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

I Glass -- how to increase the surface area

  1. Oct 26, 2018 #1
    I need to increase the surface area of the glass which will be used in a solar still with the intention of keeping the glass as cool as possible. My first thought was bubble wrap because it's transparent and I thought it would not interfere with the light but then I remembered it is a good insulator and would probably make my glass heat up, so then I was thinking of corrugated glass.

    Can someone give me some guidance on other methods to keep the glass cool without using any electricity or any form of human intervention such as pouring water on it.

    Thanks
     
  2. jcsd
  3. Oct 26, 2018 #2

    berkeman

    User Avatar

    Staff: Mentor

    Could you post some pictures or diagrams of the type of solar still you have in mind? I'm having trouble understanding why clear glass will heat up while light is going through it. There must be some other mechanism (like a greenhouse effect?) that would be causing the heating of the glass. And why does the glass need to be kept cool? Are you operating at high enough temperatures where the glass could start to melt or deform? Thanks.
     
  4. Oct 26, 2018 #3

    berkeman

    User Avatar

    Staff: Mentor

  5. Oct 26, 2018 #4

    CWatters

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    +1

    Why does it matter how how the glass gets?

    Perhaps run water over it? Added bonus is the glass stays clean?
     
  6. Oct 26, 2018 #5
    Yes that is how it might look.

    You want to have a big temperature difference from the water to the glass will increase condensation so output will be increased.
     
  7. Oct 26, 2018 #6

    berkeman

    User Avatar

    Staff: Mentor

    Are there any patents yet for innovative configurations of solar stills that have that benefit?
     
  8. Oct 26, 2018 #7
    Is there a surface coating that is transparent to the sunlight wavelengths that contain the most energy, and has high emissivity at the long wave infrared frequencies of warm glass?

    Plus what @berkeman said about patents. A patent search is an excellent way to find out what other people have done.
     
  9. Oct 27, 2018 #8
    Potentially dumb idea, but what about a glass sandwich? Two layers separated by a few mm, with open ends at top and bottom, but sealed on the sides. The idea being that as the glass directly on top of the still heats up, the gap to acts as a chimney, drawing in cool air from the bottom.
     
  10. Oct 27, 2018 #9
    Not a dumb idea at all. US Patent 4,235,679 from 1979:
    upload_2018-10-27_7-30-52.png
     
  11. Oct 27, 2018 #10

    256bits

    User Avatar
    Gold Member

    The glass heats up not from the sunlight, but from the inside temperature of the vapour inside the enclosure, ( and from the condensation of the water upon it ). Condensation has a heat of vapourization, and it is this energy transferred to the glass as the water vapour condenses to liquid that has to be expulsed to the anbient. The still performance efficiency is an inverse function of the ambient air temperature, with cooler air making the still more efficient as the glass transfers the energy it gained from the condensation, to the air through conduction and convection at a higher rate.

    The glass sandwich - the chimney effect, although not altogether that strong at a low inclined angle could be expected to improve the flow of directed air over the bottom pane, and eliminate stagnant areas. How much of an improvement - hard to say. Calculations for a single pass type of heat exchanger for the sandwich could be done to give an indication of heat flow to the moving air, and compare that for without the sandwich.

    Although if our calculations are good for a 10% error, and the improvement is only 5%, the calculations won't show that one way or another.
    Might be better to have two stills one with and one without and compare the experimental results.
    .
     
  12. Oct 27, 2018 #11

    256bits

    User Avatar
    Gold Member

    What's the guy got with his water flow going the wrong way, and the insulation on the wrong side of the enclosure at the left hand side.?
    I guess he/she is trying to do two competing things at once - heating up the seawater and cooling the condensation. hmm.
     
  13. Oct 27, 2018 #12

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    We need to examine the original scenario here. The output of a solar still will be proportional to the area and, once the air inside it is saturated, the water collected will just be proportional to the total energy that's been received and not be affected too much by the glass temperature. The whole still should be insulated from the ground underneath so that all the received energy does what's required by staying inside the still, rather than warming up the ground. If you have space behind or to the side of the still, an area with a cooler surface could produce condensation earlier. If there are motors involved, why not pour some salt water over the top of the glass? It also seems over complicated - like it grew and grew.
    It looks right to me. It's common to circulate water in contra flow when you want to maximise heat exchange.
     
  14. Oct 27, 2018 #13

    256bits

    User Avatar
    Gold Member

    Here is how I see it.
    Look at the section up top and to the left. In the present configuration,
    The water flowing in the section open to sunlight approaches somewhere between the temperature of the condensate and the ambient. It then travels down the section insulated from ambient with no heat flow to the ambient environment, but rather absorbing heat from the vapour due to conduction ( and perhaps some condensation ) and thus approaches the vapour temperature. Next, flowing through the sea water/ water heat exchanger, there is an exchange of heat from the sea water to the water. As the water exits, since its temperature closely resembles that of the liquid sea water, which is being heated by the sunlight, the condensation upon the inner glass is inhibited and performance drops.

    The insulation should be on the inner surface so that heat flow can be from the water to ambient, so that the water exits at a lessor temperature than the vapour.
    The water flow should be in the opposite direction so that the water enters the section open to sunlight at the lessor temperature.

