Wood framed partition walls same height as bearing walls? 1

[jeffhed]

I want to find out what occurs in other areas of the country on wood framed partition walls. Here all bearing and partition walls are framed at the same height. This ends up leading to drywall cracking in multiple story structures where partition walls are located in the midspan of a truss. The truss can't deflect, and ends up "bearing" on the partition wall. This causes the floor joists or sheathing to support loads that were not originally intended. The floor joists then deflect more than intended causing drywall cracking in the ceiling on the lower floor. I have done a lot of distress inspections recently where this has been one of the causes for ceiling cracking. We specify that the partition walls be framed a fraction of an inch lower than the bearing walls to make sure the trusses have room to deflect without transferring load to the partition walls. However, I am positive this is never done in the field. I want to get some input as to what others do to preserve the intended load path.

 

[DaveAtkins]

I believe it is typical in my area for non-bearing wood stud walls to be framed just the way you describe--but I have not heard of there being issues with cracking of drywall.

I think Simpson makes some sort of slip track for wood stud walls, but I have never heard of anyone using it.

However, with cold formed steel stud walls, slip tracks are used all the time.

DaveAtkins

 

[jeffhed]

Dave,
The cracking I mentioned is only on large span trusses. The most recent one being a 39'-0" roof truss. The partition wall is located right at the midspan. This particular model has been built multiple times and in every unit there are drywall cracks right at the end of the partition wall. The crack runs parallel with the existing trusses which leads me to believe that the crack is from differential deflection of two adjacent trusses. One truss can't deflect because of the partition wall located at the approximate mid span, the other truss can deflect fully where there are no partition walls in the entire span due to a hallway. The builder wants me to come up with a way to construct the future units to prevent the drywall cracks. We could make this a bearing wall with a header, but it seems to me if we used the slip track from Simpson, this could solve the problem instead of spending more money on beams, footings, etc.

Jeff

 

[azcats]

The consulting firms I've worked for had standard details for non-bearing walls that utilized the Simpson DTC clip. You may still want to specify the way the ceiling and wall drywall meet though to let the ceiling board go down relative to the wall and not crush it.

 

[dhengr]

I’d want to see floor plans, framing plans and the truss layout plans and bldg. sections so as to understand the load path. 39' is a long truss with some less predictability in terms of possible deflection. I certainly would want the first full spanning truss a max. distance (full truss space) away from the end of the partition wall. Pay attention to sht.rk. joints in this area, and attach the sht.rk. so it flexes and tolerates the truss movement in this area. And, are you sure this isn’t actually a truss uplift problem, changing seasonally, if it shows up immediately below the trusses? Alternatively, lower level joists are not usually designed to pick up roof loads from partitions, so that could be a problem. But again, I can’t see the load path from here or the crack location.

 

[hokie66]

Where I am, the top plate of the partition walls are set 20mm below the truss bearing height. Then blocks are added between the trusses to both brace the trusses and develop global bracing forces if the partition is a bracing element. There still has to be allowance for this vertical movement in the drywall detail at the cornice.

 

[JAE]

Isn't your ceiling sheathing attached to the trusses AND to the vertical interior wall sheathing as well. You could put a slip track in there but the sheathing is "connected" to everyting through the mudding. If the truss over the wall deflects downward and you have a slip track, you'll now get cracking all along the slip-tracked wall.

You would need an expansion joint between wall and ceiling all along the partition wall.

Also, if this is a home based on the IRC or something simillar, the interior walls many times serve as the empiracle shear walls that provide stability to the structure. Slip tracking these walls reduces that lateral capacity and stiffness.

Better to design the trusses to be supported by the interior walls and add bridging where necessary to minimize differential deflections.

 

[msquared48]

I use Simpson DTC clips at the roof trues for both the partition and any internal non-bearing shear walls connections. These allow for vertical movement, but do provde some lateral capacity too, just not much.

As for the floor joists, I generally do not use the DTC clips for partition walls. Technically I should. Just never have, but with no problems as long as the salls generally stack. I also spec double joists under any interior partition wall parallel to the joists that span over 50% of the joist span length.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[msquared48]

Sorry about that typo, it's roof TRUSSES.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[BAretired]

Trues are okay too.

BA

 

[woodman88]

I am more use to the framers buying all the studs cut the same length and using double top plates for bearing walls with single top plate for non-bearing walls. This gives a 1.5" gap. Where lateral support is needed at the top of a non-bearing wall, additional blocking or plywood/OSB is used to allow a wall clip to be used.

Garth Dreger PE - AZ Phoenix area
As EOR's we should take the responsibility to design our structures to support the components we allow in our design per that industry standards.

 

[jeffhed]

dhengr: I am pretty sure it isn't a truss uplift problem. The partition wall in question is in the center of the truss spa

JAE: Yes there is the dilemma of possible cracking at the joint between the ceiling and the wall, but I myself would prefer this over the crack off of the end of the partition wall. The partition wall in this instance is not a bracing element.

woodman88: In Arizona they typically provide a 1-1/2" gap between the trusses and the partition walls? Is there a lot of cracking at the joint between the ceiling and the walls as some have suggested could occur?

 

[StructuralEskimo]

Where I am from they are always framed flush to truss. Although I have always questioned this framing method as it is not the best for drywall cracking and the truss may experience some loading at panel points and in between panel points where it is not intended...but it is hard to change an entire city.

From my experience this is a builder driven practice and they are accepting the fact they may have to come back and repair drywall.

 

[beaming]

There are building code requirements preventing the common practice, but it is still the common practice. One explicit such requirement is that non-bearing partitions not be placed so as to receive load and thereby change the action of the roof structure. Design loads are seldom realized right away after a building is built, so architects, engineers, and contractors get away with paying no attention.

Making the partitions short doesn't prevent gyp board cracking unless the ceiling doesn't connect to the partition. When a roof load is realized, the ceiling commonly moves down a good part of an inch, and so cracks the joint. I have designed slip joints for such a case using picture mold, lapped trim, or a reveal to close the apparent gap. It isn't hard to calculate, but it takes time to detail. No matter what the plans show, the structural engineer wins over the architect, because his stuff gets there first, but then the woodwork wins over the engineer, showing no respect. It really is true that the structure doesn't care what we figure.

 

[spats]

I normally just carry all bearing and partition walls to the same height. For partion walls parallel (and between) framing simply add separate blocking each side to the adjacent truss/floor joist, and you get some flexing in the blocking. Perpendicular... whatever may be, may be! An extra support will not fail the truss (usually), and the wall can buckle a little if it's not strong enough.

 

[BAretired]

I have often thought that if it hadn't been for non bearing partitions in some walk-up apartments, the roof would have collapsed because the trusses are rarely adequately braced.

BA

 

[oldrunner]

Check with the truss plate institute (TPI). They have some guidance for your problem including a discussion sheet ceiling nailing and guide for non-bearings walls.

Long spans will lift up, some places more so than others. Especially when the attic ventilation is not adequate. Number wise, trusses that bear on "non-bearing" walls between panel points fail.

Also if these are pre-manufactured trusses, look at the details and recommendations that come with the design package (usually in tiny type).

There can also be other problems such a lumber shrinkage, sill crushing, placing the sheet rock inside before the final roofing is installed. Consider if a 38 foot span rests on exterior walls of a number of storeys: Count the number of plates and then check the shrinkage when the plates dry out from 15% to say 8% or 9%. The vertical shrinkage can get fairly significant.

 

[keenaseng]

I have thought about this problem and discussed it with tradesmen on site. My conclusions are 'non-loadbearing' walls that finish short of the underside of truss (to allow the truss to deflect) are not a good idea. There are just too many trades that need to be informed of the detail (that are most often not) to make it work:
1) Engineers deflection of truss needs to be accurate (often not)
2) Carpenters need to install the vertically sliding ties/clips along the top of the stud wall correctly (often not done right)
3) Cornice is not to be set to wall to allow movement (I have seen many cornice set to wall and ceiling)
4) Painters are not to paint 'wall to cornice' junction - what client would accept that?
And then what happens where loadbearing walls are adjacent non-loadbearing walls - how does cornice junctions work here? Or where trusses run parrallel to masonry party walls - do we really expect the cornice to slide up and down these walls with wind unscathed?
I think alot of the time the loads we expect to hit the roof after the finishes are applied never occur, so there is no problem. When the deflection is achieved you see the cracks as described by the OP.
Design the truss to be suported on the internal walls and follow the load path down

 

[a2mfk]

One thing to keep in mind is the actual load on a pre-eng truss is very small compared to required design loads. Once the trusses are up and the roof is decked, your live load goes to about zero (workers are off the roof). Then in construction sequencing you may have the walls going up but a good portion of your dead load is in place and the trusses have deflected X% already. The additional MEP and gyp board/ceiling weight will have an effect but how much...?

I don't think its a bad idea to spec a small gap at the T.O. wall and a vertical slip connection, as much for dead load deflection as for truss uplift action due to thermal/moisture cycling. This causes cracking and separation along wall-ceiling joints all of the time, I see it doing forensic investigations and had it happen in my own house. Ideally this joint should not be finished with gyp compound but caulked or covered by a crown molding detail that will accommodate movements.

In my line of work I see a lot of places where expansion joints should probably be located but it is not standard practice. And so you have cosmetic cracks and separations which we all know are not a structural issue, but lay people do not and they are the owners of these buildings. Therefore, it becomes a problem...

 

[jeffhed]

Thanks everyone for the good information. It seems to me like the best thing to do to prevent the cracking in the future units would be to make this partition wall a bearing wall. We will need to have a beam through the hall ceiling, but it is right at the end of the hall and can be hidden with an archway or something else. It will cost more than it has in the past, but as far as fixing the drywall cracks, I think the contractor won't mind the extra cost. He has fixed some of these cracks multiple times and is ready to try something else to prevent this on future units.