steel special moment frame

[bjb]

I have a steel special moment frame that I am trying to design according to FEMA 350. The connection type I am considering is the Welded Unreinforced Flange-Welded Web (WUF-W) connection. According to fig. 3-8, the weld of the shear plate to the column flange is symbolically shown as a partial penn. weld with reinforcing fillet all to the near side, but in the notes it says this weld is from the far side. Am I misinterpreting the weld symbol in Fig 3-8, or is there a mistake in Fig 3-8? Welding from the far side does not make sense to me because it seems to interfere with the CJP weld of the beam web to the column flange per note 3.

Any help is greatly appreciated

 

[shemp]

I see the inconsistency that you have pointed out. The 2001 errata doesn't mention anything on this issue. I would tend to think that the partial pen weld should be on the far side like the footnote says. Placing the weld on the far side would help keep the weld more in line with the beam web. I don't think you would have a problem with the welds interfering as long as the edge of beam's web was far enough from the column flange to allow room for the reinforcing fillet weld (which is as thick as the beam web.)

I haven't looked closely at this connection until now. I've got a question. The paragraph on pg. 3-28 says that "web joints for these connections are made with complete joint penetration groove welds of the beam web to the column flange." That quote makes it sound like the beam web is welded directly to the column flange. That's not true, right? I would think that the load path for shear is like a typical connection: through the beam web, into the shear plate, then into the column flange. That's how you see it, right?

Another inconsistency with the quoted sentence is that it says "complete" joint penetration welds connect the web to the column flange, but then the figure shows a partial penetration weld. (the same one that you're not sure which side it's meant to be welded.)

Shemp

 

[bjb]

Shemp

Thanks for the reply. My interpretation is that it is desired to have a CJP connecting the beam web to the column flange. I think their intent is to fully develope the beam web to the column flange to minimize shear stress in the flange. In any event, this detail is not clear, and when I go to the FEMA webiste there is no guidance there. This is all very fustrating. I wish AISC would give us step by step procedures for these special connections with solved examples like they do for other types of moment joints.

I am a structural engineer from upstate New York, and we have only been required to design for seismic since january of this year. This detail most closely resembles the moment connections that were typically in use, so I was hoping to go with it. This is for a small one story fire house. What experiences have you had with special moment resisting connections?

 

[Taro]

Yes, the wording is confusing. The beam web is to be full-pen welded to the column per note 3. The shear tab is to be full-depth partial-pen welded with a reinforcing fillet. It seems the term "far side" refers to the opposite side of the shear tab from the beam web (my interpretation only).

 

[shemp]

bjb,
Just to be sure, the shear tab and the beam web are welded to the column flange?
I suppose this makes the shear connection extremely rigid in an attempt to take shears away from the flange welds.
I read a little article this morning where the author says that this type of moment connection still doesn't solve the Northridge problems. It's at this link if you're interested:

http://www.muckraker.org/image2/RRFundamental.pdf

So you've got to use the IBC instead of the NYC now, huh? I imagine that the Use Group III pushed you up into Seismic Design Category D, right? I don't have a lot of experience designing special moment frames. A job near the New Madrid fault zone required them so I used the bolted flange plate design. Another job that I'm working on, also in the same area, requires them.

Let me know if you run across any problems. One that I remember happened while checking stresses on the beam's flange for the bolted connections. I was designing the connection for the full moment capacity of the beam. But the beam's flange was too thin. So, I needed a bigger flange, but that gives you a bigger moment capacity. So, I got caught in that circle. Then I ended up with a bigger beam which required a bigger column (because of the strong column weak beam concept. Luckily you won't have to worry about the SC/WB concept b/c you only have a one story building.) Like I said, if you have a problem, post it - I'd like to hear about it.
Shemp

 

[bjb]

Taro: Thanks for your reply. My interpretation is similar to yours, but the confusion does not have me confident enough to use this connection on my project.

Shemp: In upstate (Albany), we never even had to deal with the NYC code, but now that we're IBC based we've been thrown into the deep end. New York State had it's own building code, which was decades out of date but easy to use. I ran into similar problems with the bolted flange plate connection. Next I tried the welded free flange connection, but the web plates and welds for the beam flange were enormous. I've abandoned them and the welded unreinforced flange connection, and am now trying the reduced beam section out. My problem is that for architectural reasons I am stuck with a column (w12x58) that isn't quite as stiff as I would like, so for drift considerations I had to go with a large beam (w24x76). This puts a lot of demand on the connection. I wonder if in the case of one-story buildings where you can form a plastic hinge in the column if it would be ok to just size the beam column connection for the amplified earthquake combination. For my case I don't see the logic in a connection designed to allow for a plastic hinge in the beam when I know that I'm not going to get one there, but in the column instead.

 

[shemp]

bjb,
Yeah, I did the same thing that you're thinking of doing... sizing the connection for the amplified load instead of the capacity. The force was much bigger than would ever occur - just like you said.

 

[CWIC]

There are a number of comments regarding the FEMA 350 document whcih can be viewed at:

http://www.seaosc.org/

in the scrolling banner (Announcments) on the right when you go to their site.

I my opinion all of the FEMA 35X series of recommendations are in need of review, recommendations and commentary by more than the SAC committee members. The weld symbols are incorrect for Figure 3-8. (Perhaps comments or suggestions could be incorporated into future errata?)

Don't get me started on FEMA 353...

 

[Taro]

AISC has started the review process already. They will eventually turn the SAC recommendations into standards. Additional testing of other connection configurations is also ongoing. The FEMA 350 series is by no means perfect, but it is far better than what we had before and what just about any other structural system currently has.

 

[bjb]

Taro, I am interested in your opinion on the following. In a one-story building when you meet the slenderness guidelines and don't have to meet the strong column/weak beam criteria, you can have a weak column, where the plastic hinge mechanism forms in the column. Therefore, you shouldn't have to design the beam to column connection for the expected capacity of the beam, but should only have to design it for the amplified earthquake combination.

Agree or disagree?

 

[Taro]

Disagree. Although it is acceptable to have hinging in the column for a one-story building, it is not a good idea to have the connections be the weak link.

 

[bjb]

Taro, I see your point. The connection has to be strong enough to force yielding in the column. In your opinion though, do you think the panel zone still has to be designed based on the expected capacity of the beam? This would seem overly conservative. If you design the strength of the panel zone to be sufficient to allow hinging in the weak column, but don't meet the expected capacity of the strong beam, that would seem sufficient. The case I am thinking of is when you have a large beam to help control drift, when architecturally you are limited to a smaller column.

 

[Taro]

After reading your previous posts that describe the beam and column sizes, I understand your problem. Going strictly by the book (AISC Seismic Provisions), you would still need to provide connections (including panel zones) that are sized for the capacity of the beam. However, you may be able to convince the building official that your situation is different.

Basically, you are "turning the connection around" so that the plastic hinge will form in the column instead of the beam. This column hinge will act as a "fuse" to limit the load that can be transferred to the connection and beam. The connection should then be designed for the expected flexural capacity of the column (including overstrength factors and projected moment magnification) instead of the beam.