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

Superheated Vacuum Steam Cycle

  1. Oct 30, 2009 #1
    Ok bear with me as this sounds a bit bizarre, especially for a first post. For a while I have been thinking of how to take advantage of the fact that a black passive solar collector can collect low temp BTU's at low costs, but due to the low temperature it can not be utilized in a steam cycle efficiently.

    My idea is to operate a steam cycle at a pressure low enough so that the passive solar collected heat is hot enough to boil the water. Since I believe I can produce around 140 deg F. passive solar heat I want to run the boiler system at around 3 psia and the condenser at around .5 psia.

    I recognize that at such low temperatures the system would be extremely inefficient, since ideal Carnot efficiencies are 1-Tc/Th. So the saturated steam produced by the passive collector would then be superheated using a conventional burner. Since I plan to run a small 1-5kW steam engine the superheat would be go to no more than 500 deg F. This should produce a dramatic increase in efficiency, while not requiring significant amounts of energy. Looking at enthalpies at 3psia for 140 deg and 500 deg steam I am thinking that roughly 85% of the energy can come from solar energy and 15% can come from some combustible fuel.

    The obvious difficulty is that this cycle would have inherently low energy densities at such low pressures. I did some calculations and concluded that for a 5kW steam engine at 100 rpm that I'd be looking at a piston larger than 12" in diameter. However since such low pressures are involved it need not be built nearly as rugged as a conventional engine. I was also thinking that the boiler danger would be greatly minimzed since the system is under vacuum and the amount of energy contained in the tube on a volumetric basis would be quite low, if the tube imploded it would probably not be deadly.

    So I am hoping for any feedback I can get on this idea? Is there something basic that I am missing that makes it impossible? Thanks in advance for the help.
  2. jcsd
  3. Oct 31, 2009 #2

    efficiency ~ 12%
    mass flow rate required~15Kg/s

    mass flow rate required~7kg/s

    Too impractical for such low power output, & calculations are based on simple ideal thermodynamical relations.
  4. Nov 1, 2009 #3


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    If only free energy is used (ie: sunlight) then I don't see any problem with low efficiency.

    Why use water as a working fluid though? There are refrigerants that can work at a few hundred psi over this temperature range.

    I think the primary difficulty of any system like this is going to be making it cheap enough to be commercially viable.
  5. Nov 1, 2009 #4
    Thanks for the replies ank_gl and Q_Goest,

    How did you come up with those efficiency numbers? Here is my method, first off for the non superheated cycle I'll assume my T_low=550 Rankine (R) and my T_high=600 R. Then using carnot efficiencies as a limit and assuming that my actual cycle achieves roughly 40% (nice and conservative)of this limit I get a pitiful 3.3% predicted efficiency.

    Now for the superheat case I'll again assume a T_low of 550 R and a T_high of 960 R and with an actual eff. of .4 x Carnot ideal I get a predicted eff. of 17%, which is almost a six times increase in efficiency.

    Now if I really wanted to get great efficiencies I could just use a an LP section of a turbine and operate at the thermal limits of the turbine material and probably approach 30% efficiencies. In a steam engine though I am limited to temperatures where the cylinder lubrication will not break down and I am told this is around 600 deg F.

    Of course I agree that the power density is extremely low, but since I am interested in making small scale distributed systems that harness an inherently low energy dense fuel source (sun) then I do not see so much of a problem. In the fossil fuel era we have gotten very use to tremendous power densities and large scale centralized plants and I suspect that this could change over the next century.


    Yes I know about ORCs but one problem I see is that many of the fluids used do not tolerate high temps very well, which would not work for a highly superheated cycle such as this and would therefore limit overall efficiency. The other reason for my interest in water is the simple availablity of it, since I am trying to dream up a simple, fairly safe (another reason for working with steam in a vacuum) solar energy system that is accesble to most anyone, I'd rather not use an organic fluid.

