Unbraced length of a beam 7

[shaneelliss]

As most of you probably know, the strength of a beam in bending depends on the unbraced length of the compression flange. My question is, what defines braced? If I have a channel spanning 20 feet with another channel alongside of it 2 feet away, and they are connected to each other with a plate welded continuously along the top flange of each, are these beams continuously braced? I have another pair of these channels 6 feet away and they could be connected to the first pair on, say, 5' centers. If the first pair aren't considered braced, would this give them an unbraced length of 5' or is it still too wobbly to be called braced?

 

[msquared48]

Normally, the bracing comes from either a wood or concrete diaphragm, or wood or steel purlins framed 90 degrees to the beam you are bracing.

To rely on elements that are parallel to the same beam is not a good idea, unless they are really stiff elements. That's a judgment call.

Mike McCann
McCann Engineering

 

[AlmostPE]

or do a custom section and call it not braced.

 

[hokie66]

Your pair of channels with a plate welded continuously to both is not braced, but the assembly forms a beam with a top (compression) flange which is 2 feet wide, so is much stronger than just the two channels.

 

[DaveAtkins]

You could check this as a beam lying on its side, 2 feet deep. The only load the beam lying on its side needs to resist is the buckling force in the two channels (say, the summation of 2% of the force in each channel's top flange). The stress induced by the buckling force would be added to the stress induced by the direct vertical loads.

DaveAtkins

 

[shin25]

I would say-tie the bottom flanges of the channels every so often and call it a box shape and forget about the lateral torsional buckling. Given, that the plate that you have welded on the top is sufficiently rigid.

 

[csd72]

Dave has a good answer - if in doubt, do the numbers. There are explicit rules in the code regarding what loads the braces need to take.

I would not provide a continuous weld though, a staggered weld would be less likely to warp the channels.

If you are treating it as a built up section then you need to check the shear flow in the welds using v = VQ/It.

Bracing 2 unbraced members together does not help as they could then both buckle in the same direction. There needs to be a load path back to a rigid point of support(like your plate does).

hope this helps.

csd

 

[271828]

I'd say quit trying to home-brew the definition (just ribbing, guys! ) of braced and read AISC 13th Ed. App. 6 and the Yura/Helwig bracing seminar notes. The notes, especially, talk about stuff like the proposed situation.

 

[Lion06]

It seems most people are saying this situation is unbraced. Please see the following thread,
javascriptpenindex(450,450,'

http://www.tipmaster.com/includes/refinfo.cfm?w=450&h=450')

which has a very similar situation but with discrete stair treads instead of a continuous top plate. I suggested the possibility of unbraced in this situation and everyone else seemed to think it was braced.
Can someone please explain the difference to me?

 

[Lion06]

sorry, try this thread507-192367

 

[WillisV]

"Bracing 2 unbraced members together does not help as they could then both buckle in the same direction."

This is a common misperception. Bracing two unbraced members together does help provided the bracing is designed properly. Common Example: A bridge - two beams with cross frames between.

See Appendix 6 of the 13th Edition Manual for all the explicit requirements, but essentially if cross frames are present they are considered beam nodal bracing and may either be classified as lateral or torsional bracing depending on their placement and must meet both the strength and stiffness requirements of the Code. If the cross frames are designed to meet these requirements, then yes the unbraced length of the beams is reduced to the spacing between cross frames.

A continuous plate or slab would be classified as continuous torsional bracing and need to meet the requirements of Appendix 6.3.2a and 2b.

 

[shin25]

I agree with willisV.

It is hard for me to realize that two seperate members which are continuously connected at the top with a rigid plate will be able to torsionally twist to create a lateral torsional buckling situation!!

 

[csd72]

WillisV/Shin25,

You guys would make really good politicians!

You take a quote out of context and pick it to pieces.

Maybe you should have read the rest of the paragraph:

"There needs to be a load path back to a rigid point of support(like your plate does)."

I agree that crossed bracing would to the trick to provide torsional bracing, but I dont think that was what the OP was suggesting.

Structural EIT,

A stair is not 2' wide by 20' long so the lateral capacity is much more obvious. Also I am not saying that it is unbraced only that the OP needs to do some numbers to confirm it meets the code.

csd

 

[shaneelliss]

Thanks for the guidance. I guess I have not looked through the new AISC manual enough. I didn't realize there was such a layed out explanantion of bracing in the commentary. Has that always been there or is it new in the 13th edition?

 

[271828]

Shane, it's a fairly new inclusion. The first time it showed up was in the 3rd Ed. LRFD Manual. The 13th Ed. Manual includes the info in Appendix 6 and (especially useful) the Commentary to Appendix 6.

You'll really need the bracing seminar notes to dig into the subject, though. The entire subject is very convoluted.

 

[271828]

Beware trying to use the plate as a torsional brace. The web distortional stiffness will likely eat its lunch and make it not work.

Torsional brace may be connected anywhere on the section height, but they don't do much good without adequate web distortional stiffness.

 

[miecz]

I've faced this situation myself. I don't think the plate qualifies as bracing according to Appendix 6. It can't qualify as a lateral brace, and I don't think it can qualify as a torsional brace, for the reasons stated by 271828. At the same time, my gut feeling is that the plate really braces the channels. It's too bad no one has ever tested this configuration.

 

[271828]

"At the same time, my gut feeling is that the plate really braces the channels. It's too bad no one has ever tested this configuration."

I tend to agree, although it would be hard (for me anyway) to prove. The plate doesn't brace the channels by tying them together because both could buckle in the same direction if they're equally loaded. If only one was loaded (or loaded more), the other one could provide some bracing for the loaded one.

However, for the channel top flange at midspan to laterally displace more than that same channel flange at the 1/4 point, the plate has to bend. I think this makes it fall under the category of a relative brace, but I'd have to study the notes to be sure. I think the seminar notes show at least one case that's similar--seem to remember it being two members connected together by a horizontal truss.

 

[miecz]

271828-

Like I said, I've faced this situation before, hunted for a solution, and always came up empty. Always ended up treating the beams unbraced, even though I believe they are braced. Never went to the seminar and so I don't have the notes. I would be VERY interested in a solution, if it's there.

 

[Lion06]

Can someone please explain the difference to me between this situation and the one referenced in the thread I posted above? I thought from the time of the first thread above that this would be an unbraced condition, but it seems the views are mixed with a consensus being reached on each thread. People are saying this one is unbraced, but the stair stringer in the last one is not braced. I am not following the reasoning.