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

Moving small quanities of water in a pressurized loop

  1. Aug 20, 2015 #1
    First time here. I browsed around a bit, but did not see a topic quite like what I am about to write, so I started a new topic. If I missed an existing topic, please point me to it, and I will thank you for doing so.

    Please select a prefix? I seem to remember some of this from High School and/or first year College Physics, but that was about 35 years ago, and I haven't had occasion to use much of what I studied...so I am gonna guess that it would be High School level stuff

    Simple scenario: I want to move water from my hot water heater, up to the roof, where it will run through a solar collector, and then drop back into the water heater. I will do this using a very low (possibly solar) powered pump, during daylight hours only.

    The questions that come to mind are as follows:

    When selecting a pump, do I have to account for the pressure of the water in the loop, in addition to the pressure created by the elevation change between the tank and the solar collector, or, because it is a closed loop, does that pressure cancel itself out, coming down the return side of the loop?

    2. Since I will have a closed loop, would it make sense to place the pump at the highest elevation, thus requiring less power to operate? (I think this is a "no", but I am not a Mathematician, Engineer, or Physicist...) Does it make more sense to have it at the bottom of the return side? (push-me, pull-me, probably makes no difference, right?) My gut tells me anywhere on a closed loop is going to require virtually the same energy, but I have to ask...

    I'm sure there must be an online calculator, somewhere, but I can not find one for the layman. Can anyone tell me how many PSI (sorry for old school standards,) at the bottom of a 20 foot column of standing water.

    In the end what I have envisioned is moving a gallon (or so) of water an hour - just enough to replace heat that is leaching off the tank, and into the atmosphere of the utility room where the tank resides. I will probably use 1/4" copper tubing for the loop, wrapped in some sort of insulation, most likely fiberglass, on the feed and return, and an enclosed reflection box as the collector.

    I give the information about application so that you can envision what I am setting up. I have seen proof of concept, and I know that a collector like this does a very good job of heating water to scalding temperatures, in under a minute on a clear day, if collector volumes, and other details of the collector itself are properly constructed. However, incorporating a storage tank, and recirculating the hot water in an elevated loop through the tank are not something I have good information about.

    Any and all suggestions and raw information are very much appreciated, and should this mental exercise actually become a project that I build, I will be more than happy to share the results.

    Thank you in advance,

  2. jcsd
  3. Aug 20, 2015 #2


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    All you need is pump enough to maintain the very small pressure difference for the flow rates you describe; it will have to be robust enough to confine line pressure from your water supply.
    Pretty much anywhere on the loop --- lower is probably better to prevent "vapor lock" or necessity of priming --- you are talking about heating water.
    ~ 9 psi.
  4. Aug 21, 2015 #3

    So, as long as my pump can handle a 9psi load (I'll probably double it because I tend to overbuild any prototype,) and the volume that I decide on, I should be fine, right?

    You raise an excellent point regarding vapor lock - I have dissolved methane in my well water, and the hot water side of my faucet always "spits" because of of the gas separating, and then collecting in the top of the hot water tank, much like CO2 perculating out of a glass of soda. I'll have to make an allowance for bleeding that gas out of my loop periodically, manually, if not automagically...

    I am considering a roller pump, because they tend to be fairly quiet. Any recomendations...please support with reasons.

    Thank you for your assistance,

  5. Aug 22, 2015 #4


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    If the system is sealed very high temperatures are possible. Especially in the collector once your tank is up to maximum temperature causing the pump to switch off. It's even possible for soldered joints to fail.


  6. Aug 22, 2015 #5
    Thank you for that link. I have only had a few minutes to browse it, so far, and my head already hurts. That's usually a good sign for me, as it means that I have identified a tretise that has something from which I can learn. :) For instance, I mistakenly called my proposed design a closed loop - clearly, it is not, as it would feed from, and empty into the hot water tank in my utility room...

