Double and Bolted Welded Connection with combined shear and Axial force. 2

[amain002]

Hello fellow engineers,

I am having some trouble trying to analyze the double angle bolted welded connection with both axial and shear force acting on it. Please see attached file for connection description. I was not able to find much information in the AISC manual, so hoping that you guys could help me out.

I am using a double angle connection bolted to the beam web and welded to the embed plate. The connection sees both axial and shear force on it. I can calculate the resultant force and check the capacity for the bolt for shear, but however, I am not sure how to check the weld strength. The weld experiences an eccentric loading and the angle will experience prying effect. Angle thickness will be designed to account for the prying effect.
Can anyone refer me or guide me on how to design/check the weld capacity for this connection. Since the axial load is high, it will see more tension than shear. Generally, Shear plate is preferred for this connection, but my axial load being high, shear tabs does not work.
Any help will be greatly appreciated. Thank you all.

http://files.engineering.com/getfil...c-4061-98ae-664ca7879a5d&file=Beam_Detail.pdf

 

[rb1957]

not really my area, but that's never stopped me ...

i'd consider the welds to be in pure shear , and then the bolts transfer the shear and the offset moment into the I-beam ... combined shear on the bolts.

another day in paradise, or is paradise one day closer ?

 

[Teguci]

With high axial, I'd think you'd want to return the weld on the bottom as well to prevent a notch on the backside of the weld. If you just have positive bending on the beam, then I'd return the bottom weld the full width of the angles. It'll make your tension reaction a bit eccentric, but should hold those angles from rotating against the fillet welds on the sides. A little concerned with the bottom weld unzipping, so it should be checked.

 

[nutte]

Don't do it. This connection, double angles bolted to the beam web and welded to the support, should not be used to resist axial load. Get a shear tab to work.

We have discussed this at length in the past:

http://www.eng-tips.com/viewthread.cfm?qid=358106

See also Bill Thornton's response under "Double-Angle Connections":

http://msc.aisc.org/globalassets/modern-steel/steel-interchange/2011/092011_sept_si.pdf

 

[JoshPlumSE]

If you have axial load on that connection, the weld itself will go into tension which is problematic to say the least.

We ran into this same issue when we added the ability to consider axial force to our shear connections four our RISAConnection program. What we did was require that the entire bottom edge of the connection be welded also. That actually changes it into a very different connection.

Even with that change we got a lot of criticism from users (on one of the Eng-Tips forums) about the connection. And, we eventually added a couple more design checks and limitations that would restrict how much axial load would be allowed in the connection. I'll try to find that thread and post a link in a follow up response.

 

[KootK]

I like a seated connection with a single, stability clip angle for this if space permits. The axial load in the beam becomes eccentric but it's worth it for the connection simplicity.

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.

 

[rb1957]

i missed the axial load ... yes, welds in tension ... bad !

could you weld and bolt the flange (so the bolts take the axial) ? i know (I think) that welding and bolting is bad/disliked practice).

how good is that "embedded" plate, in carrying out-of-plane loads ?

the I-beam has the shear in the web and the axial in the caps ... separately attach each loadpath (angles top and bttm) ? you could weld the three angles to the I-beam and bolt to the plate ? I guess welding to the plate is easier ?



another day in paradise, or is paradise one day closer ?

 

[JoshPlumSE]

Looks like Nutte beat me to it....

Note:
This previous thread isn't my greatest moment. It takes me quite before I understand the basic argument against the type of connection everyone was talking about. So, ignore most of what I say before my Jan 20, 20:05 post. It took me that long to really figure out what everyone else was talking about!

I should point out that this is still NOT the type of connection that is in the RISAConnection program, RISAConnection automatically adds the bottom edge weld whenever there is axial loading. Even then, we have now added in those restrictions I talked about towards the end of the thread. An automatic fail when the axial load gets above a certain amount and a new limit state for "angle leg bending".

 

[amain002]

Thank you all for your great comments. It was really helpful.

Nutte/Josh, the tread was very helpful.

So the conclusion is that double angle welded to the column/girder/embed plate supporting axial load is not really a good idea. Reason: the weld on the OSL of the angle will be subjected to a tension force (not recommended), thus connection with "AXIAL LOAD" on DA welded to support should be avoided.
I guess I will be going back to shear tab with larger bolts

Thank you all.

 

[amain002]

However, one more question.
Say I have a shear tab connection with 10 kips shear and 50 kips axial load. Will the weld on the plate not be subjected to tension loading? How would you consider or find the capacity of the weld in this case. Just a thought.

 

[KootK]

Welds can resist tension. The issue is really weld roots experiencing very high tensions as a result of bending demands being placed on welds. This article explains it much better than I ever could: Link. For the shear tab case, it's really just a vector sum demand on the welds.

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.

 

[andriver]

See this link for a weld check subjected to an eccentric load. I think you are asking HOW TO do a weld check?

http://www.bgstructuralengineering.com/BGSCM13/BGSCM005/BGSCM00502.htm

 

[CBlaker114]

Agree with KootK on this one. It's not the tension in the weld that is the issue.. it's the fact that there's an eccentricity equal to the length of the angle leg causing torsion through the longitudinal axis of the weld (from the axial load in the beam). Without the weld return.. it holds practically zero load.. with this connection, you're asking the weld return to resolve that torsion through tension which won't be pretty.

When using the shear tab, you can just find the resultant force of the shear and axial and size the weld for that value as there's no eccentricity.