Nightmare: Soaked Roof Polyiso — Please Advise
I’m building an octagonal house in Zone 2A, doing all the work myself (with a helper). Roof assembly is: timber frame → 2×8 T&G → Adhero → 2 layers of 3 1/2” fiber/paper-faced polyiso → plywood → peel & stick asphalt underlayment → standing seam. 6:12.
I waited until the forecast showed 10+ days with no rain (highs up to 106 F). I got the 2nd layer of polyiso installed by the end of Day 2, and 4 of the 8 facets seam-sealed with canned foam + Zip tape. As I’m tarping it – boom, flash 4-hour storm, ladders fall, tarps blow away, I’m stranded on the roof. Basically in tears with my helper as I wait on a friend to drive over and get us down! Welcome to central TX.
Today, Day 4, after 2 days of 100 and no rain, I ripped up a couple pieces. It’s quite wet between the layers of polyiso (i.e. top layer’s bottom facer is still wet, as is the bottom layer’s top facer) — drumroll… *even on the facets that I seam-sealed w/Zip/foam*.
Questions:
1) Can I assume that any water that’s *under* the *bottom* layer of polyiso will eventually dry out, as the Adhero is vapor permeable, atop T&G facing a conditioned space?
2) But what about what’s between the layers? We’re in a heat wave (100+F daily) and this roof is in full sun… but the polyiso is thick, impermeable, and sandwiched. So is there any point in waiting 2 more days and doing some more sample removals? Or do I have to suck it up and tear this whole !@(*!)ing thing apart and start over?
Thanks for any ideas. I’m running out of time/money to take off from work to finish this, living in a barely-AC’d trailer on site, so it’s super stressful. My nightmare is moving in, after 3 years of killing myself to do this, and hmm, what’s that moldy smell. I’ve worked construction on and off for years, and to be frank, this building is 99th percentile of quality for residential, and I don’t want to ruin it. Just blown away.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Polyiso foam is actually slightly vapor open, the fiber facing is vapor open, Adhero is vapor open, and 2x8 T+G is vapor open. Vapor drive in your climate will usually be to the interior, especially on hot, sunny days. I would get the exterior water-tight as soon as you can, then get the interior as cool and dry as possible. If you have an HVAC system already, crank up the A/C and consider buying or renting a heavy-duty dehumidifier. If you don't have central A/C yet, get some window rattlers or rent something that will make the interior cool--moisture always flows from warm to cool and from wet to dry.
Thanks for responding Michael -- I've put a *lot* of your posts into direct practice on this building. My concern is that the fiber facing is an organic material -- mold readily grows on it. How slight is slightly permeable? Most of what I've seen here says as little as 2 perms per inch... I'm dealing with (2) 3 1/2" layers.
If you let it bake in the sun, that's going to help to drive that moisture out too while things are open. Your best bet is to probably let it bake out for a few days, then see if it appears to be improving. I'm not clear if you had a few days of rain, or if you've had a few days since it rained, from your post, so maybe you've already done this.
Ultimately I'd be concerned if the facer was saturated, but if it just seems a little damp here and there you're probably OK. If you want some extra insurance against mold, put a coat of mold killing primer (Zinsser makes this, others probably do too) for some extra long term insurance against future mold growth. I wouldn't do that until you're sure it's dry though, since the primer will help to seal the facer, slowing drying even more.
Ultimately it's best to not seal things in if there is any saturated material in the mix, but if it's just a *little* moist in a few spots, then I agree with Michael that it will probably dry out over time towards the interior.
Bill
I rarely use fiber-faced polyiso so I don't have much data on it, but I do have an email from a tech rep at one manufacturer who says it's 1-1.5 perms per inch. At 7 inches, it's probably around 0.1-o.2 perms. That's not much, but better than nothing; with sun heating up the exterior and making the interior cool and dry, I think it will dry out pretty quickly, and as soon as the moisture levels drop, mold will be held at bay. The fiber facing is essentially inconsequential.
Also, the perm rating is not a homogeneous quality; it increases as the material gets thicker. The inner layer should dry out pretty quickly, and the outer layer will dry out eventually.
Don't leave the exterior uncovered. With sun beating on the panels, moisture will be driven into the foam, not out, despite what you might think is logical. If you get another surprise rain storm, you'll be at square one again. If you were in a cold climate it would be a more serious situation, but with your conditions and assembly it doesn't sound like a huge deal to me.
