Hi,

I am looking for advice on finishing the exterior of a large shop/garage.  The building was designed to be very well insulated in Alaska.  It is 2x6 construction with a polyethylene vapor barrier beneath the sheet rock.  The stud bay insulation is 3 inches of sprayed urethane, which was sprayed against the external plywood sheathing.    I would like to put some type of insulation barrier on the outside of the sheathing to break up the heat loss from the studs.  The contractor would like to use Tyvek Drainwrap on the sheathing followed by insulated vinyl siding to accomplish this. 

 

I have concerns related to this:  First, will the air gap created by the drainwrap ribbing make any insulation in the siding worthless with regard to decreased heat loss from the building?  Second, if the Tyvek is tight enough to prevent heat loss through air gaps, will the insulated vinyl siding be so poorly permeable to trap moisture and make the sheathing eventually get wet from retained moisture?

 

Although the outlets and windows are all mounted for the insulated siding, would I be better served by using XPS covered by regular Tyvek, and then finished with conventional, leaky, uninsulated vinyl siding?

 

Thanks for your help!

I don't see how "insulated vinyl siding" can be much more than a gimmick. Put your insulation in the rigid foam outside the studs and go with the (certainly cheaper) uninsulated siding.

What fills the rest of the space in the stud bays? (5-1/2" - 3" = 2-1/2" space?)

Posted By Craig Doser on 23 Feb 2014 07:14 PM

Hi,

...It is 2x6 construction with a polyethylene vapor barrier beneath the sheet rock.  The stud bay insulation is 3 inches of sprayed urethane, which was sprayed against the external plywood sheathing...

What else is installed in the stud cavities to fill out the 5-1/2" depth of the studs?  If nothing, then why did you use 2x6 studs?

I would not use drainwrap on that location. Sheet foam has to be tight to the building to work at all, and I do not believe "insulated" siding is designed to be tight at all edges, but is just fastened above the top of the insulation, and not fastened at all at the bottom. I agree with ICFHYBRID that it is a gimmick; I do give the siding manufacturers credit for spreading the concept, however erroneous, that hanging a small amount of foam will work. Probably the same folks that brought you insulating paint and bubble wrap.

I would tend to agree with what the others have indicated/suggested. Dana is the resident expert on this subject. We have building assembly moisture analysis software that you might find useful:

Borst Building Assembly Analysis Software

Please keep in mind that this software has limitations in that it only models vapor diffusion, so please carefully read the instructions and use it accordingly.

Thank you for all of the words of wisdom. The 2x6 bay just has the wiring inside of the urethane. In retrospect, it probably could have been built with 2x4 construction. I did not start questioning the design early in the project. The only water lines run in an interior wall. The designs I have seen put forth in Fine Homebuilding put regular Tyvek on the sheathing, then lay the XPS on top of the Tyvek, and then the uninsulated siding on top of that. Is that the order you would recommend? Thanks again!!

Posted By Craig Doser on 25 Feb 2014 05:21 PM
Thank you for all of the words of wisdom. The 2x6 bay just has the wiring inside of the urethane. In retrospect, it probably could have been built with 2x4 construction. I did not start questioning the design early in the project. The only water lines run in an interior wall. The designs I have seen put forth in Fine Homebuilding put regular Tyvek on the sheathing, then lay the XPS on top of the Tyvek, and then the uninsulated siding on top of that. Is that the order you would recommend? Thanks again!!

The thermal bridging of that 3" of framing depth is nearly 2x that of the full-depth 2x6. Assuming it's closed-cell foam at about R6.5/inch,  for a 16" o.c. framed (~25% typical framing fraction) vinyl sided construction with 1/2" CDX sheathing and 1/2" gypsum on the interior, that shorter path through the R1.2/inch wood results in a "whole-wall R" of about R10.3 

That's less than R1 more than what you get out of 2x4 with blown cellulose in 2x4 framing (which comes in at about R9.5 whole-wall).

If you blow the remaining 2.5" of depth with cellulose or fiberglass it will bring the whole-wall R up to about R16.2, which is about the same improvement as adding more than an inch of XPS over what you've got.

