Scissor Truss Failure

[AnimusVox]

Hello everyone, I'm currently doing some forensic analysis of a scissor truss that has failed. The truss was built in 2008. I've duplicated the truss manufacturer's truss model in RISA 3D and have applied all of the same loads, and the vertical reactions are similar enough that I'm reasonably confident they're similar for the purpose of analysis comparison.

During my modeling I encountered something unsettling - The manufacturer modeled the end conditions as a pin and a roller. I tried this in my analysis and the lateral deflection was ~1.5 in (theirs was 0.83 in). Shouldn't the end condition be pinned-pinned, as the ends of the truss are being nailed to the wall?

 

[KootK]

Pin-roller is very standard. When the ends of the truss kick out laterally, the bearing walls usually just go along for the ride.

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.

 

[RWW0002]

A pin-pin condition for scissor trusses will yield very high thrust values (horizontal reaction at bearing) and may lead to unrealistic truss behavior in your model. The lateral deflection for a pin-roller condition would be divided between the two exterior walls as they deflect outward.

The most realistic way to model the condition might be to set the scissor trusses on columns (height = wall height). The base of the column may have to be fixed with a weak spring of some sort to make the model stable, but it should give you the best estimate of actual truss/wall behavior

I generally try to force the truss supplier to stiffen the truss such that the lateral deflection at the walls is limited to around H/200.

 

[SteelPE]

I would model the truss as it is installed in the field. If you have a pin-pin you would have to figure out a way to support the lateral thrust and since these things typically sit on perimeter wall that is almost impossible to do. SO pin-roller makes sense and is the way that I would approach the analysis.

With their deflection being .83" and yours being 1.5" are you sure that theirs is 0.83" on each side? After all .83"x2=1.66" which is almost 1.5"

 

[Guastavino]

Pin-roller, unless of course you have something to resist the thrust. I've never thought of scissor trusses as a good idea because of having to deal with the thrust aspect.

 

[msquared48]

You have to deal with lateral spread due to deflection with any truss, just a lot more with scissor trusses.

Mike McCann, PE, SE (WA)

 

[RPMG]

Pin-Roller. Otherwise, you can add springs or model walls/columns.

Their lateral displacement may be 0.67" Dead + 0.83" Live = 1.5" Total. The manufacturer might not report the dead load displacement because it's permanent and only relevant from a force-stiffness standpoint.