Wood Truss to Concrete / CMU Wall - Out-of Plane Connection

[WoodDesignCat]

There are a few other older threads that have discussed this topic, but I feel that some of them have come to different conclusions. I'm going to state some conclusions that I have come to and hope to get others input to see if there is a better connection than what I plan on doing.

I have two similar projects, a 8" CMU building with part of the roof as trusses and part as rafters and an 8" ICF building with truss roof. For both buildings we plan on using a 4x8 P.T. top plate so that they trusses are not in direct contact with the CMU/concrete and wont absorb moisture. We are connecting the Plate to the wall using anchor bolts. Now for the connection of the trusses and out of plan blocking to the plate is what I'm not sure on. Right now I am planning on using Simpson clips (A34, A35, LS70), am only using the capacity of the clips that put the plate in cross grain compression, not in tension. Two of my bosses agree this is how it should be done, as well as "Design of Wood Structures" book, see the attached image from the book.

http://files.engineering.com/getfil...aef3-8d0c62d7fd99&file=Top_Plate_Tension.jpeg

The problem I am having is that I am in a high seismic area (SDS = 1.58 for one job and SDS = 1 for the other) and in some cases the rafters are spaced at 5' on center with 10 foot tall walls. So I'm not going to be able to get clips on one side of the plate to work for the entire load. Is there a better way to transfer the out of plan load and still have the rafter bear on a plate? Or is there a way to reinforce the plate?

 

[msquared48]

You do not have a cross grain bending problem to really worry about with that connection. If this were a ledger with the depth oriented vertically, then there would be a problem. However that is not the case.

As an alternate, could you space your joists closer together at the end so that the force to the A35's is less? Or double up the end joist and add one A35 to each joist on the same side.

Mike McCann, PE, SE (WA)

 

[WoodDesignCat]

The plate is not in cross grain bending, but some of the clips would put it in cross grain tension. From what both of my bosses have told me (both have their SE license) and from what I see in "Design of Wood Structures", I should only be using the clips that would put the plate in cross grain compression, and ignore the clips that would put it in tension, as shown in the picture attached to the original post. So if I only had clips on one side of the anchor bolts I would only be able to take load in one direction...

The architect wants exposed rafters at 5', if I can't find a solution we will ask to put them closer together.

 

[azcats]

How about a thru bolt running the 8" direction of the top plate with a washer each end. Center it in the 4" side of the plate and under each rafter. It would restrain the plate against cross grain tension and drag it back thru the plate to compression. It would be a bit of extra labor, but not much material and fairly common stuff.

 

[WoodDesignCat]

@azcats,
That sounds like an interesting solution. I was wondering if there was a way to reinforce wood for cross grain tension using screws, but a small bolt with washers each side might work better. Or maybe one of the ThruLOK screws that has a washer each end.

 

[KootK]

I think that screws would actually work better as reinforcement. Easier install and less issues with shrinkage neutering the effect.

It's been a couple of decades now but I used to work in the metal plate connected wood truss industry where this is an issue at many of the plated web to chord joints. If I remember correctly, the rule was always to grab the member beyond the centriod. If there was a way to use that logic here, you could centre your clips and count on two in each direction.

The bolts really have the same problem (and same solution) in my opinion. Unless the bolts happen to land right beside the clips, they will induce cross grain tension as well. It isn't as though we can trick the shear in the top plate into remaining solely on the side of the plate where the clips are.

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.

 

[msquared48]

The plates with many teeth, or nails, such as A35's, that transfer the load to many individual grain lines are far less likely to induce lateral tension or tearing than the localized effect of a bolt on one grain line that concentrates the force.

If the edge distances recommended by the AITC and NDS are adhered to, staggering bolts and other fastener lines, I really don't see a problem here.
.

.


Mike McCann, PE, SE (WA)

 

[WoodDesignCat]

@KootK
I think two LS70 clips had enough capacity, but I'm not sure I would want to use their full capacity just because half of them are beyond the centroid.
As far as the bolts creating cross grain tension, that just sounds like a different perspective for the same problem. I guess the question would be whether it's better to reinforce the top plate at the anchor bolt locations, or at the rafter locations.

@msquared48
You don't believe that transferring out of plan load from the wall to the plate using bolts, and then transferring that load to the rafters from the tension side of the plate would split the plate? I am adhering to NDS bolt edge distances and spacing. Right now I don't have the bolts staggered. I'm not worried about the clips splitting the top plate because of their fastener spacing, I'm worried about the plate splitting in line with the bolts when relying on clips on the tension side of the plate.

I think I will either redesign with screws at a certain spacing each side of the top plate as cross grain tension reinforcement. Or I will stagger the anchor bolts so that there is a larger compression region of the top plate, but I'm not sure I will be able to get the anchor bolt breakout strength in the wall to work this way though.

 

[azcats]

How about a large plate (7 1/2" sq?) atop the top plate at each anchor bolt that has holes for screws or lag bolts into the top plate.

So the load goes from rafter - clips - top plate - lag bolts - steel plate - anchor bolt - wall. Then there's really no cross grain tension and you can utilize the whole width of the top plate.

I'd never seen the ThruLOK faster before - looks pretty slick.

 

[msquared48]

WoodDesignCat:

I can see what you mean relative to the bolt, especially if the bolts are not staggered.

However, I do believe that the problem would be reduced by 50% if the A35's or whatever were placed at the center of the plate with half the force transmitted in tension and half in compression relative to the bolt.

I just think you are making more of the problem than really exists here.

No offense...

Mike McCann, PE, SE (WA)

 

[WoodDesignCat]

@azcats
A metal 7-1/2" plate at each AB? If I were to use custom metal plates I'd probably just put one at each rafter with with clips connected directly to them... Which might be a better option than reinforcing the plates with screws.
I think I saw the ThruLOK one I was thinking of two years ago, I haven't actually used it on anything yet, but it seems like a better option than bolts for tension applications.

 

[KootK]

I think two LS70 clips had enough capacity, but I'm not sure I would want to use their full capacity just because half of them are beyond the centroid.

My point was really that you might consider moving the clips to the centroid in the hope that you could use both.

I guess the question would be whether it's better to reinforce the top plate at the anchor bolt locations, or at the rafter locations.

What I was getting at is that, if it's a problem at both locations, then reinforcement would be required at both locations. Of course, I don't think that it's a problem at all with the clips centered on the plate. At least not unless there's a code cause somewhere prohibiting the use of perpendicular to grain tension outright. And, were there such a clause, it would preclude the use of the bolts altogether as that connection depends on tension perpendicular to grain. I suppose you could say that you've allowed the plate to split and then bear only on the remaining loaded half in compression. That's another kettle of fish however.

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.