Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

What is the most efficient method of lifting water using heat?

  1. Jul 4, 2007 #1
    I post this question here as a starting point, because I am not sure which forum this question would best be answered in. If this isn't the best place for this question, please let me know.

    I have been puzzling over the lifting of water using heat, and see a few different methods that it can be done. However I am not sure what the limitations or efficiencies might be, and would like to determine the most efficient lifting method.

    The first method I envision is a model similar to the earth and how water is lifted into the atmosphere forming clouds, the second would be a model similar to a coffee percolator and a third would use something on the order of a rankine cycle pump to elevate the water.

    Any method of lifting would have to exist in a closed system where the temperatures at the lower elevation would be higher than the temperatures at the higher elevation, and that the temperatures at the lower elevation would not exceed 100C. Further, there will be a small Delta T difference between the lower and upper elevations and a difference in atmospheric pressure at the higher elevation which may be regulated between normal atmospheric pressure and 0 atmospheric pressure, and a heat exchanger to remove heat to accomodate the atmospheric vapor model of lifting water.

    Since it will be a closed system one can simply imagine two sealed tanks connected by a pipe or pipes with a heat source at the lower elevation to power the system.

    I am not asking here if this will be an efficient method of lifting water, I am simply asking what may be the MOST efficient method within the constraints of the model given.
     
  2. jcsd
  3. Jul 4, 2007 #2

    russ_watters

    User Avatar

    Staff: Mentor

    Well, you listed natural circulation via convection and the use of a steam engine. We just so happen to have a thread open where someone suggested natural convection to run a turbine and I calculated an extrordinarily low efficiency. The problem is that virtually all of the energy required to vaporize the water gets lost when you condense it. So a steam engine really is the best way.

    That said, I really don't understand your constraints: you say you want to input heat at the bottom, but you don't want to go over 100C. I'm not sure I understand why.
     
  4. Jul 4, 2007 #3
    I do not want to exceed 100C for the following reasons:

    A) Low tech not requiring special materials or pressure vessels.
    B) Input heat source.

    Another important factor is that there should be no loss of water in the system (or very very little).
     
  5. Jul 4, 2007 #4
    Ok, well it seems that the practical application is needed. I have a large ranch on which I have cattle. There is a small river below the ranch, varying from around 175 to 930 feet below the grazing range that the cattle use. Since the ranch is in Arizona, I have ample sunlight to produce heat, however there are no powerlines anywhere near and the cost of putting them in to run a pump is prohibitive. I do have access to low cost solar domestic water heating flat plates. My original idea was to feed the water into the panels and allow the water to evaporate and rise naturally through a large diameter pipe to a tank at the top where a heat exchanger would condense the vapor and drip into the tank. The water feed to the panels may be accomplished by gravity running a line from a higher elevation on the river to the panel array that would heat the water. Low tech is a must because of it's location and ease of repair should it become damaged (the property borders on public lands which see a fair amount of use). Photovoltaic systems and pumps would probably get stolen or vandalized. The water will be used for the cattle and if I can scale the system large enough, to provide a source of water for irrigation as well extending the viable range time on the property for the cattle. Well, I hope this helps to explain the problem in a more practical way.
     
  6. Jul 5, 2007 #5

    FredGarvin

    User Avatar
    Science Advisor

    175 to 930 ft is a quite a head to have to pump for any application.

    I'd be afraid that anything you do will result in such a low volume flow rate that it wouldn't be useful. Do you have any limitations on size of the set up?

    Russ, didn't you say that sunlight provides something on the order of 100 W/ft^2?
     
  7. Jul 5, 2007 #6

    russ_watters

    User Avatar

    Staff: Mentor

    125, if the sun is directly overhead...
     
  8. Jul 5, 2007 #7
    Size is not too much of a limitation unless something were to exceed 20 acres (The area that I own along the river).

    Yes, the pressures of lifting water that high was one of my concerns also, which is why I had considered using the panels to evaporate the water, and using a heat exchanger and a small vacuum pump at the upper tank to create a difference in pressure and lower the boiling point of water in the system, causing the water vapor to race upward to the upper tank, and condense back to the liquid state when it went through the heat exchanger as it enterd the tank. I could maintain a gas powered vacuum pump and heat exchange system at the upper tank without concern of vandalism. The difference in temperature between the Th and Tl would be in the range of 50 - 80C, and the pressure in the tank (vacuum) could be regulated to cause water to boil in the system at any desired temperature.

    Further, we get about 6.5 sun hours a day with an averaged input of around 250 BTU/FT2/HR. I currently have two hundred used but servicable 4'X8' solar hot water heating panels (got them in an auction at a very good price), which by my estimation should provide 10,400,000 BTU's per day or 3046 Kilowatt hours.
     
