WELDED

COI~T

Il\TUOUS FRAlviEiS & THEIR C01:IPONENTS

by

Robert L. Ketter & Lynn So Beedle

This work has been carried out as
part of an investigation sponsor$d . j bi11'cly by Jche Wel.ding

Iteset\1~cJ1

Council, and the Navy Department

with funds furnished. by the following:

American Institute of Steel Construction

Anlerican Irorl and Steel I11stit1.l.te
Colunm Research COUI1Cil (Advisory)
Irlstittlte of Research, Lehigh U'niversity

Office of Naval Research (Contract No. 39303)

Bureau of Ships
Bureau of Ya.r-ds and Docks

Fritz Engineering Laboratory

Departmerlt of Civil Engineering e.11d Ivlechanics
Lehigh'University
Bethlehem, Pennsylvania
~Tovember,

1952

Fritz Laboratory Report No. 205A.11

205A.ll

aE...~~

II o

III.
IV ()

'V

VI

VII.
Vlll

tt

rrJTRODUCT ION

NOTES ON THE ANALYSIS

~ffiASUREMENT

OF ROTATION

TESTS RESULTS AND DISCUSSION

a.

Moment-Rotation Curves

b~

Rotation Capacity

11

CONCLTJSIOJ\TS

16

ACI\l\f O\,\JLI~DGI~1Vm~NTS

18

NOi8~NCLATURE

18

RE~FI~REl\T

19

CES

!1PPEMDJ~

SAJYIPLE

NU~1ERICAL

DEF{IVATIO:f:\T OF 1v1-9 CURVIG

20

205'A.ll
If

INTRODUCTIOl\T

In advance of a more detailed report being prepared on the

sub~

ject of the load-deflec.tion and load-rotation behavior of WF' steel

columns, it is considered that current interest warrants preparation

of a short report discussing the

moment~rotation

problem.

Not only

is this needed to further describe the basic behavior of caltoons loadr

ed with combined bending and thrust, but also to evaluate energy

ab~

sorption of complete frames and to develop tentative proportions for

colunms intended f-or use in plas'tic design.

Performance of structures subjected to loads sUfficient to cal1se

plastic action is directly related to the ability of certain members

or connections to develop plastic hinges.

Moreover, these members

must allow rotation through cOTIlparatively

la~rge

development of plastic hinges elsewhere o

The pnecise requirement for

angles' to en'sure . the

the tlrotation capacitytr of a particular structural cOYl1ponent is de...

pendent on such factors as (a) type of structure (degree of restraint)9

(b) type and

]~ocation

of loads and (0) lengt11-depth ratio of members

An investigation is needed to establish a criterion of required

tatio11 capacity talcing into account the apove mentioned factors

such stUdy:(2) made for a contiUQous beam subjected to uniform

ro~

0118

load~

j,ng, suggests that a member nlust be able to sustain a rotation of

eight times the predicted elastic value to ensure the necessary

de(~

velopment of plas'tic hinges elsewhel'e in the strt1.cture.

This report will be primarily a presetitation of experimental

re~

suIts of fl1J.l scaJ.. e column tests in which as .... delivered members of

various lengths have been tested under different combinations of end

bending moment and thrust.

Four conditions of loading have been

ied and are shown diagralnmatically in Fig. 1.

stud~

..2 ...

~,)Mo

~_.~.-

i
I

II Hd"
.

Fig, 1

Each of these conditions will hereafter be referred to by its letter

designation.

For example:

condition t.tb tt would mean that the merflber

is held fixed (not allowed to rota te) a.t one end vlhile the other end

is sUbjected to an axial thrust, P, and a bending moment, Mo Tests have been carried out on two sizes of WF sections (8WF31
and 4WF13)

t)

These members viere tested in lengths of 8ft., 12ft~. and

1.6 ft. giving a ran.ge of L/r from 21 to llJL.

Table 1 surllmarizes the test condi tiollS f'or each specimen.

It

also serves to catalog four u'studies U that wiJ.. l be plon.tion,ed later"

For further reference, Table 2 shows on

~-Mo

interaction curves the

load condition of,each test, test number, section, slenderness ratio

and rolling number.
II.

l'JOTES
ON TIlE
...........,.......
*

Al~ALYS

IS
.....

The momenc .... rotation (M... g) relationship depends on -the M....

curv~e~

since it is the integrated effect of the latter which determines the

load-deformation conditions.
is the determination of

M-~

Therefore, of fundamental importance

curves above the elastic limit o

Such

Inoment... curvatul~e relationships mtlst include tIle variable P, and a sOllJ"_:

lution to this problem has been presented in a previous Lehigh

port(6).

rG~

To illustrate the effect of axial thrust, one figure from

that report is reproduced as Fig. 2.

205A.ll

p
It

==

o.

2Py-

-----

1.0

---

M
My

___

O. 4- fit

~_---------Jy--

,. =

0.6"

------~-~y

__.__ ,..~.~_
. ._.... ~.~ t B.t i en
~

" = O. 8 ~y

aXIs

8 VvF 31

t:..--...:..----:;......~ ....---:.-...;....,.t_ '---t- ....._L--L..---l.--'----L.--I-"--""""'""-.....-.J.

0.5

1.0

Another varia-ble which lna JT have a pronounced influence on the

moment-curvature relation is the presence of coolin.g residual

stress~

This variable is to be discussed in a future report; however, one

figure has been included to indicate the degree of variation due to
such trlocked-in tt stresses, Fig. 3 indicating the theoretical rela...
tionship.

Curront work suggests that the effect of residual stress

becomes more pronounced as the ratio of axial thrust to Py is increased. However, axial load has been relatively low in a majority
of the tests carried

011t

thus far in the experimental investigation,

a condition corresponding to that preseIlt in portal frame structures

(p (0.15 Py ).

