GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Window U-value comparison to wall R-value and window replacement

Keith Miller | Posted in General Questions on

We have an 1870s Victorian farmhouse that doesn’t have insulation in the walls, and to keep it simple, we cannot add insulation in the walls without going to great expense, more than we can afford.

We want to reside our house so we are going to remove the aluminum siding that covers the original cedar siding. Our plan is to then cover the existing cedar siding with 1″ polyiso board (R-6) before we install Celect siding (cellular pvc siding). We are limiting ourselves to 1″ insulation board since more than this reduces the distance the eaves overhang and impacts the limited, but present, ornamental trim along the eave. While we are installing new siding we have made the sad decision to replace our wonderful, but drafty and deteriorating old, original DH wood windows. My questions/clarifications:

1) This one seems easy to me but I just want to make sure I’m not missing something: If my wall has an insulated R-value of 6, then I assume it doesn’t make sense to spend the extra money to buy a windows that has an equivalent heat loss factor higher than 6, correct?

2) I’ve spent months researching windows and I see that most people on here recommend/prefer fiberglass windows. I am obviously not opposed to PVC since I’m wrapping my house and trimming with it so can anyone recommend a high-end PVC window manufacturer(s) that might be close to fiberglass, even if it doesn’t equal it? I just want to compare fiberglass to vinyl in both specs and price. There is so much marketing hype out there it is really difficult and time consuming to compare anything. I prefer individuals expertise/opinions to help weed through the claims.

3) I’m planning to install these replacement windows as “outtie” windows and I will treat the outside surface of the insulation board like new construction sheathing (the replacement windows made with a nailing flange). Essentially, these windows are installed like new construction but will slide in from the outside “over’ the existing window frames like a replacement window. I will either order extended window jambs from the manufacture or (more likely) customize my own to fill the gap between the interior window stop trim and stool trim. I looked at doing an “innie” window but there is a lot more involved with trimming the exterior around the window. I have read a few articles on this (including the “innie” and “outtie” article on GBA) but I haven’t heard of anyone doing it this way. Does anyone have any comments, positive or negative, to my plans?

Thank you!

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. Stephen E | | #1

    For cost wise the composit windows do very well. Would replace all windows that you can (code) to picture windows. U and R value are secondary to air infiltration in windows. In an older house I would be more concerned about air infiltration than foaming up. Keeps your original style, cost less and has higher ROI. Get to 1.5 arc at 50 pascals and you will be doing a lot better than just foam.

  2. Expert Member
    Dana Dorsett | | #2

    An R6 window has a U-factor of about U-0.17, which is a fairly pricy window. A code-min replacement window would be U-0.320-U0.40. (depending on your location - See: )

    If the windows aren't actually rotted out and can be rebuilt tightened up cost-effectively (most can), a tight low-E storm window over tight wood sashed double-hung single panes comes has a combined performance of about U0.32-U0.34, for LESS money than a new replacement window. Harvey Tru-Channel storm windows are the tightest in the biz, and they have a hard-coat low-E glazing option, but they are a regional player (northeastern USA), but the Larson Silver or Gold series low-E storms sold through the box store chains are pretty good too.

    It's almost always cost effective to squirt some cellulose into the wall cavities from the exterior when re-siding, which will reduce air infiltration losses by quite a bit, whereas with a foam-over-only those wall cavities are all potential thermal bypass paths that come with an inherent stack-effect infiltration driver.

    In cold/very cold climates polyiso may not be your best choice, whether you insulate the cavities or not. When the average temp through the foam drops below 40F or so EPS may actually out perform it. When the average temp through the foam is below 25F EPS will outperform most polyiso by a substantial margin. The labeled R6 is it's performance when the average temp through the foam is 75F. EPS gains performance fairly linearly with falling temp, whereas polyiso has a funky performance derating curve that falls in performance with temp- gradually at first, then completely off a cliff at mid-foam temps much below 30F.

    At 1", EPS with cellulose cavity fill would work well enough in climate zones 1-4 (and into zone 5) without having to worry about interior side vapor retarders, and if you use unfaced goods, provides a reasonable amount of drying capacity toward the exterior. In colder zones 1" rigid rock wool would be better, since it would provide the maximum exterior drying capacity for protection your (probably wide board plank) sheathing.