    The salt water/ water heat exchanger configuration serves no purpose. In fact it would be much better to bypass the sea water and avoid bringing the water definitely to the temperature of the salt water.

    The vapour and liquid salt water in the closed container well be at the same temperature except in the region where the vapour contacts the glass pane.
     
  15. Oct 27, 2018 #14

    LURCH

    User Avatar
    Science Advisor

    If temperature differential is what promotes condensation, then heating up the inside of the glass is better than cooling the outside, because it increases the evaporation rate. Maybe you could focus your attention in that direction. This might even bring you back to your original idea of bubble wrap. Insulation that traps heat inside the container would mean that no matter how hot it gets outside, the heat inside the still will be much greater, insuring a large temperature difference at the glass interface. Dark paint on the inside will also help, and would be pretty cheap.
     
  16. Oct 28, 2018 #15

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    I wouldn't quarrel with that particularly. The basis of the whole thing is a bit questionable though. I can't actually see any point in using the (warm) condensate for cooling if there is a lot of cold sea water available. I would have thought that a cooled condenser (glass roof) and a warm mass of water at the bottom would be what's needed and a (separate) supply of cool sea water would achieve the best condensation. The sea water here would need to be filtered (coarsely) to stop it absorbing incident energy.
    I assuming that there would be a pump involved somewhere.
     
  17. Oct 28, 2018 #16

    256bits

    User Avatar
    Gold Member

    That is a another cool idea. The OP could somehow incorporate that in his system.
     
  18. Oct 28, 2018 #17

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    On second thoughts, it isn't as simple as that. Using heat from the condensation of the vapour to warm up the new seawater gives you 'free' energy so it would be best to use that pre-warmed sea water in the evaporator as the vapour pressure would be higher. It's a common practice to use hot exhaust air or water to pre-heat incoming fluid in heating systems. (Condensing boilers, for example). I wouldn't think that a layer of seawater would affect the light throughput - except there could be more of a problem with growth of marine life in the warm, well lit water. That could be a problem over more than a week or two and clear freshwater (an indirect system) could be kept cleaner.
    EDIT: PS There is a clever dodge possible for this sort of device and that is to use a solar powered pump. It only needs to work when the sun is shining and the whole thing can be autonomous without the need for external power. I came across that idea in the context of a solar hot water system. Very smart, I thought,
     
    Last edited: Oct 28, 2018
  19. Oct 28, 2018 #18

    Tom.G

    User Avatar
    Science Advisor

    As (subtly) indicated in the drawing that @jrmichler posted, the heat exchanger circuit is completely separate from the seawater circuit. That is also claim 1.c. in the patent.
    c. said passage means in said heat-exchanger base means and in said transparent cover means being filled with a heat-exchanger working fluid;

    Addressing the concern of @256bits regarding the position of insulation on the left wall, the insulation is there to conserve the heat recovered from the condensation to add heat to the evaporator... or as @sophiecentaur pointed out that 'waste' heat is recycled to add thermal energy at the evaporator plate, greatly increasing the overall thermal efficiency of the system.

    As the evaporator plate is a counterflow heat exchanger, the working fluid is cooled approaching the temperature of the incoming seawater; which is then used to cool the condenser.

    Overall a rather clever design!
     
  20. Oct 29, 2018 #19

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    A good addition to the basic idea. However, I still have a problem with the cooling circuit. I am not convinced that the best method is to use the condensate, which is warm, to cool the condenser. The through flow will be very slow and I can see the upper and lower volumes of water being nearly at the same temperature. Is that a good thing? If there is a good supply of cold seawater then would this not be better for cooling the condenser? The system could be regarded as a heat engine, pumping water from the brine to the condenser, I think. So a bigger temperature difference would / could increase the efficiency and lift more H2O out of the seawater for a given input of solar energy. It would require a separate heat exchanger to allow clean water, cooled by seawater, to circulate and keep the 'window' clear. That water would need to have some biocide in it. Weed and barnacles are clingy little devils.
    I have tried to figure out the exact functions in the patent diagram. The separate freshwater circulation seems to be there but, however I look at it, I can't see my idea (above) of an extra heat exchanger included in it. Adding it would be pretty simple as long as there is a surplus of seawater available. Any desalination plant would need to be near the sea because this simple system will perform worse as the seawater becomes more concentrated. There would need to be some optimisation here, between brine temperature and brine flow rate.
    PS I have not read the patent because they tend to be such turgid documents and are not written for engineers but for other sneaky loophole-finding lawyers.
    PPS I believe that Reverse Osmosis systems are preferred these days - but they need to be large scale so this almost DIYable.
     
  21. Oct 29, 2018 #20
    The really turgid part of a patent is the claims. Those are written by and for lawyers. When I read a patent, I start with the drawings, then the description section. I only read the claims when absolutely necessary.

    There are 29 patents that reference 4,235,679. Anybody who's interested in solar stills might find it worthwhile to look them up and find what other people have done with that concept. Also, Google Patents found several hundred solar still patents. If the OP is interested in solar stills, (s)he would be well advised to do that patent searching.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted