twin girder diaphragms 1

[BLUEPUPIL]

Can I assume (from the picture) that the unbraced length equals the distance between diaphragms?

 

[KootK]

In the Yura paper that I referenced:

1) He acknowledges that that paired buckling mode is unlikely for most practical arrangements and;

2) He provides a simplified, conservative checking procedure that is:

a) Based on system level warping.

b) Conservative.

c) So simple you could do it on an abacus.

d) Intended to be a litmus test for whether or not more sophisticated evaluation is necessary.

I'd personally start with that and only move on to more labor intensive exercises if this hat didn't pan out.

Again, personally, I'd ditch the twin beam setup entirely. I feel that it is a clever, but impractical solution. In north america, we brace steel beams with wood diaphragms all the time, and successfully. And, even baring that, I feel that there are other solutions available to ameliorate lateral torsional buckling that would be more cost effective, and less spatially intrusive, than the twin beam setup.

 

[lexeng18]

If the cross members that connect the two girders are able to transfer weak axis moment due to truss action (like a vierendeel truss), are BridgeSmith's original assumptions true from his first post?

1) that the unbraced length between individual girders is equal to cross frame spacing
2) global buckling could be checking for combined section in the weak axis direction, and 2*Ix in the strong axis direction across entire span length?

Just want to make sure I am following this discussion properly.

 

[KootK]

1) that the unbraced length between individual girders is equal to cross frame spacing

In my opinion, this is true even without the vierendeel action. The diaphragms are effectively "roll beams" in non-bridge speak and I'd expect them to limit the individual girder lateral torsional buckling length the the diaphragm spacing and force the critical buckling mode to be one that involves both girders acting in concert.

2) global buckling could be checking for combined section in the weak axis direction, and 2*Ix in the strong axis direction across entire span length?

This one's a little more nuanced as the vierendeel action would encourage a degree of composite behavior about the vertical axis but would do so imperfectly. That partial efficacy would need to be accounted for somehow at which point I'd be looking to an elastic buckling analysis for resolution. I doubt that there's ten living humans that I'd trust to work out a closed form solution to this. Yura, Galambos, White, Ziemian, Trahair, Zalka... Obviously, I've got a north american bias. The ghost of Timoshenko would be ideal if it could be summoned somehow. These concepts are a bit complex to discus via Ouija board though.

 

[BLUEPUPIL]

Girder Span= 9м
Load [LL+DL]= 20kN/m

1.First variant- Single girder- HEA360; weight=112kg/m.

2..Second variant- Twin girder with diaphragms- 2xHEA220; weight=2x50.5=101kg/m.

2..Third variant- Twin girder with diaphragms- 2xIPE300; weight=2x42.2=85kg/m.

Single girder Mмax=101kN.м
Single girder load 20/2=10kN.м

HEA220
CTICM-LTBeam-Mмax(cr)=132kN.м- Fine!

IPE300
CTICM-LTBeam-Mмax(cr)=61kN.м- Rejected.


To recap,
In this particular case the two HEA girders have enough potential to handle LTB even without diaphragms i.e. acting as a typical single girder. If we add the bracing of timber joists the picture become pretty clear.

Again, personally, I'd ditch the twin beam setup entirely.

I agree, but investor insist to use steel beams with small „h“.

Finally, somewhat theoretical question, it’s not quite clear for me how a lateral seismic load affects LTB?

Thanks to all! I appreciate your input!