Michael -- I [think I] know what you mean about solar vapor drive, but it certainly seems that the *direct* sun dries out the facer very quickly. What I did today was tear out a piece of the top layer on each facet of the roof, hoping to get some convective drying under adjacent pieces. I could suck it up and tear off all of the top layer, in which case one day of sun would dry the hell out of everything (lots of other time/money downsides). Again, I'm happy to be further educated. And in these conditions -- seriously 106F and dew points approaching 80F, realistically, I can dry out the interior if I rent a commercial dehu, but cooling it with no windows/doors/insulation is a different story. If I bust my ass, I could make that happen within maybe 2.5 months...
Bill -- sorry about the stress-fueled lack of clarity in the post ha -- we had a night of intense rain, no rain in the ~44 hours since, 100F+ temps.
I once asked building science expert Bill Rose this question and his response was crystal clear: water vapor always moves from warm to cold and from wet to dry. Martin Holladay wrote about solar vapor drive here: https://www.greenbuildingadvisor.com/article/when-sunshine-drives-moisture-into-walls.
It's really a combination of heat and humidity that governs moisture movement, so maybe your outdoor air is dry enough and/or the foam is hot enough from the sun that it will dry into the air, at least until the saturation levels have reduced.
It's a complicated subject. This might help: https://www.greenbuildingadvisor.com/article/how-to-use-the-psychrometric-chart.
How much water matters? Does the mold just eat the paper facers where there's water, then once it consumes the water, hopelessly waits for more til it dies? Does it turn into a big colony that reaches the eaves and start eating the fascia? I tried so hard to avoid this but here I am.
What I've always seen is that when things dry out, the mold stops growing and just remains as a sort of stain on whatever was once wet. I have no idea how long it can "live" in that state, with "live" meaning if it gets wet again, it will come right back. I know that mold killing primer is intended to be painted right over mold to kill it, and to prevent it from coming back, so that's a good option for you here -- but you can't apply it until things dry out enough that the primer can adhere properly.
In terms of "how wet" is a problem, my usual test is if it drips when I lift it, or if it's wet enough that the dampness causes panels to stick together, then that's "too wet". If I see a few damp spots on the facer, but it's not drippy at the edges and doesn't stick from being wet, then I consider it "OK" to dry out on it's own. It's a bit of a judgement call when you're between a puddle and a damp mark on the facer.
I assume the polyiso itself isn't too bad? Have you tried poking your finger in to see if it weeps like a wet sponge?
Bill
Hey Bill! See above for reply to your other question. Other answers: The facers don't drip, they're just saturated. I've had some ~3/4" D chunks of foam tear off when I pulled up some pieces -- and I'd say no, the polyiso itself is *not* wet. Kilz/Zinsser feels pointless here -- I'd have to remove the top layer of polyiso to apply it, in which case I'd rather just use bleach and let it bake in the sun (which dries it out very quickly in 106F... seriously, TX is an oven right now). But see my reply to Michael above for why I can't just instantly create a cool dry environment beneath the foam.
This is tough for me because I am a best-practices perfectionist who is also a worker. That is, I am the one specifying "sorry, you left a gap in the flashing, tear it out" -- but I'm also the one installing the window. I like doing jobs (including this one for my own family) this way. But it's extremely unforgiving when you bill yourself out at $0/hr (on a good day!).
Perhaps you should up your rate -- $0/hr and you'll have to fine yourself for paying yourself below minimum wage ;-)
No problem regarding the stress confusing your post a bit. Just be glad you didn't panic and fall off the roof during the storm! We've all had things go wrong on projects before, it's part of the process. I like to joke that construction is "all F'ed up and then you have building when you're done... If you're lucky ;-)
Regarding moisture drive and covering things, I'm going to have to disagree with Michael slightly here. You DO want to cover things, but NOT all the time. The reason is that the usual materials you're likely to cover things with are going to be polyethylene sheet, or some kind of poly tarp. Polyethylene sheet is a vapor barrier, and many/most tarps don't breath either. What you end up doing then is baking out moisture, but keeping the moisture in the confined atmosphere between the roof and the covering you used, which means that you're just driving moisture in and out of the panels, without any actual drying to the enviornment. That's a problem. You DO want to cover things when a storm is approaching, buy you DON'T want things covered while you're trying to get things to dry out.
