rcar
 New Member
 Posts:16
 |
| 19 May 2010 07:30 PM |
|
Hi all—
I've been following the forum for many months now, learning a bit everyday--it's been a terrific source of knowledge so far. Still, I haven't found an answer to my situation, so here it goes as my first post.
Location: Walla Walla (SE Washington state) 99362
I'm planning for an ICF residence with a flat unvented roof built with 14" I joists, EPDM on top. A local engineer thinks I can stuff the cavities with fiberglass insulation, paint the drywall with vapor barrier sealer and call it good.
For some reason, I'm not that comfortable with this design. Many posts talk about using polyiso (or similar) on the top of the roof deck and fiberglass/cellulose in the cavities inside. Seeing how this helps with the dew point being outside the roof assembly, I thought of asking here how you think a roof like this should be designed for best performance. Or perhaps point me to where I can figure it out.
For reference, I'm including a picture to show the roof on the great room and the mini attic space (to be used for ducts, wires, etc) on the rest of the house.
Many thanks in advance for your thoughts.
Robert Ps. Sorry for the big blurb, I could not find how to split it into paragraphs ... |
|
|
|
|
|
|
cmkavala
 Veteran Member
 Posts:4327

 |
| 19 May 2010 08:37 PM |
|
rcar;
sounds like a perfect sips job |
|
| Chris Kavala<br>[email protected]<br>1-877-321-SIPS<br /> |
|
|
Dana1
 Senior Member
 Posts:6991
 |
| 20 May 2010 11:20 AM |
|
Posted By rcar on 19 May 2010 07:30 PM
Hi all—
I've been following the forum for many months now, learning a bit everyday--it's been a terrific source of knowledge so far. Still, I haven't found an answer to my situation, so here it goes as my first post.
Location: Walla Walla (SE Washington state) 99362
I'm planning for an ICF residence with a flat unvented roof built with 14" I joists, EPDM on top. A local engineer thinks I can stuff the cavities with fiberglass insulation, paint the drywall with vapor barrier sealer and call it good.
For some reason, I'm not that comfortable with this design. Many posts talk about using polyiso (or similar) on the top of the roof deck and fiberglass/cellulose in the cavities inside. Seeing how this helps with the dew point being outside the roof assembly, I thought of asking here how you think a roof like this should be designed for best performance. Or perhaps point me to where I can figure it out.
For reference, I'm including a picture to show the roof on the great room and the mini attic space (to be used for ducts, wires, etc) on the rest of the house.
Many thanks in advance for your thoughts.
Robert Ps. Sorry for the big blurb, I could not find how to split it into paragraphs ...
EPDM is a fairly low-perm product (most are well-under 0.5 perms- some are under 0.05 perms)- the assembly has to be able to dry toward the interior. But applying an interior side vapor retarder would pretty much guarantee that any moisture that found it's way in there would pretty much stay there. Without an interior vapor retarder, interior moisture would condense in the fiberglass in winter losing R-value and creating seasonal mold conditions. Since Walla Walla is fairly dry, and the average January temp is a modest ~30F you MIGHT be able to get away with stuffing the 14" bays with cellulose to buffer seasonal humidity, but you might want to model that with WUFI before committing to it. With no venting and reasonable air-tightess practices at the ceiling air-transported moisture into the cavities will be pretty low- it'll be an all-diffusion game. But if your EPDM roof ever leaks you'll have (literally) a ton of wet cellulose to clean up. (Even the dry weight of cellulose would have to be factored into your I-joist spacing.) A sure bet would be to use 6" of EPS above the roof deck (or a 6" SIP) for ~ R25, then finish it out with ~6" of better-quality high-density spray fiberglass sprayed onto the underside of the roof deck to acheive an even better