Question regarding lateral buckling restraint provided by purlins 2

[atabgc]

Hello.

I've been reading the document "Design of Steel Portal Frames for Europe" by CM King, and I found this note regarding lateral stability of portal frame rafters:

"The effective length of members for lateral buckling and lateral torsional buckling can be reduced by diagonal stays to purlins and siderails. However, the use of purlins for lateral restraint should be agreed with the checking enginer before proceding. In some countries, this may not be allowed. In many countries it will only be allowed if the purlins are aligned with the nodes of the roof bracing truss and the forces from the bracing loads must be included, when calculating the resistance of the purlins."

Would anyone be able to elaborate a little on the second part of the paragraph? What's the reasoning behind only considering some of the purlins for the lateral restraint? And does Eurocode 3 state anything in this regard?

Thanks in advance.

 

[KootK]

What's the reasoning behind only considering some of the purlins for the lateral restraint?

For the purlins to serve as lateral bracing, they need to be, themselves, prevented from lateral travel in the plane of the roof deck. And that means one of two things usually:

1) A deck that can serve as a true diaphragm rather than just roofing. Ergo not standing seam etc OR;

2) Purlins aligned with horizontal trussing nodes within the roof deck.

 

[Agent666]

I'd add/extend to what KootK noted:-
There are two types/methods of lateral restraint available in most codes for considering restraint to a section under flexure:-

1: preventing lateral movement of the critical compression flange but not necessarily preventing twist of the cross section

2: and preventing twist of the cross section, but not necessarily preventing lateral movement of the cross section

3: prevent movement and twist

Some codes may not allow the twist restraint route, which is why they may be saying the restraint needs to align with the roof structure bracing nodes.

Generally for axial restraints you would need to restrain both flanges against lateral movement (both flanges are critical if you like), and restrain twist of the cross section. Though there are ways of working out the restraint afforded for connection to only one flange under axial load. Under axial load you are trying to obviously force higher modes of buckling so its easy to correlate this with not allowing any lateral movement at the point of restraint.

 

[atabgc]

Thanks!