XPS is fairly vapor retardent, but you could get away with 1" even with the interior poly, but not more.  You can can go much higher-R on the foam by using 1.5lb density (aka "Type-II")  EPS, which is about 2.5x as vapor permeable as XPS at any given R-value.  EPS also has a much lower greenhouse gas footprint than XPS, since it's blown with pentane (about 7x CO2 lifecycle potentcy) instead of HFC134a (about 1400x CO2).   And as that HFC blowing agent leaks out over the next 40-50 years, the R-value of XPS ends up being the same as EPS of the same density, but still fairly vapor-retardent.  At 2" (R8.2) EPS is still slightly more vapor permeable than XPS is at 1" (R5).

Whether the housewrap goes over the foam vs. between the foam & sheathing is a matter of how you've mounted & flashed the windows.  Putting it between the foam & sheathing is an "innie" mount, where the glass is roughly co-planar with the structural sheathing, and the flashing can be properly lapped with the housewrap as a continuous drain-plane.  It's useful to use a crinkle-type housewrap on innie-stackups as a capillary break and micro-gap into which liquid water directed there by the flashing can follow gravity.  On "outie" mounts the glass is roughly co-planer with the exterior of the foam or siding, and as long as there's some miniscule amount of air between the siding and housewrap, plain flat housewrap goods are just fine.

In either it's key to get the flashing right- flash the rough-openings for the windows, lapping it properly with the housewrap as one continous air and water-tight layer. Gaps between the flashing and housewrap will allow orders of magnitude more liquid water migration, particularly during the wind-driven rain events that are more likely to push water past the siding.

Dana,

Thank you for your recommendations! The stud wall is on 24 inch centers. Interestingly enough, the fellow that sprayed the urethane recommended against me filling the whole 2x6 cavity with urethane. He said that the properties of urethane and how well it sealed would give me very little bang for my buck, if I used more than 3 inches in the wall. I may not remember this correctly, but he was touting something along the lines that 3 inches of sprayed urethane was the equivalent of 12 inches of fiberglass, since there were no air leaks around the edges of the bats or convection circuits within the insulation itself.

In your description of an “innie” mount for windows with external rigid foam insulation, you describe using crinkle-type housewrap. I am confused. If I use the crinkle housewrap so as to allow water to drain at the level of the housewrap covering the sheathing, doesn’t that set up a convection circuit and make any external rigid foam insulation worthless? How can you have air flow and drainage via crinkle housewrap without losing the effectiveness of rigid foam that sits further outside?

In this particular building, the windows are already mounted flush onto the sheathing, the interior walls are painted, and the building is already heated.

Thanks again!

Posted By Craig Doser on 26 Feb 2014 10:57 PM
Dana,

Thank you for your recommendations! The stud wall is on 24 inch centers. Interestingly enough, the fellow that sprayed the urethane recommended against me filling the whole 2x6 cavity with urethane. He said that the properties of urethane and how well it sealed would give me very little bang for my buck, if I used more than 3 inches in the wall. I may not remember this correctly, but he was touting something along the lines that 3 inches of sprayed urethane was the equivalent of 12 inches of fiberglass, since there were no air leaks around the edges of the bats or convection circuits within the insulation itself.

In your description of an “innie” mount for windows with external rigid foam insulation, you describe using crinkle-type housewrap. I am confused. If I use the crinkle housewrap so as to allow water to drain at the level of the housewrap covering the sheathing, doesn’t that set up a convection circuit and make any external rigid foam insulation worthless? How can you have air flow and drainage via crinkle housewrap without losing the effectiveness of rigid foam that sits further outside?

In this particular building, the windows are already mounted flush onto the sheathing, the interior walls are painted, and the building is already heated.

Thanks again!

What he's REALLY saying by that is along the lines of:

"At 17 cents per R/foot the additional performance gained by another 2"+ of closed cell foam will not be very cost effective, since half it's performance is being robbed by the thermal bridging, and the fact that in real $/R is 2-3x as expensive as fiber or exterior foam alternatives." 

