0--0----0

8'*"31
0.2 Strong Axis Bending
L/rx=55
p=o
0.1

o 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

8t END ROTATION

I
Fig. AS TEST RESULTS (A SERIES) VI
N·
 - .53

- --- - - - - Theory _::1:_

0---0----0 Experiment :.z.l:
0.7
4VFI3
Strong Axis Bending
~=O
0.6

TESTA7
0.5 P/Py=O.l6
Mo L/rx=1I2
TEST A5
Mp
P/Py =0.33
0.4 L/rx=1I1 ----------
,
, .

0.3

0.2

TEST A6
"...--,
0.1 ... ,..."" ,\ P/Py =0.502
\ L/rx =112
\
\
\
\

0.01 0.02 0.03 0.04 0.05 0.06 0.07

9, END ROTATION

•
•
Fig. A6TEST RESULTS (A SERIES)
 -54

1.0
Local Buckling
,a:::.Jb-o:=:-o:-:-~::o=..:~~~ - - - - - - - -
0.9 TEST 'A-9
P/~=01l20
"

0.8
--- -- , '",
.....

0.7 TEST A-8 \

P/Py =0.300 \
\
\
\
- -- - ---- Theory
\
\
o 0 0 Experiment \

0.5
I
I
0.4
I
I
I
I ~
~I-·
I
I 8BI3
0.3 I Strong Axis
I Bending
I TEST A-IO
P/Py =0.600 . L/r.x=52 , ~=O

0.2

0.1

• o 0.02 004
8, END
006
ROTATION
008 0.10

Fig. A7 ,TEST RESULTS (A SERIES)'

 205A.35 -55

REF ERE N.C E S

1. Austin, W. J.
STRENGTH AND DESIGN OF METAL BEAM COLUMNS,
ASCE Proceedings, 87 (ST-4), April 1961

2. Driscoll, G. C., Jr.

ROTATION CAPACITY REQUIREMENTS FOR BEAMS AND PORTAL FRAMES,
Ph,D. Dissertation, Lehigh University, 1958

3. Ojalvo, M, and Lu, L.

ANALYSIS OF FRAMES LOADED IN. THE PLASTIC RANGE,
ASCE Proceedings, 87 (EM-4), August 1961

4. Ojalvo, eM. and Levi, V.

COLUMN,DESIGN IN CONTINUOUS STRUCTURES,
Fritz Laboratory Report No, 278.4, July 1961

, 5. Ojalvo, M.
RESTRAINED COLUMNS,
ASCE Proceedings, 86 (EM-5), October 1960

6. Bleich, F.
BUCKLING STRENGTH OF METAL STRUCTURES,
McGraw-Hill, 1952

7. Ojalvo, M. and Fukumoto, Y.

NOMOGRAPHS FOR THE SOLUTION OF BEAM-COLUMN PROBLEMS,
Fritz Laboratory Report No, 278.5, July 1961

8. Ketter, R., Kaminsky, E. and Beedle, L. S.

PLASTIC DEFORMATION OF WIDE-FLANGE BEAM-COLUMNS,
ASCE Transactions, Vol. 120, 1955

9. "
Haaijer, G. and Thurlimann, B,
ON INELASTIC BUCKLING IN STEEL,
ASCE Proceedings, 84 (EM-2), April 1958

10. ASCE
COMMENTARY ON PLASTIC DESIGN IN STEEL,
ASCE Manual No. 41

11. Galambos, T, V.
INELASTIC LATERAL-TORSIONAL BUCKLING OF ECCENTRICALLY
LOADED WF COLUMNS,
Ph.D. Dissertation, Lehigh University, 1959
 -56

REF. ERE N C·E S (continued)

12. CRC
GUIDE TO DESIGN CRITERIA FOR METAL COMPRESSION MEMBERS~
. CRC, 1960

13. Horne, M. R.
THE STANCHION PROBLEM IN CONTlNUOUS STRUCTURES DESIGNED
BY THE PLASTIC THEORY,
B. W. R. A. Report FE 1/42

14. Van Kuren, C. and Galambos, T. V.

BEAM-COLUMN EXPERIMENTS,
Fritz Laboratory Report No. 205A.30, July 1961

15. Lee, G. C. and Galambos, T. V.

POST-BUCKLING STRENGTH OF WIDE-FLANGE BEAMS,
Proc. ASCE, Vol. 88 (EM1) , February 1962, p. 59
16, Kusuda, T" Sarubbi, R. and Thur1imann, B.
THE SPACING OF LATERAL BRACING IN PLASTIC DESIGN,
Fritz Laboratory Report No. 205E.11
.,
DATE DUE

·f

\.x.\\\ \iH u1~1 VfRSl/fJ \,

DEC 9 1984

.
lL...i8RA~/
~~