clear-assembly R-value than 14" of blown fiberglass (since the EPS is now thermally-breaking the I-joists). That would leave you 8" to run wiring & plumbing (even ducts) in the joist bays. The R-value of the fiberglass would be fairly stable since the roof deck would block infiltration & convective losses and the delta-T across the fiber layer that drives those losses would be half what it would be in an all-fiberglass setup. Roof leaks would be detected without accumulating water-weight, and the roof deck stays reliably above the dew point in winter. You can probably get away with 4" of EPS & 8" of spray fg, but lower than 4" that you'd want to model the humidity of the roof deck just to be sure. |
|
|
|
|
rcar
 New Member
 Posts:16
 |
| 20 May 2010 01:11 PM |
|
Chris: I'm waiting for a freight charge to see if it'll work for me ... |
|
|
|
|
rcar
 New Member
 Posts:16
 |
| 20 May 2010 01:32 PM |
|
Dana1— You provided exactly the kind of analysis I was hoping for. You mentioned WUFI, I downloaded it but it only runs on windows machines so I have no way to try that. Interesting comment on the EPDM low-perm values. I thought EPDM was a standard product. I guess I will have to look at the specs for each brand I'm considering. If I have to go with at least 4" foam on top of the roof deck, it'll make the roof profile too thick for the places that overhangs on patios. You are making me think I should pursue a SIP solution ... Many thanks for your take. |
|
|
|
|
Dana1
 Senior Member
 Posts:6991
 |
| 20 May 2010 06:07 PM |
|
Posted By rcar on 20 May 2010 01:32 PM
Dana1— You provided exactly the kind of analysis I was hoping for. You mentioned WUFI, I downloaded it but it only runs on windows machines so I have no way to try that. Interesting comment on the EPDM low-perm values. I thought EPDM was a standard product. I guess I will have to look at the specs for each brand I'm considering. If I have to go with at least 4" foam on top of the roof deck, it'll make the roof profile too thick for the places that overhangs on patios. You are making me think I should pursue a SIP solution ... Many thanks for your take.
The permeability of all EPDM roofing is low- but thicker/better versions are can be EXTREMELY low (less permeable than 6-mil poly.) It'll be in the specs somewhere, but it won't be an important factor in this analysis- they're ALL too low to be able to use an interior vapor retarder with no roof-deck ventilation. You can go with thinner foam if you A: back off from the ~R50 of total R or B: go to an insulation with a higher K value (K=R/inch). The roof deck temp is determined by the balance of interior & exterior R values and the interior/exterior temps. If you go with (more expensive, but not outrageous) iso roofing insulation you can get ~R20 out of only 3" (compared to ~R16 with 4" of EPS), or R13-ish out of 2" iso, R16-ish out of 2.5" thick goods. In your climate if you have at least ~30% of the R above of the roof deck you're pretty much good to go. You may be able to get by with 25%, but I'd want a better model before making that call. R13 of foam would be ~25%, R16 would be ~32%. Last time I priced it 3" fiber-faced iso was running ~$55 for a 4x8' sheet, f.o.b. the distributor's warehouse, the 2" goods were ~$38/sheet. |
|
|
|
|
rcar
 New Member
 Posts:16
 |
| 21 May 2010 02:50 PM |
|
Dana1— Your last paragraph above nailed it for me. The 30-70% insulation split you talk about is the spec I've been trying to figure out. So I were to drop to the total R value to 40, and put 2" foil-face polyiso on top of the roof deck, then I need to do about R26 in the cavities, correct? Would this assembly forgo the interior vapor barrier and or the latex vapor barrier paint? Can the cavities be partially filled or should I have them packed with FG or cellulose? I guess the latter will change the % ratio ... Thanks again. |
|
|
|
|
JohnyH
 Basic Member
 Posts:114
 |
| 21 May 2010 07:55 PM |
|
