I agree with the conclusions that everyone has come to. The Kelix system talked about in another thread would have 314 sq ft of pipe to ground contact with 30,000 cf of dirt 5' away from the pipe for a 300' hole. I don't see how it can work for a 5 ton system. But maybe it can....still wondering how.

Both good points

Standing column wells can work at 60ft/ton in rock or below ground water depth. But they have bleed, ground water flow and make direct use of the water.

I like the comments that have been posted so far. I was drawn to this thread while researching a similar idea.

I would like to see some more justification/explanation for the claims made so far. I particular some of my questions/objections are:

1: It seems the 180,000 cf volume surrounding the line comes from a 10’ by 10’ by 1800’ bow surround this line, when really a 5’ radius cylinder would be more accurate. That would make the volume only pi&5^2*1800 or ~141,000 cf.

2: I assume the 5’ number used so far comes from (or is the basis of) the spacing for horizontal ground loop trenches, which is 10’ (i.e. 5’ for each loop). However, it doesn’t seem right to use the entire 5’ as changing temperature the same amount. If it did, then the 10’ spacing between loops would not be enough to prevent thermal crossover. Rather, it seems that temperature change would decrease with distance, just as ground temperature variations decrease with depth. See http://www.geo4va.vt.edu/A1/A1.htm for a representative graphic of what I’m referring to. I can’t figure out the decay rate from the figure, but even if it were “only” a linear decay, the heat stored would only be 1/3 that of the full-volume calculation. This can be seen by comparing the volume of a cone to the volume of a cylinder, except in this case the “height” would be the temperature. Does anyone have a more accurate temperature decay rate?

3: Is 120 lb/cf and .2 btu/lb/F a reasonable assumption? I found one site that has similar numbers for concrete and brick. Is soil density and heat capacity really the same as concrete?

Using the changes from 1 and 2 above, I get that the heat storage capacity of the line would be 141,000*120*.2*20/3 = 22.56M btu, but still a good amount of storage.

I agree that a 2500 gallon tank is probably way too small. I was considering a much larger tank 24’ by 40’ by 10’, for 9600 cf or 72,000 gallons (I know, it’s huge). With that size tank, though, the water would store ~12M btu, using 62.5 lb/cf and 1 btu/lb/F and 20F temperature change. The surrounding soil would have a volume of 34*50*20-24*40*10 = 24,400 cf, and would store 24,400*120*.2*20/3 = 3.904M btu. This would be a total of ~16M btu, not as much as the line, but nearly as dismal the other numbers would suggest.

I noticed the use of squares vs cylinders too - but all of this is to give a *rough* idea of the principles of why a small tank wouldn't work and is absolutely not something you can depend on for real heat loads. Use proper software for that.

Yes, there is a thermal gradient and some heat does come from further away.  120 lb/cf and .2 btu/lb/F are reasonable for *some* soils.  On the other hand, manure (sort of like soil) can produce 4000 btu/lb*.


* - one dump truck load and you are set for the season!

Posted By jonr on 12 Feb 2010 09:06 AM
On the other hand, manure (sort of like soil) can produce 4000 btu/lb*.


* - one dump truck load and you are set for the season!

Well, that would change the conversation completely when someone asks, "what do you heat your house with?"

It never fails to amaze me how these loop efficency discussions allways find a way to get around to

"the poop loop"
Eric Sackett
weberwelldrilling.com

Try finding help to help you drill in that.   

Posted By jbar
I would like to see some more justification/explanation for the claims made so far.

While you're at it, suggest you consider "wall effects" in the large tank model.
Lack of turbulence (i.e., Reynolds number essentially zero) will vastly impede
heat transfer from the surrounding soil to the tank.

...Murphy was an optimist,

Looby