    Last edited: Jul 5, 2007
  9. Jul 5, 2007 #8

    FredGarvin

    User Avatar
    Science Advisor

    Looking at the practical side of things...

    You will have to use a lot of that initial energy to heat the pipes themselves. You may have some troubles at certain times of the year just getting the piping to stay at temperature. This also means you would have to do your best (if not better) to insulate every square inch of the piping in the system. If you got one cold point in there, you will condense out and it will go the direction gravity takes it. I haven't run the numbers, but I would bet that the heat transfer from the piping will kill you in the end.

    IMO I would think that your time and money could be better spent by fabricating a decent shed that could be locked and burglar proofed to create a pumping station which would be driven by a gas pump or a propane set up. That way you would have guaranteed flow with minimum hassle. The water is always in the liquid state and you could run all of that piping without insulation.
     
  10. Jul 5, 2007 #9

    jim mcnamara

    User Avatar
    Science Advisor
    Gold Member

    About the only type of pumping system that moves water those kind of vertical distances (900') are jet pumps - ie., deep water well systems.

    I'm in New Mexico - the local water table is about 190' and windmills can lift small volumes of water that far. Since your AUM's are huge, you can't have that many head needing water in a given area, so ~300 gallons per day per tank should do it. Minus evaporation. Windmills can do it.

    As to vandals, somebody is sure to shoot any kind of stock tank you put up, anyway. I used to carry rubber plugs made for test tubes, usually 3 bags - .25", .375", and .5" for magnum handgun holes in our water system pipes and tanks. About once every five years we went out and welded the holes. You just have to "ride fence" to keep up with it. Windmills are hard to destroy, IMO.
     
  11. Jul 5, 2007 #10
    I can easily insulate the pipe by surrounding it by a second slightly larger pipe and since there will be a vacuum pump on site, i can pull a vacuum between the pipes for insulation.

    Given that there are no 'cold spots' in the pipe leading to the upper tank, what is the theoretical mass of water that could be moved per second assuming the boiling point of the water is at 40C, the temperature in the heating system is 85C, a lift of say 200 meters with a large diameter pipe leading to the upper tank with minimal friction losses and virtually no loss of heat in transit and a heat exchanger at the top to re-condense the vapor at a temperature of 20C.

    Total input of energy per day would be around 10^7 BTU's, and day length of 7.5 hours.

    Jim, I desire not only to water the cattle but also if it is possible to supply SOME water for irrigation.
     
  12. Jul 10, 2007 #11
    Some comments & suggestions

    Hi Bernie,

    I don't know if this thread is still active, but just in case here's my 5 cents worth:

    1. The idea of evaporating the water seems like a good way to overcome the huge head (as much as 900 feet I gather). As mentioned condensation might be a problem though. I fear a vacuum pump would be problematic to power, as quite a lot of power will probably be needed to maintain a decent vacuum (and you can forget about 0 atmospheres I should think, because that would be a total vacuum, which is hard to obtain even in a laboratory, if not on a space station at least). However: If it is possible to place solar panels along the pipe, you should be able to re-heat the steam on the way, which would reduce the need for both vacuum and insulation. You might be able to purchase the large insulated pipes, which are used for "remote heating" (don't know the correct English term?) by which I mean: The concept of transferring heat from power stations to a large number of homes by means of pumping hot water through insulated pipes (it is used in several cities in my country). Alternatively you could make such insulated pipes yourself: Place a larger pipe outside a smaller one and fill the gap with polyurethane foam (which you purchase as two fluids which foam when mixed). The foam will set and harden and thus keep the two pipes separated and quite well insulated.

    2. You could consider using a (or several) Stirling engine(s) to drive a (or several) pump(s). The advantage of a Stirling engine is its ability to work with temperatures below the boiling point of water. However, they don't produce a lot of torque, so I guess a large engine would be needed to drive a decent sized pump, and you might need several engines along the 900 feet pipe. These, of course, might be targets for vandalism (although you could possibly place everything in the ground, with only a few, locked, access holes?)

    3. You might use an alternative fluid in the solar panels and the steam-engine, with a lower boiling point than water (alcohol, ether or similar). This way you should be able to operate the engine even on cloudy days etc.

    4. I think one of the largest problem you need to overcome is the large head. 900 feet is a lot of pipe (introducing resistance) and if it goes straight up it is a lot of gravity as well. You will probably need to use a displacement pump, since a dynamic pump (such as a centrifugal pump) would require a constant and high speed, which any solar-powered system can't guarantee. I would suggest a gear-pump, since it is a rotary device, and thus simpler and smaller than a piston-pump.

