Truss Frame - Sports Arena Model in SAP2000

[TB_STR]

I am working on a feasibility project to determine if the existing framing in a sports area can handle additional loading of a solar PV system on the roof.
I have created a model in SAP2000. The building is 150 ft wide by 250 ft long. The main trusses span the 150-ft direction and are supported by columns. Transverse trusses provide horizontal bridging between the main trusses. These are fairly heavy members. The top chord of the main trusses is a WT6x110.5 and the bottom chord is WT6x120.5. The roof is metal deck with joists to transmit roof loading to the main trusses. Due to the bearing depth at the joist ends, the transverse bridging is 2.5 inches below the bottom of the metal deck so is not transmitting roof loads.

The transverse bridging trusses consist of a top W16x36 and bottom W14x34 with L3x3x 1/4" diagonal angles at each bay. The L3x3x 1/4" angles are failing and I am not sure why. They fail under dead load only. The building is standing and being in New England has handled snow loads. I am inclined to think that the L3x3's aren't contributing to the load carrying system and were maybe used to stabilize the W-beams during construction. Would it be reasonable to take out the L3x3's from the model and if the other members pass than the system can handle the additional loading. These aren't zero force members though. This is an analysis of vertical loading only. There is a separate lateral bracing system for wind loading. I have not been able to determine why the L3x3's are failing and a couple W16x36's randomly are coming up with errors. All loads are modeled as distributed along the main trusses and joists and metal decking checked separately. They can handle the additional loading. The main trusses pass under additional loading. L3x3's failing under dead load can't be.

I have attached the model if anyone has SAP2000 and can look at the stresses and error messages - L3x3s diagonals are failing in buckling under dead load. So I am confused as to what to tell the client as to whether the system can handle the loading and/or whether or not the SAP2000 model is correct.

 

[BAretired]

Can't read your file. An L3x3x1/4 is not a very robust member. Perhaps it is just too long to be considered suitable as a brace by SAP2000.

BA

 

[KootK]

I didn't review your SAP model. Too lazy. I'd check out some screen captures if you posted them however. In what year was the facility constructed?

For now, my thoughts are as follows:

1) the original designer may have intended the bridging system to act simply as a load distribution member(s). It sounds as though you've come to that conclusion yourself as well. Since the question of just how much distribution is needed is somewhat arbitrary, so then is the calculated demand on the bridging system. In a model containing an infinite number of primary trusses and uniform loading, the demand on the bridging would be zero. So something about your model must be causing primary trusses to deflect differentially relative to their neighbors or to boundary conditions such as perimeter framing lines. Does that resonate with any of what you've got going on?

2) If your cross bridging system entails a X-braced panel between each primary truss, then the webs in compression could all buckle and you'd still have a valid truss in play to distribute any differential loading or movement in the primary truss system. Of course, not having viewed your SAP model, I'm not sure if you've actually got X-braced panels. Might just be conventional trussing.

I am inclined to think that the L3x3's aren't contributing to the load carrying system and were maybe used to stabilize the W-beams during construction. Would it be reasonable to take out the L3x3's from the model and if the other members pass than the system can handle the additional loading.

In general, I agree with this sentiment. I'm skeptical of the erection bracing hypothesis however. It all sounds pretty robust for erection bracing.


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.

 

[JedClampett]

I haven't looked at your model, because I can't. But an easy error to make (because I've made it myself) is to model a double angle as a single angle.

 

[TB_STR]

At BAretired - it is too long to be considered a brace.
I am getting the following error "fa > Fe (ASD H1); Design can not proceed" and a warning message: "kl/r > 200 (AISC-ASD B7, AISC-ASD SAM 4)"

I uploaded a file of the main truss with an overstressed angle. This, I believe is due to sloped ends of the building as only the two end trusses show this angle failing - but only on one side. The trusses are symmetrical.

Also uploaded the transverse truss with the brace failures and some 16x36 member failures.
They span between joints in the main truss frame and the members of the main truss vary.
The 16x36 are framing into the main WT compression member - I wonder if it is modeling it as the vertical angle beneath the truss joint as a support.

 

[TB_STR]

Here is the truss frame with one angle failing.