Design of steel beam to beam connections for lateral shear 1

[jochav5280]

Good day,

Does anyone have any design advice or design reference materials for designing steel beam end connections for lateral shear? Specifically, I'm looking for guidance on how to check beam-to-beam clip-angle connections for lateral shear capacity, (are there any interaction effects that need to be considered in situations where axial, lateral shear and vertical shear are all present simultaneously?) Is it as simple as taking the resultant shear from vertical/lateral shear and checking that the web of the incoming beam has sufficient shear area? Are there any buckling considerations? I can't seem to find any guidance from AISC on this topic.

What alternatives are there to clip-angle connections for resisting lateral shear? I've seen colleagues weld a plate to the flanges of the support beam, and then end plate and bolt the incoming beam to the support beam plate, (doesn't this introduce torsion in the supporting beam due to the half-flange eccentricity?) This connection is nice as it eliminates the need to cope the incoming beam, so the beam doesn't have to rely on the web to carry the lateral shear.

An additional related question is how do other engineers eliminate lateral shears when modeling their structures? Our structures typically do not have a concrete floor diaphragm, instead we use braced floors. Releasing the lateral shear degree of freedom results in an unstable structure, so I've heard that recommendation and am not a fan. Are there any good approaches for limiting lateral shear forces from a modeling perspective; we use STAAD.Pro if anyone has any specific advise.

Thank you very much for your time and help!

Best regards,

jochav5280

 

[BSVBD]

Wow! There's a lot here to consider!

I'm looking for guidance on how to check beam-to-beam clip-angle connections for lateral shear capacity

Do you have proposed details and shear values for us to consider?

What alternatives are there to clip-angle connections for resisting lateral shear?

A section or detail to refer to would be more helpful to provide an applicable suggestion for an alternate.

doesn't this introduce torsion in the supporting beam due to the half-flange eccentricity?

Depends on magnitude of shear and total eccentricity. The typical shear may be negligible assuming the load is parallel to beam in question. Details would be helpful.

Before i was forced to submit to IBC in 2001, i was raised to not consider shear at all when designing the low-profile commercial structures that are typical of my firm. On one hand i respect your thought of not wanting to neglect; on the other hand, ALL depending on the profile of the structure, for low-profile structures, within reason, i can conclude by looking at a building, considering profile and percentage of openings, among other variables, whether wind shear need to be considered at all. Also understand that i am in a virtually non-seismic area.

Details and values would be helpful.

 

[JoshPlumSE]

Since no one else is responding, I thought I'd pipe in with my thoughts.

1) Bolts on the supporting side / girder: The design of these should be simple. You've got a resultant shear from vertical and horizontal shears. If you have a tension in the bolt as well, (from eccentric moment or axial force) then you do an interaction check with the AISC bolt formulas.

2) Outstanding leg of the angle which connects to the supported beam / joist. The angle will have weak axis bending on it. The moment capacity for weak axis bending should be based on the plastic section modulus. Easy to calculate. If you've got shear and axial force on the connection as well, then you can do an interaction check using AISC chapter H.

3) Effect of weak axis moment on the web of the girder. In my opinion, this is probably not likely to control. But, it can be handled with yield line. Something similar to what was done in the paper "Bending Under Seated Connections" by Abolitz and Warner. This was for shear tabs. The same concept applies, just you have to consider the thickness of the connection.
I believe the soon to be published update to the AISC manual will have the following formula in it:
Mn = (tw^2 * Fy / 4) * [2*T/L + 4*L/(T-c) +8*sqrt(T/(T-c))]*L

4) Effect of axial force on the web of the girder. Same concept of yield lines. I believe the soon to be published update to the AISC manual will have the following formula in it:
Rn = (tw^2 * Fy / 4) * [4*sqrt(2*T*a*b(a+b)) + L*(a+b)]/(a*b)
where
tw = thickness of girder web
Fy = yield stress of girder
T = clear distance between flange fillets
a = distance from top of angle to flange fillet
b = distance from bottom of angle to flange fillet
L = horizontal length between of double clip angle connection
c = depth of clip angles

I think that covers most everything.

 

[jochav5280]

Thank you for your responses,

JoshPlum, your process seems reasonable, yet I can't get any by in from AISC on what to do; it appears that there hasn't been any testing on these connection types.

I design industrial structures and in most cases, we are unable to use concrete floors to form our diaphragm, so we end up using braced floors. Unfortunately, we are only able to get bracing in some of the floor bays, which leads to other areas acting as a "lean-on" type of system where relatively high lateral shears can result. Per AISC, it appears that a lot of practice is to ignore these forces, but they can be quite significant in our structures, (i.e. 30-40 kips.)

Thank you,

jochav5280

 

[KootK]

Practically, I consider these kind of connections to fall into one of two categories:

1) Lateral loads are so small relative to other effects that they are not worth considering explicitly.

2) Lateral loads are big and and I'll go with "real" lateral connections like this: Link

Where the dividing line lies is obviously a matter of judgment. I suspect that AISC's silence on this issue refelcts not just a lack of research on these details but, for reasons similar to mine, a lack of demand for research on these details.

Frankly, I'm surprised that you're getting such significant forces in non-braced joints. But, then, I know little of your situation. If you post your framing plan, I'd be happy to review and comment.

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.

 

[JoshPlumSE]

This topic tends to be one that gets relatively little attention. My concern on the subject stems from the software side. Currently RISAConnection ignores weak axis forces completely. Mostly because of lack of demand from our users. It would seem that engineers don't often design shear connections with large weak axis loading. It certainly surprises me that the Connection design reference (Tamboli, or AISC examples) don't really consider the subject much.

Obviously, RISA Technologies (my employer) needs to have a more rational basis for these loads if we want to add horizontal bracing connections into RISAConnection. Similar to what KootK linked to in that MSC article, these brace connections are where those forces can start to get very large. Since there seems to be a good amount of user demand for these brace connections, I have probably put more thought into how to handle these loads than most engineers.

Some final additional thoughts:
1) Connection design can certainly be tricky. But, it sometimes helps to think of it in terms of what limit states you need to investigate rather than what formulas you need to use.

2) In that respect, you get back to first principles. Follow the load path through the connection. Then, for each element in the load path, you try to identify the limit states that could govern.

3) If you do this, then you will have a good idea of what you're easily able to check given standard code formulas and where the limit states may not have direct code provisions.

4) In this case it's the web failures where you don't see directly relevant code provisions. However, the failure modes for the web will end up being very similar to what you see for shear connections to a rectangular HSS Tube section. So, that is a good place to start.... Trying to understand the underlying theory behind those HSS code provisions and if they can be modified to work for a wide flange web instead.

5) I have a bit of an advantage in this as I have attended many of the AISC committee meetings over the years. So, I've got alternate code formulas that never got adopted or proposed new formulas that will get adopted in the next code cycle. Other people have done most of the heavy lifting and I just need to review / understand what they have done.