Module 05, Columns
Module 05, Columns
Module 05, Columns
MODULE-5
ANALYSIS OF COLUMNS
A compression member may buckle locally or globally. Buckling failure of a column is a sudden
failure.
Buckling is different from bending as the former is caused by the compression load and the later
is by lateral load.
In other words, if this load is applied, and a slight displacement occurs at A, the bars will
tend to move out of equilibrium and not be restored to their original position.
At the verge of buckling: Pcr = kL, (Neutral equilibrium)
d2 𝑣 Pcr
⇒ 2 + λ2 𝑣 = 0 −− −(2) where λ =
dx EI
As we know that sine of any angle equal to nπ is equal to zero, where n = 0, 1, 2, 3…..
Pcr Pcr 2 2
n2 π2 EI
λL = nπ ⇒ 𝐿 = nπ ⇒ 𝐿 = nπ ⇒ Pcr =
EI EI L2
The value of n cannot be zero as it will result in zero load which is not possible. n = 1, 2, 3….
Correspond to buckling in first, 2nd , 3rd …. modes. Which is possible when the column is
laterally braced at zero, one, two, … locations along the length of column.
𝛑𝟐 𝐄𝐈
For n = 1 𝐏𝐜𝐫 = 𝟐 −− −(𝟒)
𝐋
This is the Euler’s buckling formula for critical load of long columns. The critical load of
column at buckling is directly proportional to the flexural rigidity (EI) and inversely
proportional to the square of column length.
Ratio of effective length to the actual length of column is known as effective length factor.
A plot is drawn between the slenderness ratio and the critical stress. As 30
the buckling formula is developed for linearly elastic material, it is 25
applicable to up to the proportional stress. The minimum slenderness
20
ratio up to which the Euler’s formula is applicable is given by:
15
π2 E 10
Le /r min = −− −(7)
σP 5
For steel having proportional stress equal to 30 ksi and E = 29000 ksi:
π2 E π2 29000
Le /r min = = = 98
σP 30
π2 E π2 10,000
Le /r min = = = 60
σP 28
Where PR , Py and PE are the Rankine, Yield and Euler’s capacity of column. Writing the above equation in terms of
stresses:
1 1 1 σy σE σy σy σy
= + ⇒ σR = ⇒ σR = σy ⇒ σR = σy ⇒ σR = 2
−− −(8)
σR σy σE σE + σY 1+ 1 + 2 Le /r 2 1 + a Le /r
σE π E
σy
Where a = 2
π E
36 1 28 1
For steel a = 2 = and for Aluminum, a = 2 =
π 29000 7950 π 10,000 3525
The value proposed by Rankine-Gordon for mild steel is 1/7500 which is very closed to the calculated values.
2π2 E
Cc =
σyp
12π2 E Le
σw = 2
−− −(1) For > Cc (2)
23 Le /r r
Le /r 2 σyp Le
σw = 1 − −− −(2) For < Cc
2Cc2 FS r
(1)
3
5 3Le /r Le /r
Where: FS = + −
3 8Cc 8Cc3
(3)
d2 v d2 v d2 v P d2 v P
EI 2 = M ⇒ EI 2 = −Pv ⇒ 2 + v = 0 ⇒ 2 + λ2 v = 0 −− −(2) where λ =
dx dx dx EI dx EI
Where C1 and C2 are the integration constants. These constants can be determined from the boundary conditions.
At x = 0, v = e Equation 3 ⇒ e = C1 sin 0 + C2 cos 0 ⇒ C2 = e
1 − cos λL 2 sin2 λL/2
At x = L, v = e Equation 3 ⇒ e = C1 sin λL + e cos λL ⇒ C1 = 𝑒 =𝑒
sin λL 2 sin λL/2 cos λL/2
λL
C1 = 𝑒 tan
2
λL
Equation 3 ⇒ 𝑣 = 𝑒 tan sin λx + e cos λx −− −(4)
2
λL λL λL λL
For maximum 𝑣, 𝑣max = 𝑒 tan sin + e cos = e sec
2 2 2 2
λL
Therefore maximum M, Mmax = P𝑣max = Pe sec
2
λL
P Mmax P Pe sec 2 P ec λL
Therefore maximum Stress, σmax = + c= + c ⇒ σmax = 1 + 2 sec −−− −(5)
A I A I A r 2
P ec λL P ec PL P ec P L
σmax = 1 + 2 sec = 1 + 2 sec ⇒ σmax = 1 + 2 sec −− −(6)
A r 2 A r EI 2 A r 4EA r
P ec P L 10 3 2 10
σmax = 1 + 2 sec = 1+ sec 10 = 1.0 1 + 3.375 1.000 = 4.38
A r 4EA r 10 1.332 4 29000 10
L 10 3 2 10
For = 200 σmax = 1+ sec 200 = 1.0 1 + 3.375 1.201 = 5.05 ksi
r 10 1.332 4 29000 10
For smaller L/r the secondary effect is negligible as shown by the secant term equal to 1.0 For larger L/r the effect is
significant.
Note that the term inside the secant is in radian. The value should be converted to degree before calculating secant
or the calculator shall be set to radian mode. To account for boundary condition effective length shall be used.