    If you can strip the existing siding you would be better able to make it air-tight by detailing a vapor-permeable housewrap as an air barrier, and that would give you another 1/2-1" for exterior side insulation. Unlike R6 polyiso, both R6 EPS and R6 rock wool would deliver BETTER than a true R6 in cold climates, since both increase in performance with falling temp. Type-II EPS at 1.5" tests at about R6.3 @ 75F mid-foam temp, but that improves to R7 @ 25F mid-foam, which is about 2x the performance of 1" polyiso at that temperature profile. R7 isn't great, but it's way better than R3.5-R4 (the true performance of polyiso at climate zone 7 winter extremes) or R4.7 (the true performance of EPS at those temps.)

  3. Keith Miller | | #3

    Thank you very much for your quick and informative replies!

    S E : I understand the benefits with your recommendation but since this is an old house I am hesitant to eliminate DH for picture windows. I like the function of DH windows and I guess I need to put up with the fact they are less tight and therefore worse at keeping heat in.

    Dana : great information - I will have to digest a good bit of it. I didn't know about EPS vs Polyiso. That is misleading to rate the Rvalue at one temperature if that value changes drastically over a range of temperatures (real life).

    With regards to spraying foam into the wall cavities I dont know how much that will benefit my situation. The walls are filled with brick (called brick nogging) and the mortar was not smoothed out but rather squeezed out when put in. So the existing air cavity is very irregular and varies from .5" to 1.5". I only know this since we removed a small section of a porch wall to install some HVAC duct. If we had the money we would remove the cedar siding, the brick nogging, insulate the wall cavities, resheath the entire wall , WRB and then siding. And I have looked into restoring our old windows and the pricing I received was just as much as replacement windows. I would say 20% of the windows have a broken piece of glass that is large. This glass is very expensive to replace (wavy glass which I love). And then the sashes/jambs/sills/stiles are obviously covered in what seems like 5 layers of lead paint, and all of the paint has alligatored and deteriorated to certain degrees. The sashes can be removed and restored offsite (nobody does them around here) but to strip the jambs and repaint is very time consuming, difficult to do correctly and therefore expensive. Some of our windows are fairly big ( 30x82) so you are right in that it isn't cheap to buy new ones, either. but it looked like it was going to cost us about $800+/- to restore one large window (incl. new, custom storm window but does NOT include new glass, which can be almost $200). Then we still need to use a storm window, which doesn't thrill my wife, and repaint every 'X' years (which doesn't thrill me). I can't say replacing them is the right decision but we have so much to do with all of the buildings on our old farm property that if the decision isn't clear to me one way or the other I lean toward lower maintenance whenever possible. If more people in my area (NJ) restored their windows then I believe the price could be much more persuasive.

  4. Charlie Sullivan | | #4

    On the question of whether Rwindow > Rwall makes sense, the window heat flow and wall heat flow are actually independent. The amount of heat the leaks through your windows depends on on their U factor and the temperature difference between inside and out. It is not affected by your wall insulation. So if it's worth the cost to get U=0.2 windows vs. 0.3 windows, it's worth it, and if it's not it's not, whether you have an R-5 wall or an R-30 wall.

    On the other hand, if you have a fixed budget, you of course want to spend it where you get the biggest bang for the buck, and that's rarely on windows. But if you've already decided to spend the money on replacing the windows based on maintenance, not energy, that's the biggest part of the cost, and the difference between different performance levels in the windows is generally a lot smaller than the overall project cost of replacing the windows.

    It would be kind of fun to end up with Rwindow> stay warm, you' want to sit next to the windows rather than the wall! But there have been some good suggestions about how to help your wall R-value, so perhaps you won't end up with that situation.

  5. Keith Miller | | #5

    Thanks Charlie. I know what you are saying but as you mentioned it seems kind of silly at face value to end up with the window's Rvalue greater than the wall's Rvalue unless a good bit of the wall surface is windows. It is unfortunate that the price jumps so much to insulate my exterior walls but that is the reality of my house and simply not even close to feasible with our budget.

