Welcome! Here are the website rules, as well as some tips for using this forum.
Need to contact us? Visit https://heatinghelp.com/contact-us/.
Click here to Find a Contractor in your area.

Help with Design - Earth Floor over radiant hydronic

Options
gillysoco
gillysoco Member Posts: 4
We're working on putting a system in an old farmhouse that we're renovating. Planning on putting in an earth based floor 1 7/8" thick (it's a mix of clay ~30%, sand ~60%, and chopped straw ~10%) I've been told that it should act thermally similar to gypcrete

I hired a DIY radiant company that said they included design... turns out they sold me a bunch of stuff and a tubing layout based on some rule of thumb guesses... now they are asking for more $ for an actual design, and I'm leery.

I'm concerned about the floor temps / heating loads / tubing layout, and am looking for a recommendation of someone that I could hire to do the design portion to make sure we're on the right track.

Thanks in advance for any recommendations!

Comments

  • Jamie Hall
    Jamie Hall Member Posts: 23,344
    Options
    Hmm... well, that's a little out of the ordinary.

    Earth floors tend to work best when they are very slightly damp. Obviously, wet and they are muddy (and slippery). Dry, they tend to dust rather ferociously. So that's a consideration. I have an idea that running radiant heat in one would dry it out too much...

    That said, the other question is what is to be under the floor? Radiant floors almost have to be insulated from the ground beneath them (I'm presuming this is at ground level?). Otherwise, a good bit of the heat you will be putting into the floor mass will actually be heating the ground beneath, rather than the building. Over time the earth beneath will warm up -- but there will always be a large loss there, and it may take weeks to get the temperature up to where it is stable.

    So I guess my first real question is... why? Mind, I'm not opposed to earth floors in principle. I'm just not at all sure that combining that with radiant heat is a particularly good idea.

    All that said, as far as the radiant tubing layout and heating load, it would be very conventional -- loops spaced about 8 inches on centre, nothing under and cabinets of built-ins, that sort of thing. Loops of approximately equal length, and no loop over about 200 feet. I'd use oxygen barrier PEX, to avoid corrosion problems.

    Also, Id keep the heat load down to no more than 20 BTUh per square foot, and plan to keep the floor relatively cold so that it doesn't become totally dry...
    Br. Jamie, osb
    Building superintendent/caretaker, 7200 sq. ft. historic house museum with dependencies in New England
    GGross
  • GGross
    GGross Member Posts: 1,050
    Options
    I think whoever is providing the earth based flooring needs to sign off on a radiant design and take liability. I do several designs per week, and I personally would not sign off on an earth based floor without the providers taking responsibility for the floor material based on a design that assumes gypcrete. I am not an engineer however, so you may have luck with one, but then you are talking big $$$. Best thing to do would be come up with a design, and get the floor provider to sign off that the floor can handle the design conditions. And everything else @Jamie Hall said, you must insulate below the floor
    GroundUp
  • gillysoco
    gillysoco Member Posts: 4
    Options
    Well I'm the homeowner, heat installer, and floor guy :/ ... As far as the floor construction goes, I'm going based off the book Earthen Floors: A Modern Approach to an Ancient Practice... It will be over a wooden subfloor.

    Some areas are inaccessible crawl space, those I won't be able to insulate, or should I say would be much harder to insulate.

    Without getting into more details...

    I'm really looking to connect with someone that I could hire to help with the heat load calcs / design

    If you know someone I should be talking to, or are them let me know!

    Thanks!



    GGross
  • hot_rod
    hot_rod Member Posts: 22,158
    Options
    Certainly possibly. Sounds like maybe are building your own floors?

    I remember reading about a job where tube was under a sand bed in a back yard to snuggle your toes into🤗

    Coming up with a R value for  that floor build up is part of the answer

    Floor mass works both ways, while it can store heat, there is also a longer warm up time, and temperature over-shoot to consider

    Its basically an earthen flywheel
    Bob "hot rod" Rohr
    trainer for Caleffi NA
    Living the hydronic dream
    GGrossRich_49PC7060
  • hot_rod
    hot_rod Member Posts: 22,158
    Options
    gillysoco said:
    Well I'm the homeowner, heat installer, and floor guy :/ ... As far as the floor construction goes, I'm going based off the book Earthen Floors: A Modern Approach to an Ancient Practice... It will be over a wooden subfloor. Some areas are inaccessible crawl space, those I won't be able to insulate, or should I say would be much harder to insulate. Without getting into more details... I'm really looking to connect with someone that I could hire to help with the heat load calcs / design If you know someone I should be talking to, or are them let me know! Thanks!
    The load is the load regardless of the floor material. So that is the first number you need to calculate

    Radiant floors can comfortable emit 25 btu/ sq ft of exposed floor   So divide the load number by available floor space. Its that simple 

    Earthen floor wouldn’t scare me, concrete and gyp is earth based🥴

    What type  of wear surface do you get? Like a clay tennis court? Is it sealed?  That would be my concern.
    Bob "hot rod" Rohr
    trainer for Caleffi NA
    Living the hydronic dream
    GGross
  • gillysoco
    gillysoco Member Posts: 4
    Options
    Hey Bob, supposedly it's as durable as a wood floor, has a similar hardness at least... it get's sealed with linseed oil after it dries.

    It's a mix of clay soil, masons sand, and chopped straw (or another fiber to add some flexibility) we're in the process of trying out different mix ratios

    If it didnt come across yet, Im not a heating guy... I used cool calc to calculate the loads, is that relevant for hydronic??

    If not is there another place you'd recommend for calculating the loads?

  • hot_rod
    hot_rod Member Posts: 22,158
    Options
    I would just call it a 2” gyp floor for the calc

    Cool Calc is ACCA approved, so if you input all the data correctly it should be accurate

    If a local supplier will not help I think all the radiant manufacturers will do a load calc , check their website

    Rehau, Uponor, Mr Pex, Watts Radiant, Viega

    A room by room load on radiant specific software would be nice.


    If you have the Cool Calc results , post them

    If all the rooms have a load of 27 btu per sq ft, or less radiant should cover it. In high load areas you may need supplement heat

    Do they make odorless  linseed oil?  I used some on my truck bed, sure did stink. Maybe after it soaks in the smell will go away.

    Bob "hot rod" Rohr
    trainer for Caleffi NA
    Living the hydronic dream
  • Tim_D
    Tim_D Member Posts: 128
    Options
    You need to know what the space requires in terms of btu/sqft. Then you need to work out the thermal conductivity of the flooring material. ASHRAE Fundamentals is a great place to start. You can then work out the fluid temperatures and tube spacing. I would probably start with 3/8" tubing at 4" on center to get the lowest possible water temperature. It would seem that you would want the lowest possible water temperatures to minimize the thermal expansion and contraction of the tubing and to minimize localized over drying of the material at the tubing surface. Of course, you will also need to account for downward and edge losses. Just an FYI, straw was added to adobe and other earthen type construction to add R-value and I am pretty certain that this system will not have the thermal characteristics of gypcrete or lite weight concrete.
  • psb75
    psb75 Member Posts: 836
    Options
    How about getting a dry weight measure of a cubic foot of your material, compacted to the expected density you plan to build with--then you can begin your calculations. Heat-loss calculation of the building is also a critical number. I'm leery of the insulative value underneath your floor. That is fairly critical too.
    The moisture level in your initial installation of the flooring material will be an interesting factor to consider.
  • hot_rod
    hot_rod Member Posts: 22,158
    Options
    There are probably different bends of floor products like that, There us info available should on typical adobe material R value. Maybe compare the floor mix to adobe to get an idea

    I’m thinking it will be lower than carpet and pad for heat transfer, which goes over radiant often
    Bob "hot rod" Rohr
    trainer for Caleffi NA
    Living the hydronic dream
  • gillysoco
    gillysoco Member Posts: 4
    edited July 2023
    Options
    Thank you all for your suggestions so far and your interest... I've realized that we made some changes to the floorplan since the cool calc and I'll need to make some adjustments... I will work on that in the next few days and post it here.

    Overall the BTUH/SF for the entire house is 31.9 (this was based on having no floor insulation... I will update that in the cool calc for the areas that I can access under the floor) I'm thinking 2" of polystyrene (R-10) but can up that if needed and am open to input there.

    There are 2 rooms that I cannot get to under the floors (without tearing them out) The current plan is to use radiant panels in those rooms and carpet the floors with a thick pad in them.

    As far as contacting a manufacturer for the design / calcs... that's what I probably should have done!
    I've already bought a bunch of stuff from Blueridge...they said design was included, turns out it was simply a tubing layout... I'm just concerned that room by room heat calcs weren't done.. they want more $ to do it but I don't trust them now that they've pulled a bait and switch on the design.

