This is an idea I've been playing around with for a while, even though many people tell me it's impossible. However, as far as I can tell, it does not actually violate the laws of physics. I am talking about creating a temperature differential using a heat pump, then passing ambient (warmer) air over the cool side of the compressor. Then, passing the hot side of the compressor over water or some other substance that will be boiled, extracting all of the energy. Potentially even some hydrocarbon that boils at low temp. Passing this boiled fluid through a generator turbine, and then using the left over heat to preheat the fluid going into the compressor in addition to the atmospheric heat.

Simply put, the machine would run the same way as your home heat pump does, except instead of heating your house, the output heat would be used to generate electricity.

It may require leaving the device to run for some amount of time in order to build up enough heat in the machine to be able to deliver usable temperature heat.

With all of current technologies limitations put aside, would this be possible at all? Every heat transfer IS in fact going from hot to cold differential. Yes there is some input power required, but it seems like it would effectively multiply what ever amount of power you put into the compressor.

After all, isn't it more efficient to heat something using a compressor cycle instead of restive heating directly?

I can add images or more information if I was unclear. Thanks for any help!

  • $\begingroup$ Yes there is some input power required, but it seems like it would effectively multiply what ever amount of power you put into the compressor. How does it seem that? Often such ideas of perpetual motion machine for infinite energy lack to show this key point: that required input-energy is less than the usable output. Would you elaborate on your thought that it will "effectively multiply" the input energy? Images and sketches are very welcome as well. $\endgroup$ – Steeven Apr 22 '16 at 20:33
  • $\begingroup$ Not exactly a duplicate, but has good numbers showing why this doesn't work. physics.stackexchange.com/questions/229350/… $\endgroup$ – BowlOfRed Apr 22 '16 at 21:15
  • $\begingroup$ Your idea is similar to OTEC, but on a much smaller scale. The thermal efficiency will be WAY LOW, so any capital investment in equipment to carry out your scheme will not be economical. For more info, see en.wikipedia.org/wiki/Ocean_thermal_energy_conversion $\endgroup$ – David White May 28 at 19:10

The laws of thermodynamics state quite clearly that the increase in entropy at your power plant supplying the input power must be greater than the decrease in entropy in your machine which is pumping heat from a cooler region to a warmer one. This implies that the input power is greater than the output power.

You might be interested to know, though, that power can be produced from differences in ambient temperature. For instance, if your groundwater is 15 K warmer than your atmosphere in the winter, you could build a power plant which requires no input power once it is running. This is made possible because the required increase in entropy can be satisfied by allowing large amounts of heat to flow from the groundwater into the atmosphere in the process of extracting power.


The machine you would like to build is a slightly messy version of a "perpetual motion machine of the second kind" about which you can read here:


This might also be good to read:


  • $\begingroup$ Thanks for your explanation. I am aware of such generators, but that is not what I'm trying to build here. Can you possibly answer why my 10 amp @ 120v air conditioner has an output rating of 3500w? This is what seemed like it would make the power generation possible, the fact that heat pumps seem to create more energy than simple resistive heating. $\endgroup$ – Colby McDowell Apr 22 '16 at 21:04
  • $\begingroup$ You are equating thermal energy and mechanical work (or the potential to do so via electrical power). I will edit my answer to include a link that might help. $\endgroup$ – Duncan Harris Apr 22 '16 at 21:09
  • $\begingroup$ So if I used this 1200w heat pump to pump 3500w of heat into a chamber filled with water converting it to steam, and I had a steam turbine capable of 66% efficiency, would it not still be capable of delivering 2.3KW? Even if the cold side was submerged in some geothermal sink? $\endgroup$ – Colby McDowell Apr 23 '16 at 0:46
  • 3
    $\begingroup$ The heat pump only achieves that impressive heat output for very small temperature differences (eg 5 K or so between inside and outside on a hot day). If you avoided this issue by using a heat engine that operated nearer to the ambient temperature, you would find it impossible to get a sufficient heat engine efficiency. These two effects complement each other perfectly. There is not a loophole. Thousands of engineers have devoted their careers to finding one ever since the industrial revolution. $\endgroup$ – Duncan Harris Apr 23 '16 at 1:11

Someone here said " Thousands of engineers have devoted their careers to finding one ever since the industrial revolution."

I'm sure they are busy complying with energy sellers instead of looking for real engineering solution. Let me go back to "drinking bird". If you are still guessing - it DOES convert ambient heat into mechanical energy. It DOES it 24/7, without direct solar radiation, wind, waves, fission or fusion, etc.
Total volume of "drinking bird" toy is 20 cubic cm. It generates 0.001 Watt. 1000 of them will generate 1 Watt. Therefore one cubic meter of properly designed "drinking bird" will generate 50 Watt 24/7. Average home energy consumption is 10-20 kWh. Therefore 20-40 cubic meters of the "device" will give give home all energy it needs. 24/7, no batteries, no payments to centralized power plants. Shell we fence gentle people?

  • $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$ – ACuriousMind May 25 '17 at 20:12
  • $\begingroup$ Nioko, your idea is totally impractical. In addition, no one has been able to design a device bigger than a toy to do what the drinking bird does. $\endgroup$ – David White May 28 at 19:14

protected by Community May 25 '17 at 16:26

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

Not the answer you're looking for? Browse other questions tagged or ask your own question.