This is a hot topic … people have been weighing in both for and against floor insulation by email and on the blog. In fact you might even say the discussion has been heated … alright I’m going to drop the puns from this point forward!
As near as I can make out the basic earthship design has evolved somewhat since the publication of Michael Reynold’s third earthship book, and some of the newer design details do not seem to be readily available (unless I have missed them in my reading).
In his newer work Mike Reynolds refers to a thermal wrap around the earthship and it took a fair bit of digging (I can’t help it) to find out what this thermal wrap is. It turns out the thermal wrap is a layer of rigid insulation placed around the bermed walls of the earthship after an initial backfill of the walls has been done.
I am assuming the point of this thermal wrap is to increase the amount of insulated mass attached to the earthship, and consequently to increase the efficiency of the ‘thermal battery’ that is the basis of the earthship design.
I have upgraded my initial drawing and include a cross section view of this whole idea at the top of this article. I believe that this thermal wrap is supposed to address issues that the earthship design has had in colder climates, specifically that it did not retain enough heat during the winter, and thus was not successful in regulating its temperature. Also, I have read complaints that despite a given earthship staying warm enough during the winter months, the floor temperature was simply too cold.
So, on the one side of this argument is the original earthship concept. The basic premise of this argument is that the minimum temperature of the earthship should not dip lower than about 55 degrees F (stable earth temperature), and the warming action of the sun during the day should in fact keep that temperature in an acceptable range of 65-75 degrees F.
On the other side of the argument are the reported failures of the earthship design in cold weather climates. Reading between the lines I think the failure of the design stems from the ground under and to the north of the earthship still being affected by seasonal temperatures. In other words, the ground under the earthship settles out somewhere between stable earth temperature and a value above freezing. I suspect this failure is also tied to quality of winter sun in various regions. We live in a valley and I have some concerns about the amount of sun we get in December.
So where do I weigh in on this entire argument …
I have now been in an earthship in Ontario on December 20th (2nd shortest day of the year) and it was -20 degrees C outside. The inside temperature was about 65 degrees F and the only additional heat was provided by a wood cook stove when food was being cooked. The basic premise that the earthship should maintain a reasonable and consistent temperature was being met. The floor in this earthship was not insulated, and the builder agrees with Micheal Reynolds that the floor should NOT be insulated.
I have to admit, although not intolerable, I did find the floor cooler than I would have liked. I tend to agree with the basic statement that an uninsulated floor is going to be cooler than your desirable room temperature of say 20 degrees C simply because stable earth temperature is below this temperature. It seems to follow that in a colder winter climate the floor will be that much colder than desired room temperature.
I intend on having a backup heating system in my house. I am actually in favour of radiant heat as opposed to direct wood heat, and I would prefer a warm floor. It would be possible to run the radiant heat in the tire walls and/or the floor. I do not see any need to insulate or heat the floor under the planters and a few other locations. Beyond this I have not come to any conclusions.