For the 8WF31 members tested, Fig. 3 can then be con-

sidered as representative of the influence of this variable.

Since

-the magnitude and distribution of rosidual stres;ses in rolled sec-

tions is undoubtedly a function of type and size of the section, such

varation may not be representative of all sections.

(Fig. 3 is the

result of an analytical study based on measured residual stress distribution values.)

\""--( P

= 0)

\
~..".--

1.0

~~

/\~
I

-Neglecting Residual stress IP

Including

Re.s j du a 1St res s

= O.~2P.)

~.....

A'~:;A

Assumed

Residual

5tr~ss Pattern

(8 WF 31)

0.,

1.0

':y-

Fig. 3

Given the particular M.. 0

reJ~ation,

obtained by numerical integration.

Newmark's method(?) is adapted

to a solution of problems of this type.

pendix A and in fig. 23

moment-rotation curves can be

An example is shown in

Ap~

the theoretical curve (including the in-

fluence of residual stress) is compared with experimental results.

The experimental set-up, techniques of experimentation, etc.,

have been described in a previous report(l).

In all tests, end ro-

tation measurements have been made with a level bar, illustrated in

Fig. 4.

-5-

bubble

Fig.

This apparatus is connected to the base plate of the test specim~

as shown in Fig. 4(c).

It is possibJ.. e with differen'c brackets to

place level bars along the length of the member&

Good reproducib-

ility has been obtained using this instrument,

L_JJ.Qmg!TI=]..9J1~~t1-9!l_.Qurv~s

The influence of four variables are presented in this report.

These are:

In anyone case, the other three variables are maintained constant o

Tl1.is j.s summarized in the folJ.. o,'!ing table which also shows the

numbers applicable to each study.

,fig',r[':"~"

STUDY

LOADING

SLENDERNESS

AXI AL

CONDITION

RATiO

THRUST

The behavior of each test specinlen 1.3 shown in Figs.

5-15.

In

each of these curves the end bending moment, Mo , has been plotted
against the end rotation, QA. The ordinate has been made nondimensional by dividing by

MY,

the theoretical yield moment for the

rolled shape in the absence of axial thrust.

For each individual

curve, the computed yield and plastic monlents are shovln, due rega..rd

being taken of the load condition and axial thrust.

are detel-'mined from either reference lr or

(These values

5.)

Exeellent agreement between experimental and theoretical result2

was consistently observed in the elastic range 7 although the

theo~

retical curves are not shown. (Fig. 23 is an example.)

By way of explaining the figures, the particular variable is

indicated in parenthesis after the test number in each case.
in Study 1 (influence of loading condition),
and load condition

T~12

Thus

(c) means Test 12

!tOfte

It is apparent, of course, that in the early tests the defor

nJ

mation was not continued through a sufficient _angle change to give

~7~

205A.ll
a definite indication of the collapse pattern.
every effort is made

~to

In current tests,

determine the uunloading tt curve.

l_TlQ-l.Llillg.-ginE Conq.i:tJon

As indicated in Table 2, three series of tests have been compJ.eted or are

under,~ay

vlhich indicate the influence of this val'liable,)

In each seI'ies the ratio of axial thrust was held constant at

p

= O.12~y.
~_ 5:_._U31l.F31--,-1:L~ . .; 5.5...,..

PLRy:..:: 0.12.1 Elastically,

the stiffness increased in the order T-12,

corresporlding loading conditions being

T~13, T~4

and

expected?

T~14,

eft, ltd", "btl and "au.

the
Tests

T.lt- and T-14 were not carried to their ultinlate load, although all

current and future te'sts are bej.. ng contiIluecl to cOlnplete failure

T~12

(t

carried an appreciably smaller end moment than did tests using

any of the other three loading conditions and the failure involved
lateral-torsional buckling.

In ectch of 'the other three cases local

buckling was observed.

increased~

in the order

tr c tt ,

tt'd tt a11d Bb u and there

been carrj4ed out for this slenderness ratio.

buckling is more pronounced o

a correspond--

No condition "au tests have

ing improvement in rotation capacity.

were more slender than those of Fig.

\vE:lS

5,

Since these coltmms

the tendency toward lateral

ThiS supposition is supported since

both T.. 26 (ltett) and T.. 23 (ttd'~) failed due to lateral buckling.
T~24 CUbit ) had not reached the maximum loa~d
however~

(~,:)111mn
~f

Test

when the test was stoppec j

at that time there was noticeable lateral twisting of the

as well as a local wrinlcling of the compression flange.

Bot11

these conditions could be observed by eye; however, it was im-

possible to determine which developed first.

..8..

205A .11
F.i~L_Qt![[1.h.jJLr ~=..54J'.L4-=-~=2Q

(lle lf ) whieh failed

due to lateral buckling, had less rigidity and failed at a lower

load than T-17 (nb H

,whicll. failed d1.le to local instability.

lliI!l].j.~_:h~:ID~h.:hs
__.. ttl<;b:.t

5,

Figs.

6 and 7 indicate

that stiffness of a

dependent on the condition of loading.

beam~coluw1

For those conditions tested,

stiffn.ess increased in t11e order tie", ud u , ttb U and Ha tt

being the weakest.

is

the first

Furthermore, the type of failure seems to be

markedly influenced by loading condition.

In each case the mombor

with tIle less stiffness, (conditj.. on "c u loading), faj.led due to lateral~torsional

buckling, while the stiffer members showed a tendency

toward local. instability.

f33~VJ)X_g:~

__SleI)fternJ?'Lltill_io

Da.ta may be drawn from three series of 'tests to study this

iable.
It

var~

Two are condition Itc" loadirlgs, wllile the third is condition

b"
Ej.1k.-1-.lYlE31-,-_901lg..i.tj.9.LL_~_9~y_1:. 0.12) Failure in each

case was due to lateral-torsional buckling.