Or maybe he hasn't really thought about it that deeply, and is just repeating foam-guy marketing lore.

It doesn't take 3" of foam to air-seal a stud cavity- it's done every day using cheap acoustic sealant caulk and NO foam at a much lower price point.  It would have to be the world's sloppiest carpenter & batt installer to make 3" of  foam in 2x6 construction the thermal-equivalent of R38 fiberglass in 2x12 construction, especially after you  see the performance hit you're suffering from the 3" thermal bridge of wood with the high-R/inch foam.  Even at a 20% framing fraction (24" o.c. framing) more than half of the heat loss of the assembly is going through the R1.2/inch wood when you have R6.5/inch foam in the cavities- the center cavity R stops mattering very much.

In low-cost energy areas with far fewer heating degree days the cost effectiveness argument breaks down at 2" or less.  It's not that the extra "bang" isn't worth something, only that it isn't worth the "buck" steep up-charge achieving that R with closed cell polyurethane, especially when it's performance is being cut off at the knees by the thermal bridging.  Closed cell spray foam is SO expensive that it's generally not worth installing more than a flash-inch for improved seal and providing a non-wicking condensing surface, then filling the rest of the cavity with cheap fiber, after which you can apply the difference in cost toward insulation on the outside of the sheathing. 

In a zone 8 AK climate with a flash-inch of foam in the wall cavities (about R6) with 2" of EPS on the exterior (about R8.5) gives you sufficient R value outside of the 4.5" of fiber that you would not even need the interior poly. The ~1 perm vapor retardency  of the inch of closed cell is more than sufficient for protecting the sheathing from wintertime moisture accumulation when it's kept warmer by the exterior rigid foam.  It's not exactly to the very letter of the code-prescriptive when you split the foam-R between exterior EPS and interior polyurethane, but it definitely works.  The condensing surface is the interior face of the polyurethane, which doesn't wick moisture toward the sheathing. With ~R15 of EPS/plywood/polyurethane between the 4.5" of (~R16.5) fiber and outdoors condensing events are non-persistent. 

That stackup delivers ~R25 whole-wall performance (more than 2x the 3" of foam alone) for about the same or less money up front and as well as higher moisture resilience by having a drying path toward the interior.  Done as new-construction EPS runs about 12-15 cents/R-foot installed, cellulose on the order of 5-7 cents/R-foot, installed, but as retrofit it'll usually be a bit more. Even though 15 cents R-foot might seem like it's getting up toward spray polyurethane on cost, the fact that it's not thermally bridged means you're getting the full-value out of that R instead of only about half. The true comparative when applied to the whole-assembly performance is that it usually comes in less than half the $/whole-wall-R to go with exterior EPS than going with closed cell foam (any thickness) as cavity fill.

BTW: Moving the stud spacing to  24" o.c. instead of 16" adds only about R0.8 to the whole-wall R estimates for 3" of closed cell foam (without the 2.5" of fiber to fill up the void.)

The air-resistance of micro-gaps created by crinkle type housewrap is significant, so the volume of air convecting- miniscule in fact, far less air than what you'd normally see being convected through R19 batts in a 2x6 cavity that hadn't been well caulked or foam sealed. That tiny energy use penalty is worth paying for the resilience gained by the better drainage/moisture-purge. It's not nearly as purging as a 1/4" rainscreen gap behind the siding, but it doesn't need to be, since the bulk moisture intrusions that reach all the way behind the foam are going to be orders of magnitude lower than what blows by the siding.

Dana,

Thank you again for your thoughts, experience and calculations. I wish we would have had this discussion a year ago . . . . It sounds like adding the 2 inches of EPS on the outside may be the way to go. Although I would love to add blown fiberglass to fill the cavities and pull out the polyethylene, it would be a retrofit now as the interior walls are textured and painted. I do not have a good feel for the costs of such a retrofit now, although my wife's patience might be tested if I found out.

Thanks for your insights about the thermodynamics of the crinkle housewrap. I have been trying to figure out for a year how it would not ruin external insulation efficiency. Knowing that it is just an issue of capillary resistance due to narrow diameters promoting water flow and resisting airflow answered that.

Now, to discuss all this with the contractor . . .

The only case I can think of where insulated yet ventilated siding makes sense is in the rare case where you only want to block solar gain - say a tropical country where you are not using AC. Then insulation plus ventilation can outperform tight air sealed insulation.

What is the proper way to attach the 2 inches of EPS and noninsulated vinyl siding to the sheathing? What kind of fasteners should I be using for that? Thanks!

Posted By Craig Doser on 28 Feb 2014 12:29 PM
What is the proper way to attach the 2 inches of EPS and noninsulated vinyl siding to the sheathing? What kind of fasteners should I be using for that? Thanks!

Use 1x4 furring through-screwed 1.5" into the studs 24" o.c. with pancake head timber screws (eg FastenMaster Headlok, available ins some lengths at box-stores), and attach the siding to the furring. With 1x furring 2" foam and 1/2" sheathing that's a 5"  screw at about 55-60 cents per screw.

More details about hanging exterior foam can be found here, with links on that page to even finer details about the particulars around window & flashing, etc.

A discussion about hanging vinyl siding on furring can be found here.

Super! Thanks again.

I posed basically the same questions to the Cold Climate Research Center here in Alaska, and their response was quite a bit different. It appears that the climate here makes things different. Honestly, it leaves me scratching my head about all of this. Here is what they said:


I think you have two main paths forward that are very different. It will depend how high you want to shoot in terms of R-value for the walls. So I'll give a brief description of each and will let you take it from there to see what best fits your goals.

Assumptions
1) You installed closed cell polyurethane foam. The answers will change if it's open cell polyurethane.
2) Your shop is located around Kenai in terms of climate. If you're in somewhere like Homer, the second answer will change a little bit.

Option 1 (R-20)
No additional insulation, just focus on optimal air tightening. Add Tyvek before siding and detail it very well. From the vapor retarder, polyurethane and Tyvek acting in unison, you should have a very airtight assembly, and it should perform better than what people think of when they consider a typical 2x6 wall.

Your nominal R-value, without considering the studs, would be roughly R-20. The whole-wall R-value won't be substantially less if you have 24" oc stud spacing without a lot of extra framing, which are details we can go into if you want.

Option 2 (R-50)
Add sufficient exterior insulation to make the wall assembly follow REMOTE wall principles. For the Kenai, this would be roughly R-30. Typically we find that people balk at adding that much exterior insulation in the form of EPS and XPS, although it's possible to do. A different solution would be to create some form of an external framed cavity so you can fill it with a blown-in or fibrous insulation (such as fiberglass or cellulose). If this is a new idea you are unfamiliar with, take a look at this short report to get a rough idea for an example for new construction:
http://www.cchrc.org/docs/snapshots/arctic_wall2013.pdf

Why such a wide range from R-20 to 50 with nothing in between? Because of concern about moisture control. We've found from our experiments that adding just a little exterior insulation poses long-term risks for moisture accumulation and mold growth in wood framing. So of course you can install 2" of EPS on the exterior, but it's not something we can recommend as best practice. If you want to read more about the moisture control concerns, there are several short reports at the bottom of this web page:
http://www.cchrc.org/safe-effective-exterior-insulation-retrofits

Besides helping you out with your shop, please feel free to pass along to your contractor that CCHRC is available to help figure out the most cost effective insulation strategy given specific budget constraints. We'd be happy to work with him from the start for future projects. We've done that with various builders and housing authorities where we model designs being considered to compare the initial cost versus cost for heating over time. We also can help with advice on trying new building techniques such as REMOTE, double-stud walls, etc., as that's very much part of our mission, and we're happy to help.

