Make:

I don't see how this would work if the temperature is above -20 degrees. If the refrigerant boils at "about 20 degrees below zero", it can only condense if it were colder than -20. He says "It takes about 10 days when the temperature is between zero and 20° F to completely freeze the block of ice." This does not make sense. The only way for this to work is a refrigerant that boils at a high temperature, like 20 degrees.

Reply to this comment

I don't see how this would work if the temperature is above -20 degrees. If the refrigerant boils at "about 20 degrees below zero", it can only condense if it were colder than -20. He says "It takes about 10 days when the temperature is between zero and 20° F to completely freeze the block of ice." This does not make sense. The only way for this to work is a refrigerant that boils at a high temperature, like 20 degrees.

Reply to this comment

This is very neat, a modern variation on the "Ice house" of old. You just don't need quite as much space for this one, though.

syntaxinflux, the actual temperature of something staring up at a clear sky at night is far, far below the temperatures you would expect. Get an IR thermometer and try pointing it upwards, and it will drop away to 50C below zero. This means that some of the fins will cool further than you would otherwise expect as they radiate heat into the sky, letting the coolant liquify and run down to freeze the water.

A nice modification to this might be to have another one that used saltwater as the temperature store, so it can drop further below freezing and be used as a freezer compartment at, say, -10C. Obviously this would depend on the freezing/melting point of the saltwater, but could be looked up or determined empirically.

Reply to this comment

Oh, I see. I did not get that the heat sink was 50 feet above the ground.

Reply to this comment