As

expec~ed,

increased with decreasing length: (compare T-12, L/r : 55;

1,/r

= 41;

and

T-19,

greatest strength.

L/r

= 27)~

stiffness
T~16,

The shortest menlber (T~19) had

In comparing rotation capacity, it would seem

that the longer members carried their load through a larger rotation?
however, such a comparison is misleading since the slender members
allowed greater rotations in the elastic range.
}~&)--3:_..Q~~.Q.illl<i.itionnb~y:Z-Q.,1.?) Relative stiffness

increased in the order

(L/r =

55).

T~9

(L/r

= Ill),

T-24 (L/r

=84)

and T-17

T..17 failed due to local buclcling; T-21.r was still

increasing in load when the test was stopped (rotation was over 8
times the elastic rotation) but there was noticeable lateral-torsional
deformation as well as a marked local wrinkling D
a pronounced

lateral~torsional movement o

T~9

failed due to

Since T-24 was from a

dif~

ferent rolling than T9 and T-17, it is possible that a difference

in magnitude and distrlbution of residual stresses accounts for the
different behavior.

Measurements of residual stresses in each of

these members are planned for the near future.

-=

,J:.L'Ey O.B). The slenderness

ratios in these two tests are 55 and 84. As would be expected, the
Fig. 1 0 C

Q+WF13, Con..ditio~c"

shorter, stiffer member

due to

~ad

lateral~torsional

SYmm~~izing

the greater strength.

Both members faileq

buckling.

tho trends observed in study 2, stiffness in the inelastic

as well as the elastic range is influenced by the slenderness

ratio~

As would be expected, the more stocky members were the more rigid.
One further observation that can be'made based on this study is, that
members tested under condi tj.on

If

en loading failed due to lateral-

torsional buckling regardless of slenderness ratio; whereas, the type

of faj.ll11'e for condition Itb" loading seerned to depend on the slender?l'-

ness of the member.

'~1W~-3L.~~. 111e_ I~J:l111~Jl.QLQf_ Axj_aU!rr_\l~

F~ig !l.wl1.t..._-lIDiF3~lli:li.t1on_.~lb:r4-_!i1'-=....2.2l

undoubtly

wou~d

Since T-lt (p :: Oc1 2 Py .)

have failed due to local instability had the test

been carried far enough, and since T-3 (p

lateral~,.torsionaJ
. bucl~ling ~

failed due to

axial thrust plays an important role i,n

Q8termining the type of failure B

. formation on rotation

= O.lt9 Py )

capacity~

These early tests give little

in~

205A.ll
Fi~.K~.1..b.JlQDli.;it1-.Qp._.~tp~,'.,L/r,:: 111)

expected to carry less load than T-9 (p

served condition.

~':

T-7 (p

O.lOP y ).

= O.27Py )

\"a8

Such was the ob-

Failure in both cases was due to lateral-torsional

buclcling.

= 551

!11.f~_.llL~-i4W"."E13~Co1].Q1.tiQ;tL.~I

T-17 (p
bilj.ty.

= O.12Py )

and T-2l (p

= 0450Py)'

failed

Both of these members,

~e

to local insta-

The "hump" in these curves seems to be characteristic of

those conditions in which failure was of the local instability type

(se(.~

also Fig. 11, T-4).

Similar behavior has been observed in tho

continuous beam tests and has been discussed in Progress Report 5(3)0
U1 tima te

moment~

carrying capacity of a beam--column is directly

related to the amount of axial thrust applied to the member. The same

correlation is not observed when considering the ability of each

member to plastically deform thru relatively large end angle

changes~

Based on these few curves it would seem that increased axial thrust
does not have the pronounced detrimental effect on rotation capacity
that it has on moment carrying capacity.

STUDY 4:
~~,

..

""~.~~&&;"~"'-~""_

The Influence of Size of

;.:c::.~-~~~-.,~-.,-~---"-......".r.._~~~~

Cross~Section

.. _ _ ~""""""'''-~~~,j..B;;;r.~~~~~.:..JIC.P~~:r.:~~
.. ~ ~ ' r " ' "

A question frequently asked is, "Will the results of these tests

describe tho behavior of geometrically similar sections but of

ferent size from those

tested~tt

Therefore, correlation between the

two sections tested under identical conditions of loading and

derness ratio is next

dif~

slenf~

examined~

Ej...g.,.J~..L....J.gJW.
..g j. t ~,-Id.r-.,:: .i2~n.y-=-.9.,L12.2.

tll.o same general trends were

observed~

in ea.ch test.

indi cat es that

The 8WF31 secticr.i.

yioldcd at a proportionately lower load, and this indicates a

latively greater magnitude of residual stress.

re~

Both tests indicated

....11...

205A.ll

the tthumptr or sUbsequent incrGase in moment strength ctfter a certain

amount of plastic
130

straining~

Tl1is was mentioned in discussing Fig.

Each of the members failed due to l08al instability.

FJ~;.<,J~2.;_..i.Q..on<lL~J.91L H cl~..J-lr

di tion

11

= 52',. )~Ly_';=--Q'p.12).

For the con-

en type of loading, failures were of -tile la teral-torsiol1al

buckling type.

Here again the

8~inch

section yielded first, but

there was no sUbsequent recovery of moment strength and consequently

the 8WF31 member carried less ultimate mOInent than -the comparison

test.
The above comparisons indicate that the influence of size of
member is of little importance in the elastic' range.

