GeoThermal with 2000 Gallon Tank?

Discussion in 'Hybrid Systems' started by Guest, Oct 20, 2008.

  1. Guest

    Guest Guest

    I am at a point where I need to replace my furnace and most likely my AC unit as well. I was very interested in a geothermal system, but the 25k price tag for a vertical closed loop system was just to far outside my budget. Then I had a contractor come out and suggest installing a 2000 gallon cement tank in the ground and using that to make a version of a open loop system. I live just to the west of Chicago, Illinois and I know we are very close to bedrock here so that is why this installer has used the tank systems.
    I am just wondering if this system sounds feasible. His price was half of everything else I was quoted which while got me interested, but also got me questioning if it would work well. My home is about 2400sq foot and it is fairly well insulated. I spent about 2500 last year for utilities and that is with a 20+ year 80,000 btu (~60% efficient) gas furnace and 15+ year old 2 1/2 ton (6 seer) ac unit.

  2. Palace GeoThermal

    Palace GeoThermal Well-Known Member Industry Professional Forum Leader

    Ask him to show you a system like this that works. If it was that simple, it would be a common practice.

    I am very skeptical about the idea.
  3. Guest

    Guest Guest

    Tank heating system

    It won't work. The surface area of the tank is too small to absorb enough heat. The only heat you can get is the heat from freezing 2000 gallons (16660 pounds) of water or 2.3 million BTU, if you can freeze ALL of the water, which you can't because if you did the tank would break open because water expands when it freezes. 2.3 million BTU or 23 therms might heat your house for two or three days.

  4. GregG

    GregG New Member

    how deep is your tank going to be?

    How deep will your tank be? How many feet of loop between the geo and before you get to the tank? Assuming you aren't heating the tank (or bringing it to ground temperature) it would take approximately 75 hours to freeze the tank from 33 to 32 degrees with a three ton geo. But, if your tank is deep enough to be in average ground temps then the delta with the tank and the ground is going to get large enough that heat transfer to the tank would be huge. I'd have to assume more than 36000 btu's per hour. I'm just guessing here that your average ground temp at 7 feet or lower would be 50 degrees. 18 degree delta through a concrete wall with no insulation would be pretty high. I'm not saying it would work but I'd have to see the math either way. A three ton closed loop would probably have 60 gallons of water in the pipe, right?
  5. engineer

    engineer Well-Known Member Industry Professional Forum Leader

    This thread is over 2 years old...why resurrect it?
  6. GregG

    GregG New Member

    sorry, I just joined the forum

    I'm a new member. Its not an old thread to me. They are all new.
    We have a water to water geo in Northeast, Indiana. We use a hybrid system on our geo. I'm taking heat out of the water I'm using to water our chickens and ducks and if they aren't drinking enough water it can then dump water. So, I'm not saying a two thousand gallon tank could be used for a 3 ton geo but there definately is a size of tank and especially if its temperature coupled to a pump and dump system it could work.
  7. engineer

    engineer Well-Known Member Industry Professional Forum Leader

    If the system pumps and dumps there is no need of a tank.

    I'm still having trouble understanding what benefit might be derived from including a 2000 gal tank in a geo system
  8. GregG

    GregG New Member

    bedrock was their issue

    Didn't he say bedrock was a problem at their site? I would think that could make it difficult for a horizontal or a vertical loop. Also might make it difficult to find a place to dump a pump and dump system. Finding a spot deep enough for one tank might be easier than a trench long enough for the slinky?
    I'd personally still want to see the math because I think if the tank was concrete and the bottom and top was deep enough the rough math in my head says it might work. Like I said before, a loop for a 3 ton might only have 60 gallons of water in it. Without figuring the heat transfer to the earth it wouldn't take very long to freeze all the water in a loop with a geo. A concrete tank is going to take and give heat to the earth fairly easily I would think. If you could get the whole tank (top and bottom) to a depth that would be below the average soil temp for an area I still could see how it might work. I'm trying to remember back from my agronomy classes at Purdue but I think depending on soil type the temps start to average out at 4' of depth and by 7' there wouldn't be much to gain by going deeper regardless of soil type. At 20' to 30' depending on soil types there is no annual variation in temperature regardless of soil types. The applied science behind geos is kind of cool to me.
  9. urthbuoy

    urthbuoy Well-Known Member Industry Professional Forum Leader

    Heat transfer across a concrete tank is easy to calculate. It is parallel to our basement heat loss calculation. So if you want a quick estimate - if you had a tank as big as your basement, you'd have 1/3 of the heat your full house requires around here. That assumes the thermoconductivity of the ground is sufficient to keep providing that amount.