Intersesting read, I met with an ICF builder/contractor today and we discussed flat roofs. It is one thing I had never considered in the design of our retirement abode! Living in Ottawa our winters can be long and cold! The house we are in now has R60, blown in Fiber Glass and I would like to get the same R value in our new home! Is R60 in a ceiling something I should not be trying to attain in an ICF home? John |
|
|
|
|
Alton
 Veteran Member
 Posts:2164
 |
| 22 May 2010 12:55 PM |
|
Chris,
Who sells FRP SIPS that is close to Montgomery, AL and Orlando, FL? Can you furnish me the web site address? |
|
Residential Designer & Construction Technology Consultant -- E-mail: Alton at Auburn dot Edu Use email format with @ and period . 334 826-3979 |
|
|
Dana1
 Senior Member
 Posts:6991
 |
| 24 May 2010 03:29 PM |
|
Posted By JohnyH on 21 May 2010 07:55 PM
Intersesting read, I met with an ICF builder/contractor today and we discussed flat roofs. It is one thing I had never considered in the design of our retirement abode! Living in Ottawa our winters can be long and cold! The house we are in now has R60, blown in Fiber Glass and I would like to get the same R value in our new home! Is R60 in a ceiling something I should not be trying to attain in an ICF home? John
In Ottawa R60 (or more) is a good thing, but how you get there makes a difference. With a ventilated roof structure and (Canadian code-required) interior side poly vapor retarders you'd be fine. With a CONCRETE roof deck (no wood) and at least 4" of closed cell foam at the roof deck, and the rest in spray-fiberglass (no vapor retarder) you'd also be fine. Or 6-8" of EPS between the membrane & wooden roof deck, spray fiberglass underneath, no vapor retarder, also fine, but you'd probably have an argument with the building inspectors on the latter two. Rigid EPS is pretty cheap stuff compared to SPF- if you have room for it, 15" above the roof deck could deliver the whole R60. I'm not sure a flat roof is a very good idea in Ottawa though. It's possible to ventilate a flat roof, but the vent stacks have to be reliably above the potential snow line (which can look pretty silly in a place as snowy as Ottawa after it's atypical once in 50 years snowstorm.) |
|
|
|
|
JohnyH
 Basic Member
 Posts:114
 |
| 24 May 2010 10:12 PM |
|
Thanks for the response, how is a flat roof ventilated? John |
|
|
|
|
Dana1
 Senior Member
 Posts:6991
 |
| 25 May 2010 03:48 PM |
|
Posted By JohnyH on 24 May 2010 10:12 PM
Thanks for the response, how is a flat roof ventilated? John
Flat roof ventilation These can be active (powered), passive, or wind-operated (venturi or turbine types), but the stacks need to be waterproof and reliably higher than the highest potential snow depth. The pressure boundary of the structure then occurs at the ceiling level. Unvented stackupClosed cell foam against the roof deck, fiber below. |
|
|
|
|
JohnyH
 Basic Member
 Posts:114
 |
| 26 May 2010 07:29 AM |
|
Thanks for the information, a co worker just gave me his copy of the Canadian Wood-Frame House Construction. They do a show a flat roof and unvented stack up so I would have to assume that it is OK to do it that way in Canada? Now the question would be do I want to? What would be the life expetancy of a properly built flat roof before it would need to be redone?
John
|
|
|
|
|
rcar
 New Member
 Posts:16
 |
| 08 Jun 2010 08:03 PM |
|
Okay, I'll be doing a SIP roof (and walls too instead of ICF). I wanted to use steel SIPs but freight costs killed the deal so it's OSB. The question now is, do I use any insulation at all on top of the 10" EPS roof panel or do I use something like Dens-Deck before applying EPDM? Googling does not seem to pop up a lot of information on this field, so if anyone has links, tips, research to share, I would certainly appreciate it! |
|
|
|
|
cmkavala
 Veteran Member
 Posts:4327

 |
| 08 Jun 2010 09:12 PM |
|
EPDM specs will vary between mfgs. , but most require a base sheet of some type. Let your roofer guide you |
|
| Chris Kavala<br>[email protected]<br>1-877-321-SIPS<br /> |
|
|