air pocket beneath earthen floor

wanted to see if anyone had any creative ways to best insulate under an earthen floor. I like this approach, but open to other suggestions to reduce heat conduction right into the earth! https://i.pinimg.com/736x/1a/29/a7/1a29a7500468da9c77ee37fef2036846--floor-insulation-plastic-bottles.jpg

Glass bottles are frequently used underneath cob ovens for insulation and would probably work for an earthen floor. Though bottles have a larger amount of air space which could cause convective currents, this is why windows use gaps less than 5/8 of an inch, and reduce the r value. It is tempting to use free material, especially waste materials, but insulation is the most crucial element in a cold environment and you may end up spending much more or burning more wood fuel for something less insulated.

I've heard people using a layer of light clay straw under their earthen floors, but i think that is more suited for less extreme colds. Poly iso board is going to be the most effective insulator at r 6-8 per inch. You can find it on craigslist all the time for half the cost of new. Maybe a cross between a layer of iso board and a layer of bottles so as to reduce the amount of board you need to buy, but i don't think a single layer of bottles is going to be as effective as you want.

A lot of people use pumice or lava rock under their floors if they have access to it. Where I live though it is $60 a yard, so not very economical.


Just went on to this thread https://permies.com/t/5906/Super-Insulated-Floors-Thermal-Mass which has someone praising their bottle floor performance. Check it out there are some more ideas.

How funny, I had just read through that thread the other day. My project is going to be on a pretty small scale (8x12 cabin with a loft) but I am trying to approach it's design with a model of "maximum efficiency" whatever that really is. Anyway I was also hoping to use natural materials but it seems to me as well that the iso-board would be the best material for insulating under the earthen floor. Realistically it is more my style do sort of a combo approach, so I will probably use a little bit of all of your suggestions, iso board if I have to, thanks!

With a little bit of 'outside the box' thinking, you can probably find a few ways to increase the insulation value of the floor. It really depends on what you have available, what is affordable, and what goals you have in mind. Some people may be fine with using waste streams like used foam, cardboard, carpeting and such. Some people may have an abundance of natural materials to use. Others still may have a lead on someone nearby with an abundance of something they could use.

One really interesting example of an insulator is in this video below:



Buying materials to make your own carbon foam might not make sense, but if you can get free materials and have everything on hand then it might be cost effective. What I am imagining is laying down these 'foam tiles' in a grid spaced apart from each other a bit, covering with a layer of your earthen floor material, laying another grid offset a bit from the first one and maybe repeat a few times. This would minimize the path from the floor surface to the earth and the pockets of air trapped by the foam can insulate very well. You wouldn't be able to use it in a complete sheet because it is too weak and brittle.

However there is another similar material that may be readily available and much stronger that could add some insulating value - natural hardwood charcoal. Again, it wouldn't be cost effective to buy a lot of it, but if you can find some cheap or already have plenty of wood and tools on hand to make your own then it could work. It could either be put down as a layer or possibly broken into smaller pieces and mixed in with the earthen floor to raise the insulation value. Perhaps a number of materials could be used at different layers to make use of materials on hand vs buying new materials.

I can also imagine certain fungi, mosses, and other natural materials could be dried out and used in a floor. Creating pockets of low heat conduction and lowering the overall thermal mass of the floor should reduce the amount of heat that could be lost through the floor. You could build an insulating base and then have a solid floor on top so you have a thermal mass insulated from the ground that helps moderate the temperature inside. Even using large gravel with pockets of air being trapped and sufficient drainage could be beneficial as a base layer that is strong but not as conductive as a solid mass. There are lots of different options to explore. Good luck!

Hey thanks, that video is something else! I do love the charcoal idea! a little put off by the argon gas thing but that guy got some serious readings there, super cool. Also, even though I'd like to use all natural materials, I am going to cut corners somewhere along the line if I know myself at all. Still, I am very intrigued by the carbon/charcoal angle, I wonder how this could be further experimented on? Also, I am painfully ignorant but isn't compaction an issue when insulating under an earthen floor?

Yes, compaction would be an issue depending on what you use and how you use it. Digging down to the subsoil and building up a strong base should minimize settling and compaction. I've learned a bit about that from trial and error, but my sandy soil probably behaves differently from other soil types.

I was mostly spit-balling some ideas to get the creative juices flowing. The guy from the video listed does a ton of work with metals and machining, so his idea of using argon to make an inert atmosphere was in line with what his target audience may have available. He does tons of 'outside the box' thinking, which is what lead me to think that wood charcoal could be an option. I would still use a top coat of whatever earthen material, but it could be a decent insulating aggregate. I was thinking of houses I have worked on where they use styrofoam beads (like a bean bag chair) inside prefabricated concrete panels to slow thermal transfer. If I remember correctly, they can use something around 8% styrofoam aggregate by volume without significant loss of strength.

