Insulating a steel post-and-beam house
I am using a steel post and beam building for the shell of my new home. The wall girts are on six foot centers, are 10 inches deep and have an 8 inch flange. Attached to the in side of those girts will be traditional two by four framed walls. My plan was to put 3 in of eps insulation and 7 in of blown-in cellulose behind the two by four frame wall and leave the stud cavities empty. I would use 1″ EPS furring strips between the siding and the girts. This would reduce my thermal transfer through the girts as well as producing a rainscreen of sorts. How much of a difference would there be if I put the EPS insulation in between the girts and the siding? If I do this then I have to put 10″ of cellulose behind the stud walls. It seems that 13″ of insulation is overkill.
Thanks
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Tony,
I'm guessing that your steel framing looks something like the framing shown in the photo below.
Steel-framed buildings are hard to insulate, because thermal bridging through the framing is so great that the heat flow through the framing undermines the value of any insulation inserted between the framing.
Ideally, you'll take one of two approaches: either (1) install all of the insulation (in continuous, uninterrupted layers) on the exterior side of the steel framing, or (2) install all of the insulation (in continuous, uninterrupted layers) on the interior side of the framing.
Because of the difficulties associated with insulating a concrete slab on grade (and making sure that your sub-slab insulation is continuous with your wall insulation), option 1 is much easier than option 2. You might consider using structural insulated panels (SIPs) or nailbase to insulate your building.
By the way, it would be helpful to know your climate zone or geographical location.
.
I am in Denver, Colorado. Climate Zone 5. Our building is similar but only has 3 wall girts. I recognize that all of the rigid insulation on the outside of the framing would be best. I'm curious as to how much of a difference it would make in the performance of the whole wall assembly. If it makes a small improvement in r value but requires 20 extra bags of cellulose to fill in the walls, is it worth it?
Tony,
According to "Thermal Design and Code Compliance for Cold-Formed Steel Walls", Table 6, the actual whole-wall R-value of a steel-framed wall insulated with R-21 fiberglass batt insulation, and sheathed with 1/2-inch OSB, is R-7.4. (See the table reproduced below.)
In other words, thermal bridging through the steel studs reduces the R-value of the wall assembly from R-21 (labeled R-value on the insulation) to R-7.4.
According to another source (Peter Yost), "Steel framing typically reduces the in-cavity R-value by as much as 50%, while wood framing reduces in-cavity R-value by a bit less than 10%."
You implied that you are worried about the expense of purchasing "20 extra bags of cellulose to fill in the walls." But I'm not urging that approach. That would be a waste of cellulose. What you need is a thick layer of exterior rigid foam.
.
Tony,
Here's another table that tells the story of thermal bridging through steel framing. Click on the image to enlarge it.
.
In Denver simply changing the spec on the continuous 3" foam from EPS to polyisocyanurate would get you to code-min performance on a U-factor basis (< U0.060). The U-factors in Martin's first table makes it with 3" EPS, but is only valid if there is R13 between the steel framing- it'll come up a bit shy if the cavities are empty unless the foam is polyiso.
If you want better than code-min, put the cellulose in the wood 2x4 framing cavities, not the steel framing. A non-structural 2x4 wall will have a much lower framing fraction than a structural 2x4 wall- filling it with cellulose would add another ~R10-R11 of "whole-wall-R" after factoring in the thermal bridging. If the 2x4s need to be structural (say, to support floor joists for a second floor) filling the bays with cellulose would still add at least ~R9 to the whole-wall R.
With 3" of exterior foam (any type) on the outside of a 2x4/cellulose wall there is PLENTY of dew point control at the foam/fiber boundary, eliminating the need for interior side vapor retarders in a Denver climate.
I have seen one metal building company replace the steel girts with wood. If your frames are close enough you use 2x wood girds. They could weld a tab on the frame for the girt to bolt to. This would allow a thick wood framed cellulose wall.