    In the main book that I'm going off of, there is a chapter on radiant heating... she writes "Most installers will not be familiar with the specifics of an earthen floor but the system specifications can be designed as if a concrete floor is being used'

    Hot Rod Bob... the linseed should cure within a few weeks. It polymerizes more than dries, I've used plenty on outdoor stuff and my ratty 67 rambler and after a few weeks it doesn't seem to stink.

    Tim D - My understanding from tons of reading and youtube and a little practical experience with natural building / adobe / mud plasters ... is that the fiber portion is mostly to add flexiblity / tensile strength to the mix... I'm sure that it adds some insulative value but my bet is that it is minimal.

    This is a really geeked out article on adobe insulation / thermal conductivity. https://greenhomebuilding.com/QandA/adobe/mass.htm

    The most I can understand is that adobe (very similar material clay, sand, straw) is a poor insulator but has a high thermal capacity, and is slow to heat and cool... so I'm thinking of it a really big heat flywheel.

    For what it's worth weight wise it's about 10 PSF per inch (so 20 PSF with my planned thickess) They recommend 1" minimum over the top of the tubes.
  • psb75
    psb75 Member Posts: 836
    Options
    Radiant panels + carpet and thick pad ≠ "good".
  • hot_rod
    hot_rod Member Posts: 22,158
    Options
    The very first plan should be to get the structure as efficient as possible. Insulation, air sealing, etc. Anything that gets the heat load lower. This reduces the load you have to cover with the radiant. Andcourse the operating cost of heating and cooling.

    31 BTU/ sq ft is not doable with just radiant floors. You need to have loads more around 25 BTU/ ft.

    It that high load is just one room, adding supplemental heat will be an option. Panel radiators, radiant walls or ceilings for example.

    Hard surfaces will transfer the heat better, carpet, pad, any floor covering will lower the floor output. Tile, wood, earth, bricks, etc are the better floor coverings for heat transfer, maybe some throw rugs.

    You need an accurate room by room load before you go any further.
    Bob "hot rod" Rohr
    trainer for Caleffi NA
    Living the hydronic dream
  • pewischmeyer
    pewischmeyer Member Posts: 3
    Options
    I built a strawbale house and ended up going with a concrete slab with radiant heat, but looked into earthen floor pretty extensively. The reason we didn't go that route was just the labor cost and not have a good local source of clay.

    From what I've learned, start with well compacted gravel, put down a vapor barrier, then a layer of 2" extruded polystyrene insulation, and then "staple" your pex tubing down to that with pex clips. The just put your earthen mix over that. I wouldn't worry too much about your heating load because they clay is going to build up and store heat really well. If your tubes are spaced farther apart, it will just take a little longer to heat up, but once it's warm it's going to hold heat really well.

    I ran three zones through our domestic hot water heater which is turned up to 140, our slab is 10" thick and my lines are all about 10-12" apart, so it takes about a day and half to come up to temp, but then it holds heat really well. The house is really well insulated, so it stays pretty warm even without the floor heat turned on.

    I used a TACO 3-zone controller with NC solenoid valves and a single circ pump pushing upstream of the manifold.
  • hot_rod
    hot_rod Member Posts: 22,158
    Options

    I built a strawbale house and ended up going with a concrete slab with radiant heat, but looked into earthen floor pretty extensively. The reason we didn't go that route was just the labor cost and not have a good local source of clay.

    From what I've learned, start with well compacted gravel, put down a vapor barrier, then a layer of 2" extruded polystyrene insulation, and then "staple" your pex tubing down to that with pex clips. The just put your earthen mix over that. I wouldn't worry too much about your heating load because they clay is going to build up and store heat really well. If your tubes are spaced farther apart, it will just take a little longer to heat up, but once it's warm it's going to hold heat really well.

    I ran three zones through our domestic hot water heater which is turned up to 140, our slab is 10" thick and my lines are all about 10-12" apart, so it takes about a day and half to come up to temp, but then it holds heat really well. The house is really well insulated, so it stays pretty warm even without the floor heat turned on.

    I used a TACO 3-zone controller with NC solenoid valves and a single circ pump pushing upstream of the manifold.

    A 10" thick slab! Am I reading that correct? is the tubing placed at the bottom of the slab?
    I've always been intrigued by straw bale construction, I hope it works well for you.
    Bob "hot rod" Rohr
    trainer for Caleffi NA
    Living the hydronic dream
  • psb75
    psb75 Member Posts: 836
    Options
    10" thick slab--THAT is some "thermal MASS"!