Further, these

.tests suggest that once yielding occurs, tl1en two members of

dif~

ferent size will not behave the same unless the loading condition is
favorable or unless the cooling residual stresses are of nearly the
same magnitude.
~.ROTATION

CAPAClI1

Mention was made in the Introduction of the importance of

ro~

tation capacity, the ability of a member to deform thru a relatively

large angle change while sustaj_ning a mome,nt approximately equal to

the plastic hinge momont.

Study of figures

5-15

reveals that few

of the columns exhibit a behavior that approaches the idealized

U

plastic hinge U case.

(:B"'or tIle idealized case rotation would contintle.

indefinitely at Mpc ).
The test curves previously examined fall into four general types
that are indicated diagramatically in Fig. 16 and in Table 30
A might be considered as the basic type.
(T-14 is an example)

ill

CurVG

Gurve B applios to a test

vThich t~he loading was stopped before the

maximun

It

collapse" mon1ent vias reached.

In ctlrve C the member con-

tinued to carry increased moment through a large angle change.

the test

~las

stopped pl"ior to corl1plete "unloading".

A member re ...

suIting in curve D actually has no rotation capacity since the

imum momont does not

Again

max~

Mpc
Rotation capacity, R, will be defined as the ratio of observed

end angle change,

yield,

appr~aCh

eA'

to the theoretical end rotation at initial

eye, or
R

Fig. 16 indicates that

= eA/eyc

non~dimensionaliz~ng the

coordinates provides

an abscissa that is a dirGct measure of rotation capacity.

Because of the nature of the experimental curves, rotation
aci ties will be compaI'od according to soveral

It

cap~

cases tl or cri teria o

The trends will be examined as a function of slenderness ratio, ratio

of axial load 9 and loading condition.

TYPES 'OF CURVES

1.0

PROPERTIES OF CURVES
1.0

e Ie

I( lye

eIfIe yc
~

Fig~

16

205A.ll

The nomenclature used in this particular section is as follows:

MO

"

= Mc)men.t

appli(~d

at the and c:f a boctmr.-.-column

:: Predict8cl e.lastic Ij.nli t

mOln8~1.~;~

tn~J.llcling

the effect of

r~!~j.a.l t~J::rllst, If

= Observed

=PredictGd
=beJ~:rC)L1SO

end rotation corresponding to Moo

end rotation corresponding to Myc
j~n

moment bela"'T that predicted

linlit rotation o

at

the elastic

- In.crease of Inaximum mome11t over predicted elElstic limit

rotation.

= Increase

in rota'tion over preo_ic'ted. elastic limit ro'"

tation at the elastic limit moment.

= Rotation
~

Rotation

Case

z1

1l

capacity at the maximum observed moment.

c~pacity

at the elastic limit moment on the

unloadi.Jlg curve.

Incr8ment of rotation capacity at predicted

yield moment CaRl) and reduction of moment at the

elastic limit rotation

(~Ml).

Such a comparison is primarily

an examination of the tlinitial

yield" charactoristics of the member.

17 and 18.

The results are shown in

Figls~

The variable here is slenderness ratio and testresults,

in general, indicate that the shorter the member the greater the
departure from predicted behavior o

The maximum reduction in moment

is about 10% whereas the increase in rotation is more than

50%.

is considered that the discrepancies should be the greatost for

CU .loading and there

is some agreerIlcnt here since the

ditiorl

tt

d_.,1.tj OJ:J

Hb tt tests sho'\V' less redllction.

It
con~

con~

-14..

205A .11

pase 2=

Rotation capacity at the maXilTI11m moment

(R2 ) and increase in monlent above the elastic limit

value (AM2 ) ~
This case, which

descl~ibes

the ultimate strength of the various

beam" columns , is shown in figures 19 and 20.

plotted against slenderness ratio.

slenderness ratio.

In Fig. 196M2 is

Fig. 20 is a plot of R versus

2

It is interesting to note that'those tests ex-

hibiting greatest strength and rotation capacity (as defined by R2)

failed by local instability; whereas, lateral-torsional buckling resulted in weaker, less stiff, members,

This is to be expected since

lateral buckling involves general yielding over a comparatively large

area whereas local instability is usually confined to a smaller

region that is often conducive to

strain~hardening.

As further work

is carried out, it may become possible to define the condition above

which local buckling will occur.
In no case does the rotation capacity approach the value of
eigh'c which was mentiOl1.ed in the Introdu.ction (page 1).

In fact, a

rotation capacity of about two seems to be about all that most of

these columns can achieve at their maximum moment.

The influence of

loading condition is pronounced and Fig, 20 suggests that condition

ub" and ttd" tests w01.11d provide nlore satisfactory performance in

columns of L/r less than

50.

Fig. 21 was prepared to S110W the influence of axial thrust on

rotation capacity, and the latter function is plotted against the
ratio P/Py.
condj.~tion

Comparison was only possible in the case of loading

ttb u tests.

Lat(~ral~torsional buckling \'las

observed

il1

the case of T9, T7, and T10 and R2 decreases gradually as the axial

-15...

205A.ll

load increases.

However, the trend indicated by tests

T~17

T~21

and

suggests that when the mode of failure changes to local buckling,

then the rotation capacity is actually improved by an increase of

axial thrust.

This trend requires

Ccg>JL_~ :

confirmation~

Rotation capaci"ty, R3 , at t11e elastic limit

(unloading curve).
The graph of Fig o 22 shows trends similiar to those observed
in Fig. 20.

The rotation capacity measured at the elastic limit

(on the unloading curve) is plotted against slenderness ratio.

An

increase in slonderness ratio is accompanied by a docrease in the

rotation capacity; however, the influence of this variable appears
to be less pronotmced than that of loading condition.

