I think I've found Maxwell's Demon, however the demon is quite large...

[DreamThinkBuild]

Just throwing out ideas but would the idea still work if you have it built in pipe sections? Use a copper or aluminum on the top and bottom of next pipe. This way it could be modular just add more on top.

[Nabo00o]

It could, but you see what you done here is really just building one big pipe.
What matter is height and not the volume, and I think by placing several different layers in one pipe, or just stacking one pipe on the next one will only make it harder, and unnecessary so.
You can just as easily use one long pipe, but since it is unpractical (and hard to find) in such lengths it would be easier to stack several of them side by side, and connect them in series, if that description fits here.

I think at least for now if no better ideas surface, that the normal 120 mm wide 2 meter long pvc tube would fit the job perfect, of course any variant in length and thickness wouldn't be any big problem either.

Naboo

Edit: Btw is it Blender you used for the image, I'm just starting to learn the program myself 

[lumen]

I have an 80ft deep well that I don't use and that is why I was thinking of PVC pipe with vacuum insulation.
I could stack a column of 10" dia. pipe 80ft long and just supply some ambient heat to the bottom. The column should separate to a reasonably high temperature at the top with the vacuum insulation along the entire length.
I would still need to do a single test segment and get the numbers first to determine if any real value exists in this concept.

This may end up being a good way to extract geothermal heat even from a 50F ground temperature. You could have hot water year round for free!


     

[Nabo00o]

Wow thats pretty deep, and yeah I can understand why you would want to test if this works at all to begin with.

Also I would recommend that you first try everything out with pure air, because air does truly have a tendency to separate into temperature layers when not disturbed.
And in a way you could call it geotermal heat, but remember, its not from the ground, its from the normal air around you that the heat is extracted. Call it gravitermal or buoyatermal heat instead 

Also if you had the time and proper measuring devices, you should test what the difference in temperature between the bottom and top of two equally long and insulated pipes can be, one which is filled with water and the other with air.

I have a feeling that the one with air may give a higher temperature but a lower supply rate than the one filled with water. I don't really know how oil would behave but I know it would be expensive and risky to fill a well full of it


Btw I get what you're saying now, if it is in the middle of the winter a four time amplification wouldn't do you much good if the temperature was at -10 degrees Celsius (14 degrees Fahrenheit).

Naboo

[lumen]

It may be possible to calculate the heat separation by using some charts I found.