Nail withdrawal capacity for wind uplift

[RacingAZ]

Is this allowed under the IBC? There is an(2005)NDS table that shows nail withdrawal values. I remember under the old UBC, it's not allowed to rely on the withdrawal of the nails for support.

 

[woodman88]

It depends on how the Building Official interprets the IBC section 2304.9.6 Load Path. The NDS (2001 may be the same in 2005) table 11.2C gives nail withdrawal values.

Garth Dreger PE
AZ Phoenix area

 

[LowLax]

I believe it's not allowed if the withdrawal is from the end grain. Like a top plate to a stud for example. I think that's in NDS somewhere.

 

[bdruehl]

i agree with nds no withdrawl allowed from end if memory serves right... but not only that, i think it is good practice to not use nail withdrawl for load resistance, if only available connection swith to screws

 

[RacingAZ]

The reason for the question is that I'm reviewing a solar panel system installation in a wood roof and the manufacturer came up with equipment hardware with standoffs (legs) that directly and randomly screw into the plywood sheathing with no regards to the joists or trusses below.

Typically and what I have done in the past, the supporting legs would be installed directly on top and along the existing joist or truss lines for support.

The manufacturer provided sealed calcs stamped by a Civil PE showing their justification. The way they did it is that each leg reaction (gravity or uplift) approximates a man weight (200-250 lbs), which I get. The part I'm skeptical with is the use or reliance in the existing plywood nailing into the existing joists/trusses to keep the system in place.

So let's say the legs of the panels are screwed into the existing plywood sheathing, between the joists/trusses. The plywood is supposed to span between the 2 joists/trusses to support the legs then the plywood nailing transfers the uplift force into the joists/trusses.

This is for a location with no snow load consideration. I'd like to hear what you guys think of the mfr. claim?

 

[woodman88]

Per the IBC table 2304.9.1 with a 1/2" sheathing requires at a minimum 6d, 2.375"x0.113" nail or 1.75" 16gage staple at 6" edge and 12" intermediate spacing. Now is the mfg sealed calcs stating that this connection works (of the sheathing to the joists/trusses) or does it just states that the panels connection to the sheathing works? If they are stating it works (and have the calcs for all three or at least what is stated in the plans) to the joists/trusses and does not add addition uplift than the building can handle, then I say they are taking the responsibility of it. If they have calcs for just one of the connections of the sheathing (and not the one stated in the plans) to the joist/truss, then I would require that they connect to the joists/trusses or blocking between them and have them detail it. Make sure they used the correct sheathing values also.

Garth Dreger PE
AZ Phoenix area

 

[RacingAZ]

Thanks Garth. My main issue is the reliance in the nail withdrawal capacity to support the system. Isn't the withdrawal capacity nothing but the frictional force between the wood and the nail? I'm just skeptical because I was trained early on not to rely on nail withdrawal for support even if the numbers calc out.

 

[BAretired]

A connection to plywood deck between joists is not a proper connection for reasons already given. Framing members between joists should be added to get a reliable result.

BA

 

[TXStructural]

The IBC nailing schedule for sheathing is mainly based on wind uplift on the raw roof. Gravity considerations aside, adding anything which increases uplift (and uplift point loads in particular) would necessarily require additional nailing and/or framing.

 

[woodman88]

RacingAZ, Per your note above, the standlegs are screwed to the sheathing. I personally would not rely on just sheathing for withdraw value and would want a joist/truss or blocking at the screws but I am not sure, as they sealed the detail, if I would reject it. Depending on the point load on the sheathing, blocking may be needed to prevent sagging of the sheathing. The calcs for the uplift of the sheathing to the joist/truss, I would require it to assume the load is adjacent to the joist/truss and have no problem with them using the minimum required connection. It really depends on how well they have covered things in their calcs.

Garth Dreger PE
AZ Phoenix area

 

[Ron]

Wait a minute guys...let's get some common sense injected here. Sheathing is attached to trusses with...nails. In uplift, the sheathing is a critical structural component. It is attached with nails in direct tension when subjected to uplift. That's what the code requires. What's the issue? The only other option is to screw the sheathing to the trusses. Not usually done.

