Diaphragm shear vs boundary nailing

[masterdesign]

Consider:

Diaphragm capacity
2021 SDPWS Table 4.2c
Wood sheathing
10d common nails (0.148" x 3")
1 1/2" framing penetration
19/32" sheathing thickness
2" nominal framing (G = 0.50)
6" boundary nailing
Case 1 = 800/2 = 400plf
Case 3 = 600/2 = 300plf

Boundary nailing capacity
10d common (0.148" x 3")
19/32" sheathing thickness (G = 0.50)
1 1/2" framing penetration (G = 0.50)
Z = 95# (2018 NDS Table 12Q or dowel equations)
Z' @ 6" = 95*1.6*12/6 = 304plf

Now I can understand if the diaphragm capacity is less than the design value of the boundary nailing, but how can the diaphragm capacity be more than what the nails can transfer?

 

[DaveAtkins]

You did not include the diaphragm adjustment factor (1.1), but that still doesn't get you to 400 plf.

My guess is the diaphragm capacities are based on testing.

DaveAtkins

 

[driftLimiter]

Well there is a diaphragm factor Cdi = 1.1 that you can include in the strength calculation of the fastener but that doesn't make up the difference.

I think this is an interesting question.

 

[masterdesign]

Good catch fellas.

Z' @ 6" = 95*1.6*1.1*12/6 = 334plf

The comment on testing makes sense and I'm thinking you are definitely right.

 

[driftLimiter]

Also something about the nature of wind vs seismic according to SDPWS, for seismic diaphragms this observation doesn't hold but for wind it does... Curious.

 

[volcomrr]

Also something about the nature of wind vs seismic according to SDPWS, for seismic diaphragms this observation doesn't hold but for wind it does... Curious.

The SDPWS table is based on wind and the diaphragm values include the following: Per IBC 2306.2, you get a 1.4 increase for diaphragms, which can apply to the fastener. To convert the table to seismic you need to divide the table values by 1.4.

334plf * 1.4 = 468plf.

I can't remember the ESR report off the top of my head, but it shows the same tables as AWC/NDS except without the 1.4 factor.

 

[masterdesign]

Volcomrr

Thanks for your comments.

Starting with the 2021 SDPWS, the diaphragm tables do not differentiate between wind and seismic. They list nominal values independent of force type. [For ASD] These values are then adjusted by a factor of 2.0 for wind and 2.8 for seismic, which accounts for the 1.4 factor.

You mention the 1.4 factor "can apply to the fastener". I agree this factor applies to the assembly as a whole, of which the nail is part of. I have never seen this factor applied to the design of an individual fastener; however, if we could assume it did apply to the design of the boundary nailing within the diaphragm assembly (which is what I believe you are saying), that would resolve the question I have posed. I want to believe, but I don't see this in any literature. It would need to be an engineering judgement call. Maybe it is the right one.

 

[phamENG]

I suspect this is an example of empirical testing showing us that the system is greater than the sum of its parts. I wouldn't want to try to quantify it mathematically, but consider the sheathing as a whole. For the wall to fail due to fastener failure at the boundary of the sheet only, what else would have to happen to get the required differential movement between the end studs and the sheet of plywood/OSB? Either the end studs would have to move, or the sheathing would have to move.

End studs moving:
If the end studs moved while the rest of the wall remained in place, then the top and bottom plates would have to flex quite a bit over 16". Probably leading a failure in bending of the plates.

Sheathing moving:
For this to happen, the sheathing would have to shear and essentially rip. Along the fasteners would be the most likely spot. You'd have to look closely at the capacity of the sheathing in the connection to check this one...the Z' is just looking at the dowel connection in isolation. I don't believe your calculation is considering group effects, block shear, etc.

In reality, for the sheathing to separate from the studs and allow a racking failure, the entire sheet would have to. That includes field nailing. Again, I wouldn't want to quantify it, but they are there, they have resistance, and they will have to fail along with the rest of it. Lab testing seems to have proven this out.

 

[volcomrr]

You mention the 1.4 factor "can apply to the fastener". I agree this factor applies to the assembly as a whole, of which the nail is part of. I have never seen this factor applied to the design of an individual fastener; however, if we could assume it did apply to the design of the boundary nailing within the diaphragm assembly (which is what I believe you are saying), that would resolve the question I have posed. I want to believe, but I don't see this in any literature. It would need to be an engineering judgement call. Maybe it is the right one.

You are correct that I was referring to the boundary nailing for the "applies to the fastener". I went down this same rabbit hole a month ago, just proving to myself that the nails I put in a collector would indeed hit the shear values given in the tables (so I ended up calcing my nail as Z' * 1.6 (Cd) * 140%). Using the 40% increase got me to the values in the table.

As a matter of fact, I misspoke, the 40% increase in IBC Section 2306.2 is applicable to only the staple charts listed; the sentence right after referring you to ESR-1539 for nails.... Looking at ESR-1539, there is a 1.4 difference between the wind and seismic values. And, like you mentioned, the 1.4 is covered in the omega factor in SPDWS. It'd be nice to see some published literature speaking of the 1.4.

 

[masterdesign]

Pham: Thanks for typing that out. It is a good description of why values for an assembly would be higher than values for an individual fastener in the assembly.

I x-posted to this to reddit and got a good comment that helped me see this differently.

When you do calculations, you make a lot of assumptions and simplifications. Boundary conditions, connection fixity, friction, etc. And since they're assumptions or approximations, we tend to err on the conservative side. Real life testing doesn't make any of those approximations, it just shows you what really happens under real conditions. In an ideal world, testing results would always be higher than calculated results because, like I said, that means that our assumptions are conservative. Making unconservative assumptions without knowing it could be very dangerous.

This made me realize that my original premise - that the boundary nailing values (calculated) should be higher than the diaphragm values (tested) - may actually be backwards! We should want the tested values to be higher than the calculated values, so we know that the calculated values are lower than what it will fail at.

(to be clear, I recognize that the above comment may or may not apply here specifically since we are comparing an assembly to a fastener)

 

[AaronMcD]

That's just the edge nails. I mean, look at case 3, if the framing is 24" o.c. you only have 5 nails along that side of the plywood. But the whole sheet has 27 nails - 27*1.6*1.1*95 = 4514 lbs = 564 plf. Only the 5 edge nails is 5*1.6*1.1*95/8 = 105 plf. Reality for a diaphragm is somewhere in between.

 

[masterdesign]

So let's take stock of where we are at.

Diaphragm values (tested)

WIND (from original post)
v strong = 400 plf
v weak = 300 plf

SEISMIC (didn't include these in the original post)
v strong = 285 plf
v weak = 214 plf


Boundary nailing values (calculated)

WIND OR SEISMIC - including diaphragm construction factor
v = 334 plf

WIND - if 1.4 factor which is permitted in diaphragm assembly is applied
v = 468 plf

 

[masterdesign]

Aaron: I believe you are making a comparison between the total nails in the sheet versus the total nails along the transfer length? The force transfer occurs along the boundary edge, where the nails are 6" oc. So along 8 feet you will have 17. Please correct me if I mis-understood your comment.