GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Earth Tube vs. Glycol Ground Loop

StrongRoots | Posted in General Questions on

I have read:

Do Earth Tubes Make Any Sense?

All About Earth Tubes

My Earth Tube Story

From these articles, it was clear Martin really loves earth tubes. : )

Some seem to still support earth tubes, others don’t.  Confusing.

Using a Glycol Ground Loop to Condition Ventilation Air

There seems to be less uncertainty.


Heat pump in an earth-banked greenhouse?

Clearly not a good idea.

We are a full-time, Central PA vegetable farm on a bootstrap budget.  Propane costs crushed us last year.  My goal: partially heat (34 deg. F) a 34 x 96 foot greenhouse (6 mil plastic, double wall with inflated air between layers in ceiling and walls, roll-up sides).  Rigid foam board installed as a thermal break 12 – 18 inches deep and attached to baseboard.  Zone 5.  Silt loam soil.

Excavation work will be done by myself and my neighbor.

My annual propane costs for partial heating from Nov through April would run 2 – 5k dollars, depending on the price of propane, external vs internal temperatures, age of my plastic, etc.

If you were me, what sort of ground based heating/cooling system would you install?  A full-on vertical geothermal system costing 50 – 60k is not commercially viable given our cash flow/budget.  Our budget is $500 – $2k.  More like 500 – 1000. Bootstrap style.  Is  this even realistic?

This is a DIY project.  The earth tube seems less daunting, confident I can get it done.  Want to get the idea down first, then do material pricing.  I worry about performance, but really like the idea of low carbon heating and no propane tanks/heaters.  Can’t afford to get it wrong.


GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. paul_wiedefeld | | #1

    Take a look at compost heat systems, that'd take advantage of your farm's materials. Or maybe a cheap unglazed pool solar thermal collector designed for glycol could get you some cheap Btus.

    1. StrongRoots | | #3

      I don't produce as much compost as I would like. I do remember seeing something out of Cornell and U. of Vermont years ago. I don't think my compost is at scale for this. I briefly looked into it about 5 years ago. Happy to be wrong.

  2. StrongRoots | | #2

    I should add, there are lots of youtube videos on earth tube greenhouses. Not confident about their results over time. A lot of them just show the process without results over time. There is the famous example of the Nebraska greenhouse growing citrus using an earth tube.

    I wanted to see what this community said before plunging ahead.

  3. charlie_sullivan | | #4

    The downsides of earth tubes don't worry me as much as for greenhouses as for living spaces for people. But I see it as being more useful or tempering ventilation air than for supplying heat.

    1. StrongRoots | | #5

      I was picturing 45 deg. F (ish) air flowing through my greenhouse. While that is far from indoor air temp of 65 deg F, for me, I would consider that heated air. Or am I missing something? Tks.

      1. charlie_sullivan | | #10

        To keep your greenhouse above freezing all winter, you need to flow a lot of 45 F air into it. With a high flow rate of air through a tube in in the ground, you have two problems. First, the air temperature will be lower the faster the flow. Second, you will deplete the heat in the soil around the tube over time. If you only have enough flow to supply fresh air, both of those problems are smaller.

  4. Expert Member
    BILL WICHERS | | #6

    I would probably go with a liquid system. If you KNOW FOR SURE the liquid won't ever get below freezing, you don't need glycol and can use straight water. The glycol is for freeze protection, and actually reduces the ability of the water to transport heat a little bit (i.e. the water/glycol mix is slightly less efficient than water would be alone). I would use a water system because you can use smaller/cheaper pipe, and transport more thermal energy more easily. I would probably use 1" or 1-1/4" polyethylene "sprinkler" pipe. Don't go much bigger than that since you want surface area, and larger pipe has less surface area per unit volume of contained fluid. That's a fancy way to say you'd be better off running several small pipes slightly spaced apart in a trench compared to one large pipe in terms of heat transfer ability due to more ground/pipe contact area.

    Inside the greenhouses, you need a small circulation pump and regular water to air heat exchangers. You might be able to use "fin tube", which are the same linear heat exchangers (copper pipe with aluminum fins) that are commonly used to heat a greenhouse with a hot water boiler.

    Before you go to all this effort and expense, I have an idea you might try. You might try using reflective film stretched between your hoop houses (which is what I assume you're growing inside of), setup to reflect sunlight into the hoop house. This will allow you to collect more sunlight from the areas on the side of your hoop houses that would normally be "wasted", and potentially raise the interior temperature that way. The usual black barrels full of water could be used for thermal storage inside to help mitigate nighttime temperature drops.