Let me know what more information we can provide and best of luck with your shop,

--
Colin Craven
Building Science Research Director
Cold Climate Housing Research Center
www.cchrc.org
907-457-3454

---------- Forwarded message ----------
Date: Tue, Mar 4, 2014 at 8:45 PM
Subject: Form submission from: Contact Us
To: [email protected]


Submitted on Tuesday, March 4, 2014 - 20:45
Submitted by anonymous user: [166.147.88.16]
Submitted values are:

Name: Craig Doser
Comments:
Hi,

I'm looking for advice. I am trying to complete the construction of what is hoped to be a very well insulated large garage/shop that is 40x44. The contractor talked me out of following your typical REMOTE external insulation plan because he felt it was cost prohibitive. Instead, what we have to date is 2x6 construction with the wall built as follows: polyethylene vapor barrier underneath sheetrock, and the stud bay has 3 inches of urethane sprayed against the external sheathing. The remainder of the stud bay is empty except for wiring. We did not get the outside completed before winter, but it was tentatively set up (lights, outlets, faucet, windows) to use insulated vinyl siding Since I have had the winter to research this, it appears that not filling the stud bay completely with some insulation was a mistake. Unfortunately, the interior walls are textured and painted now. The building is heated now with infloor heat via boiler and natural gas. Considering all of that, I am wondering what I can safely do on the outside to further insulate the walls without causing condensation and rot issues? Would I be able to use Tyvek drainwrap against the sheathing, then 2 inches of EPS, followed by noninsulated vinyl siding? Or would you recommend some other way to reduce heat loss from the studs? Any advice you could give would be greatly appreciated, as my contractor seems to be not terribly familiar with superinsulation of structures, even though he raised here on the Kenai Peninsula.

Thanks so much!

Note that the CCHRC did things that one wouldn't do in a new house:

. An outlet box was installed on each wall to allow for sensor wire passage and to
create an air leakage path. The stud cavity and 6 mil polyethylene sheeting were installed to mimic the quality of
construction typically found in older homes in Alaska

They also positively pressurized the house and ran one winter at 40% humidity. Fiberglass (vs cellulose). Overall, they proved that you have to deliberately work at it to get mold to grow, even with 2" of external EPS.

First even with 24" spacing on the 2x6 the thermal bridging of the R3 studs is huge, more than half the heat loss of the wall assembly (not counting windows), which can be cut measurably by filling the remaining 2" with cheap fiber.  The nominal R20 of 2" of ccSPF underperforms the R20 of full cavity fill and there's no getting around that.  The framing fraction of a 24" spacing wall is about 25% framing. When that 25% of the total wall area is R3.6 (3" of cc foam) it's a dramatically higher heat loss than when that 25% fraction is R6.6 (full cavity fill). You can't cheat the physics- heat will follow the path of lowest impedance.  Bumping the stud spacing to 24" only reduces the framing fraction to about 20%.

Second, the vapor permeance of EPS is pretty high compared to other foams, and at less than 4" provides adequate drying toward the exterior for an assembly that already has 3" of closed cell foam &/or a poly vapor barrier on the interior- it's not a moisture trap.

Did you even look at the video in his link? (http://www.cchrc.org/safe-effective...-retrofits ).  A few issues, one pretty major:

1) They were using 2" FOIL FACED EPS which is as much a vapor-barrier as polyethylene sheeting, and absolutely the wrong stuff to put on the exterior, since it blocks vapor diffusion drying completely, and forms a moisture trap in combination with the interior vapor barrier.

2) They also used flat Tyvek up against the foil facer which interferes somewhat with the bulk moisture drainage issue too.

3) It wasn't clear what the siding had been installed, but note there was also ZERO roof overhang, which means wetter siding. 

It all points to the fact that if you make three design errors on a single assembly sometimes it's not going to be very resilient.  In this case the foil facer is probably hands-down the biggest mistake. 

If you use unfaced EPS, crinkly housewrap, have at least some roof overhang, then use inherently back-ventilated siding like vinyl your sheathing won't ever look like the moldy one in the video.  Also, they talked in the video (but didn't show) about how ugly it looked on the interior side, and indicated a fault in the poly vapor barrier. That won't happen to you since you have air-impermeable foam between the poly and the wood- the air-transported moisture only gets as far as the foam, after which it's all vapor diffusion via 0.3-0.4 perm polyurethane which is sufficient to block 2/3 of the moisture from reaching the Tyvek compared to the plywood or OSB alone in the event of a leak in the interior vapor barrier.