  6. Expert Member
    Dana Dorsett | | #6

    " That is misleading to rate the Rvalue at one temperature if that value changes drastically over a range of temperatures (real life). "

    There are many climates where the average winter temps remain above the knee in the curve. The FTC requires that R-value be labeled at it's performance at 75F average temp through the foam using ASTM C518 test methods. Manufacturers of EPS will usually include the +40F and +25F ASTM C518 tested performance in their specification sheets, since it is substantially better than the legally labeled performance. (eg: ) Polyiso is typically derated to R5.5-R5.7/inch for sheathing applications by the manufacturers, but SFAIK that is not an FTC requirement, and that R5.5/inch may still be optimistic for average performance in climate zones 6 & 7, depending on how much other insulation there is to the interior & exterior sides of the polyiso layer.

    If there is effectively no R-value to the materials on the interior side, the average mid winter temperature in your climate zone 7 area is 10F and your average interior temp is 70F, the average temp through the foam is still 40F or so, and the stuff will still be performing OK, if below it's labeled rating. If you have an inch of polyiso on the exterior and have R15 fiberglass or rock wool in the 2x4 cavities, the average mid-winter temp through the foam is going to be about 25F-40F, and it will perform miserably. But if you live in climate zone your average mid-winter temps are well above 30F, and the average temp through the foam would be well into it's pretty-good performance range.

    In cold climates you can mimic the warmer climate performance by layering different foams in stacked layers. You can still get polyiso's higher R/inch to advantage by being conscious of where you place it in the stack-up. Using EPS on the exterior side of polyiso at roughly equal thickness provides better average performance in climate zone 7 than an equal thickness of EPS alone, and much better coldest-temperature performance than polyiso alone.

    "With regards to spraying foam into the wall cavities I dont know how much that will benefit my situation. The walls are filled with brick (called brick nogging) and the mortar was not smoothed out but rather squeezed out when put in. So the existing air cavity is very irregular and varies from .5" to 1.5". I only know this since we removed a small section of a porch wall to install some HVAC duct. "

    I was recommending cellulose, not foam for the wall cavities, but given the brick nogging situation you might actually do better with non-expanding injection foam than cellulose. The brick nogging is not air-tight which leaves it as a injection foam will find & fill the spaces between the brick no matter how convoluted the shape. If there was more space than 0.5"-1.5" to deal with cellulose would do nearly as well for tightening up the cavities, but there will be gaps due to clogging during installation if it narrows to a half-inch in places. Injection foam (eg TriPolymer, CoreFill 500 etc) is an updated, much less problematic variant of the disaster-stuff that was around in the 1970s, and it will work better than expanding polyurethane foams in this application. With injection foams (unlike polyurethane) there is no chance of wall blow-outs, and a much reduced chance of voids that would occur where expanding foams would create blockages before the cavity was able to completely fill. Injection foams are only under pressure during installation, and they will shrink a bit as it cures, but those thermal bypass passages will shrink to something less than 0.01" compared to the 0.5-1.5" you currently have. I'm not a big fan of the stuff for most applications, but this is one of the exceptions- you have very few alternatives for improving the air tightness those wall cavities, and if you don't, the air leaks will undercut the performance of your exterior rigid foam.

  7. Keith Miller | | #7 is a picture of the wall with the cedar removed (inside our enclosed porch). I removed all of the excess mortar that "squeezed out" of the brick joints since at one point we thought we might use the nogging in this wall decoratively.

  8. Keith Miller | | #8

    Wow - simply awesome! Thank you so much Dana. I guess I should bite the bullet and drill a bunch of holes in the cedar siding and have the bays filled with foam, even if it is only 1/2" in some areas. Given the irregular nature of the brick & mortar I think it will be challenging to figure out how many holes to drill in each bay to get a complete fill. Even if we aren't 100% successful I agree that it will only improve the wall's performance - 85% foam is much better than an air space. And if I can average almost 1" of injection foam then that adds significant Rvalue to the wall assembly. BTW, I live in zone 6a (-5F to -10F) so it looks like EPS is what I should use.