    1. StrongRoots | | #8

      Bill, thanks very much. This system would be the only source of heat. If a pump died, or power went out, then the water would stop circulating and I would definitely be at risk of freezing. That makes me lean to glycol-water mix.

      When you say polyethylene sprinkler pipe, do you mean orchard tubing? I use this when setting up irrigation in the field.
      I also have easy access to 200 psi polypipe or other pipe materials:

      Is this better or equivalent to PEX?

      I am thinking 6 foot depth.

      How many trenches/lines would you bury?

      How many feet of piping would you install?

      What would be the system flow rate for a house of this size? What size pump would you recommend?

      I am picturing straight runs, but the GBA glycol article link above shows loops in the foundation. I could certainly dig out the soil in the footprint of the greenhouse and run loops if I needed that for the heating/cooling load then cover back up with the soil. I would certainly prefer to err on the side of caution than corner cutting.

      I did some looking and did find duofin, specifically designed for greenhouse applications. I've also seen people running pipes through their planting beds. I would prefer water to air heat exchangers, then pumping 'warm' air through the house using convection tubing hung about 8 or ten feet off the ground. My preference is based on weeding, soil prep etc. I am not hydro, I grow in soil.

      Any feedback or guidance most appreciated. I am clearly at the information gathering stage. I'll do more research, and share a potential DIY design here. Thanks for any sage advice.

      1. Expert Member
        NICK KEENAN | | #11

        Glycol is really expensive. I buy it in bulk and the best price I can get is around $25/gallon. That's for a five gallon bucket of pure glycol. When you think of how many gallons a system like this will need it's going to be a significant expense.

        1. Expert Member
          BILL WICHERS | | #19

          It's not as bad when bought commercially from mechanical supply houses. We use it in 55 gallon drums when filling large systems, and its several hundred bucks per drum, so maybe $2-3/gallon. That was several years ago though, so it may well have goine up in price since then.

          Note that we don't necassarily run a 50/50 mix of glycol to water, either. We tailor the mix for the conditions we expect to see, so we just put in a little more glycol than we need. This usually ends up more around 30/70 or 40/60 glycol/water.

          BTW, propylene glycol is less efficient than ethylene glycol, but it's also nontoxic. I recommend using propylene glycol in ground loops for that reason -- it's safer if you have any leaks.


          1. Expert Member
            NICK KEENAN | | #21

            I'm seeing prices of over $1100 for a 55 gallon drum, or around $20/gallon. I don't use enough to make it worth the difference over 5 gallon buckets, which are a lot more convenient.

            In my climate (DC) I do a 30% mix so it comes out to about $7/gallon for the mix.

          2. Expert Member
            BILL WICHERS | | #22

            DC, looks like glycol prices have gone up a LOT since I last ordered any! Wow. We probably got better rates, since we'd usually oder pallets of 4+ barrels at a time, but still. Wow...

            Note that for 1" pipe you get about 24.5 feet of pipe per gallon of fill water, and even if you go with a 50/50 water/glycol mix, that's about 50 feet per gallon of mix. You won't need much glycol, a 55 gallon drum would be good for about 2,.695 feet of 1" ID pipe!

            My guess is this would still come out cheaper than an air system though, since the larger corrugated plastic pipe is significantly more expensive than the smaller sprinkler pipe.


  5. walta100 | | #7

    When I see people post ground source heat pumps loop water temps late in the heating season the numbers generally seem to be in the low thirties. That tends to make sense to me as they have cooled the ground around the loop.

    It seems to me if you want to transfer much energy you need about 10 ° differentials at each transfer and I don’t see how you can get the number of BTUs you need into and into the air.

    When I do some quick math it seems $5000 of propane = 2100 gallons = 192 million BTUs. That is a lot of energy even if 40% went out the flue pipe.

    My wild guess is your loop length will be measured in miles of pipe to collect a million BTUs without a heat pump.

    1. StrongRoots | | #9

      Hi Walter,

      A $5,000 propane bill would be high for me, for sure, but if I plant tomatoes last week of Feb, and they are heated until mid May, and it's cold, it is possible. I spent close to 10k total on propane last year for 3 heated greenhouses (different ambient temps). The difference here tomatoes require 50 to 60 deg F rather than the 34 deg F. for a partially heated house.

