Non-Load Bearing Partition Wall Top Connection

[Greatone76]

We have a project with non load bearing wood partition walls on and under wood trusses. We detailed the Simpson Connection that would support the wall out of plane, but had vertical slots to allow for deflection of the roof and floor trusses, so the non-load bearing walls would not become load bearing. The arch/contractor has come back and said this is not standard framing and want it removed from the drawings set. I know in a house you would just attached the non-load bearing wall. This is a 3 level multi use (retail, office and residential). What is your opinion on needing and/or using the vertically slotted top connectors for non-load bearing walls in what application and in what types of buildings. Thanks in advance.

 

[jdgengineer]

I would defer to the truss manufacturer, who will likely require it. That way you aren't the bad guy...

Unless their trusses can handle the point load from the partition walls, and you have a load path below the partition walls that can handle it.

 

[Lion06]

I would ask for it. Even if the truss manufacturer says they will design the truss for the third support point, you now have to design the floor framing to take that load.

 

[DaveAtkins]

I would not require the slide clips. The contractor is correct--this is not typical in wood construction. And if a non-load bearing partition tries to accept some load, is there really a problem? I have not heard of any issues (such as cracked GWB) in this type of structure.

DaveAtkins

 

[emmgjld]

I have seen too many small problems with multistory residential & commercial traced back to this specific detail (more properly lack of detail). I have even seen a couple of nasty failures (trussed roof & floors), requiring extensive repairs.
I would recommend you get the truss manufacturer involved. Between the two of you, the argument is MUCH better.

 

[mike20793]

I will disagree; you absolutely need the truss clips at the top of the partition walls or some other way to brace the walls without letting them take load. This is very common for multilevel wood buildings in my area. You must brace your walls and they cannot be loaded up by the trusses. A single 2x4 stud in a partition wall cannot handle 3 levels of load, even with minimal tributary area. Woodworks has a seminar about this very issue and how to properly deal with it.

 

[manstrom]

It is not common practice, but it is "best practice". The detail will prevent ceiling cracks due to differential movement, but it won't cause a failure.

Trusses won't really pick up the extra support since they don't have a node there. If a non bearing wall hits between 2 nodes in a bottom chord, the chord will flex and not transfer as much load.

You can CYA by saying, "okay you don't have to do it, but don't come back to me with any drywall cracks in walls or ceilings".

When I am working on a problem, I never think about beauty but when I have finished, if the solution is not beautiful, I know it is wrong.

-R. Buckminster Fuller

 

[Guastavino]

I would leave it as you have it. Load goes to stiffness. The "non-load bearing wall" doesn't know that it's not load bearing. With all of that said, I agree with Manstrom, it's best practice. Sometimes in these situations I email the contractor (so it's in writing) that they can do it how they want to, but I would recommend the other way. Then, if anything ever happened (in a serviceability sense) then I would point to the email and say, I didn't recommend it, and it wasn't a life safety issue so I gave them the option. In my opinion, it puts the liability on the contractor, and I'm OK with that.

 

[KootK]

Some additional thoughts:

-In the past, I worked in a market in the US where the slip connections where unheard of. No problems.

-I now work in a market where the slip connections are common place. So I take advantage of that.

-I worked for the wood truss council of america for a while and the only reported problem that would crop up was roof trusses actually lifting connected walls up into the air. Nifty.

-I'm skeptical of the argument that says that bearing walls won't end up at points of truss stiffness. Walk through a building in progress and you'll see all kinds of situations where non-bearing walls fall within a foot or so of truss bottom chord panel points. With floor trusses, your panel points may only be 14" apart so it becomes quite impractical to try to dodge them.

-As structural engineers, we love do drone on about how wood structures are redundant in all kinds of miraculous ways that, thus far, remain unaccounted for. Gap your non-bearing walls and a lot of that redundancy goes away in my opinion. Just sayin'...