If you can find waste streams it can provide you with a layer of insulation then that diverts waste, saves you money instead of buying new, and gives you the insulation value you are after. Using some sort of bottles like you originally mention could also work. I know in the Pantheon in Rome they supposedly made a bunch of little clay pots they threw in with the cement (not like todays cement) to reduce the weight of the dome, which was the largest freestanding dome for many centuries. I'd imagine the air pockets created would at least slow down heat conduction.

As far as experimentation, maybe you could make up some test pieces. Perhaps a pair of wooden boxes with one being the control and the other with a layered or mixed batch of what you want to test. Set it out on a sunny day and then measure the temperature of the bottom layer in each box. The temperature difference between the ground under the box and the sun on top should be analogous to the difference between the heat from a heater and the cool ground below. The more heat energy that can pass through, the less insulating it is.

I have seen designs with the beads - even a conceptual one where black glass beads are actually kept in a void behind the south facing window pane during sunlight and then "drained" into a void in the floor. To then be somehow moved back into sunlight again the next day and so on. As far as waste streams, I work in a restaurant so I can get as much cardboard as I want from the boxes the shipments come in. I have thought about shredding this cardboard somehow and experimenting with some type of clay slip application to make insulating blocks. Either way I am definitely on board with outside the box ideas. Oh and I started another thread about Medusahead Grass, which is so common it grows in droves on my property. I can't wait until spring so I can do some testing! Too much snow and no shop to work in.

Daniel Schmidt wrote:One really interesting example of an insulator is in this video below

About a year ago I actually made some carbon foam after watching that video. I used a biscuit tin with a small hole punched in the top, put some bread slices in it, and threw it in a camp fire. It actually worked. It was just for fun though.

I'm not a builder - but I believe the thing about insulation is that if you don't do it properly, you may as well not bother at all.

Daniel Ray wrote:Glass bottles are frequently used underneath cob ovens for insulation and would probably work for an earthen floor.

Bottles are somewhat suitable for clay ovens because solid glass has a reasonably high heat tolerance which is an important consideration for ovens. You don't need that characteristic in a floor. Solid glass is actually a terrible insulator - about the same as solid cement according to this site I just found. Glass bottles will have a lot of thermal bridging around the walls of the bottles.

Has anybody got scientific test data on glass bottles set in clay? I very much doubt it, and my hunch is that it is not worth the time and labour invested.

The small amount I do know about insulation is that it's all about SEPARATION of materials and restricting the movement of air. You want as little physical contact (bridging) between the ground and your walking surface. There's a reason why cavity-brick walls are two skins of brick with almost no contact between them - that achieves the separation. Then fibrous insulation stops the convection currents in the gap.

Aaron Tusmith wrote:I am painfully ignorant but isn't compaction an issue when insulating under an earthen floor?

Yep, you definitely can't use materials that are both load-bearing and susceptible to settling. That will be a disaster.

@Daniel: From the link you posted:

Irene Kightley wrote:The insulation has been fantastic, cool in summer in warm on bare feet in the winter when the stove is on and when the winter sun heats the mass up during the day.

I believe the above comment is conflating two different characteristics. Cool-in-summer and warm-in-winter is a characteristic of high thermal mass, not high insulation. High insulation feels warm all-year-round.

We insulated some floors as you see in the pictures. We used leftover ends of plastic from the greenhouses that we (solely) heat our buildings with, or old retired greenhouse plastic that has a few holes. Sewed it into long tubes on the sewing machine, and stuffed them with sawdust and shavings from construction. Laid over a gravel bed, with cob in between and over the top. The hope is that the vaulted shape of the tubes will keep those puffy pillows standing under there. To be honest, I can't really tell if the building is insulated by this. In one of the spaces insulated this way (in these photos), it a space we keep our shoes on, in the north side of the building, and the space doesn't stay spectacularly warm anyway. In another space we did like this, it's our dining hall, where we sit on the floor, so we lay foam matting down with carpets over, so again I don't know how it really is.

In our latest building, we put a straw-cob layer underneath, then gravel, and a wood floor on top.

We have used glass bottles in the floor of a homemade low-tech solar water heater.

Nowadays there's so much plastic garbage in our region, that if we have to do it again I'd consider stomping down plastic bottles to reduce the size of the air pocket inside (as Daniel Ray mentioned above) and using those. I think it would work to lay down a layer of mud, press a layer of flatted plastic bottles into it, lay down another layer of mud and work it into the gaps, and then either do the finish layer or do another layer of plastic bottles and mud.