When you attach something to the sheathing, between trusses or other structural support, the sheathing then has to transmit the load to the structure through...nails.

I would not allow any equipment subject to wind uplift or overturning to be attached to only the sheathing. Two problems exist...one is the fastener pullout of the fasteners attaching the device to the sheathing. The second is the secondary effect of the additional load on the sheathing fasteners. Attach such devices only at the support and make sure the fasteners have length and capacity sufficient to handle the pullout from the support level.

 

[woodman88]

Ron, Common sense says that if you do the calculation of the allowable uplift, of the sheathing to the joist/truss, that you will get a very small number. The most I would allow is the use of three nails/staples for this connection. If they use more than this I would have to take a very good look at the calculations.
So just what kind of uplift is there from the solor panel?

Garth Dreger PE
AZ Phoenix area

 

[beton1]

Careful with the solar panels and what the manufacturer/engineer for the frame/panels states. We are going through the excercise of reviewing a industrial building with added panels on a OWSJ/steel deck roof system. (Another thread).

 

[par060]

I've framed many houses and apartment buildings without useing anything but nails for connections. Although the sill plate was bolted down to masonry foundation by J bolts.

 

[BlastResistant]

I don't have any problems with nailing directly into plywood sheathing or using NDS nail withdrawal values, but I do have a problem with the actual nail withdrawal values themselves. If you nail directly into plywood sheathing without any regard for trusses or framing members below, then in the calculations you must assume that supports will be anchored into ONLY plywood. I don't have the NDS in front of me, but the withdrawal values are PER INCH of penetration. For a 1/2 or 5/8 inch piece of plywood and nail withdrawal values on the order of 35 pounds per inch of penetration, you have only about 17 pounds of withdrawal per nail. I don't know the exact circumstances of your particular project, nor do I know the layout of the PV panel array or the anchor point layout of the racking/mounting system, but even in the lightest of conditions, I can't imagine each standoff having less than 200 pounds of uplift force. It's probably more like 400 to 600 pounds. I just can't see how the stamping engineer justified using nails with 17 pounds of uplift resistance to anchor 200-1000 pounds of uplift force.

 

[BlastResistant]

After reading your second post, I see you say the standoffs are randomly "screwed" while your first post refers to relying on "nails" for pullout values. If I understand correctly you are actually asking if it is okay to randomly
"screw" anchors into plywood and rely on the withdrawal values of "nails" at the framing members below into which the plywood is anchored. My response to that would be yes, if the screws used have sufficient achorage values into the plywood to resist uplift at the PV system anchorage points. You are not changing the overall uplift pressure on the roof by adding PV panels -- you are only changing the distribution of the loads from uniform wind uplift to concentrated point loads at the standoff leg anchor points. By the time the uplift force at the standoff anchor points is distributed through the plywood sheathing into the trusses and/or framing members below via nails, the nails can't tell if the forces are coming from concentrated standoff point loads or uniform wind uplift.

 

[TXStructural]

I think you will be potentially increasing the uplift. By adding a shape to an otherwise smooth roof, you invariably change wind forces. You will also be creating a couple which must be resisted differently than a relatively uniform wind on roof.
The change is most likely cladding-related, rather than MWFRS, but needs to be considered for both.

 

[RacingAZ]

Thanks for all the replies. As I said previously, my main issue is with the reliance to the nail withdrawal capacity
between the plywood and the truss/joist members to provide the uplift resistance (assuming a complete load path). I wanted to bounce it off to other engineers to get a feel if some of you guys share my concern. Thanks again.

 

[omoreno1]

I would verify the uplift of the solar panel attachment, this should probably be treated as roof-top structure per ASCE 7, this should yield higher uplift pressures that you would typically find in zones 2 and 3 for C&C loadings. Even for simple span pressurs for trusses at 16" oc or 24" o.c at higher wind speeds 100+, the 12" oc field nail spacing is not adequate for uplift. I cannot imagine the solar panel uplift load being resisted by nails fastened only into sheathing, you can not even achieve the minimum nail penetration to achieve any load value. I would fasten to a blocking member or a solid structural member and just dump what ever the solar panel engineer specified.