    Pretty much the reason we don't need to take the math much further unless you want to build in a creek.
  10. GregG

    GregG New Member

    I've never seen a loop in person

    I've never seen a loop in person but I hear talk of people putting in 250 feet of slinky per ton of geo. I don't think that would be the space of a basement.
    I assume we could use Ohm's law to calculate heat transfer through a concrete wall. The resistance to heat transfer would be about non-existant. The tank walls themselves would probably be about 1/3 the btu holding capacity of the water in the tank. (btu's per volume of a cubic foot of concrete compared to a cubic foot of water and not total quantity of btu's in the tank compared to the walls.) Just a thought, if the tank was buried deep enough to get into ground water which might be reasonable near enough to Lake Michigan or one of it tributaries a 3 ton geo would struggle to ever change the tank from ground water temperature.
  11. docjenser

    docjenser Well-Known Member Industry Professional Forum Leader

    I would call that borderline for a pond loop, way too short for anything else. 800ft is minimum for our climate, with great ground conductivity.
    The whole Tank idea is questionable, for ponds we need around 1/2 acre, and 12 ft depth. Way more than 2000 gallons. Entire surface area is not enough to exchange enough heat for 4 tons. It cannot beat the physics!
  12. Looby

    Looby Member Forum Leader

    Ohm's law becomes rather ...uh, "problematic", when R goes to zero.

    Riddle me this, Batman:

    What's the difference between a buried 2000 gal tank and a "conventional"
    horizontal loop with a single ten foot length of pipe ...six feet in diameter?

    ...and BTW, what's the Reynolds Number for a 6 foot ID pipe at 9 GPM?

  13. engineer

    engineer Well-Known Member Industry Professional Forum Leader

    Make the buried 2000 gallon tank really long and skinny and put the inlet and outlets at opposite ends of the long skinny tank. Then it will be in contact with sufficient soil for proper heat transfer

    Oops - that's called a pipe!
  14. Reiner

    Reiner New Member

    And chances are you can make it 200 gallons or less instead of 2,000. :)
  15. GregG

    GregG New Member

    estimated r-value of poured concrete tank?

    I'd place the r-value of a pre-cast concrete tank around 0.1 per inch. But I'm no engineer. I don't really have a dog in this fight. I do know that it would take significantly longer than 3 days to freeze the tank. And especially if its buried so the top was below 6-7 feet and probably couldn't ever be froze if its in the water table.

    The difference between a concrete tank and a single six foot diameter pipe would be the conductivity of pex versus concrete. Pex actually has pretty good r-value per inch say compared to copper or concrete.
  16. docjenser

    docjenser Well-Known Member Industry Professional Forum Leader

    hcool down = (2000 gal) (8.34 lbs/gal) ((50 oF) - (25 oF)) (1.0 Btu/lb oF) / (24 hr)

    = 17374 Btu/hr

    It takes 17 KBTU/h to cool 2000 gallons of water from 50F down to 25F within 24 hours. The heatgain through the wall will be insignificant.
  17. GregG

    GregG New Member

    what about phase change btu's?

    Didn't the water change states between 25 degrees and 50 degrees? Wouldn't the btu's required between 50 and 25 be the same as a 195 degree delta? Isn't that how it would take a 3 ton geo about three days to freeze the water assuming no heat gain through the concrete walls?
  18. urthbuoy

    urthbuoy Well-Known Member Industry Professional Forum Leader


    You're asking us to prove something we know doesn't work when the onus should be on you to change our minds. We all know it doesn't because the rate-limiting step is the amount of heat able to transfer out of the soils. Make your tank out of copper if you want, it will still freeze.

    I'll give you some points if you now argue balanced loads switching between 8 hours cooling 8 hours heating could make use of such a setup but it has no application for an entire season of heating.

    Salts, phase change gels, waxes, etc... it has been looked in to.
  19. Looby

    Looby Member Forum Leader

    R-values per inch are stated ...uh, per inch! So, a thick slab
    of low-r material can easily offer more thermal resistance than a
    thin layer of high-r material. However, whether pex or concrete,
    the pipe wall is NOT the thermal bottleneck.

    Be aware that non-turbulent (laminar flow) water is a horrid thermal
    conductor. Also, most of the 9 GPM flow in a 6' dia. pipe will travel
    down the middle of the pipe (google: "parabolic velocity gradient").

    So, you'll get virtually NO radial mixing; and there'll be a 2+ foot
    thick insulating boundary layer of still water between the inside
    wall of the pipe the 9 GPM flowing down the center of the pipe.

    Again, what's the Reynolds # for 9 GPM flow in a 6 foot ID pipe?

    Last edited: Dec 9, 2010
  20. docjenser

    docjenser Well-Known Member Industry Professional Forum Leader

    A BTU is defined as amount of heat required to raise the temperature of one 1 pound (0.454 kg) of liquid water by 1 °F (0.556 °C) at a constant pressure of one atmosphere. So if you take one BTU out of the tank, you cool one pound of water down by 1 degree.

    Yes, I forgot the phase change, I just used the formula to heat up or cool down pool water.

    Alright, to freeze one gram of ice takes 79.7 calories, each pound has 454 grams. One BTU is 252 calories, and each gallon is 8.34 pounds of water. So each gallon should release 1,197 BTUs while freezing. 2,395 MBTUs are needed to be extracted to freeze 2000 gallons. Add the 300,240 BTUs extracted to get the water from 50 to 32 degrees. With the lets say the 36,000 BTUh extracted by your heatload, it should take you roughly 75 hours to freeze that water. Not accounting heat transfer from the ground into the tank. You can add some heatgain for the tank through the ground, but it will be only a matter of days until the surrounding ground will be frozen.
    Last edited: Dec 9, 2010

Share This Page