RBS or WUF-W Flange Force 1

[T_Bat]

Hello everyone,

I have what may be a simple question. I'm looking at designing some moment connections for an Intermediate Moment Frame. I'm between using and RBS or WUF-W connections (I've asked for fabricator input). One thing I've noticed is that the design examples I've found don't check the beam flange's ability to carry the flange force from the design moment. For example on a standard R=3 direct welded moment connections with a bolted shear tab I always verify that the beam flanges will be able to actually carry and transfer the flange forces resulting from the design moment to the column. The RBS example in AISC 341-10 does not consider this failure mode. This may be, in part, since the web of the beam is CJP welded to the column. If a bolted web connection is used in an RBS wouldn't the flanges be required to develop the couple from the "probable maximum moment"? Same question applies to the WUF-W.

Hopefully this is a simple answer and I just am missing something.

Thanks in advance!

 

[T_Bat]

Well after some thought and discussion with some other engineers I think I have reasoned out how to approach these. The typical connection design checks must all be satisfied for the actual applied loads on the connection. Then in order to prequalify the connection I need to satisfy the requirements of AISC 358. If the prequalifications do not explicitly require the flange yield check then it can be neglected for capacity design.

Basically design for applied loads then modify only as required by the explicit prequalifcation requirements of AISC 358.

Thoughts/comments?

 

[sandman21]

If you did not weld the web you would not have a RBS or WUF-W. Other configurations have different requirements, a BFP will have a check of the plates for the moment. I dont understand your second post. To utilize a pre-qualified connection you are requirement to comply with the connections in 358. The connections will develop the proper plastic moment of the beams. If you did not provide the welded web you would not comply with the pre-qualified connections and would need testing to determine acceptance. You are not checking the applied loads you are designing the connections for the plastic moment of the beams.

 

[Deker]

Seismic detailing requirements aside, a welded-flange bolted-web connection will generally be able to develop Mp of the beam (Link). If you reduce the beam section, even better since you get a corresponding decrease in the moment you need to develop. With seismic, however, we also need to consider ductility. Welded-flange bolted-web connections have a mixed track record of being able to achieve the 0.04 rad rotation required for SMF connections (see the commentary to AISC 358-10 5.6). As such, the RBS with a bolted-web connection is only permitted for IMF. If you are designing the frames as IMF, the RBS with a bolted-web will generally be cheaper than the WUF-W which requires the welded web. Be sure to check with your local fabricators, though, as the preferred connection type tends to be regional. In SoCal, the RBS is the most prevalent non-proprietary connection, whereas the WUF-W seems to be more prevalent in NorCal.

 

[T_Bat]

Thanks - these are IMF's which do not require welded webs to be prequalified. The fabricator's preference is the RBS with bolted webs. But this goes back to my issue - if I am supposed to design the connection for the "maximum probable moment" it seems that there is almost no scenario where a couple between beam flanges only can develop this moment. I understand that the load is decreased by the flange cut at the RBS but the moment at the beam/column interface is plastic moment at the RBS plus Vrbs*Sh (EQN 5.8-6). This is in almost any reasonable scenario much higher than what you can get from the flange yield only.

I guess I'm saying there are two design processes to satisfy:

1) Design the entire connections for any applied load from analysis. This would include any checks for the flange force resulting from all moments from design loads.
2) Analyze/verify the connection from (1) satisfies all requirements of 358 (which do not include a check of the flange force. At least from what I can see (pgs 9.2-16 through 9.2-19 from AISC 358-10).

 

[Deker]

I understand that the load is decreased by the flange cut at the RBS but the moment at the beam/column interface is plastic moment at the RBS plus Vrbs*Sh (EQN 5.8-6). This is in almost any reasonable scenario much higher than what you can get from the flange yield only.

You are permitted to use the full Zx of the beam, not just the flanges. The rationale is explained in the link I posted above. If you still can't make it work, you need to adjust the dimensions of the RBS cuts.

I guess I'm saying there are two design processes to satisfy:

1) Design the entire connections for any applied load from analysis. This would include any checks for the flange force resulting from all moments from design loads.
2) Analyze/verify the connection from (1) satisfies all requirements of 358 (which do not include a check of the flange force. At least from what I can see (pgs 9.2-16 through 9.2-19 from AISC 358-10).

Step 1 is unnecessary for the design of the connection (with the exception of panel zone shear). Once you size your frame members, jump straight to the mechanism design outlined in AISC 358.

 

[T_Bat]

You are permitted to use the full Zx of the beam, not just the flanges. The rationale is explained in the link I posted above. If you still can't make it work, you need to adjust the dimensions of the RBS cuts.

I agree - the whole point of capacity design is to utilize the full plastic capacity of the section. My issue is if I have to design a connection to develop the full plastic capacity - I can't do that with only welding the flanges.

Step 1 is unnecessary for the design of the connection (with the exception of panel zone shear). Once you size your frame members, jump straight to the mechanism design outlined in AISC 358

I think we may actually be agreeing here...? For clarity I'm saying that regardless of the seismic design provisions, my beam and connection have to be able to adequately carry the actual applied loads from analysis (combinations of dead, live, wind, and seismic loads from an equivalent lateral force procedure or other EQ analysis). The capacity design from AISC 358 is in addition to this analysis. I believe you confirmed what I'm saying - once I'm doing the AISC 358 stuff the only "conventional" connection design checks are for the panel zone shear and column flange stiffeners.

 

[Deker]

My issue is if I have to design a connection to develop the full plastic capacity - I can't do that with only welding the flanges.

Did you read the link I posted above? Or the commentary to AISC 358 5.6?

 

[T_Bat]

I have (or just did whoops). The article by Dowswell/Muir is exactly what I was looking for. I guess I still stand by my previous statements then. The beam connection has to be able to develop the loads from a "regular" analysis and then I have to make sure it's prequalified. Although I guess the design check for a flange yield limit state never really applies, at least according to the MSC article.