Torsional Stiffeners

[Althalus]

I need either confirmation or correction.

The situation I've got is a skid with C12 girders and C8 joists with an all around fillet at joist ends. The 4" below is space for cables and conduit to be installed. The floor plate above also ties it all together.

The person checking has asked about torsion because the shear center is not at the joist/girder interface.

The school of thought I was taught was that if you have torsional restraints that are at least 2/3 the depth of the beam (which I have) then that was adequate torsional restraint. I'm being challenged on it now to provide a source for that.

Well, I've looked and I can't find anything that speaks to the minimum depth of torsional restraint required at any depth. Does anyone know?

 

[EPCI-Steel]

I've never heard that before, I looked in AISC-13 and Blodgett because I was curious and saw nothing that eluded to that. Usually for that type of connection I provide stiffeners that extend to the bottom flange.

 

[JAE]

Well if you have C8's at 4" o.c. (a ridiculous extreme) then yes, you have torsional restraint. If the C8's are spaced further (you don't adequately describe your configuration) then depending on how the load is applied, then you could possibly have torsion between the C8's in the C12's.

The C8's, fully welded as you suggest, certainly resist torsion in the C12 at the locations of the C8's. But not between them.


Check out Eng-Tips Forum's Policies here:
faq731-376

 

[WARose]

[blue] (Althalus) [/blue]

Well, I've looked and I can't find anything that speaks to the minimum depth of torsional restraint required at any depth. Does anyone know?

For the purposes of bracing, Appendix 6 of AISC (13th edition) addresses this. IIRC, there is no required minimum depth or mandatory location for attachment. But torsional bracing can be negated if the web lacks distortional stiffness. You also have to run a variety of localized checks on the sections involved.

In any case, whatever "restraint" you have.....you still have to check the member for the torsional forces developed. At the very least, there will be compatibility torsional stresses as the members displace.

And by the way, your subject line mentions "stiffeners".....be advised: plate stiffeners do virtually nothing for a beam in torsion. (I know you are using channels as your "stiffeners" but I wanted to make that point in case I am misunderstanding something.)

 

[KootK]

I was taught 0.6d which is, for all intents and purposes, the same thing. I think that your reviewer's being ridiculous and pedantic. C12/C8 welded all around? Come on. Any engineer with an ounce of judgment is going to OK this scenario without wasting precious time trying to "prove" it. I'm pretty sure that I have seen this in print somewhere but I've no idea where. And that's fine. Engineers are expected to apply their judgment and don't need to defend every last decision with a printed reference.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[KootK]

Also, in this situation, your joists are delivering the shear load outboard of the girder shear centers. That means that your torques are actually opposing your primary joist flexure. Pretty great.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[BAretired]

There is nothing special about the joist/girder interface. In fact, that is not the point of application of load. The girder may be considered to be loaded at its shear center if the only load is coming from the joists. Its torsional rotation is limited by the end rotation of the joists which are moment connected to the girder.

BA

 

[Althalus]

Thank you for all your replies.

JAE, I always like reading your comments.

The C8 joists are at 24" o.c. But the only C12 are at the end points (all perimeter members are C12). So, his argument would make the torsional span about 30 feet long. This is a ridiculously long span for a channel to resist torsion. But if we count the C8, then the torsional span is only 24" long. Not only that, but there is no other load other than from the joists. So, if the joists themselves provide torsional restraint, there is virtually no torsional load.

WARose,

Unfortunately I only have the 9th and 14th editions. But that idea makes sense. I'm pretty sure other minor details at that connection will be sufficient to diminish the extreme fiber load to the point the web will be ok.

KootK,

Yes, that was my knee jerk reaction. But I always have to think in terms of "if it ever goes to court..." And we did take into account the opposing loads. But with a 30ft torsional span, that simply didn't do anything.

BARetiered,
If it were only a shear load, that would be true. But if it is a moment ocnnection, then it is not, correct?

 

[canwesteng]

https://www.aisc.org/steel-solutions-center/engineering-faqs/4.3.-stability-bracing/#9534

 

[WARose]

Unfortunately I only have the 9th and 14th editions.

This is addressed in the 14th edition as well. (Same Appendix # IIRC.)