Insulating existing steel frame house
I have one problem room, a double car garage that had been converted into a family room. House was built in 1999, not sure of the construction/insulation method that was used. Total wall thickness is about 8 1/4 inch including the brick veneer, which looks like a conventional 3 inch brick.
That converted garage room is cold and difficult to heat. I had some additional heat outlets added and a small return to that room, better but not OK. I currently have that room shut off and closed as I do not currently need that space, but would like to be able to use it.
I had an insulation vendor come in and suggested spray in expanding open cell foam installed from the outside. Based on building practices in the Knoxville TN area, he is quite sure that outside wall in the former garage has no insulation. After discussing the cold walls in the laundry, kitchen area, he also suggested doing that for the whole house, I have a laundry room and kitchen on outside walls and they seem to radiate cold as well.
Not sure if this is a good idea for this type of construction. Need to know what questions to ask and if we need to do some limited demo to see how the house was built before I know what to do. He did agree to come back with a thermo imaging camera to take a look at the whole house and look at walls, windows and doors. I know I have a couple of issues there as well.
Would appreciate any input on this project.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
![](https://www.greenbuildingadvisor.com/app/themes/greenbuildingadvisor/dist/img/modules/detail-library/thumb1.png)
![](https://www.greenbuildingadvisor.com/app/themes/greenbuildingadvisor/dist/img/modules/detail-library/thumb2.png)
Replies
User-6976262,
First of all, can you tell us your name?
Second, it shouldn't be too hard to tell whether there is any insulation between your studs. One way is to remove the cover plates from electrical boxes (receptacles and switches) on exterior walls. Expand the crack (carefully and safely) between the drywall and the box, and look into the cavity with a flashlight, or snag a little insulation with a crochet hook. Warning: turn off the electricity at the circuit breaker before doing this investigation.
One big caveat: Installing any type of insulation between steel studs is almost useless, because of the serious thermal bridging through the studs. The only effective way to address the thermal bridging through steel studs is by installing a continuous layer of exterior insulation (usually rigid foam) between the wall sheathing and the siding, or by installing a continuous layer of interior insulation (again, usually rigid foam) between the studs and the drywall.
Are the studs steel, or just the large structural beams?
Steel studs
Appreciate the response. The house has full brick veneer all round, so pulling that off is not very attractive to do and uneconomical. It appears that just that wall in the converted garage is not insulated, confirmed by prodding and noting the wall temps. The other exterior walls I checked have pink fiberglass insulation.
I know it is not an ideal scenario, but from what I read, insulating the cavity does provide R value. My concern is creating a moisture issue in the walls when I fill them up with foam. I am stuck with what I got for the time being and trying to make it work as best I can, without causing another problem.
Bruce Harkness
The studs may already have fiber insulation in them. If that's the case other than cutting into the air leakage, open cell foam would not improve the thermal performance AT ALL. With a hole-saw find some locations on the exterior walls to make some exploratory 1.5-2" holes in the wall board so you can get a good look at what's inside the walls, and see what there is for sheathing (if any) on the exterior side of the studs.
A 2x4 wood studwall with R3.7/inch foam or fiber insulation has "whole-wall R" of about R11. But the high thermal conductivity of steel makes the equivalent wall about R6. Displacing the fiber with expanding foam of similar R value doesn't change anything- it's the conductivity of the steel that's killing it here.
Adding an inch of foil faced polyisocyanurate foam behind the wallboard, adds about R6 to both the steel stud fraction and the fiber fraction, which would bring the assembly up to the R12-ish whole-wall range, which will make a noticeable difference.
I can't probe the outside walls with large holes, it is brick veneer on the whole face. Redoing all the interior walls would be equally painful considering all the trim and door/window openings would need to be finished to accommodate the thicker walls, in addition to the wall sockets/switches. It might be a little more tolerable if I could apply this new interior insulation and wallboard, where most needed, over the existing drywall.
The walls, with the exception of one wall in that converted garage, appear to have fiberglass insulation in them.
Bruce,
The choice is yours. But I agree with Dana -- If you can't install a continuous layer of rigid foam, it's hardly worth making any changes.
If you decide to install rigid foam, it can be installed on the interior side of the existing drywall. You don't have to remove the drywall first. Of course, you'll need to extend the electrical boxes, install new drywall, and change the trim.
Thanks a bunch for your inputs, looks like I will just address the one wall without insulation. That is actually the easiest as it has no openings, except for the main breaker panel.
I was suggesting making the exploratory holes on the finished interior walls (say, under cabinets or in closets where the cosmetics of the repairs matter less), not coring through the brick veneer.
IRC code-min for steel framing in lieu of 2x4/R13 has long been R5 continuous insulation on the exterior, with R11 minimum in the stud bays, at least since IRC 2003. I"m not sure about older versions. It would be good to know if that is what indeed got built, or if it was R11s without the R5, or something else. If the stud bays are in fact EMPTY, blowing cellulose or fiberglass into them drilling from the interior side isn't that big of a deal (hole saws make clean, easy to patch blowing holes), and it would be quite a bit cheaper & safer than half-pound polyurethane foam pours, which can bow or even pop the wallboard, and it effectively glues it all together making repairs a real PITA.
Got it Dana, I guess I should probe from the inside, that makes much more sense.
The contractor is proposing a non-ISO product called Retrofoam. When he comes back with the IR camera, I can ask him about other options for doing that empty wall from the inside. The IR should confirm that the wall is empty, as expected. Will also be able to confirm the stud spacing.
