Trusses at Non Load Bearing "Partition" Walls 3

[conradlovejoy]

I have always had an issue with my company's lack of detailing the interaction (or what I think should be a lack thereof) of floor and roof trusses passing over non load bearing (NLB) walls. We dictate which walls in a structure are to be LB and, without making any special recommendations, seem to assume that if the contractor erects the truss in a way that it is in contact with a non structural element like a partition wall that there will be negligible effect. I tend to disagree. If there is a 30 foot truss span, the space is split down the middle at by a partition wall, and the truss bears on all three walls, the middle reaction at the partition wall is greater (about 25%) than the two reactions at the walls that are supposed to be LB! Even in the case that the truss was erected such that it was slightly above and not in contact with the NLB wall, if it deflected enough after the building was occupied and loaded, then the reaction will immediately occur upon their contact in my opinion.

In attempt to remedy this, I drew up some details that could be used to ensure the framing elements are not in contact with partition walls but are still connected to the NLB walls via slip track connectors and someone above me "poo-poo'd" the idea insinuating that not designating those walls as LB is good enough and that somehow the lack of vertical members in the truss directly above NLB walls basically causes no reaction to occur. When I pushed on that notion (because I completely don't agree) it changed slightly to the notion that if they were in contact the worse that would happen was that the bottom chord of the truss would deflect a little bit. I remember walking through truss design in Structural Analysis but I don't remember anything like that notion. I wouldn't treat a truss exactly like a beam for precise design, but for something conceptual like this I think its safe to assume that the reaction in a three span condition occurs similarly to a more typical beam.

Another issue I have with it is the "internal" forces within the truss. The truss web members have a very specific "layout" of tension and compression members. I am fairly positive that introducing a reaction that wasn't intended/included in design can screw that entire system up.

Am I getting this wrong, somehow? I feel this is a bigger deal than its being treated.

 

[HK2013]

We have had this discussion in our office as well... Following.

 

[KootK]

The theory: you're 100% right.

The practice: problems arise infrequenly enough that detailing for the movement may be viewed as a costly solution to an imaginary problem. That's how it's been in the markets that I've practiced in so far.



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.

 

[mike20793]

I always use details to show separation and clips between non load bearing walls and trusses. On a recent project with precision end trimmed studs, the contractor suggested using a single top plate at the non load bearing walls so all the studs could be the same length and not have an issue at non load bearing walls. We approved with a few caveats, but it worked well. There are a wide variety of slotted connectors that allow the walls to be braced off the trusses, yet let the truss deflect without loading up the wall. It should be a typical detail in all wood buildings, in my opinion. In multistory wood buildings, the lack of load path under non load bearing walls often becomes a serviceability issue leading to cracks in drywall.

Note: The fact that the contractor proposed an alternate solution that he had used many times before, shows that it is fairly common practice in my area to have separation between trusses and non load bearing walls.

 

[XR250]

I'm with KootK on this one. I have never seen this cause a problem in practice and I have inspected a a huge number of buildings in the last 20 years.

 

[msquared48]

I have seen problems with this in differential deflection issues where not all trusses of the same span see the non load bearing walls though.

This is not an imaginary problem.

Mike McCann, PE, SE (WA)

 

[TXEng-USA]

My advice would be to include the detail and let the contractor propose an alternative if they like. In our market, the good framers will ask for this even if it is not shown.

 

[OldBldgGuy]

It is standard practice in our area to have a slip connection between NLB partitions and trusses, but the reason is the opposite of what is discussed here. Deflection of the trusses is upward in the winter, sometimes very drastically, and if the trusses are connected to the partitions they will lift them off the floor.

 

[XR250]

Deflection of the trusses is upward in the winter, sometimes very drastically, and if the trusses are connected to the partitions they will lift them off the floor.

Now I have see that happen many times in practice. It either picks up the walls or gaps at the top of the wall. No one in our area does any special details for this. My guess is that even with a slip detail, unless the sheetrockers are on board and the trusses do not move too much, cracks will still occur at the roof wall junction.

 

[John2025]

XR250,
I'm curious what the sheetrockers can do. I've always wondered what happens in these situations where the truss can move on a clip, but the drywall can't.
John

 

[KootK]

[link support.sbcindustry.com/images/publication_images/ttbpartsep.pdf]Link[/url]
Link

The main ingredient is, of course, a sympathetic drywall team.

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.

 

[John2025]

Learn something new every day. Pretty simple solution once you see it.

 

[KootK]

I'm actually more intersteed in the finish details for the case of downwards truss movement. Is holding all the mouldings down 1.5" the only lever available for pulling.

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.

 

[John2025]

It seems like it would still work if they keep the clips no higher than the top of the top plate.

 

[VTtoVB]

The vertical load on door headers in NLB walls can cause doors to stick during high roof loads, such as heavy snows. In one case with sticking doors, the owner discovered the CFS installer screwed through the slip track. Even thought there was no structural damage, it caused building occupants to worry about the strength of the roof.