Eccentrically Loaded Lag Screws in Shear

[cancmm]

I'm looking at a ledger design for a residential deck where the contractor wants to connect via lag screws to the rim joist. The problem is that the house is sheathed on the exterior with 1-inch rigid insulation. Unless the contractor removes the insulation (not preferred), there will be an effective 1-inch gap between the ledger and rim joist. Therefore, the screws won't be designed in simple shear and there will be flexural stresses induced into the screw as well. Per the IRC, only a 1/2" gap is permitted when using their tables. Is there any guidance in NDS or similar documents to address this design issue? I see this done with leveling nuts (i.e. light posts) all the time so I figure there must be guidance somewhere.

Thanks!

 

[spats]

Not a good idea in my opinion. You're bending the heck out of the screw. In order to work, you'd probably have to go with a fairly large diameter screw. I guess it depends on the size of the deck. It terms of what the bending is in the screw, that's not an easy question to answer. The moment arm to compute bending is definitely more than 1". Think of it like a laterally loaded pile in soil. The maximum bending moment in the pile is not at the ground surface.

 

[cancmm]

spats - I hate to say that I agree with you but I'm hoping to be proven wrong.

If this doesn't work out and if removing the insulation is not an option (contractor worried about water penetration), I'll make the connection vertically non-supporting and keep a nominal connection to the house for lateral stability. However, this would require footings near the house, which would then require them to be ~8' deep to meet the house foundation level where undisturbed soil theoretically exists (20 year old house). Alas, this complicates things and increases costs for the owner, so I'd like to be completely sold on the idea if this is the best path.

I'm just throwing ideas out here. Any suggestions welcome.

 

[SteelPE]

Could you run the framing int he other direction allowing for fewer penetrations into the house?

 

[Ron]

Agree with spats. You'll have to calculate the bending and shear based on screw properties and forget the tables....you'll find a huge increase in the size of screw needed.

 

[msquared48]

I would seriously recommend exterior columns at the wall, integrated into house appearance, to support the ledger and take away the eccentricity. If not done, the eccentricity will only induce rotation of the ledger and future leaks.

Mike McCann
MMC Engineering

 

[cancmm]

Thanks all for the feedback and sanity check. Looks like exterior columns and footings are the best way to go.

 

[BAretired]

Ledgers are a questionable detail at the best of times. Separate foundations for the deck with a nominal lateral tie is the best way to go in my opinion.

BA