I'm building a shed/barn and apparently had a bit of a brain fart when nailing down the plywood floor sheathing. I nailed it down parallel to the joists (like t1-11 sheathing on an exterior wall), with a few 4x4 squares that I ended up rotating correctly. I've already nailed down a few exterior walls so it would be a big pain to pull everything up to rotate them.

How big of a deal is this in terms of structural integrity? The floor joists are 16in o.c. Plywood is 19/32. This is a 16x16 shed to store tools, lawn mower, atv, etc. The floor feels perfectly stable, but I'm not sure if there are any long term effects I need to be worried about.

  • 2
    I probably would not use use a pogo stick in it, but beside that it should be okay.
    – crip659
    Oct 5, 2021 at 12:04
  • Do you know how many layers the plywood has? Since plywood has alternating grain directions, it should have more structural integrity in the direction it has more grain in, but if it's 3 layers out of 5 then it would matter more than if it's 5 layers out of 9, for example.
    – kaya3
    Oct 5, 2021 at 13:30
  • 1
    @crip659 dammit... Oct 5, 2021 at 13:38
  • @kaya3 My best guess is that 5/8 plywood has 9-11 layers. So that would mean 4-5 of them are in the "correct" direction. Oct 5, 2021 at 13:41
  • Are all the long plywood joins on the joists? if so then there won't be much difference. However, if there are long joins in "free space" that will be the issue - perhaps extra support...
    – Solar Mike
    Oct 5, 2021 at 13:42

2 Answers 2


Assuming your Plywood is Grade 1 (Birch/Pine), at 19/32 you can run the joists at 20in perpendicular (floor) or 24in parallel (roof) for 50psf, so at 16 o.c. you're well within both limits.

I am mixing roof and floor rating, but comparing at 50psf total, since there is no floor rating for parallel installations. This matters for how the load is distributed along the sheet: for a roof it is assumed that the loading is more even. Also note that with floors all sheet edges must be supported, as you have done correctly. Unsupported edges are allowed for roofs.

Where you store the heavier equipment you could always add another layer of plywood or some other lumber to spread the load across at least two joists and achieve an even higher load rating, plus prevent long term sagging of the subfloor. This would also be effective along the perimeter of the shed where you'd have any storage racks or tables which have legs that form concentrated point loads.

enter image description here

enter image description here

Ref: https://codes.iccsafe.org/content/IBC2012/chapter-23-wood

  • This answer is wrong, wrong, wrong. Roof diaphragms get better strength when the long edges of the plywood sheets between supports have blocking. Such blocking allows the long edges to get fastened with the same frequency as the edges running parallel to supporting members. This is where the "with edge support" versus "without edge support" distinction comes from. Search the internet for blocked diaphragm for pictures and articles.
    – popham
    Nov 20 at 21:30

Running the 40/20 sheathing with an incorrectly oriented strength axis as you have, the bending stiffness of your floor will be at most (56000#-in²/ft)/(250000#-in²/ft)100% = 22% of the correctly oriented stiffness. As the number of plies goes down from 5, this percentage goes down. The numbers come from page 52 of the AWC's Manual of Engineered Wood Construction. Your sheathing was unnecessarily thick, though, so maybe the reserve strength can salvage the installation.

For something small like a shed, the goof shouldn't impact structural stability. Technically your floor sheathing is also a floor diaphragm that could be transmitting some wind load depending on how this shed sits on the ground. With only 16 ft of width, I'm sure it's fine.

Floor strength and deflection for people walking around is a problem, however. Technically there isn't a spec for incorrectly oriented sheathing. The span ratings for sheathing are defined by performance tests prescribed under NIST PS 1-19. Among the tests, there's impact testing based on ASTM E661 for correctly oriented sheathing, but no impact testing for incorrectly oriented sheathing. Without impact performance data for your 40/20 sheathing, there's no way to verify its adequacy against impact requirements.

Apart from the impact performance, I can look in Table M9.2-1 of the AWC's Manual of Engineered Wood Construction to find the minimum bending strength and stiffness for correctly oriented */16 sheathing (385#-in/ft and 86,000#-in²/ft). I can look up the corresponding values for the non-strength axis of your 40/20 sheathing (270#-in/ft and 56,000#-in²/ft). Comparing values, I conclude that your bending demand is at (385/270)100% = 143% of capacity and that your deflections are at (86000/56000)100% = 154% of maximum. In your case, then, I can conclusively say that the incorrectly oriented 40/20 sheathing is out of spec.

Remember, even if the incorrectly installed sheathing satisfied bending strength and stiffness requirements, I would still conclude that your sheathing is out of spec because of the impact thing.

  • You write "maybe the reserve strength can salvage the installation" and "I'm sure it's fine" but base that -after some calculations- ultimately on handwaving. This answer is useless useless useless. We already know it's out of spec, and so does the OP, simply because floor installations must be prependicular. In contrast, my answer provides some guidance as to whether it's still strong enough and how to mitigate bending. This answer doubles down on the problem and then handwaves the solution. In any case, it seems the OP is happy with his shed.
    – P2000
    Nov 24 at 18:50
  • @P2000, you've misinterpreted a table from the code. People subsequently following the schematic of your answer will arrive at incorrect conclusions. People searching the internet for information on these tables will find an incorrect interpretation. I mentioned the diaphragm part so that somebody with a larger structure wouldn't incorrectly assume that the floor is just a gravity system. Without additional information, the diaphragm can't be checked. If the OP comes back and asks, then I will check it, but for a 16' shed I know that it's fine.
    – popham
    Nov 24 at 19:13
  • @P2000, your answer is a "baffle with BS" answer. It's grossly irresponsible of you to provide structural engineering advice without understanding the source material that you're citing. The D/C ratios that I provided answer the "how big of a deal" part. If another person with the same problem has D/C ratios below 100%, a building official could demand that he orient it correctly, and an engineer would have to take on liability to overrule the building official. This is the purpose behind "doubling down on the problem." You've doubled down on irresponsible with your response to my critique.
    – popham
    Nov 24 at 19:26
  • it will withstand the anticipated load, but would fail inspection if applicable (e.g. where required for structures larger than 10x10). your calculations won't change that and your handwaving won't make it pass. the Op is fine with his build. if you are a structural engineer kindly provide your credentials in your profile. nobody with "a larger structure" (your comment) should take individual engineering advice from a post on stackexchange.
    – P2000
    Nov 25 at 3:23
  • @P2000, this isn't about credentialism. It's about long-winded, BS nonsense. If the anticipated loads are the loads prescribed within PS 1-19, then his floor won't withstand those loads. The D/C ratios that I provided quantify the relative loading that the floor will support and service. Your roof table has nothing to do with incorrectly oriented sheathing. It's for correctly oriented sheathing with blocking installed below the long edges between sheets (or for blocking installed on the incorrectly oriented sheathing's "oc" layout).
    – popham
    Nov 25 at 4:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.