    I hope this is of some use to you and wish you good luck with your endeavor.


    Regards,

    Inventus.
     
  13. Jul 10, 2007 #12
    Hi (again),

    Just a few clarifications to my previous post:

    In 1. I am obviously talking about the idea of evaporating the water, without the use of a pump. However, In 3. I am of course assuming a system where you use a steam engine to drive a pump, which I kind of forgot to mention. Likewise, in 4. I am also assuming a system where you use some kind of solar-powered engine (steam, Stirling or otherwise) to produce rotational power.

    Regards,

    Inventus.
     
  14. Jul 11, 2007 #13
    A perfect vacuum is not neccessary to make water vaporize at a temperature of less than 100C. At one atmosphere (sea level) water vaporizes at 100C but at higher elevations for example, water will boil at a lower temperature. My idea was to monitor the panel temperature and to adjust the vacuum to make water boil at that temperature. Insulation of the piping is relatively easy also using a vacuum. I built some parabolic concentrators and surrounded the heat pipe with a larger glass tube in which i pulled a hard vacuum (>10^-9 Torr). I monitored the vacuum for a couple of years and the vacuum held. A vacuum is the best insulation that I am aware of, and for this reason i am not worried about cold spots causing condensation on the way up.

    The most important thing I am trying to accomplish is to get an idea of theoretical maximum efficiency of different methods. I have had it suggested to use photovoltaic and electric motors, but, the efficiency of photovoltaics is about 15% of the total solar energy available. A copper solar panel with a black chrome coating will absorb about 97% of the total solar energy available. So one begins with a higher amount of total energy to work with, but using heat to lift water, by powering a stirling engine or using evaporative lifting ... amounts to a significant loss of energy out of the system. What I am trying to discover, is using different methods of using heat, what would ultimately be the most efficient system with the least loss.
     
    Last edited: Jul 11, 2007
  15. Jul 13, 2007 #14
    Well, I suppose we should consider this thread dead because either there is no one qualified to answer my original question or there is no interest.
     
  16. Jul 13, 2007 #15

    Bystander

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Oh, there's interest; just a matter of mulling over the question(s). Lifting vs. evaporating? Lifting kg/s 300 m requires 3 kW; evaporating kg/s requires 2.5 MW. At which point, you lift. How much do you want to lift? 5k head @ 10 gal./day? 500? 1000?
     
  17. Jul 13, 2007 #16
    I would like to lift move 25000 gallons a day, but your reply did make something I had been overlooking much more obvious. So perhaps now I should focus more on mechanically lifting the water instead of using evaporation, but I would still like to employ the solar low grade heat source to do this, so I guess that will leave me with a stirling or rankine cycle engine to do the work.
     
    Last edited: Jul 13, 2007
  18. Jul 13, 2007 #17

    Bystander

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    25,000 gal, 100 tons, 6 hrs., or 5 kg/s --- 15 kW from 500 kW of flat panels plus head losses in the plumbing (double it, it's farm machinery), so 30 kW delivered mechanical power --- should be doable.
     
  19. Jul 14, 2007 #18
    Owing to the fact that flat plate black chrome hot water solar heating panels will be used, the temperature will not always reach 100C at all times of the year, so I suppose I need to approach the heat powered motor to run the pump. The only options that seem to jump out at me are a stirling motor or a steam motor that uses a fluid that boils at a lower temperature than water. If i use some other fluid than water I would need to be sure it was 'environmentally safe' in case of accidental spillage in some sort of system failure.
     
  20. Jul 14, 2007 #19

    Bystander

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

  21. Jul 14, 2007 #20
    As I have been researching this problem, it occured to me that part of the problem is the temperatures attainable in the system and the boiling point of water. Then it occured to me that it may be possible to use a different fluid in a rankine cycle motor, such as ethanol instead of water. Since cold water is available at the site, it seems that a standard steam engine could be used and at the outlet, cool the ethanol (which boils at around 175F if I remember correctly) and return it to it's liquid state in a condenser, thereby maintaining a low pressure on the outlet side of the steam engine while keeping a closed system so that ethanol is not lost. The temperature of the river is about 65F in the summer time.

    I believe only a small quantity of ethanol would be required and since ethanol is 'edible' should not pose much of a risk environmentally if there should be a leak or spill. Since there will be no immediate sources of spark or open flame, likewise, almost no risk of fire or explosion.
     
    Last edited: Jul 14, 2007
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: What is the most efficient method of lifting water using heat?
Loading...