The additional

tests planned for loading conditions "d u , ttb tl and Ita" Shotlld provic1e

information covoring a more complete range of slenderness ratio.

Two things are yet needed to establish the sUitability, of plasti0
design techniques:

(a) how much rotation capacity is required in order that all

plastic hinges be developed in building frames, and
(b) is the trend. suggested by loading condition

n cit

tests

borne out for the other conditions of loading.

Four of the seven tests whose reslll ts are plotted in Fig. 22

hav~e

rotation capacity of about 2, the slendernGss ratio ranging from

to

40

84. Adlnittedly condition "c" is the worst caso, but it is

heartening that the few results available for the other loading
ditions show improved behavior in this respect.

If the roquired

rotation capacity for typical structl1res approaches the value of

con~~

-16..

205A.ll

eight, mentioned earlier, then these tests indicate that rotation

capacity may be more serious as a limitation to plastic analysis

A rotation capacity of 8.0

and design than previously anticipated.

is

kno,~

to be a severe requirement since it was developed for the

case of fixed-ended beams.

In actual framos, the columns themsolvos

supply less than complete restraint, and i.n the plastic range they
will also contribute to inelastic rotation.
capacity of

T~lS

4.0 might not be unrealistic for

It

a required rotation

cOlnplete restraint tl

and this value would be further reduced since every column allows

some flexibility.
V

(I

CON Q L-1L S ION S

Based. on the few tosts thus far carI'iecl

Ollt,

the following ab...

servations are made:

1.

The stiffness of a

of loading.

beam~column

is dependent on the condition

In increasing order of stiffness the loading

conditions investigated are "e", tid", Itb" and

ttau.

(Figts o

5.... 7)
2.

Failures, have been of two types:

lateral.torsional buckling

and local wrinkling of the rlange elements.

tested under condition

"e"

For all members

loading, failure was of the

lateral--torsional. type rega1'dles s of slender11ess; hO'\'Jever,

for the other loading conditions there is a tendency for

the stiff, short members to buckle locally. (Fig.'s

8~lO

and Table 3)

3.

Axial thrust deC1 eases the ultimate moment carryi.ng capac::.ty

1

of a beam-column but does not" necessarily decrease its ro(.I"'

tation capacity.

Further tests are needed to establish

205A,11

definite trends. (Fig's.

11~13)

4. The influence of size of member is of little importance

in the elastic range; however, when yielding occurs, two
members of different size will not behave the same unless
the loading condition is favorable or unless the cooling
residual stresses are of nearly thesame magnitude. (Fig's.

5.

There seems to be a limit

or

ultimate carrying capacity above

which local buckling is the type of failure and below which

lateral~tqrsional

instability takes place.

Further ana-

lytical work is needed to establish this transition zone.

(Fig. 19 and Table 3)

6.

Increased slenderness ratio in general decreases both the

moment value and rotation capacity.

However, in many cases

the variable of loading conditions has a more pronounced

effect, (Figts. 17-22)

7.

Only a few of the members had a rotation capacity greate'r

/'.

than

4.0, and a

conunon value ,vas about 2.0.

If the re ..

qUired rotation capacity for typical structures approaches

the value 8.0 (the ttideal tt requirement for a fixod...ended

beam, loaded at the

third~points)

then rotation capacity

may be a more important factor thanprevio1.1s1y anticipated.

(Fig. 22 and Table 3)0

-18-

YL__~_AS;I\N01J~~~MIG1I1S
TIle work is being carl~ied

Ollt

unc1er the direction of the Lehigh

Project Subconnnittee of the Strllctural -Steal Committee-, Welding

Research Council, at Fritz Engineering Laboratory of which Professor

William J o Eney is Director.

Young's modulus of elasticity

Total length of a beam-column

Moment
Moment applied at the end of a beam-column
Moment at which yield point is reached in flexure

Predicted elastic limit moment including the effect

of axial thru.. st
Plastic hinge moment, modified to include the effect
of axial compression
Decrease in moment below that predicted at the elastic
limit rotation
Increase of luaxj..lTIUln 1110ment over predic'ted elastic lirnit
rota~tion

Concentric axial load

Axial load corresponding to yield point stress across
entire section
r

Radius of gyration
Increase in rotation over predicted elastic limit ro~
tation at the elastic limit moment
Rotation capacity at the maximum observed moment
Rotation capacity at the elastic limit moment on the
u1110ading curVG

11

e'yc

cry

Observed

811(:1

rotation corresponding IVlo

Prodj.ctocl cnd rotation, corresponding to l/Iyc

Lower yield point stress

Curvature at a section (the reciprocal of the radius

of curvature)

-19-

205A o ll
Y.~~Jl_~~~~~FE11EN~

Reports in the Lehigh Series:,

1.

!'I..<h=k UTests of Colunll1s under Cornbined

Thurst a11d MOlnent tt , by Lynn S. Beedle 9 Joseph A. Ready,
and Bruce Go Johnston, Proceedings of the Society for

pr..Q.g~L,B&ILort

Experimental Stress Analysis

2.

4f)

tt

6.

for 1tJelded

Continuous Portal Frames-Test Results and Requirements

for Connections", by A.A.Topractsoglou, Lynn S. Beedle
and Bruce Ge Joln1ston, Welding Journal Research Sup.,
0

Pt9.E!..:ss~BQ.nos~-..~o .~..

