!"#$%&%' )*+,-" .

,/0"$ 1
2-#34&5 6*-,/% 7,58-&%' #%+ 49" 2::"54 *: 2%+ ;"34$#&%4

<="$=&">
uslng a crlLlcal load analysls, Lhe elasLlc flexural buckllng sLrengLh of a column wlLh varlous degrees of end resLralnL
ls lnvesLlgaLed.

!"#$%&%' <0?"54&="3
Lmploy compuLaLlonal sofLware Lo verlfy Lhe LheoreLlcal elasLlc column buckllng soluLlons obLalned from
Lhe use of dlfferenLlal equaLlons.
Cbserve how Lhe load carrylng capaclLy and buckled shape vary wlLh dlfferenL Lypes of end resLralnL.
Conflrm Lhe LheoreLlcal effecLlve lengLh (!) values LhaL appear ln 1able C-A-7.1 of Lhe CommenLary on Lhe
AlSC "#$%&'&%()&*+ '*- ")-.%).-(/ ")$$/ 0.&/1&+23 (2010).



)"49*+
repare a compuLaLlonal model of slx parallel 40-fL long sLeel W14x82 columns. SeparaLe Lhe columns by 10-fL. for
clarlLy and provlde Lhe end resLralnL condlLlons shown ln 1able C-A-7.1. 8e sure LhaL Lhe Lop of each column ls noL
resLralned ln Lhe verLlcal dlrecLlon. Subdlvlde each column lnLo 8 elemenLs and apply a 1-klp load Lo Lhe Lop of
each column. uo noL lnclude Lhe self-welghL of Lhe members. unlLs of klp, lnches, and ksl are suggesLed.

erform a planar frame (2u) elasLlc crlLlcal load analysls
1
and compuLe Lhe flrsL 10 buckllng modes for Lhe sysLem
of columns. vlew Lhe resulLlng 10 buckllng modes and correspondlng crlLlcal applled load raLlos. noLe LhaL you
may need Lo lncrease Lhe scale of Lhe deflecLed shape Lo beLLer observe Lhe shape of Lhe buckllng modes.


1
An elasLlc crlLlcal load analysls ls also referred Lo as an elasLlc buckllng or elgenvalue analysls.
(7) Joint restraint is distributed to the column above and below the joint in propor-
tion to EI/L for the two columns.
(8) All columns buckle simultaneously.
(9) No significant axial compression force exists in the girders.
The alignment chart for sidesway inhibited frames shown in Figure C-A-7.1 is based
on the following equation:
(C-A-7-1)
The alignment chart for sidesway uninhibited frames shown in Figure C-A-7.2 is
based on the following equation:
(C-A-7-2)
Comm. 7.2.] EFFECTIVE LENGTH METHOD 16.1511
Specification for Structural Steel Buildings, June 22, 2010
AMERICAN INSTITUTE OF STEEL CONSTRUCTION
Buckled shape of
column is shown by
dashed line
(a)

(b)

(c)

(d)

(e)

(f)

Theoretical K value 0.5 0.7 1.0 1.0 2.0 2.0
Recommended design
value when ideal
conditions are
approximated
0.65 0.80 1.2 1.0 2.1 2.0
End condition code
























) a (























) b ( ) c (























) d ( ) e (























) f (























d e l k c u B
i n m u l o c
l d e h s a d























f o e p a h s
y b n w o h s s
e n i l




























































































i t e r o e h T
m m o c e R
h w e u l a v
n o i t i d n o c
m i x o r p p a
n o c d n E























l a c K e u l a v 5 . 0
g i s e d d e d n e m n
l a e d i n e h
e r a s n
d e t a m
5 6 . 0
e d o c n o i t i d























7 . 0 0 . 1
0 8 . 0 2 . 1























0 . 1 0 . 2
0 . 1 1 . 2























0 . 2
0 . 2























n o c d n E























e d o c n o i t i d





































































TABLE C-A-7.1
Approximate Values of Effective
Length Factor, K
G G
K
G G K
K
A B A B
4 2
1
2
2

/
/
tan /
ta
( ) +
+

( )

+
nn /
/

2
1 0
K
K
( )
( )
=
G G K
G G
K
K
A B
A B

/ /
tan /
( )
+ ( )