      My math and your math don't jive given this lower temp point.

      I ran some calculations:
      Q = Qo + Qv, gave me about 30 - 35 kW if my indoor is 34 F and outdoor is -5 0r -10 F. I assumed a leaky greenhouse. This is equal to about a 105,000 - 120,000 BTU/hr. I am cool with about a 20 kW system as described below.

      Qo = UA (ti - to)
      Qvent = pNVcp(ti-to)

      U W/(m2 deg. C) 3.85
      A surface of greenhouse (m2) 316
      ti desired internal air temp ( C) 2.2
      to external air temp ( C) -23
      p greenhouse air density (kg/m3) 1.281
      N infiltration rate (s-1) 0.07
      V volume of greenhouse (m3) 815
      cp specific heat air 1.005
      ti internal air temp 2.2
      to external air temp -23

      Lh = qh (COPh-1/COPh * (Rp + RsFh)/(Tgmin - Tewt,min))

      qh = 25 - 35 depending on COPh and other assumptions

      COPh design heating coeff. of performance 3.86
      Rp pipe thermal resistance (C/W) 0.38 (poly pro, .41 or .51 if PEX)
      Rs soil/field thermal resistance (C/W) 3
      Fh 0.32
      T gmin minimum undisturbed ground temperature ( C) 7.22
      T estmin minimum water temp entering heat pump ( C) 6.5

      Lh = loop length = about 50 m if qh is 35. I personally think that is an underestimate for pipe length.

      Since I used a lot of literature assumptions, I would run three trenches about 3 feet wide and 3 feet deep along the footprint of the greenhouse. Do a loop in each trench, which I think equates to 200 feet of pipe per trench, or 600 feet total, roughly 4x my estimate of 50m. Sorry to mix and match US and metric units. I've heard a rule of thumb is 500 feet of pipe per ton for horizontal, which would be 3,000 feet if I did 6 tons. One factor is my greenhouse does get warm in the day, and that helps recharge the heat of the soil, which is why there is a fairly high COP (according to some EU studies). This isn't a bare soil design. Either way, even if I am way off, 3,000 feet is way less than miles and miles of pipe.

      The kicker to me is not the length of pipe and excavation--which I highly doubt is miles and miles (high side 6000 feet for bare earth). But 35 kW = about 10 tons of geothermal cooling/heating. I vaguely remember there being a difference between cooling vs heating and the conversion factor, but couldn't recall for this calc. If I am more generous, and say my crops wouldn't mind a few degrees below freezing if it was -15 outside, then my heating loads drop to about 23 kW or roughly 80 - 85,000 BTU, or about 6 - 7 tons heating. Since we very rarely get that cold, and these are cold tolerant plants, I'm good with calling it 6 tons which is slightly larger than the 5 tons used in the average house.

      Either way, very quick pricing on my end suggests about about 14 - 15k for a kit, or about 4,500 for the ground loop and about 5 to 8k for a 6 ton GHP.

      I could be very wrong, I only spent a few hours on these calcs.

      I am surprised the heating load was so high given the target temp of 34 deg F. If you all see my math is wrong, I would love to hear that because I think this could be a cool project.

      An order of magnitude wrong would be great : )

      I consider this a first stab at GHP for heating, next up I'll look into earth tubes.

      Am I just waaaaay off here? Seems really high to me. I just can't imagine an earth tube will cut it, but perhaps I am just seriously making a mistake on this. First time running the numbers. I know they use an earth tube in Nebraska to grow citrus, so perhaps my heating load is just wicked high, or I seriously underestimate the ability of an earth tube. Or maybe their greenhouse is way smaller than mine.

      Corrections most appreciated.

  6. Expert Member
    NICK KEENAN | | #12

    How deep is your frostline? My intuition is that you're better off digging down below frostline and/or insulating than trying to pump heat out of the ground using air or water and with a low temperature differential.

    There's a guy in Maine who is big advocate of passive greenhouses, I don't recall his name. His model is that the floor of the greenhouse is below frostline, every part of the greenhouse except for the south roof is highly insulated.

    A rule of thumb is that soil has an r-value of about 4 per foot. So if your frostline is 3-4 feet you need R12-16 or so to keep soil above freezing.

    1. StrongRoots | | #13

      When I bury frost free hydrant I go six feet. I believe it is more like 3 or 4 feet. I like to err on the side of caution. I could bury, I do worry about water and moisture. I will certainly check out the design. I typically vent with roll-up sides in the summer. Maybe burying would make this un-necessary? All thoughts appreciated.