2,

ttResidual S,tress alld the Yield

strength of Steel Beams tr , by Ching Huan Yang, Lynn Se

Beedle and Bruce G. Johnston, Welding Journal Research
Sup. 9 April 1952.
:er.QEr.-.q,_ RG.P_Q~r~--liQ.,, __~_~!).. ttColU~lnn strongth Under Combined

Bending and Thrust tf

,
by Robort L. Ketter, Lynn S. Beedle
and Bruce Go Johns'ton, vJeld.:ing Jor!rnal Research Sup.,

DeC81nbor

Vol. VIII, No.1, -1950

P~2E,r,~LJ3.53J)...9-r
__LEo~.f=<~.:..~,Par.LI,t!Connections

Jll1y 1951

3 c.

1952.

L, It IntoIl(1ction Curves f'or Columns tt by

Rober~t L. I(etter, and~ Lynn S. Be,':)dle , Fritz Laboratory
Report No. 205A.4, April 1951~
Pr-9Eress

cB~t

PrQRr__ess~_B1I!...9rt.".-l:,ttThe I\1oment--Cur\lature Relation For WF

Colulnns lt by Robert L. Ketter and LJTnn So Boedle, Fritz

Laboratory Report No. 205A.IO, September 1952.

Additional References:

7.

Ne\nnarlc, N. M., ttl\T1,11n(1rical Procedure for COll1puti11g De-

flections? Moments? and Buckling LoadS",. Transactions

of the A.S~C,Ea, Vole 108, 1943, pge llbl--1234 o

APPENDI~

A:

Numerical Detormination of Moment-Rotation Curves

OUTLINE of' the general method of solution (using the numerical procoSf
of Ne~nark(7) is as follows:
1.

Assume a deflection curve.

2.

Determine the moment acting at equally spaced points

along the member.

3.

From the M~~ diagram (for example Fig. 3, page 4) read

value of corresponding to the moment values at each
of these points along the column.

4.

Numorically integrate these

flection curve.

5.

If this new deflection curve does not check that originally assumed go thru steps 2,3 and 4 against using

0 values, and obtain a de-

the ne\v defJ.. ection curve.

6.

When t118 defl.ection curve is sUfficientJ.y accura'cecorrect slope values to the end of the collU1In (using

conventional finite difforence equations).

An example using this procedure follows.

Determine the value of end rotation for the 8WF31, 12 ft. column
loaded as shown in the following sketch.

O.12P, -= 41.1-.9 k

No := l050"J:~

-21--

]MUltiPlication

Factor

._------_.j
-----~--,._.-'.

Total Moment

---r--:;JIt"--.----- --..

6corr

= -(7L +
24

6o -cPR)

A-(,1.771:+ 1.740 - 0.317)

24

Iherefore,
[1.119

+ o. I33] A. ITy IE

0.0302

inches/inch

-~--- ~-.,----

TEST
NUMBER

LO'ADI NG
COND IT f ON

. \ - - ... ~--_._.~.~ f - - - ' ,_......."__ '

p
p

SECTION

L
-r

.~~.o._

_ _ -_ _- . . . - _ _

~_...... _ . _ _

STUDY I

STUDY 2

STUDY 3

STUDY 4

Loading
Conditien

Slenderness

Ax i a'

Size "f

Ratl.

Th ru s t

Member

_ _... _

_ _

~-_:-...-."-_~-

i=~~=-:~~=~~~=---=~~it=~l~;---=~~t~~!-~~~==.~~~~~=.-~.~_
=
~t:-:-~:
.
t
~-_-~it-~_~;~]
I
I
.
T-7

'---'->~----.. --~-

,T-9
'I--"-~'_'_'_

b
4WFI3
. --,-. _ -.-_. t-._ ..--~ ,~._--,

--'-','- b

4WFI3

c _ . _.'. _ _ . ,. . . .

T-IO

T-13

T-14

r.

-.---.----~-

~_~

" . __

T-17

T-19
T-20

.. ...
~

. . . . _ ....... .---....

~_

~.

8WF31

.,.._._ ..........

.r.~~

__..

..... _-~ .... ,~r _ _" r---

..... _

.....

...

~~'"

8VvF31

~-

......

55
..... _ . .

--...-..~_

.... _ .... ' . . - .

0.12

~---~.-_.--t--~

. , -

~...

Fig. 12

__._,.,.--...---.__.'--- .... _~--~_ ... , v-.__"'_"_

'''~'''''_'~_''_.''_'b.

12

Fig. 8

.. _.. -.....

~_.

-_.-"

..' _._ .......

,.~_.'

~ ~-

.__ ...

-~- ~..-P._-

~--

4WFI3

... ~..II".._

+,.~

. _ ......._.,._ __

.~.

__ .

J~

..._ . _ .

"",.'_'_'_._'.'

_.

...__.....,

_
~
_
~
.
_

,.--__.. _I

4WFI3

--

Ir----,---+------.-- .-------.--.--.-.. . _,,

d
T-23
4WFI3
o. 12
84
- . - . _ . - -__..__..

T-24

T-26

_.~._f-

_________ ._,,-~--~- . ,---.,.. -.-

..

..... __

Fig. 7
......

r ...... 4~

.. _........

~-

Fig. 7
I--......... ----.~

.-._........_-..

_-..". ..... _ -

._~.~

4WFI3

0.12

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

0.12

I~====!:t====~=========..~~

84-

_._--.. . _84

__..__.. . . .
~

. _

..

. . _ . _ ,

Fig.

.......... __ ................. _~ ...... _

Fig~

13

.-...-.....~----.-

...... - ...............

14

_ _ . . . . . ""'-...-:.- ..'1.1Io

__...4 _ _

............

_.1

"

---

..

Fig. 10

io------------- -.-.--.--+--..- - - -

Fig, 13
1-_~

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

Fig.

