1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Heat Pump Powered Steam Engine

  1. Nov 15, 2008 #1
    I have been working on an idea for a new energy source that I would like to get some input on. I took physics in college about 10 years ago, so I have a general understanding of the concepts, but I am definitely not an expert on thermodynamics.
    Here is my idea:
    1. Use a heat pump(ground source, or air source in the right climate), to concentrate heat.

    2. Use the concentrated heat in a boiler to generate steam.

    3. Use the steam to drive a piston steam engine.

    4. The steam engine then drives an electric generator.

    5. Hopefully, the generator would produce more electricity than the heat pump and associated equipment need.

    6. The exhausted steam from the engine is collected in a condenser, cooled to just below it's boiling point and injected back into the boiler. Well water would be used to provide cooling in the condenser, and then be routed to the heat pump to recycle the heat. This would increase the temperature of the water entering the heat pump and hopefully improve it's coefficient of performance(COP).

    At first glance this looks like a perpetual motion machine. I would argue that it isn't, because the system will take heat from an external source(the ground, air, or anything else that is warm enough), use a heat pump to concentrate it, and turn it into electricity. For example, in a ground source version the water going back into the well, will be colder than it came out.

    Another idea I have to improve efficiency is to use something other than water to produce the vapor(steam) that powers the engine.

    Unfortunately my knowledge of heat pumps and steam engines is limited so I have a few questions:

    1. How high of a COP can a ground source heat pump achieve? It would only be used for generating heat and not for cooling so I would think that it could be more efficient than heat pumps that are used in homes or businesses.

    2. What sort of efficiency can really be obtained from a multi expansion chamber, piston steam engine?

    3. What happens to the performance of a steam engine when a liquid other than water is used? Does using a liquid with a lower heat of vaporization affect the performance of the engine.

    4. How high of a temperature can you achieve with a heat pump?

    5. Is there something better than a steam engine to convert an external heat source into electricity?

    Using the limited information I could find online I came up with these estimates for efficiency:

    Input Heat Pump COP Boiler Efficiency Steam Engine Generator Inverter Output
    1 unit 5.5 .95 .34 .9 .9 1.439

    I think this idea is either genius or ridiculous so I would love to hear any opinions or suggestions that people may have, even if you think it's ridiculous.

    P.S. I am working on a drawing to better explain, I will post it when it is finished.
  2. jcsd
  3. Nov 15, 2008 #2


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Let's see: you're going to operate between ground water temperature and ground water temperature; trick the ground water into heating the engine and cooling itself; exhaust ground water that has chilled itself, and extract useful work from the process.

    Perpetual motion of the first kind.
  4. Nov 15, 2008 #3


    User Avatar

    Staff: Mentor

    The efficiency of a heat pump and power producing thermo engine are inverses of each other. So the best you could hope to do if there were no losses is exactly break even.
  5. Nov 15, 2008 #4
    My original idea was simply that a heat pump with a COP of 5 could take 1 kw of electricity convert it to 5 kw of heat, and as long as you could produce electricity from that at an efficiency of greater than 20% you get a net gain. The gain in electrical energy would be less than heat energy taken from the well water. I didn't think it would be a perpetual motion machine, because the well water would be leaving the system with less heat energy than it entered with.

    Thanks for the input.

    Since I have the diagram I will go ahead and post it.

    Also, even if my perpetual motion machine won't work, does anyone have answers to some of my questions about steam engines.
    3. What happens to the performance of a steam engine when a liquid other than water is used? Does using a liquid with a lower heat of vaporization affect the performance of the engine?

    Attached Files:

    Last edited: Nov 15, 2008
  6. Nov 15, 2008 #5


    User Avatar

    Staff: Mentor

    Like I said, the efficiencies are inverses of each other, so if you are able to create a heat pump that operates at a COP of 5, the maximum possible efficiency of a steam engine powered by that heat would be 20%. It takes a little bit of thermodynamics to explain why this is so, but basically, a heat pump gets more efficient as the temperature of the source and sink get closer together while a steam engine gets less efficient as the temperature of the source and sink get closer together.

    To actually figure this out mathematically requires constructing the cycles and analyzing the thermodynamics. It's a little involved if you've never taken thermodynamics, but with a week or so of study, you can learn what you need to get started. If you are really interested in this, I highly recommend you buy a thermodynamics textbook and start learning it.
    It is entirely dependent on the particular properties of the working fluid. And there are lots of choices. What you would want to do is pick the one that performs the best under the particular conditions you have available.
    As I said above, the COP is dependent on the delta-T between source and sink. In other words, the hotter you try to make the hot side, the less efficient the heat pump gets. A typical heat pump can produce 95F air from 40F air at a COP of around 4. Decrease the cold side or increase the hot side and the COP drops. A lot.
    Around 45%, though turbines work better than pistons. But again, the efficiency is dependent on temperature difference: getting an efficiency like that requires very hot and high pressure steam. Say, 500 F and 500 PSI. If you want to use 215 F steam, you'll get much, much lower efficiency.
    As hot as you want, within reason. I doubt though that people have designed real high temperature heat pumps before. And again, as I said before, the hotter the output, the lower the efficiency. Above perhaps 150F and you'll find that your heat pump is nothing more than an electric resistance heater with a COP of about 1.0
    Not at the temperatures you're talking about. Again, as the input temperature of a thermodynamic engine rises, the higher the efficiency goes. Fuel burning engines can exceed the efficiency of steam engines for this reason.
    If I won the lottery, I'd have $100 million. My number has exactly the same basis as your numbers do: I pulled it out of my head. It isn't based on anything real.
  7. Nov 15, 2008 #6
    Thanks for the explanations. I would be interested in reading a book on thermodynamics, if there is a particular title or author that you would recommend I would appreciate it. Do you think I would be able to dive right into thermodynamics without first learning some of the other fundamentals of physics?

    Thanks again.
  8. Nov 15, 2008 #7


    User Avatar

    Staff: Mentor

    You said you took some physics in college, which implies some scientific thinking - the first level of thermodynamics doesn't require anything special beyond that, so I don't think you should have a problem with it. It is mostly conceptual, not mathematical. I used an older version of this book: https://www.amazon.com/Thermodynamics-Engineering-Approach-version-1-2/dp/0072549041

    I loaned my copy of that book (2 years ago!) to a designer at my company who is intelligent, but had a pretty mediocre high school education. He's asked me a bunch of questions about it, but doesn't seem to have much trouble learning the concepts.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook