TJI Floor Joists as Floor Diaphragm @ Basement Wall - Compression Issue?

[AndBre44]

Looking for a second opinion on something - I have a basement wall that's retaining 12ft of soil for a basement, with the first floor being an 11 7/8" TJI floor system and the basement slab being present at 11.5ft below grade. Now, whenever I specify TJI joists, I don't consider them being useful as compression members based on the below linked design bulletin put out due to the statement, "A conventionally framed (compression) roof relies on a non-structural ridge board and collar/tension ties to resist the roof loading. This puts the roof rafters in both compression and bending, which TJI®'s are not designed to do. TJI roof joists are designed as bending members only". I'm aware that this is regarding roof rafters, but the statement still stands about TJI's not capable of compression, in the same vein that they're not used as wall studs.

https://www.techsupport.weyerhaeuse...s-in-a-Conventionally-Framed-Compression-Roof

Considering the height of the wall, I would already be less inclined to have my wall rely on the wood floor system due to the high load placed onto the wall (roughly 1150 lb/ft based on a 0.35ka and 125 pcf soil), however I'm receiving some pushback from the architect regarding the resulting footing size necessary for overturning resistance since I'm effectively designing this as a cantilevered retaining wall. I'm certainly aware that the presence of a plywood diaphragm would be what would help with taking the lateral load into the floor, but is the fact that the TJI's are not able to act as compressive members enough to warrant my decision? As someone who's only been in the industry for about 5 years, I'm getting a bit of the "I've been building homes for 30 years and I've never seen this" answer, and just wanted to see if I have a proper leg to stand on or if I'm in the wrong.

 

[TheDW]

I think you could add blocking etc to be reasonably confident in the TJI's transferring load to the diaphragm, then you just have to make sure the load has somewhere to go from the diaphragm.

I'll insist that the wall be raised up flush with the t/joists, and the joists attached to a ledger. I know these are built all the time with bottom bearing joists, but toenails+sillplate bending etc usually don't even come close to working...so I do hold my ground on that, and usually don't get much argument

 

[DaveAtkins]

If it was an 8 ft wall, I would say something like "this is one of those cases where empirical evidence indicates the floor diaphragm and joists can resist the lateral load from soil."

But since it is a 12 ft wall, I think it warrants some additional consideration. Perhaps calculate how much axial compression will be induced into each joist by the soil pressure. It may be a small value compared to the flexural compression, and you can say OK.

DaveAtkins

 

[AndBre44]

DW, the plan was for the framing to be dropped to be aligned with the concrete foundation wall (i.e. top of joist matches the top of wall), so that should be in line with what you were saying yourself. I think more what I was thinking of/fearful of is the idea that TJI's are not meant to be used in compression, but I can see how the blocking would alleviate some of that. I think my trepidation is in the idea that even if the diaphragm is what ultimately takes the load, if the TJI's fail/buckle under the applied compressive load, do they not effectively become the "path of least resistance" in the process?

Dave, would you mind clarifying for me what you mean by flexural compression? As in check the approximate equivalent compressive stress applied to the TJI via its gravity loading and compare that to the stress applied from the compressive loading?

 

[DaveAtkins]

Compression in the top flange of the TJI joist due to flexure = the maximum bending moment at midspan divided by the distance between the centroid of the top flange and the centroid of the bottom flange. This will be a value in pounds. Divide that number by the area of the top flange, and you have the flexural compressive stress in the top flange.

Divide the axial compression in the top flange of the TJI joist due to the lateral soil pressure by the area of the top flange, and you have the axial compressive stress in the top flange.

If these two values added together are less than what the manufacturer recommends, you are OK.

DaveAtkins

 

[TRAK.Structural]

I agree with others here that blocking seems like a prudent step for assurance against compression buckling of the TJI. You may also consider upsizing the TJI to give you extra capacity and/or adding some sawn lumber oriented the same direction as the joists and attached to the ledger so you have members that you can count on in axial compression until the load is transferred to the diaphragm.

 

[TheDW]

If there is a US version of this bulletin, I can't find it, so here you go, you may find this more helpful than the roof structure bulletin :

https://www.weyerhaeuser.com/woodproducts/document-library/document_library_detail/tb-155/#download

I think the shear transfer you're looking at is the same mechanism that wind loads on the first floor walls would transfer loads to the diaphragm, albeit higher loads.

 

[DoubleStud]

You need to design your wall like a concrete tank using FEA or tank design chart. If you have a long straight wall on the foundation, then you need to add counterfort. You need to rely on horizontal reinforcement to span from perpendicular wall to perpendicular wall (or counterfort). I hope it makes sense.

 

[ChorasDen]

I'll take a different stance on this, but I agree with DW in that the best option is to drop the joists and directly attach the diaphragm to the top edge of the concrete wall. Consider the stiffness of the entire diaphragm, and compare that to the axial stiffness of a long, slender joist flange that may not be braced, which the bottom flange is not unless you have specific detailing to provide the needed bracing. I would argue the diaphragm will take the load and ignore the axial component of loading onto the i-joist, strictly on a stiffness basis.

 

[RontheRedneck]

I think the OP is misinterpreting what the document says.

The top flange is virtually always in compression when it's used as a conventional floor joist. So compression in the flanges isn't a problem.

The issue with them being used as rafters (with no ridge beam) is that the entire I-joist is in compression - Not just one of the flanges.

I think you could use either the top or the bottom flange to brace the wall. But if you use the bottom flange, obviously it would have to be braced so it didn't buckle.

 

[lexpatrie]

As to buckling, usually there's a skirt board or a perimeter joist running along the length of the wall.

More to the point, on the lateral load question itself, are you trapping the TJI between the two walls, or is the TJI on top of the two walls.

One could potentially detail the wall itself to act as a cantilever (or horizontally span), with the top of wall detailed to allow the retaining wall to deflect, although wind loads might be tricky, I would expect in the direction of "slip" along that connection there's a floor diaphragm for the wind load and that takes it to the perpendicular wall where there's effectively no slip (the slip is going the wrong direction for this load case), and when you switch to the opposite direction it works the same way.

If one designs for certain soil pressures, it's inherent that the wall deflects. If that deflection is not permitted to occur, the wall pressure is wrong, is usually a higher value, and then the top reaction is larger as well.