Sub-Slab Insulation and In-Floor Radiant Heat
I have a customer who wants to install electric radiant slab heating in a garage addition. The garage perimeter footings will be 5′ down and the walls will sit on top of 8″ ICF frost walls 4′ tall. The manufacturer of the floor heat says it is better to not install sub-slab insulation and instead use the ground as a heat sink, but I can not find a real argument for one way versus the other when floor heat is concerned. Will it be more efficient to do it the manufacturer’s suggested way or to install 4″ of foam board underneath the slab first? The addition will be done in North Dakota where we encounter many sub-zero days in the wintertime.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Without insulation under the slab your heat will will certainly sink away. I would go with a different manufacturer because that is very bad advice.
Their theory is that the heat only goes so low and then you have a big mass of heat under the slab to help out. In their studies, they say that the garage will heat up faster with insulation installed, but the overall cost will be higher without all the extra mass. In my mind, I can see the argument for both, but without knowing how low the heat will go into the ground, I can not make an informed decision. If I knew that the ground would heat up below the bottom of the footings and dissipate outwards, I feel like it would be an easier answer. I have yet to find an article on this exact subject. Anything that I have found pertaining to sub-slab insulation does not include the use of in slab floor heat.
That's junk science, I'm sorry. Basically they're saying the ground is a better insulator than a layer of foam. It isn't.
There's a reason when you do a Manual J you don't put in the mass of the building. Because mass doesn't help provide heat. For some people it's an intuitive idea that it ought to, but that intuition is wrong.
It might work if the deep ground temperature was higher than room temperature, but that's about 1% of the US and those places don't generally have heat at all.
Let me give the manufacturer the benefit of the doubt and assume you misunderstood what he was saying.
The ground tends to stay around the average year-round temperature. In about 99% of the US that is below room temperature. In the summer, an uninsulated basement floor reduces the need for cooling. If you live in a climate that is cooling-dominated -- you spend more on air conditioning than on heat -- you might save more energy with an uninsulated slab in the summer than you lose in the winter. But it has nothing to do with the mass of the floor improving efficiency.
Note that floor heat tips the scale toward insulation. Heat flow through a material is determined by the temperature difference. With conventional heat a slab floor is going to be few degrees below room temperature and the soil might be ten or fifteen degrees below that so there's not a huge heat flow. With floor heat you have to heat the floor to fifteen or twenty degrees above room temperature so you see around double the heat loss.
> might work if the deep ground temperature was higher than room temperature
Which in a garage, might very well be the case. How warm will the floor be heated to?
They keep their existing garage at about 60 degrees.
If this was a hydronics job, Siggy would recommend a minimum of R15. Does matter if it is hydronic or resistant heat, heat flow is heat flow. Go with the R20 and watch the depth of the mat. Hot rod had good you tube video on the effect of depth on effectiveness of a heat source.