    A couple questions:
    1. Do you think I should make a closed loop anti-freeze system, as detailed in the document? Either way, could you weigh in on pros vs cons?

    2. Do you think that my proposed (non-closed-loop) system would be subject to heats extreme enough to liquify standard plumbing pipe solder? Imagine the worse case scenario, with a pump failure on a hot, clear summer day. Please bear in mind that the hot water tank has it's own pressure relief safety valve, so extreme pressure is not a likely issue for the new system's integrity.

    3. Can anyone estmate a maximum temperature possible in the coils in the collection box, again, given the above worst case scenario? It would be easier to design this, if the collection box did not have to be integral to the circulation loop. The math and science to make that calculation are way over my head, but I think I know what to do with the number, once calculated.

    4. Speaking of math, mine is pretty rusty.
    I need to calculate the volume of the liquid inside a tube.
    Do I remember correctly that the area of the cross section of the tube, multiplied by the length of the tube, gives me the volume of the tube?
    I found a site (http://www.calculatorsoup.com/calculators/construction/tank.php) that would have me believe that 100 feet of 1/4 inch tubing would hold only a quart of liquid. Granted, the site is set up for tanks, but I assumed that my "tank" was upright, had a 1/4 inch diameter base, and was 100 feet tall. That number just seems real small to me, and has me wondering if I should increase my diameter to 3/8" or even 1/2", in order to get enough heat from the roof to the tank? BY stepping up to 3/8 inch tubing, I can more than double the volume, and with 1/2 inch, it's 4 times the volume. (I understand from automotive cooling systems that moving the water very fast is not always the most efficient way to transfer heat... so I am thinking larger volume, slower velocity?) Feedback appreciated.

    I am very happy to have found this website. It is frequently difficult to get an educated, and mature response to questions on almost any topic. The free flow of information and ideas is someting I consider a valuable commodity.

    Thank you,

  7. Sep 5, 2015 #6
    If my research is correct, standard plumbing solder liquifies at a lower end of around 430 degrees (F)

    The document that CWatters linked says that this kind of solder can not be used when the working temp exceeds 230 (F)
    ref (http://www.copper.org/publications/pub_list/pdf/copper_tube_handbook.pdf) page 28

    It looks like the solar collector may see temperatures as high as 305 degrees (F) or so. ( I can't find my source for that statistic. If you think I am wrong, please offer a better answer...in the meantime, I am still trying to find the link that lead me to that number...)

    200 degrees (F) working temperature is a huge difference between what they say is ok, and what the melt point of the solder is - why the extreme? Is that just over-engineering, or have I overlooked something?

    Since my expected max is only 125 degrees (F) lower than the solder melt point, I am a bit concerned about my ability to "make this myself", as I do not have experience, or the tools to do brazing, which is the next highest temperature jointing process. I'd prefer not to spend a lot of money on tools I won't use very many times, and I don't really want to enroll at the local community college to learn how to braze pipes, for something I will probably only do once or twice in my life.

    Then again, the prospect of having a joint burst, and fill my utility room with boiling water doesn't have much appeal, either...

    Additional information worth considering:
    I have no idea what static tap pressures usually are, but if I have the collector in a loop that picks up from the tank, and dumps back into the tank, the safety relief valve, which is typically rated at 150PSI, will limit the top end of the pressure range. Also, the typical relief valve releases when tank temperatures exceed 210 degrees (F), so I will have to figure out a way to regulate the flow of the heated water from the collector, to keep the temperature under 210 degrees (F). That may include the installation of a larger, or secondary tank. One thing at a time, however...

    As always, constructive feedback is very much appreciated.

  8. Sep 8, 2015 #7
    17 days since anyone posted a reply...

    Is it just slow, or am I doing something wrong?

    If I am doing something wrong, please let me know, and I will make every effort to correct my mistake.


Share this great discussion with others via Reddit, Google+, Twitter, or Facebook