If you're still nervous about it want a much more vapor-open exterior than EPS there are pretty good 2" thick rigid rock wool options to consider, and it's less work than adding exterior trusses etc. It's extremely air-retardent wind & convection, but it's very air & vapor permeable from a moisture transport point of view- it doesn't block drying AT ALL, yet liquid water hitting the exterior surface beads up rather than wicks. The installation is similar to that of hanging rigid foam, but with rock wool the furring will compress it a lot more, so you end up having to adjust the timber screws to achieve a flat exterior, but it's otherwise similar. In my neighborhood rigid rock wool is more than 50% more expensive than virgin-stock Type-II EPS, but I've heard recently of local markets where it's actually cheaper than EPS.  See also:  http://agreenhearth.com/stone-wool-...-the-home/

BTW: Got a ZIP code? (for local climate data purposes.)

Hi Dana,

Yes, the zip code is 99669. Wow. It was never my intent to cause such a storm of controversy about how to finish this garage. I truly do appreciate all of the input I am receiving.

Posted By Craig Doser on 10 Mar 2014 06:38 PM
Hi Dana,

Yes, the zip code is 99669. Wow. It was never my intent to cause such a storm of controversy about how to finish this garage. I truly do appreciate all of the input I am receiving.

There's no storm or controversy, just some confusion about the facts and some mis-applied notions.
The Kenai area? That's the Alaskan Tropics, isn't it? (OK so it's still just a WEE bit cooler there than in the Aleutians...  )

But it's silly to design for Fairbanks-conditions in ZIP code 99669. In fact the winter mean temps are warmer in 99669 than in Grand Rapids MN, though Grand Rapids has a warmer sheathing-drying season.  But from a sheathing dew point control point of view, what works in Grand Rapids (US climate zone 7A)  works for you (zone 7C) due to the mid-winter moderating of the winter averages due to the warmer ocean temps.  You get about 6 months where the mean-temp is below freezing to Grand Rapids' 5, but your December-January averages are higher, which lowers the rate at which moisture accumulates in the sheathing.

With unfaced EPS you have a sufficiently long drying season that 2" would not be a problem under vinyl siding, as long as you have an interior side vapor retarder (like the 3" of closed cell foam that you already have.) To get there without an interior side vapor retarder you'd need to put at least 2.5" of EPS on the exterior, and 3" would be better. But unlike the cold AK interior your winter mean temps are in positive double-digits F rather than negative single digits with a month shorter drying season. Your winter averages are more than 15F warmer than Fairbanks (more than 5F warmer than Fargo ND!), which annual heating-degree-days aside, is like being in a whole other climate zone (no wait, it IS a different climate zone- Fairbanks is zone 8!).

I would have no reservations about putting 2" of unfaced Type-II EPS on your place- it's still well over 1-perm which is plenty.  But as I pointed out, 2" of rock wool would be more than 50 perms, and the drying rate of the sheathing to the exterior would probably only be limited by the vapor permeance of the housewrap (Typar is about 10 perms, most Tyvek is 30+ perms). It's still worth squirting in some cellulose or fiberglass into the 2-2.5" cavities on the interior side the spray foam, which will add about ~R5-6 to the whole-wall performance. With ~0.4 perm ccSPF between the cellulose and sheathing it the extra interior-R won't affect the rate of moisture accumulation of the sheathing in a meaningful way, but it's the thermal equivalent of adding 3.5" of foam instead of just 2", and a significant percentage of the total. Unless you're heating with wood you cut yourself, at AK fuel prices there will be reasonable pay-back on the additional cavity fill (and the foam.)

Yes. We're sort of in the Banana Belt of Alaska here. . . . this winter especially.

I have received tremendous advice and education on this forum in response to my question. I wanted to thank everybody so much. I will have much more peace of mind (and lower utility bills) going forward.

Thanks again!