Regarding baking in the sun, if you had a conditioned interior, you'd have issues with condensation forming on the back side of your insulation stack. With a non-conditioned interior, that won't be as much of a problem, since the dew point should be about the same on the interior and exterior, and the temperature on the interior won't be as cold compared to the outside as it would be if you were air conditioning. The sun should heat the exposed material up to pretty toasty temperatures, which will help with drying IF the moisture has a way to get out (i.e. no vapor barrier covering). I suppose you could cover with something vapor open like housewrap, but if you think you have a week or two with no chance of rain, you can probably get away without covering things at all.
On exposed commercial roof decks, I've seen the sun bake out moisture as long as there is nothing "in the way". If, for example, there is a roof membrane (usually a material like EPDM or TPO) in the way, the sun heats everything up and the polyiso sits in a 100% RH enviornment, and you can't have any drying under those conditions. If the roof membrane isn't on yet, the polyiso does dry out. Air movement between layers is a plus too if you can manage that somehow, possibly by putting some furring strips between layers temporarily. I'm not sure if that's really workable in your situation, but if you can do something to allow air to circulate between the layers for a while, that will greatly speed the drying process.
Bill
Can't afford to overpay myself with material costs these days! Luckily this was reclaimed polyiso. Thanks for the encouragement -- at this point, I'm just trying to avoid starting another branch of "Done Twice, Good Enough Construction LLC" haha.
RE: "The sun should heat the exposed material up to pretty toasty temperatures, which will help with drying IF the moisture has a way to get out (i.e. no vapor barrier covering)." Well, the polyiso is still, to me, pretty damn close to a vapor barrier. Personally, I've never seen even totally rain-soaked polyiso act like a sponge.
"water vapor always moves from warm to cold and from wet to dry."
Water vapor always has a net movement from wet to dry, which generally coincides with warm to cold. If the two conditions are at odds, diffusion is from wet to dry. *(see note at bottom)
I think sun beating down on a wet surface is actually a case where intuition can serve decently well. The suns radiation will impart its energy to the liquid water molecules, bringing them to the higher-energy vapor state. The vapor is not shown any signs as to which way to travel; it travels in all directions even though there may be some net flow (akin to how heat radiates in all directions equally regardless of the temperature of adjacent objects, it is just the net flow that can be described as having a flow from hot to cold).
So there will be vapor traveling up and away from the wet face of the foam baking in the sun and it will also be driven into the foam as Michael says. We essentially have vapor drive in both directions since the highest vapor pressure is at the hot, wet face.
But whether to cover or not... there is always the concern that rain will unexpectedly come again, no?
Whether you will dry out: I won't say with certainly, but given your drying capacity at your inner layers and the big Texas sun beating on the roof, it seems you will eventually have dry innards. Even though the foam is fairly low perm, it's not nil.
*footnote. A mental experiment to help see how net vapor drive is always from wet to dry and not necessarily warm to cold:
What happens to a dish of 50F degree water in a 60F degree room at an RH of 50%? 60F at 50RH is a dewpoint of 41F. The dish of water is colder than the room air but the water in the dish will evaporate (even if the water temperature was maintained at 50F).
The equilibrium vapor pressure (condensation equals evaporation) at 50F for water is 1.22 kPa. The actual vapor pressure at 60F and 50%RH is 0.873 kPa. The bowl of water, being in a saturated state at 50F, has an actual vapor pressure equal to its EVP (1.22 kPa) while the air has an actual vapor pressure of 0.872 kPa. Net flow is from high VP to low VP: water dish to air.
While I am using a bowl of liquid water as an example, we can replace it with a container with 50F air at 100% RH, with an air impermeable but vapor permeable membrane covering it, and the story is the same.
Thanks for adding some scientific clarity to the intuition Tyler, I appreciate it. I wasn't sure how to technically legitimize what seems to be the clear reality that the sun beating down is drying out the facer, not only driving the moisture into the foam.
jamesboris,
No good advice, just wanted to say I sympathize. You will get through this and better days are ahead. Chin up!
Thanks Malcolm! Trying to see it that way.
Would there be any significant benefit to setting furring strips between the top layer of insulation and topmost plywood? Might a small gap help with continued drying over the long term?
Only if the gap was effectively vented. Venting the top is very improbable on a hip roof like this. Even if the furring strip "bays" terminated at an opening at the top, there's a very large amount of roof surface area terminating at a small "port" in that case. This is why you'll see a lot of advice on this site pointing toward going unvented w/hip roofs -- for reasons that apply even more strongly in the case of an 8-sided hip roof.
It will help, but you need to balance that against how difficult it would be to temporarily put that vent space into the stack.
Bill