Strut Slenderness

The minimum slenderness limits for compression members previously given in BS 5950: 1990 have been removed from BS 5950-1: 2000.

I personally disagree with this change to the design code and would ask all structural / civil engineers to review the following warning: -

The shortening or lengthening of any member subject to axial load is generally taken to be equal to (P.L / (A.E)). This formula, used by most linear elastic computer programs, ignores shortening due to strut lateral buckling deformation. When a strut is not too slender the error in ignoring shortening caused by strut buckling is trivial, but as a strut becomes more slender this secondary shortening effect can become dominant.

The following table provides an example of the magnitude of error (Dv / x) for various strut slenderness values. For L / ry = 100 the classic assumption of strut stiffness would be in error by 10% whereas at L / ry = 200 the error increases to 280% !

Horizontal and Vertical Deflections for a Strut at full stress
203UB x 30kg in Grade S355:-

Length
(mm)

L / ry

Dh
(mm)

Dv
(mm)

Ratio of Dv / x

1585

50

3.8

2.3

1

3170

100

19.9

2.7

1.1

4755

150

53.6

3.3

1.8

6340

200

101.5

5.4

3.8

7925

250

162.9

9.4

8.1

9510

300

237.6

15.6

16

11095

350

325.5

24.4

28.9

12680

400

426.6

36.2

48.6


The infamous Strut:-
26.9dia x 3.2thk in Grade S355 x 10m long!

Length
(mm)

L / ry

Dh
(mm)

Dv
(mm)

Ratio of Dv / x

10000

1182

1322

436

6239


x = P.L

 i.e. The classic assumption for strut shortening when displacements are small.

A.E

 
Dh = Mid length lateral buckling displacement (mm)
Dv = Total of (end shortening due to Dh) + x  (mm)

The above tables show the worst possible condition with a strut at full stress. For a strut at 90% of full stress and L / ry = 200 the error in assumed strut stiffness is reduced to 80%!

I would be interested to hear the views as to the solution from other engineers (reinstate slenderness limits or another?).

Home

Page 1
Page 2
Page 3
Page 4
Page 5
Page 6
Page 7
Page 8
Problem with Strut Deformation
Response from News Groups

 

Email - Comments

CADOSS (c) 2004