15

" ' " - - - , - -...- - r - - -__ ~-'__,_o,_...._

Fig. 6

">_- _ _ ......... _ _ ~ . , _ ~ " ~ . , -

4WFf3

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

Fig. 8
Fig.

f--- ..- - - - . - ..- ..- -

,~~

._~

_.--_._~-----~*---_.-

55

0.50

I
'-"~

Fig. 8

55

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

_ _

Fig. 5
1-- - - - - . -..- - -----

27

0.12

->.---.-~-.

......_-_._-_.-.--~~ ..' i

- .' . . ."-_----,

~ __........_r~.~.----... .._ -" .---.~.~-----~ ....-.- -.----.-~--~---,-..--_..~ .1---....- - - - - - - - -........- - . - --~,_.. - - - - . _ - - - - - - - - - - - '

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

Fig. 5

u_ _ ..... _ _ . _ . ,_ _

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

0.12
- . . ..... ' ....

..- ,-_....

4r

__._.._. .

4WF13

_ _ ' ..1 ..... _ " # o _ _ ~ _

Fig. 5

._.'-"'-_...-.--..,

-'--.--'-.-.

Fig. 12

.' -~-- - . . - --""--------- --~-, -.--,-..... '.-'- _

55

0.12
~'"._--_

,~

Fig.

55

. _.. _ _ .......' . _ , _ ,

_ 4 " , . , -

II

55

0.12

--~----- :----.-~_..

O. 12

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

-.-.---i----.----.... . . . .- - -._-..
T-2f

.... ~

,_ _ " . _ . _

8WF31

--

Fig. 9

~-_

0.12

~,_

8 _

~~

........ ... ~~~.......".A

. - . _....-

-.-_.__'

_,......

-<'7_~

... _-...--....

__ _ _

1---._-_

T-16

._.... ~._ .. -

8WF31
~~

III

--- . - . __._. ,._-,_.--.. '_'__

0.51

8WF31

d
. _ ...... _ _

.-._._~

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

~_.

0.10

4WFI3

l-~"'--'---"-'~'- f--~----------""-"-

-~.--

I"

---~-- --._.- .- _~- .. -.-..- - - - . . . - .-.--~.- - -

-f---. --- .. ---.----.>-*-,-~-

T-12

a.V

-, ..-

Fig. 6

.~._~_._--

Fig. 6

_. _ _

...-...._

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

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

""'~--"'-'-'.

Fig. 9

. . . . --------.. . . .-.. . .-----~to_,-- . ~------I

Fig. 10

"":""_~~. ~~~~.-~~~

..

. . .. .

~~. ~_~ ~._ ~._,~~~~~~=:===.J.

TAB L E

..1.

205A.ll

-~-=
r 27
T -l9 (8-

~::: 4 ~
T-16 (5- 2)

~''',~.\,

rl

'01

~
OJ

T-3(A

1=
r 55
T-14( A-3)

T-It( A- 2)

.~= 55

T-12( A-l)

~,

T-l)( A- 2)

~~

"~

T-17(A-2)

t Y .)

rl
..

~
~

J::.r =84

'.
T-24-(B-l)

r -26 (B-1)

T-23(e-l)

205A.ll
.1h~~Q--U~~h ATI 0 tL:::2:K.:::._.-~-,X~[.) -~_~P~J_l1_~_N ~~~~~~_.Ii_~7~. U-f2
t---i---~-' ..----~2_--1.0 1
(

n.

_"",_?-l

~~t

-t7.t
; ]

6t'l~" 1

__ .. +__ .__-R~_...__ ... ------::-i

I

M::;

yc

t1'\nax.

PR-geERIJ~~~: OF <;UJ:L~~~
1.0

BA
......--~"III

'....

eye

T-21
r------

!I . T--23
~_

",._~_

b
~-----,.-._~-

._~

! T-24

._. 4WF130.12
.

84

.~

4WFf3

b
_h

Locat

Buckl i'ng

0.181

).10

0.04-4

4.93

5.35

_-~

_----...-.~

b-~----- --~--L~st

55

. d ._._

i-- -T=26-j- _
....--;--*

0.50
.._.-.4WFI3
. . _.. .----

:::

O. 12

_~

------_._---.A.

8~1-

Ben.ing+Twist
~-_._.. _

_1' .. __ -

8ttM"'tiJng,

_ - __ ~,

Twist

~4WFIL3'-O~I;t'-'0:" ~---A--:t:~d ;~~~C~;-;:~~

.:.-.-_

recorded value -

--.J=c=-

=---

~~-

-1'

. .-.~~~ '~~o~- ~::;. ~~~j

-.-:--..=:.--=:-:.:--------

'(Test stopped prior,to unload'ing)

o
0
0.3;8 2.07
.-_.- .------- --._- ---*-_ ...

205A,.11

8 WF 3.1
L

= O. 12Py

p
I

0.02

= 55

-,-,I_ _~_-l-_-L

O.O~

0.06

0.03

0A

0.05

,.!....._ _--t- _ _--L

0.07

0.08

0.09

-,L-_
O.JO

(inche~/inch)

Fig. 5

1.0

4 WF J3

?L --- 8-4
P

= I.

