Composite Wood Column

[bigmig]

This seems like a fundamentals question, but I think I'm making it more complicated than it should be. How would you approach the capacity design of a 3 pack wood column that consists of 1 3/4"x 3 1/2" LSL stud members, nailed side by side, combined with a 2x4 Hem Fir stud, rotated 90 degrees and nailed narrow faces of the 2 LSL's? The material properties vary significantly (manuf. lumber v.s. dimensional)but load follows stiffness...does anyone have any takes on this? I'm assuming unbraced in both directions.

 

[bjb]

My understanding is that you have an I shaped section? If so this is very unusual. I don't think I would consider this an I shaped member in my calcs, I would consider it as a bunch of individual boards. I question that you could have a reliable shear flow to consider this shape as a true I section.

I think you could determine the lowest allowable compression stress of an individual board-the Hem Fir should govern. Then you could multiply that by the area of all boards. If your LSL members are connected together correctly you could treat them as a built-up shape, but even then there would be a reduction in allowable stress.

Why the unusual construction for this column?

 

[civilperson]

What is the finished dimension cross section and what allowable flexural stresses are you using for the various species? How is the composite made, screws or epoxy or resorcinol or what?

 

[Lion06]

I would only count on the LSL's. They are making up the bulk of the member in a meaningful way. Adding an additional 2x4 sticking out of the side seems odd and I wouldn't count on that in the composite section properties.
For the built up column checks, see the column chapter in NDS and 15.3 - note the additional factor of Kf, which accounts for the efficiency of the section.

 

[bigmig]

The overall shape of the column is 5" x 3 1/2" exactly (not an "I", but just a plain rectangle). I'm assuming the connection between all planes would be adequate to transfer shear flow. My first approach was to look at just the LSL's, but the column didn't check because it was failing in the weak axis. The contractor had already added a 2x4 HF Stud to the outside for furring, so I thought I could just use this and nail it properly to transfer shear flow and increase my section. When I went to run my Cp values (NDS 15.3.1) I immediately noticed that equation 15.3-1 didn't quite work because I had two species of wood. Bjb's "composite" approach makes sense, basically using an "n" value like you would with composite concrete. This would fall into the "load follows" stiffness fact too...

 

[bjb]

I wouldn't use a composite approach such as using "n" the modular ratio. I would calculate the allowable compression stress for the weakest link and use that only for posts that are in direct and full contact with the beam. I would not rely on nails for proper shear flow or just assume that you have proper shear flow. These could be dangerous assumptions. The problem is that wood connections are typically the weak link.

If this column is in a wall then the finish material may provide bracing for the column. Sheet rock is sufficient to brace a wood post against weak axis buckling. I'm guessing that the LSL's are oriented parallel to the beam. If so, and at least one edge of the post will be attached to some sort of sheathing (plywood, OSB, sheetrock), then your post will be braced against weak axis buckling.

 

[bigmig]

BJB,

Sorry for the misinterpretation of your posting the first time. So you're basically saying that I should consider the entire column Hem Fir (the weakest material present). We just finished this particular problem, and the final solution was to replace the 2x4 stud with an LSL stud so all members were the same. The contractor ended up bracing the column with an incoming wall (weak axis support) anyways.

But for discussion, I think your approach is too conservative. Couldn't you just take the EI for both the paired up LSL's (adequately connected for shear flow) and the EI of the 2x4, and consider that a percentage of the load goes into the 2x4 (load following stiffness).

Like you pointed out, I would need to have enough connection to transfer this load. Then I would check allowable compression stresses, perp to grain etc.

Connectors are usually adequate to transfer shear flow for wood to wood because the EI values are similar (wood to steel is another story). You just need to spec out the right # and kind.

The second type of approach is the "composite approach", using the n modular value and basically trasnforming the materials into indentical species. Why would you not use this approach...we use it composite steel and concrete all the time?

 

[bjb]

One problem with using connectors for shear flow in wood is nail slip. In application, the connection may not perform the way you think it will based on calculations.

I think using a composite approach is fine in wood for something like a flitch beam, which I have designed many times. However, for a flitch beam you are not using bolts for shear flow. Providing a load path for shear flow is totally dependent on connections, which for steel and concrete are much more reliable and stiffer than wood connections. Connections are typically the weak link for wood. For example, I would never try to fabricate a wood I joist by using metal connectors to attach the flanges to the web.

I am uncomfortable with this approach for a wood column, but that is just my opinion. I would also be very concerned about the contractors quality control, and I also wonder if the two different kinds of wood would have significantly different creep characteristics. If they do, the amount of load carried by one of the wood types would change. I believe that the LSL would creep less than the Hem Fir, which would result in even more of the load being taken by the LSL.

 

[Lion06]

bjb-
NDS addresses your concern about the "tightness" (for lack of a better word) of the connection between the laminations with the Kf factor in the column stability equation for built-up columns. I don't see anything wrong with using the composite column.

That being said, depending on the properties of the wood, I might or might not count on the odd 2x4.

 

[bjb]

structuralEIT-

Maybe I'm mistaken, but I'm under the impression that the described column is a TEE shape built up out of 2x4 boards, and that not all of the boards are in bearing with the beam.

 

[Lion06]

That was my impression at first, as well, but he did correct us and say it is a rectangular column. If it were a TEE then I agree that I would only use the pieces that the beam bears on.

 

[bigmig]

BJB,

Sorry for the confusion, but the shape is a rectangle. Picture the 2 lsl's (1 3/4" x 3 1/2") nailed side by side to create a 3 1/2"x 3 1/2" square. Now add the 2x4 to the edge to create a 5" wide x 3 1/2" long rectangle. The beam above is 3 1/2" wide (a double micro) and runs over the pair of LSL's only, so the beam is running perpendicular to the 5" dimension of the overall shape.

 

[bjb]

ok, my bad