Hi Tim
On (11 Oct 97) Tim Hutzler wrote to Roy J. Tellason...
 TH> The problem is containment. If you put heat into the earth, it would
 TH> radiate outwards, and eventually to the surface and lose it.
I used to have to calculate the ratings for buried power cables, on power 
rid
circuits. Depth of burial and soil thermal conductivity, were among the vital
parameters.
Most of the heat escapes upward not outward, and this is easily verified by
site measurements of soil temps. The reason being, that resistance is
proportional to distance and the shortest route is upward to the cool air.
The small amount of heat that flows sideways or down, faces a huge resistive
path hence the soil there gets real hot, and heat is not keen to travel from
one hot body into another hot body!
But! the soil above the source cools rather well being close to a heat sink
[air] and heat runs happily from a hot body into a cool one.
 TH> any rainwater that soaks into the ground, or pass through it
 TH> (ie. a shallow water table) would carry away heat.
 TH> The heat would have to be contained in some sort of tank, a very
 TH> large tank. If you happen to have one, that's fine - most folks don't.
Cheaper yet, just insulate the surface ie the air/ soil interface. Make a
laminate of junk steel sheets, with any sort of waste insulating material
packed between- could be scrap accoustic tiles, expanded polystyrene, egg
cartons, glass bottles....
If the heated mass was [say] 5 x 5 yards then a cover of size 7 x 7 yards
would be effective. But if the heated mass is very deep down [ say, more than
2 yds deep] then more overlap might be justified. More overlap also desirable
if long term heat increase was desired while none was being recovered for 
se.
In this case the underside of the laminate would "cook" and a sideways flow 
f
energy would seek release beyond the suggested cover. In which case, hot- 
oot
it over the dirt and go get more junk steel sheets!   Cheers.......ALEC
... ........When I contemplate the moon, my head aches [Galileo]
--- PPoint 1.92
---------------