I looked at some articles on R value, if the studs as expected are 24 inch spacing, I lose about 40-48% due to the studs. I have not gotten a written quote yet, but I think it is in the neighborhood of $3300 to do the whole house with the non-iso open cell. That seems like peanuts to try this approach to optimize what I have now to get max insulating value and eliminate any air leaks. I can always go back and do some more drastic measures later. I doubt the more radical approaches are going to have attractive investment recoveries.
According to ASHRAE tables, at 24" o.c. with R13 in the cavities you'd be at ~R7.2. With R15 in 24" o.c. cavities you'd be at R7.8. Most houses are built with 16" o.c. framing, which with R13 would deliver ~R6.0.See the chart on page 3:
http://www.taitem.com/wp-content/uploads/2011/01/TT-NC-Calculating-U-values-Nov-2008.pdf
Blown fiber does a substantial amount of air leakage reduction too, especially cellulose. With fiberglass it has to be dense-packed to 1.8lbs or higher, but cellulose is pretty air retardent at any density. The air pressure of the blower push the insulation into all air-escape paths- it's not bad at all! The differences in air tightness between 1.8lb fiberglass, 3lb cellulose (inherently denser than fiberglass) and half-pound polyurethane is fairly academic. Even damp sprayed cellulose is pretty tight, though not as tight as cellulose blown under pressure inside enclosed stud bays. The insulation can only air seal the framing bays, and doesn't touch the (usually significant) air leakage under the studwall plates, etc. Blown fiberglass comes in at about R4.1-R4.2/inch @ 1.8lbs density, but it's the additional air tightness rather than the center-cavity R that would be important here.
Note the differences in air tightness of various approaches in this bit o' bloggery:
https://www.greenbuildingadvisor.com/blogs/dept/musings/air-leakage-degrades-thermal-performance-walls
Wall #4 is damp sprayed cellulose, which is leakier than any blown cellulose, and still doing OK, if somewhat less well than foam. Unfortunately they did not run comparative tests with blown fiberglass or cellulose, or dense packed fiber.
Thanks again, for the excellent references. I will punch a hole behind the cabinets to see what I have. Wait for the IR and return of the contractor.
At this point, I know what I need to do on the empty wall. Get the cavity filled and build a new interior wall layer. That wall is easy, 16 ft or so with one penetration currently. On the rigid foam, would I use one with a vapor barrier or not?
Still debating with myself on the rest of the house, if the quote is cheap enough, why not? Any concerns on condensation in the wall? I think it would be a non-issue, but as you can tell, this is not my current area of expertise.
Just one more question, the house also has steel rafters. It has 10 inches of blown insulation currently. Is there a big gain by putting another 10-12 inches of blown insulation in that space? I searched around an found no comparitive data on insulation in steel roofs and what works best.
One option the contractor could do was suck out the old and cover the roof with SPF, the ISO closed cell. He claims in a wood frame they can keep the temps much lower as that attic space is now part of the thermal envelope of conditioned space. . It is a relatively expensive option, ~ $10K vs blown in at 1/4 the cost. Not sure what works best with steel frame.
Bruce,
Are you talking about rafters? (That would imply that you have an insulated cathedral ceiling or an unvented conditioned attic.)
Or are you talking about joists? (That would imply that you have a vented unconditioned attic.)
If you have an insulated cathedral ceiling, the best solution to your thermal bridging dilemma would be to install a continuous layer of rigid foam on the exterior side of your roof sheathing. For more information, see How to Install Rigid Foam On Top of Roof Sheathing.
If you have a vented unconditioned attic, the best solution is to pile very thick insulation on your attic floor -- well above the tops of your steel joists.
Sorry, I meant joists, vented unconditioned. It has 10 inch blown now. So R19ish would be about right for a wood joist roof. I am assuming less R with greater conducting steel, I never found a reference where this data was comparing steel joist vs wood for attics. Just looking for geuidance on how much more to put in. If I recall, new homes are much higher R than what I have now.
Bruce,
Q. "It has 10 inch blown now."
A. So, maybe R-24 in the middle of the joist bay. And an absolute thermal disaster at the steel joists, undermining the value of the blown-in insulation.
Q. "So R19ish would be about right for a wood joist roof."
A. Maybe in 1965, but not now. Building codes call for a minimum of R-49 in Climate Zone 4 or anywhere colder. If I were you, I would ignore the insulation between the joists (which is just about worthless because of thermal bridging). What you want is at least R-49, measuring from the top of the joists upward.
There is no advantage to sucking out the old blown insulation unless it's full of squirrel crap and mouse nests, even if you insulate at the roof deck.
If there is sufficient clearance to install cellulose to a depth of 6" or more over the joist edges (and over the existing fiberglass) it would more than double the thermal performance of the ceiling by putting an R20-ish thermal break over those steel joists, and at 16" total depth (fiberglass + cellulose combined) it would be pretty close to current IRC code performance, despite the thermal bridging of the (now buried) steel joists. Blowing an air retardent fiber like cellulose over low-density fiberglass increases the performance of the fiberglass by blocking convection into/out of the fiberglass layer. It will compress the fiberglass a bit (which is why tracking the total depth is important) but that improves it's R/inch.
At 16" total depth center-bay the R-value of the fluff would be in the mid R50s, but the aspect that gets you to a code-max U0.026 (R38.5 "whole-assembly" is the R20+ break over the studs.
If there are ducts & air handlers in the attic it's more efficient, but a heluva lot more expensive project, and unless the budget is open-ended, the money might be better spent on air-sealing the ducts & air handler and further improving wall performance.