Once you do that, you can then run a thermal generator on the resulting temperature differential, to generate even more energy, and now you've created a perpetual motion machine, oops.

(without an existing cool place to dump heat, you can't generate energy, otherwise you could just extract heat from everything everywhere and cool it all down to absolute zero, solving both the energy problem and global warming with a single stroke)

We just need to make heatpipes all the way to space.

It's been considered!

https://www.technologyreview.com/s/608840/a-material-that-th...

As others have said, you can't use ambient outdoor heat to power air conditioning, because of thermodynamics.

It's possible to use solar heating to directly power indoor cooling, without converting to electricity as an intermediate step. My rough understanding is that this is potentially more efficient, but it adds mechanical complexity and loses flexibility (since it only really works in direct sunlight). And it's probably becoming less cost-effective in comparison as PV prices continue to drop.

I actually thought about using solar to drive AC

Yeah, there are ammonia-absorption refrigeration devices that can run on any source of concentrated heat, including concentrated sunlight (though not merely the ambient temperature). This is considerably more efficient than going through 16%-efficient photovoltaic panels first. Ammonia refrigeration has some safety issues (upon overheating, they release a giant cloud of caustic, inflammable gas) but in this case they would be substantially reduced by putting the unit out in the yard, not inside the house.

Ammonia absorption does not require violating the Third Law as some commenters have suggested below. It is less efficient than the Carnot limit, so you cannot use it to get a perpetuum mobile.

This is possible only when you have a temperature _difference_ to work with. If it’s equally hot everywhere, you’re SOL. (Think heat death of the universe)

Because E < (T2-T1)/T2

To get energy from heat, you need a hot end and a cold end. The more difference between the two, the more energy you can extract. Extracting energy from small temperature differences is very inefficient. Temperature is measured from absolute zero for this.

Exactly, this is why steam boiler engines became obsolete, because you can burn gas at a much higher temperature than boiling water.

Lower temp can only be viable if you reduce the work needed to run the engine, and can still output the needed work/energy.

This is related to why work and heat are both measured in joules.

No, this is not right. Most power generation is done by steam in the US (85%). Steam can get much hotter than boiling water, and it is cheap and makes no waste.

That doesn't have to do with temperature, it has to do with the efficiency of the engine.

Gas burns at a much higher energy than steam, so it's much easier to extract more energy with less fuel, making for a smaller form factor.

I should clarify I was talking about an engine here. Not a power generator.

What I said is correct though. You can compensate for lower temperatures with a more efficient engine & energy conversion, and of course more fuel/input.

Neither the “engine”/“power generator” distinction you are attempting to make, not the statement “gas burns at a higher energy than steam”, is physically meaningful. It is true that supercritical steam turbines operate at a lower temperature than gas turbines, and if the cold reservoir of the Carnot engines in question were at the same temperature, that would indeed make gas turbines potentially more efficient. But in fact gas turbine outlet temperatures are much higher. This is why adding a steam engine to the output of a gas turbine makes it more efficient. (This is called a “combined cycle power plant”.)

The actually relevant distinction is that gas turbines, like other internal combustion engines, have a much higher ramp rate than external-combustion engines like a steam turbine. This makes “peaker” gas turbines a crucial resource for establishing power grid stability.

Thanks for taking the time to write this informative post. I am grateful that HN is home to real engineers. :)

I am not a real engineer; I have never built a working heat engine, unless you count rebuilding a Volkswagen engine. But I'm glad my comment was helpful!

That's not possible, but geothermal energy is a passive way of cooling that actually works (to some point). Install pipes sufficiently deep under the surface and drive air through them and soil will cool the air for you - and because its temperature is almost constant throughout the year you can both cool and heat the greenhouses and barns and even homes this way with very little extra energy (just electricity for fans to move the air, and even that can be reduced to work passive for heating as the hot air climbs up the pipes on its own)

You have to have a temperature delta to make this work. (Thermodynamics 101.) A ground source heat pump is one way to do that: In Texas in the summer, a few meters underground is always cooler than the air.

https://en.m.wikipedia.org/wiki/Geothermal_heat_pump