    Just to be clear; your recommendation for my specific wall profile would be as follows (from interior to exterior):

    1) ex. Plaster
    2) ex. Brick Nogging
    3) new injection Foam
    4) ex. cedar siding
    5) eps rigid foam board to max thickness I can do (1" for me)
    6) 3/4" wood straps lined up with studs (needed?)
    7) WRB (window will be "outie" & I put WRB over straps to protect them)
    8) siding fastened to wood straps through WRB

    Now I just need to figure out which brand windows I'm going with and we should be good. The injection foam is going to eat into our budget and I think that is money better spent on the wall than the windows so I will have to go with a lesser rated window - any recommendation for a mid-price higher quality window!?

    Again, thank you - thank you - thank you for your advice!

  9. Keith Miller | | #9

    Remaining 3 questions (hopefully):

    1) While I was putting together my materials list I started to question if I need the 3/4" wood strapping/furring strips. Since I am only doing 1" of EPS rigid board and Celect is a lighter product Celect's installation specs do not require the furring strips for support. The only spec is that the fasteners penetrate the studs a min of 3/4". Thoughts?

    2) As mentioned above the injection foam will take from our budget (I don't know how much the cost will be yet) so I need to go with a mid-level window. Any recommendations?

    3) Regarding injection foam insulation (not expanding foam): Is anyone familiar with a DIY kit (not the small cans bought at mass retailers)? I will be residing the house/replacing windows in many sections over a period of a year or two so my guess is it will be expensive to have a contractor come out multiple times for "small" jobs to inject the foam.

  10. Stephen E | | #10

    A possible solution that worked on one of my retrofits is interior foam panels. Caulk the top and bottom of the panels. Tape and mastic the seams. Put drywall over it. 1/2 foam panels worked well with normal 2 1/2 inch drywall screws and adjustable electrical outlet boxes. Caulk those as well. You are able to wrap the interior box of new windows with foam panel in this configuration as well. Possibly old windows. Gives you continous insulation, an air barrier, and reduces shrinkage of foam.

    A bonus of interior foam panels is the ability to use code compliant plywood window and door headers. This is on the interior side without removing a part of the exterior wall. Most older house are missing headers so this technique eliminates the header heat sinks.

  11. Expert Member
    Dana Dorsett | | #11

    Is there no plank sheathing behind the cedar siding?

    If that's the case you'll be better off injecting the injection foam ONLY from the interior side, leaving the small air-gap as a capillary break & drying path for the cedar.

    Having the foam crew do the whole house in one shot will be cheaper than a section at a time, but you don't have to wait for the EPS and siding from a building-science point of view, but it might be a cash flow issue.

    If you're willing to take some risk of potential blow-out should you over-fill, it's possible to use "slow rise" polyurethane pours using kits from Tiger Foam Fomo-Foam et al. With a 1" average depth to the cavity a 600 board foot kit would do about 900 square feet of gross wall area. The kits run about $700 for a 600 board foot kit, and you'll need to buy some protective clothing and spare tips for the gun, etc so it could be as much as $850 with shipping and extra materials, but probably not more than $900. If you can get even a half inch of closed cell polyurethane between your interior plaster and nogging (if there is indeed a space there it would be an excellent air-barrier

    It's much easier to get satisfactory results if you do it in the spring, when the wall temps are above 60F (Read all of the instructions. They have smaller kits too, as well as fast-rise formula stuff. if you stripped the cedar siding as you go and fatten the depth of the studs by adding 1x furring it doesn't take a lot of practice to lay it on in 1.5-2" depths without having to come proud of the furring and need to trim it back. You'd still be able to put an inch of EPS on it (figuring the clapboards stack up to 3/4-1".), and you wouldn't have less room for air to move in that the lateral channels you'd have with flat EPS up against the exterior profile of the cedar siding.

  12. Keith Miller | | #12

    S E: Thanks for the suggestion! We do not want to eliminate/cover up our plaster walls right now and it becomes a much more time consuming & messy job with removing the plaster/trim and reinstalling the interior trim (in addition to redoing our siding). If our plaster was in poor shape then working from the inside would be a great suggestion and we could maybe tackle 1 room at a time. But if we were going to redo our interior walls we would probably go as far as removing the plaster and brick nogging and insulate the wall cavities.

    Dana - thanks again! This is never as simple as it seems it should be. Correct, the cedar siding is fastened directly to the house studs without any sheathing. Unfortunately, the brick nogging is essentially against the plaster walls making it impossible to fill the wall with foam from the inside. So are you saying that I shouldn't fill that air space between the cedar and brick nogging with foam? I've used Tiger expanding foam kits about 6 years ago. It works but very messy stuff and it didn't expand nearly as much as what the pros' foam does. I did have to heat the tanks to keep the stuff flowing properly.

  13. Expert Member
    Dana Dorsett | | #13

    You can (and SHOULD) fill the space between the cedar & nogging, but only when you are installing the exterior foam & WRB.

    If you filled the space without putting up the exterior layers the cedar will warp, split and fail in about 1-2 years.

    Thinking it through on my drive home last night- if you are willing to strip the cedar as you go you can fill the space with split batts thick enough so that it's a compression fit for the next layer, that would be cheaper and a more complete fill than trying a DIY spray foam job. With the thickness of the cedar gone you can probably fit 1.5" or 2" rigid foam in there, which would have lower thermal bridging than the spray foam solution that I discussed earlier.

    If your siding is super-light you don't need 1x furring to hold the foam in place. You could instead rip down some half-inch OSB and through-screw it to the studs 16" o.c. (spaced that tight for flatness), then use shorty ~1" ring-shank nails to hang the siding onto the OSB furring. That gives you another 1/4" of depth that can be dedicated to foam thickness, another R1 or so of whole-wall R (which would be a double-digit percentage improvement in this stackup.

    If you take the split-batt approach it's going to be worth air-sealing the foam layer to the studs & framing at each stud bay with either a bead of acoustic sealant or foam-board construction adhesive. With a powered caulking gun that's less work than it might seem. But any air leakage that gets behind the foam cuts into performance.

    With 1" average of compressed batt and 1.5" of Type-II EPS as sheathing (covering the framing as well as the compressed batt) your whole-wall R comes in at about R10. If you can fit 2" of EPS that brings it up a shade north of R12.

  14. Keith Miller | | #14

    If I remove the cedar then wouldn't I have to put OSB or plywood sheathing over the studs? I originally got several siding replacement quotes that included removing the cedar and replacing it with plywood prior to siding and that price ($110,000 not incl windows and insulation) is what led me to doing it myself. We have 40+ windows so that's not a small expense (is it ever?). If I was going to remove the cedar then I would go ahead and remove the brick nogging and start insulating from scratch.

    Your ideas/recommendations are really excellent and have made me think about how to best tackle this project. I want to do it right but it seems I've got too many variables that, if addressed, snowballs the project and puts us way over budget. At some point I just need to do the most I can with the money we have.

  15. Expert Member
    Dana Dorsett | | #15

    From the picture you showed with the clapboard stripped, the framing has cut in cross bracing, which would by itself eliminate the structural need for plywood sheathing. And the nogging also provides substantial rigidity and support against racking loads even if the cut-in bracing wasn't there.

    The way cedar splits there is no WAY that the cedar clapboard siding is structural! Under racking loads over time the side pressure on the nails would have split the cedar, and it wouldn't still be hanging on the house 144 years later!

    Worst-case you could install plywood racking panels at the corners only- it doesn't need full coverage even if it didn't have the nogging. But if the photo is any indication the structural capacity of that cut-in bracing is WAY better than many much-newer homes from the 1950s and later with fiberboard sheathing.

    If any of the nogging seems loose it's probably wort gobbing in some pointing mortar in the big gaps or a quick & dirty hard-mortar parge, (making it a bit more air-tight too) but it doesn't warrant the full-pointing treatment that weather-exposed brick would. If the bricks don't wiggle and the mortar isn't falling out, don't bother.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |