Web compression buckling 2

[edled]

I'm currently working on a bolted flange plate moment connection for a single story parking garage. I'm just out of college so I've been cutting my teeth with this connection; it's the first time I've had extensive experience with any real connection design, actually. We were originally were utilizing IMF's, which was some serious teeth cutting, but have since reverted to our standard R=3 criteria. I've gotten to the last step of creating my BFP excel spreadsheet, in which I check the need for stiffeners or doubler plates. Thanks to Design Guide 13, I have a good grasp on what's physically going on.

I have a question about the web compression buckling limit state however. At many of our connections, we have between 200-300 k-ft of moment coming in from one side, with sometimes <100 k-ft coming in the other side. As far as I understand, the difference in compression flange forces is made up through web panel shear. What "P" load should I be checking versus the design strength of the web? Conservatively, I realize I could use the larger flange force. Just wondering what your opinions were on the matter. I was surprised I could not find any information on the internet, I assume that in most cases you have different strength flange forces opposing one another?

Thanks,

Ed

 

[rb1957]

i've really very little idea about what you're talking about. however, if i understand your 2nd para ... you have a bunch of moment loading the beam on one side, and some opposite moment on the other ... so what happens to the difference ? to me, it sounds like the nett moment is torque on the beam ?? which could be reacted as a couple between the caps

 

[edled]

Right, there is a net moment, which then goes into the column itself and is resolved down at the base plate. What I'm concerned with is checking the buckling limit of a web loaded with two different axial loads coming from two different flange forces.

 

[JoshPlumSE]

Web compression buckling applies to a "pair of compressive forces applied at both flanges of a member at the same location". So, I think you can argue that the code allows you to use the lower of the two forces....

That being said, I have always used the higher of the two flange forces. I don't mind being conservative. I'm in seismic country, so I'm pre-disposed into thinking that we're going to want continuity plate stiffeners anyway.

 

[hokie66]

For the situation you depict, I don't check it. I just put stiffener plates in.

 

[TLHS]

Yeah, I'd say it's good practice to include stiffeners in this kind of connection. Then you know you have a nice stiff load path, your checkers know that there's a load path, it's clear what the connection does for anyone who comes along later and you've got those stiffeners there to hold your geometry in case a hinge actually ends up forming there.

While I know it *can* work given the right loads, I'm not a huge fan of the load path involved in a W section to W section 90 degree moment connection without stiffeners.

 

[edled]

Thank you all for the advice. Still getting a feel for what's standard practice.

-Ed

 

[SteelPE]

I guess it depends on what your definition of standard is. In my area many engineers don't bother to design their connections, they leave this up to the fabricator. They provide a bunch of standard details they want you to use. Then, if you can proved that their column will work for the loads they provide w/o reinforcement they will still want you to use their detail.

From what I have seen 99% of the time engineers require continuity plates in instances like this. I have only seen one engineer on a recent project even considered these forces when he designed his building.... and he increased his column size to eliminate the need for reinforcing at these connections.