0.10

eA (Incheslinch)
Fig. 6

12Py

0.15

0.20

205A.l1

1.0

~
M

--:I-

0.5

4- WF
=

1-r

~
55

p = O. 12Py

0.01

0.02

0.03

O.O~

0.0,

eA ( 1n ch e s J inc h )
Fig. 7

0.06

0.07

0.08

1.0

8 WF 31
p O. 12PY

0.01

0.08

L-...r-0.09

1.0

--=1f--. P ~- o.WF I 2P13

12."'?J' .....T.d

"'---__,'-"-'----.lL_.........--._l,_-L._-L.-_l---'~._, __1_ _t_~L__ l-..L._L-l._

0.0;

O. 10

eA

(inches/inch)

o. J..5

1.0

(Eending l' TuJist)

4 VIF 13
~) p::: O. 12P y

,
I

~_~

0.01

L l_ _- L_ _...-...-_l-_

0.02

0.0)

_-J--

O.Olt

J.-

0.06

eA (inches-inchJ
Fig o 10

L--

-----,J-

0.08

205A.l1

1.0

8 WF 31
L/ r = 55

03

0A

(Inches! inch)

Fig. 11

1. 0

4 WF 13
L/ r = I' f

o,

0.01

0.02

0.03

0.04-

0.05

eA

0.06

(inches! inch}

Fig. 12

0.0.7

O. 08

0.10

2051l.1l

1.0

4 WF 13
L/ r

0.01

0.02

0.03

0.04

8A

0.05

0.06

(inches/inch)

Fig. 13

0.07

0.08

0.09

55

0.10

2051t .11

'\

~T-17 (4WFI3)

..~~~.:~,~./

.~=ti--P

'~-'"

~~~'

1.0

r ,>';:/'.

\"

YC,' \

I "- -

~~

(Local InstabilitlJ)

'\

Instability)

T-4 (SvvF31) (Local

/
~-

y
L/ r

= 55

p :::: O. 12Py
~~~

(Benrlin(: f TIJ)isiJ

L/ r

L
i
O.5~

\J.

= 55

~ p = O. 12P y

' 1'../

t-

L
j

J.\

-<-~---":_--'--'-_J

L.-l__L-':'-1--a1

__

1.0

L __ .L.__ J
2.0

Fmg.

15

._l_ _I_ _..1.-1- 1 _ . - . \ 0 - - , . . ,

_-L-1

.a-,,!

3,,0

4.0

205A.ll

0.10

0.05

10

20

ltO

60

70

80

90

100

110

O. 7

i"~"C"

\!

~O

0.6

T- f9

\!

,
I
I

05

L\R,

.~

M 1M
o

yc

O.12P y

\\T-16
1\,

0.4-

! \
1

!
;

03

GWF31 Sections

.~

-+-----.~~ : 1----- 4VIF13 feet ione

i
\ i
\1
b "b T- 12 (8
f

0.2

II

1'\ .
\~\

tI )

\\

'bT-4 (4 11 )

0.1

10

20

I
30

T-17
I
LtO

(4 fJ

50

60

70

L
r

Fig.

18

80

T-24

100

110

205A.ll

= O.

12P y

'------..---~--

eIf/elye

lid'!?

~IT-13
'

,tt) r~

~;,b T- 4 (8 f1) (Local

(8""')

<

(Local Instabiltty)

?~ I

Instabi l- i tIl)

0.2

~_~!-17

(A.If)

(Local Instability)

!--",--.,....
.,~~

'-"'-~~'~:t.~::::..____

Itb If

~~

~~~~
"-~~~

~T~O

(IJending f TVJist)

(4

ft

1~~~lf'lf

~.:

0.1

.'-L~,~

(Bendtn~

8VJF31 Section

-------I '

~-- <-l:VJF1~3

(Bending +
TIJ)ist)

T_ 19

4,0 T-

L-._--L_---l_~. . JI

10

O=.~CI1 T-11 ~~

LO

30

---!.'

=:::rr:____l':':_.

,-10

I 2 (8")

L
r

Fig.

(DenriinrJ

~.

T-26
t

"J-

m
't)
J.uJZ,9,
:

Thitst)

Section

(Bendin~'fi Twist)

-l-i_ _~_!_I_ _ ~_.....\.___---1..-_~_-'---_--L-

60

I;

~~~

~~
~~.

19

70

80

90

JOO

r 10

5.0

M 1M
a

yc

p= O. f2P y

e A/e yc

4.0

"el" \

~~,~ IT-13

"
fib

(8 t1

~~

'~? T_ 4 (8")

II

3.0 -

Ii

! T-17 (4")

~-'=========.==-

lIb II

T-9

.='-={)-t

ltd 'II

2.0

T-20 (4")

~l

-==-=~"
I
~T-26

'~oT-t2

l.a

8WF-31 Sect ion

T-23
i

(8")

---1

...

I
4WF13 Section

fa

40

-~
r

Fig. 20

70

[30

90

100

110

205A.ll

4.0

-:(--

4WFI3 Section

21

1T-24

(I astl reco rded reading 8.9)

I
\

5.0
lIdl

T-13(Oll)

I;

l~

R.3

--------._'

~---_._--' _.-.->~ I

1.

4.0 -

M 1M

.~,._-~-~ ..

P..

"=.

30

.....

yc

/
8.18
T~+17

- ...

yc

(4")

=-=- "b"

T-20 (1 ff )
2,0

'=-~~0'4_T-26
i

~T-16

i~~

!i
8\MF31 Sect iO!l

!!.i;

!
1.0

T-12 (8")

~
_

I
\_

r- 4\'lF13 Sectifni
I

!i

i
:

..

10

2J

I
30

Ii

40

50

,
60

+k
r

Fig.

22

70

80

90

lOa

110

205A.ll

st resS
_L_
_._-dua\

1.0

\ t
SOU

.\on Neg\ECting Res \

s----.::;;_-:I~~;:ding

,..,I
My c

/
7

II'

fsidual Stress

,,-'

//

""",,"

,,// ..........-.n

u---

--------:----

Lateral-Torslcnal ~ I
Displacement first
I'
indicatt'<d ty lateral
9
deflection gages.
~I

,,,'

---"Twistlng of specimen first observed by eye.

0.5

~J1(j

P
8 WF 31
L
r

-0---

41

Expe ri men,t a I Resu Its

(TestT-16)
0.01

0.02

Fig. 23

0.03

0.04

0,05