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

How fast can you cool water, given a large enough area/run?

  1. Oct 17, 2016 #1
    Given a lot of room, talking hundreds of feet of piping, and potentially hundreds of square meters for equipment, how fast can you drop an input ranging from 50-100C down to 5-10C? It can also be partially underground (shallow). In this case, I could use the waste heat, especially if it's in a couple specific locations (i.e. a solar water heater, and a refrigerator/heat exchanger).

    In a closed loop system, I would assume that dropping the temperature down to ~25C wouldn't be difficult given some distance. But with flowing pipes, how fast could you feasibly 'flash cool' a fairly large volume of water/heat transfer fluid from ~25C to ~5C, or potentially 50C to 5C given a shorter distance?
     
  2. jcsd
  3. Oct 17, 2016 #2

    rbelli1

    User Avatar
    Gold Member

    How much money do you want to throw at the problem? Can you pin down what fluid you want to cool? How much of it? A 1mm tube of fluid is a totally different story than a 5 meter tube.

    BoB
     
  4. Oct 18, 2016 #3
    Yes, you're right about the different pipe widths. That really is the essence of my question, how much could you theoretically cool that fast as far as flow/volume is concerned. This is basically the order of importance of the different parts of the system:

    (least important)
    -Size of cooling components
    -Fluid type (Not too important as it is a closed loop, though volume is rather large)
    -Cost (It's a rather small part of a large building idea, so not extremely important as long as it's not made of solid gold)
    -Power efficiency (less electric is better if passive methods can be used)
    -Cooling speed/capacity (likely to be a fairly slow flow, but medium-large diameter pipe)
    (most important)

    The cooling potential is definitely the most important. It's looking at the possibility of being both a heating and cooling fluid within the same closed loop. Heating it is not a problem. Shedding most of the heat before having to cool it shouldn't be bad, either. But the efficacy of the system as a heat sink at that point is the main question. I could potentially do two separate loops, but I get free water pressure from it being all in one loop, which removes the need for a pump. So really, the question is whether it's more efficient to cool one system from a potentially higher temperature, or cool it from a (likely) lower temperature, with an extra pump required. But of course, that's a moot point if cooling the system down from a higher temperature is impossible or extremely expensive.
     
  5. Oct 18, 2016 #4

    russ_watters

    User Avatar

    Staff: Mentor

    How are you cooling the water, exactly? And by "how fast" are you literally mean the time or do you really mean the capacity? Since water goes in and comes out at the same time, it is effectively instant. Typically engineers ask "how much?"

    Further, usually your load dictates your capacity, not the other way around. So how much cooling do you need? Because as vague as the question is, the most complete answer possible is "a lot".
     
  6. Oct 18, 2016 #5
    I'm currently waiting on the specs from a reliable source. But if anyone happens to know by experience, it is essentially radiant cooling for 5000 sq ft.
     
  7. Oct 18, 2016 #6

    jim mcnamara

    User Avatar

    Staff: Mentor

    If you want 5°C then what are ambient temperatures?

    Your problem, as presented, appears not to be handled by radiant cooling without additional specialty cooling equipment. Point being: either I fail to see what you are doing, or your description of the problem is really incomplete.
     
  8. Oct 18, 2016 #7
    I meant as a radiant cooling system (hydronic floorboard/panel cooling). It's in a hot climate, but most of the run would be buried shallow, slightly below room temperature. It would head through room temperature for a bit (nominal change with insulation) to the roof area with temps around 60C.

    I know there are potential issues with condensation at the lower temperatures, and they aren't strictly needed to cool temps of 60C. Room temperature water would have some effect. But with a relatively slow flow, and a large load to deal with, I wanted to look at pre-cooling the liquid. I can assume immediately that spending energy to cool the attic would result in a loss overall compared to a simpler solution. But I am looking at both reducing that load, and cooling PV panels, giving me more efficiency from my power source. And it's a large PV array (~75kW per building), making even small percentage gains pretty substantial.
     
  9. Oct 18, 2016 #8

    russ_watters

    User Avatar

    Staff: Mentor

    There is no source that can provide you with the cooling load. You have to calculate it based on the specific details of what you are trying to cool. Are you designing a system to cool your house?
     
  10. Oct 18, 2016 #9
    I meant someone who deals with plumbing systems for these applications, so I could get the expected flow rate and pipe size for one of those systems. The system is designed to cool BIPV panels, with the side effect of also cooling the attic, and lowering the load on the house through convection. In that size PV system, every degree C it is cooled equates to about 100 kWh/month of power (standard PV panels get a linear 1% benefit per degree cooled). It is not meant to be the main cooling system of the building, but it should reduce the load. I know there will also be some energy lost from cooling a small amount on the outside.
     
  11. Oct 18, 2016 #10

    russ_watters

    User Avatar

    Staff: Mentor

    Ok, that's clearer. So we'll need to know the cooling power (btu/hr or watts) and flow rate and delta-t of the water and the solar panel vendor should be able to tell you. Then we can figure out how to reject the heat (hopefully you already have some idea about how you want to reject the heat).
     
    Last edited: Oct 18, 2016
  12. Oct 18, 2016 #11

    rbelli1

    User Avatar
    Gold Member

    I had an idea similar to that where under high solar influx you heat the residence by cooling the solar panels.

    I have never done the math on that for two reasons. Firstly the quality of the heat is low and the efficiency gains are low as well (given just passive heat movement aside from fluid flow). Secondly I don't have the funds to build it so the answer is not exactly useful to me.

    You need to do a cost benefit analysis to see if the added complexity is worth the effort. I would guess that a do it yourself scrap built type arrangement would be sensible. A project built from new components under contract would depend on the outcome of the calculations.

    BoB
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: How fast can you cool water, given a large enough area/run?
  1. Large Viewing Area (Replies: 1)

Loading...