    You are quite right about keeping the cost down, wich is again why I am trying to utilize passively collected solar heat as my main energy input. I believe that when material costs, technical complexity and all other factors are taken into account a simple black passive collector that utilizes convenction currents as the mode of heat transfer will prove to be the cheapest and most readily available mode of collecting solar energy.
  6. Nov 1, 2009 #5
    I assumed ideal Rankine cycle & a superheated Rankine cycle, & calculated the efficiency using state point enthalpies.

    I am not sure if that's a nice approximation. And why use such approximation when you got the steam table ready?

    I am afraid I am not understanding what you mean by this, can you explain the setup once more?

    The method suggested can (if feasible enough) be beneficial for very small (micro) scale applications. Do remember that the big consumers are always the industries, they do need large scale power sources.

    Another thing to be noted is that fossil plants burn fossil fuel very very efficiently, burning fuel at home might not be very efficient.


    Normally available water is not used in steam turbine, its demineralized water.

    Whats the problem with solar panels?

    Welcome to PF, by the way!
  7. Nov 1, 2009 #6
    Thanks for the welcome to PF. I'll have to recalculate using ideal rankine cycle and state point enthalpies. I agree with your superheated calculation, but I think that the nonsuperheated cycle is overestimating it, because that steam is saturated and you do not actually know the quality of it. Thats why I estimated with carnot cycle effic.

    I think I need to communicate that i am trying to build a really simple low cost small (5-10KW) combined heat and power system. Cost and simplicity are my main concerns, not power densities. Now having said that I know that this idea may not be the best way to go about obtaining those goals, but that is why i wanted to discuss it here.

    The problem with other solar systems is that they really are not cheap. PV panels for example are quite expensive on a kwhr/$ basis and only return their investment in very sunny areas. They also require significant energy inputs to manufacture (purified silicon). Passive solar heat at low temperature is however very easy to generate and relies on low tech cheap materials. I am trying to figure out a low cost way of transforming that heat into electricty and heat on a small scale.

    Again the problem I see with using an organic fluid that has a low vaporization temperature is that they often times become unstable above 500 deg F. Also again the reason i am sticking to 500-600 deg F is that this is the limit of what a steam engine lubricant can withstand. A steam turbine does not however have this limit since the bearings for the shaft are located outside the steam path, which means that it can withsand steam temps of around 1010 deg F, which significantly increases thermal efficiences, but at the expense of relying on more complex machinery.

    It also has occured to me that perhaps higher pressures are possible in a two stage process where heat of vaporization is applied from passive solar heat under a vacuum to generate saturated steam. This saturated steam then rises vertically into a valve controlled high pressure steam chamber. The chambers would then close once full with the saturated vacuum steam. Then fire is applied and heat is added in a isometric process, once the maximum temperature is reached and the steam engine cylinder is ready for an expansion cycle the upper valve on the tube is opened and the pressurized superheated steam is released to the steam engine. The chamber is left open until the steam engine cylinder uncovers the exhaust port and produces vacuum in the steam chamber. Then the upper valve closes and the lower chamber valve opens and is filled with saturated vaccuum and the cylce begins again. I think though i need to draw that one up for it to make much sense....
  8. Nov 1, 2009 #7
    You can calculate the quality of the steam at the turbine outlet by assuming some isentropic efficiency, or even consider the expansion as an isentropic process. Anyways, efficiency isnt a major issue, because you are only using free(solar) energy. The problems I see are majorly commercial, as Q_goest rightly pointed out.

    Cost analysis for the proposed system should be interesting.

    On the other hand, why transform it to electricity? There are other (efficient) methods to use this energy, eg. a vapor absorption cycle. You plan to extract ~5kw energy, that is you have atleast 25kw energy available assuming 20% efficiency, though this would require huge heat exchanging surfaces. One can use this energy to run a VAR.

    Keep in mind that any type of machinery involved is very complex, & then there is another thing called "control".
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook