Truss to Beam Connection

[PatBethea]

I've got steel trusses currently sitting on top of a concrete header. The trusses span 60 feet. The roof is curved. The bottom chords of the trusses are 4" channel oriented so that they are sitting on their 'legs'. The bottom chord currently rests on the concrete header - the two are not connected.

I had a SE design a new steel beam and column system, but he failed to address the connection of trusses to the beam. According to his numbers there is an approximate 7300 lb uplift reaction per truss while the deadload is about 5300 lb. Two-thousand pounds of force isn't much to counter. I figure I can just run a fillet weld down one side of the truss (the beam has a 10" flange)and be done with it - the only shear should be from thermal expansion.

Is there anything else to consider? What bothers me is the fact that the truss is not currently connected to the header. The building was constructed in 1927 and I realize they didn't have the same building requirements, but is there another reason why the truss would be allowed to 'float'?

 

[structuralaggie]

"floating" is not a good idea, but allowing lateral movement can be. Is this a scissor truss?

 

[structuralaggie]

Or maybe a truss with a lot of lateral defelction?

 

[PatBethea]

Structualaggie - no it is not a Sciccors truss.

I also don't think it sees any lateral deflection. This is Florida, so there are no seismic conditions. I would therefore assume that lateral loading would be due to wind. The front and back of the building are constructed of concrete block which extends above the roof line by a couple of feet. I assume the walls take the lateral load and would be surprised if the trusses see any lateral deflection. But then, I'm a ME and not a SE, so I may not understand this quite right.

 

[MillR]

If the roof is curved, it's likely the trusses do have lateral deflection.

 

[twinnell]

They are probably barrel trusses....the bottom chords are flat....no thrust.

 

[structuralaggie]

PatBetha - if they have a flat bottom chord like twinnel says then there probably is not much lateral deflection. If the geometery of the truss is such that horizontal deflection can occur, vertial loads on the truss will cause the truss to push horizontially against the top of the wall under vertical loading. It is not a desireable condition.

If there is not lateral deflection due to gravity loads, weld that sucker.

 

[PatBethea]

Yes, I Googled "barrel truss" and it appears that this is what I have.

Will welding the "barrel trusses" to the beam be a cause for concern? If so, are there any suggested alternatives?

 

[PatBethea]

Thank you, StructualAggie.

 

[structuralaggie]

The only problem I can think of is if your old truss was made out of weldable steel. Are the connections bolted or riveted, or are they welded? I don't know much about old steel, except that some of it had lower carbon content than todays stuff and may not be weldable or may need some special type of welding.

 

[twinnell]

One thing to be careful of is that nobody in this forum has seen any structural drawings for your building, nor have they gone to the site and perfomed an investigation. They only answer what you ask....can you weld the truss. This building is in Florida and you have barrel trusses supported on a steel beam. How is the roof diaphram load transferred to the lateral system? What is the lateral system? Welding the truss to the beam is probably fine, but, how is the lateral load transfered out of the roof?

 

[PatBethea]

All of the truss connections (within the truss) are welded.

I'm a little concerned about the age of the steel, mostly for the same reasons you mentioned. Welding seems to be the most attractive option, though. I don't want to cut holes in the bottom chord for through-bolting, and I really don't want to design (and have fabricated) a unique hold down clamp that would bolt to the beam. If there's another (viable) option, I would like to hear it.

 

[MillR]

I think you ought to discuss this with the structural engineer that designed the beam. He'll understand all the complexities and possibly have an explanation for not welding down the trusses in the first place.

 

[UcfSE]

Note that the design uplift will be more than 2000 pounds. You need to use the corret load combination to calculate this.