( )
( )
=
2
36
6
0
Learnlng Module number 1 2
@,"34&*%3
1) AfLer recordlng Lhe buckllng loads (producL of resulLlng applled load raLlo and Lhe glven axlal load of 1 klp)
for each glven end resLralnL, compare Lhese resulLs wlLh Lhe equaLlon !
!"
! !
!
!" !"
!
uslng Lhe
LheoreLlcal effecLlve lengLh (!) facLors provlded ln 1able C-A-7.1.
2) Slmllarly, back-calculaLe and compare Lhe effecLlve lengLh facLor ! for each end resLralnL uslng ! !
! ! !" !
!"
, where 4
%-
ls each of Lhe column's crlLlcal loads based on compuLaLlonal analysls.
3) SorL Lhe buckllng loads from smallesL Lo largesL. WhaL can be concluded regardlng Lhe degree of end
resLralnL ln relaLlon Lo Lhe magnlLude of column buckllng load?
4) ln revlewlng Lhe 10 buckllng modes, are hlgher buckllng modes shown for any of Lhe columns? Why are
Lhese modes of llLLle Lo no lnLeresL? When mlghL Lhey be useful?
3) CommenL on how Lhe acLual end resLralnL may dlffer from Lhe ldeallzed ones glven. Why ls your response
dlfflculL Lo model uslng compuLaLlonal analysls? Pow would Lhese dlfferences lmpacL Lhe !-values and ln
Lurn, Lhe correspondlng crlLlcal loads?

)*$" A,% >&49 6*/B,4#4&*%#- C%#-D3&3E
1) AL Lhe Lop and boLLom of Lhe Luler column (plnned aL Lhe boLLom and roller aL Lhe Lop), provlde
addlLlonal 3-fL long horlzonLal sLeel beams Lo Lhe rlghL of Lhe column. rovlde a roller (verLlcal resLralnL)
aL Lhe free ends of Lhe beams.
a. uslng Lhe same bendlng momenL of lnerLla (5) for boLh beams, vary Lhe value of 5 from small Lo
large and compuLe Lhe resulLlng buckllng load for Lhls column. repare a Lwo-dlmenslonal ploL,
wlLh Lhe value of Lhe beams' 5 on Lhe horlzonLal axls and Lhe column's crlLlcal buckllng load on
Lhe verLlcal axls. ln addlLlon Lo or alLernaLlvely, prepare a slmllar ploL wlLh Lhe back-calculaLed !-
facLor on Lhe verLlcal axls (lnsLead of Lhe crlLlcal buckllng load). ln general, dlscuss Lhe lmpacL
LhaL Lhe flexural sLlffness of Lhe beams has on Lhe crlLlcal buckllng load of Lhe column.
b. 8epeaL Lhe above exerclse, buL no longer requlre LhaL Lhe Lop and boLLom beams have equal
values for Lhelr momenL of lnerLla 5. erform many crlLlcal load analyses for varlous
comblnaLlons of Lop- and boLLom-beam 5 values. 1ake advanLage of Lhe facL LhaL LhaL Lhe
problem ls symmeLrlc, LhaL ls, reverslng Lhe values of Lop and boLLom 5's wlll glve Lhe same resulL.
Show Lhe resulLs of Lhls sLudy by preparlng a Lhree-dlmenslonal ploL, wlLh boLLom-beam 5 value
on a horlzonLal axls (x-axls), Lop-beam 5 value on Lhe oLher horlzonLal axls (y-axls), and Lhe
columns crlLlcal load and/or !-facLor on Lhe verLlcal axls (z-axls).
c. 8epeaL elLher or boLh of Lhe above exerclses, provldlng beams (agaln, wlLh rollers aL Lhelr free
ends) aL Lhe Lop and boLLom of a column LhaL ls pln supporLed aL lLs base and free Lo LranslaLe aL
lLs Lop (noLe: 1hls case wlLhouL Lhe beams was noL one of Lhe orlglnal cases lnvesLlgaLed. Why
noL?)
2) 8epeaL parL or all of Lhe above exerclse uslng Lhe comblnaLlon of (l) a rlgld beam (beam wlLh a very large
momenL of lnerLla), and (ll) beam-Lo-column connecLlons of varlous roLaLlonal resLralnL (roLaLlonal sprlng
sLlffness).

F,''"34"+ G#$/H,B 2I"$5&3"
repare a model LhaL only has one column and Lhe Luler end resLralnL condlLlons (plnned aL base and roller aL
Lop). SLarLlng wlLh one elemenL, compare Lhe buckllng load compuLed by an "-#34&5 5$&4&5#- -*#+ #%#-D3&3 wlLh
Lhe LheoreLlcal value. Lxplaln why Lhe resulLs are noL ln close agreemenL. Subdlvlde Lhe column lnLo Lwo
elemenLs and compare Lhe resulLs. uslng an accepLable error of 0.3, keep repeaLlng Lhls subdlvlslon process
unLll Lhe adequaLe number of elemenLs ls obLalned. repare a Lable LhaL shows number of elemenLs and Lhe
correspondlng percenL error. Pow many elemenLs are needed? ls Lhls resulL a funcLlon of Lhe degree of end
resLralnL? lf so, Lry some oLher end condlLlons and deLermlne whlch end resLralnL condlLlon should be used
and whaL ls Lhe correspondlng number of requlred elemenLs ls for 0.3 accuracy?

C++&4&*%#- ;"3*,$5"3
AlSC "#$%&'&%()&*+ '*- ")-.%).-(/ ")$$/ 0.&/1&+23 (+1 6*77$+)(-8 (2010)