Member Lateral Restraint Loads

[Ussuri]

We have been having a discussion about the origin and the method behind the use of a nominal load for designing member lateral restraints. The UK code BS5950 states that restrainst should be designed for a force not less than 1% of the axial load. The old code BS449 states this should 2.5% of the axial load. I believe other codes have similar approaches although the percentage varies. I have included the BS5950 text below:

A restraint should have sufficient strength and stiffness to inhibit movement of the restrained point in position or direction as appropriate. Positional restraints should be connected to an appropriate shear diaphragm or system of triangulated bracing. Positional restraints to compression members forming the flanges of lattice girders should satisfy the recommendations for lateral restraint of beams specified in 4.3.2. All other positional restraints to compression members should be capable of resisting a force of not less than 1.0 % of the axial force in the member and transferring it to the adjacent points of positional restraint.​

I have done a literature search to try and find a discussion or a reasoning behing this approach but I have been unsuccessful. Is anyone aware of the reasoning behind this approach? Test data? Empirical rules of thumb?

 

[DST148]

Although not directly related to British Codes.....
Ref AISC 360-10, Specifications for Structural steel Buildings- Commentary:
For column bracing the reference quoted is : Yura, J.A.(1995) - "Bracing for Stability - State of Art" Proceedings of ASCE Structures Congress XIII, Boston
The beam bracing requirements are based on the recommendations of Yura(2001) : "Fundamentals of beam bracing", Engineering Journal, AISC, Vol. 38, 1st Quarter.

 

[BAretired]

The Canadian code has varied quite a bit over the years. The current version of CSA S16-01 allows three methods, namely 9.2.5 Simplified Analysis, 9.2.6.1 Second Order Method and 9.2.6.2 Direct Method.

The compression member is assumed to deviate from a straight line in accordance with tolerances prescribed elsewhere in the code. Axial compression in the member produces a moment due to that initial eccentricity but, under the action of the axial load, the eccentricity grows.

If the brace is not stiff enough, the eccentricity will continue to grow without limit. Thus the brace must have adequate strength and adequate stiffness.

The method used in S16-01 is credited to Dr. George Winter.

BA

 

[charliealphabravo]

Some good discussion in an old post...although it starts out talking about plumbness.

http://www.eng-tips.com/viewthread.cfm?qid=240591

 

[BAretired]

Not sure why the spaces between paragraphs were so large in my earlier post. They didn't look that way in the preview.

If the member to be braced is a column, out of plumbness is a proper consideration. Again, the tolerance for plumbness is specified in the code. If the column is pinned top and bottom, the lateral force required to brace the top of column varies directly with the out of plumb dimension. If the brace is very flexible, the column becomes more out of plumb and the brace force increases accordingly.

BA

 

[saad73joss]

In Eurocode, it s 1%

 

[dik]

BART... S16-09?

Dik

 

[Ussuri]

The AISC commentary is useful. I see in appendix 6 they also use 1% for nodal bracing stating that the "winter rigid model" gives a brace force of 0.8%. I assume this is the same fella mentioned by BA above.