  7. walta100 | | #14

    When I read your first post I did not notice anything about a heat pump I thought you plan was to move air or water thru underground tubes and scavenge the grounds heat directly. A direct system would need to be miles long to keep a green house above freezing. I agree a GS heat pump would need a shorter loop.

    If you do a ground source heat pump what will it do to your electric bill?


    1. StrongRoots | | #16

      I've read something like, for every 1 kW of elec for GHP, you get 4kW back. I think if all was installed right, I would have about a $500 elec cost. I did not calculate this number. That is a guesstimate. I'm buying 1,200 gallons of propane next week, I'll know for sure the price then. But if I assume 2.50/gal, I've estimated that 34 degrees will burn through about 3,000 gallons, or roughly $7,500 for Nov - Apr. This is a fair point of comparison to the heating provided by $15k for geothermal, indicating a two year payback for equal management. Anything less than 5 years is a sound investment to me. In practice, I can't afford this given prices paid for internationally imported food or Cali imports. The Nebraska greenhouse is a very different design than my 34 x 100, quite a bit smaller and more hobby or homestead than commercial scale. If glycol direct heat would require lots of vertical or horizontal piping, and a GHP was not part of my earth tube design, I think an earth tube would take miles of pipe--if it worked. Silver nitrate pipe would help with mold. I think I will pursue the geothermal GHP with glycol, it does give me a competitive advantage over local growers because the propane costs are prohibitive. However, I will install the system in two stages. Next week, my neighbor and I will be excavating for the frost free hydrant and the baseboard thermal break. I may as well just dig three trenches for 600 feet of pipe, connect to a manifold, then cap it when it enters my greenhouse for the 2022 season. Then in 2023, taking the money I get from the crops in the greenhouse, invest in a GHP. So invest over time, laying the pipe now rather than after the fact.

      I'll use:

      Does 600 feet of pipe seem reasonable for a 5 - 6 ton system, or is my estimate way off? Given this is not bare earth, and instead gets warmed to about 60 during the day.

      3 foot trench + 12 feet + 3 foot trench + 12 + 3 = 33 feet. My house is 34 x 96, that would be close to 6oo feet of pipe inside the house itself.

      Am I way off on my 600 foot estimate?

      Would definitely like to know now before I start digging.

      Any time and insight is supremely appreciated.

      Thank you all.

  8. walta100 | | #17

    In my opinion ground source heat pumps generally do not make financial sense. Air source heat pumps may make financial but that is dependent on your local fuel costs. Since you are installing the loop yourself your numbers may work for you.

    Even in the best case it seems likely to take years to recover the cost of any fuel switch.

    When you go DIY you are forgoing all the experience and expertise the pros bring to the job. Seems to me loop design is a very local thing in that what works in wet clay soil will not work in dry sandy soil. How much loop you need per ton depends on how long your local heating season is.

    If you decide to go ground source you may want to ask a few questions on the GeoExchange fourm.


    1. StrongRoots | | #18

      You're the man, thanks, I will do that. This is a major investment for us, as long as the payback is 5 years or less, to me, that pencils out. We also really like the idea of low carbon heat energy. We can't pay someone to put it in, so it will have to be DIY, which is cool, that way I can fix it too, and learn for the other houses if/when this works. Thanks again.

  9. frankcrawford | | #20

    check out
    they have detail designs and a planning package for passive solar greenhouses using annualized heat storage that work in northern Canadian climates.

    1. StrongRoots | | #23

      Thank you, I would, but it appears all the information is for a fee of a few hundred dollars or at least a hundred dollars Canadian.

      Looplink is free, this is a design package that some people in the geothermal industry use:

      So too are the igshpa GHP design and installation standards. When I am finished with my design and build, I will make the results known, for free, so others can choose for themselves whether it is worth it or not.

      Professional design and build books are also available.

      We are a commercial scale farm, from what I've seen, most passive houses are much smaller than the house we will be installing, require half of it to be hard roofed and often partially buried, and frequently have hard walls. All of this gets pricey quick. I am happy to be wrong, and will do some passive geothermal designing, but I suspect it won't be practical at my scale, and the amount of piping will exceed my available field space.

      There is a lot of high quality free information in the scientific literature and webpages. Books are also a lot cheaper.

      Thank you for sharing though, I genuinely do appreciate that.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |