Insulated From the Real World

As part of a major renovation to our house in Seattle, my wife and I are going through the process of evaluating many major system components, from heating to plumbing to flooring. One part of this evaluation is to look closely at insulation and trying to separate what "contractor default" from the "This Old House" behavior. In contractor default behavior, the contractor specifies <foo> because <foo> is what s/he's always worked with. In "This Old House" behavior, I specify <bar> because I saw <bar> on a $10M remodel on "This Old House" and thought I'd like it in my own house.

Insulation is one of those things. We have a ~1200-foot 1920 Craftsman-style house in Seattle. As is typical of those houses, the framing is 2x4 studs on the exterior walls with ship-lap sheathing under cedar siding on the outside and lath and plaster on the inside. No insulation (who needs insulation in Seattle?). Our plan is to add a full second story to the house -- about another 1300 feet of space. Code requires us to insulate the second story, and engineering requires us to remove nearly all the ship lap and replace it with skip sheathing (basically 4x8 sheets of OSB or plywood) to incease the structural integrity of the house. The first part means we need to choose an insulation material. The second part means that we can install just about any kind of insulation because the house is going to be totally open from the outside.

The contractor default behavior is to use fiberglass batts. The This Old House behavior is to use expanding foam. The medium behavior is to use blown cellulose. We've gone through a bidding process on each of these materials and, as expected, the costs are about foam = ~2x cellulose = ~3x fiberglass. In other words, foam is about 3x more expensive than fiberglass. So the question is, is it really that much better?

So first a quick lesson. We’re probably all familiar with fiberglass batts – the orange or yellow stuff you see crammed into walls and ceilings or stacked to the tops of the aisles in Home Depot. It’s good stuff – pretty inexpensive, doesn’t require much expertise or special tools to install, good insulator, reasonably resistant to fire, water, etc. You may be less familiar with blown cellulose, a gray, odorless, fluffy loose fill insulation usually made from recycled newspaper often treated with a variety of chemicals to help it resist fire and insects. Unlike fiberglass batts, cellulose is blown into cavities using what looks like a big fire hose and is held in place on walls with a thin membrane.

Icynene expanding foam insulation is rather like the stuff that comes in a can at the hardware store, only… well, only different. Instead of being a single compound, it’s actually a mixture of two compounds that use water as a propellant. You spray it into a cavity, wait until it expands until about 100x its size, then you shave it off with a really big hacksaw.

In evaluating each of these, there are several things to consider. One gross factor to consider is the R value of insulation per inch. It's not a perfect measure of insulation performance by any means, but it will give a rough guideline on what you get and in a house with 2x4 construction downstairs, getting in a bit more insulation in the constrained space can make a difference. The R values vary depending on a host of factors, but it came out roughly that the open cell foam we were looking at (Icynene) boasts about an R value of 3.6/inch. Cellulose is about 3.4/inch. Fiberglass is about 3/inch. Interesting side note: the data on the Internet vary hugely (up to 20%) in some of these numbers, and a lot has to do with the specific type of material and how it's installed. But in a nutshell, fiberglass isn’t very good and cellulose and Icynene are very good.

Other factors include how well they plug air leaks. Expanding foam is basically going to fill every nook and cranny. Cellulose does an OK job. Again, fiberglass isn’t very good. There’s also how well the insulation does when it gets wet. Again fiberglass doesn’t fare particularly well.

When everything nets out, the cost of the expanding foam is just too high based on what we get. If I had another $10,000, I’m not sure even then if I’d spend it on the foam. Instead, it’s probably going to be blown cellulose.