Gusset bolts and welds

[CTSeng]

Butting heads with a fabricator over the horizontal bracing connection shown on the attached detail. The gusset is welded on one edge and bolted on the other. I’ve told them they cannot mix bolts and welds on the gusset. Am I off base here?

 

[nutte]

Yes. There is nothing wrong with what they're doing.

 

[HgTX]

There's nothing wrong with what they're doing, because, if I'm reading your drawing correctly, they're not mixing bolts and welds in the same connection. The weld connects the gusset to your horizontal member, and the bolts connect the gusset to the intersecting member.

If I'm reading your drawing correctly.

Hg

Eng-Tips policies: faq731-376

 

[JAE]

I think there is something wrong with the connection if you are in a high seismic area. The weld vs. bolt thing is OK (I agree with the others).

But the gusset should be detailed with a distance back from the ends of the gusset connections per AISC. This allows out of plane plastic buckling to occur. See the attached detail from AISC. It doesn't appear that your detail does this. But as I said, if it is low seismic, then it isn't required.

 

[DRC1]

I think you are right not to accept the detail. The bolts supporting the vertical moment will strin more under a given load than the weld sifting more load than anticipated to the weld. Bolting the diagonal member is fine.

 

[CTSeng]

Bolting the diagonal is not my concern, of course I'm ok with this. But looking at the gusset which connects the diagonal to the node, all the load would go thru the stiffer element (the weld) with little or none to thru the bolted edge? Those who say it's ok, how would you distribute forces between the welded edge and bolted edge?

 

[WillisV]

Gusset connections utilizing combinations of bolts and welds at the different interfaces are VERY common. Take a look at page 13-4 of the 13th ed. manual - in particular (b) gusset free body - bolts on one side, weld on the other. This is done for field fit up (weld the gusset to the beam in the shop, bolt to the column in the field). You appear to have a horizontal gusset, but the idea is similiar. I would distribute the forces based on statics - not stiffnesses.

It's when you have combinations of welds and bolts to transfer force across the same plane that you have to be careful (for instance if on one side of the gusset you had three bolts and 6 inches of weld - then you need to be careful to use SC bolts etc. to share force between elements.

 

[bootlegend]

AISC's Uniform Force Method gives you equations to calculate interface forces for gusset plates based on the stiffness. See page 13-3 of the 13th edition. You can then size the weld or bolts for that force.

 

[nutte]

WillisV, that's the diagram I was picturing. This situation is as common as dirt, yet there are people here saying it should be avoided?

Slightly off-topic, you mention using slip critical bolts to share the load with weld in a common shear plane. The 13th edition specification took this out. That has been the practice for a long time, but it was deliberately changed, as it has been shown to yield unconservative results in certain cases.

 

[WillisV]

@nutte - yup I forgot that was pulled for new connections - I have generally used that kind of distribution only in existing structures anyway where its still allowed though.