@Manstrom/njlutzwe: with these written disclaimer emails, are these situations where the contractor is also your client? I would hate to find myself having to explain to an owner/architect client why I gave the contractor the liberty of choosing a construction method that led to shabby building performance.




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.

 

[Guastavino]

@kookK, great question.

Most of the time it's been in design build, where I work for the contractor. If I was working for the owner, I would inform the owner and let the owner make an educated decision. I would tell the owner that they paid for it and I recommend it, but if you want to ease up on the contractor and aren't worried about it, that's fine.

With that said, I only do this with what I deem to be serviceability issues. In those events, when I am working for the contractor it's his or her call. For example, I could argue it's best practice to design a floor for L/480 live load deflection to keep them from being bouncy, but the contractor may not care. It's not a code requirement, it's not a life safety requirement. The contractor can tell me he wants the code minimum and it's totally ok. It may not perform as well as the L/480 floor, but that's their call to make. Same with slab-on-grade control joints. I show a location that works. If they want to change it up, by all means, go for it. Ultimately, the contractor can decide on some of those things if I'm working for them. And, in reality, the owner is part of that decision, albeit passively. If you hire a contractor for design build, you do it assuming you want the bare minimum, except where you specify elsewhere. Contractors aren't in the business of giving things away, and owners know that. So if they want it designed based on best practices, they probably aren't going the design build route.

Just my two cents. I'd love to hear your response and how you deal with these issues.

 

[KootK]

Great summary njl, thanks for providing it.

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.

 

[Greatone76]

emmgjld - do you have pictures of a place I can get some pictures of such issues???

 

[Triangled]

On a job right now where some kind of connection as described above may have prevented the excessive deflections at the tip of cantilevered mezzanine joists supporting a "non-bearing" wall hard-framed to the underside of roof members.

 

[BS2]

I have seen a similar detail cause some fairly significant issues. In one case, a high end two story home had roof trusses that loaded a non-load bearing wall to the point that it caused notable and excessive localized deflection in the floor below. The contractor (also the homeowner) built the walls tight between the trusses before installing the concrete tile. The fix for this situation was not simple. I would definitely say this is best practice and even an essential practice if you anticipate significant dead load. At the very least, installing the non-load bearing walls after most of the dead load is added would help the situation. Although, I know this is not how they would build it.

 

[XR250]

I have never worried about in wood construction. As KootK mentioned, the only issue I have ever witnessed is the truss picking up the wall.

 

[emmgjld]

Greatone76;
I will look. The best/spectacular failure was in 1977, I may still have photos.
Most cases I have seen do not photograph well & many other excuses are produced, clouding the issue. Too many times our expansive & collapsible soils are blamed for any cracking or structure movement.

In my experience, most cases are related to heavy (tile) roofs & heavy (tile or concrete surfaces) floors. Complicated trussed roofs seem to factor in. The problem has decreased some since the mid 90's, when the trusses were 'better' designed (better programs), constructed better (better equipment) and the snow loads in Colorado were better appreciated in some of the low load areas.

 

[thaidavid40]

We have seen problems for commercial, multi-level construction, where no provision was made for floor deflections above non-structural partitions. The biggest effect we've noticed is service-level cracking of wall finishes and floor at these levels. For that reason alone, we ALWAYS detail for floor and roof movement independent of the partition walls (while providing for bracing them at the top, of course).
Dave

Thaidavid

 

[medeek]

After reading through all of the comments on this thread I am no more the wiser on this subject, it seems like the opinion on how to handle this sort of thing is all over the place with no overall consensus.

Logically it would seem to me that the slotted connectors would allow for relative movement, would that not increase the likelihood of the GWP cracking?

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[dcarr82775]

So long as the drywall contractor provides little isolation joints between the top of the partitions and the ceiling it theoretically works fine, but looks a little weird. It they don't, you probably get cracking worse than you might get from no gap. I think it really comes down to local practice.

 

[mike20793]

I'm with thaidavid40 on this one. I've seen it crack up drywall in several multistory buildings, so we always provide a detail showing the slotted truss clips at partition walls.