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

Home made pool heater.

  1. Oct 26, 2008 #1
    >>>>>>>Home made pool heater.<<<<<<<<<

    Ok, i know this sound a little stupid.

    I would like to make a home made hot tub for my mates for this Xmas.

    I have put together a manifold from 15mm Dia copper tubing and is 8 Meters long.

    I will be heating this manifold within a beer keg that will hold a coal fire.

    The pool i have bought is a 10ft dia 30" high pool holding approx 3500lt of water.

    My pump 650 watt, max pump rate is 11500 Lt/hour.

    So my question to anyone who can answer it (or give a little guidance)

    Can you work out how long the pool will take to heat up to a nice comfortable temperature.

    Now i am aware there are probably 1000s of variables in this equation, those i would like you to guess/over look.

    The water that i will be heating up with firstly be straight from the tap (through the heater and into the pool) and then i will be circulation the water (so it will be cold to start with).

    So if someone thinks they can tackle this i would be very grateful (and might help me work out how much coal i will need and how long before people arrive i will have to fire it up, I'm thinking 20hours)

    also if this is in the wrong place i am sorry, but couldn't see any thermodynamics pages.

    Any other information you need i can give.

    Last edited: Oct 27, 2008
  2. jcsd
  3. Oct 28, 2008 #2
    I don't have an answer to your question, but a FYI, my buddy had a similar contraption. They had a long copper tube that was painted black in wooden box that twisted and turned to give a lot of surface area. Water continually circulated through it and the tubes would get very hot from the sun. I'm not sure how well it ever worked, but someone had told him that you get a bigger net heating effect if you pump in water that's slightly higher in temperature fast than much higher temperature slower. I'm not sure if that's true, but that's what he heard. Good luck.
  4. Oct 28, 2008 #3


    User Avatar
    Gold Member

    Unfortunately, I can't help with the calculations either. I would suggest, however, that you add some sort of filtration/demineralization system to the circuit to minimize scaling and clogging.
  5. Oct 28, 2008 #4
    This is NOT rigorous but will get you in the ballpark (soccer field, if you prefer).....

    One btu will heat a pound of water about one degree farenheit in one hour.

    The real difficult part is figuring out how many BTU's a coal fire will produce hourly and how much of that output would be transferred via your manifold to actually heat the circulating water.

    My guess is that your 15mm tubing is far too small...you need more diameter!!! a 650 watt pump should be able to pump a LOT more water than the tube will handle...I suggest tubing of 3 or 4 cm at least.... (650 watts compares with about 746 watts/hp, 650/746 is about 0.87 HP...BIG pump....more than enough

    Weight of 3500 liters of water:

    1 gal = 3.8 liters and weighs about 8.4 lbs, so 3500 x 1/3.8 x 8.4 is 7737 lbs

    let's assume 25,000 BTU is used to effectively heat the water....this is about five gas stove top burners worth of heat...quite a lot...

    25,000 would heat 25,000lbs of water one degree/hr and you have 7737 lbs, so it would rise
    about 25,000/7737 or 3.2 degrees farenheit per hour...

    I'm confident a coal fire in a regular size US beer keg could produce enough heat, but I don't know how much would pass to the water....

    With the larger tubing I recommend, I'd GUESS maybe 2 degrees F per hour temp rise....

    My gas heated pool at home: about 30,000 gal water, a 250,000 BTU heater raises the temp about one degree farenheit per hour....I know because I have measured it many times using a thermometer...using a 1.5 HP circulation pump....

    so : 30,000 x 8.4 lbs/gal /250,000 IS about the one degree I measure!!!

    (Somebody check my math, got to go...)
  6. Oct 28, 2008 #5


    User Avatar

    Staff: Mentor

    I recommend keeping the 15cm tubing, but manifolding 8 one meter pieces toegether for a multi-tube heat exchanger. That will improve heat transfer rate by increasing flow while keeping the surface area large. Bigger pipes have a higher inside area to surface area ratio.
    Actually, it's more like 3 - 9,000 BTU is pretty typical for a medium-sized range burner:

    Coal has a heat capacity of 9,000 to 14,000 btu. Figure your combustion efficiency to a terrible 50% and you'll need about 4 lb an hour of charcoal to get that 25,000 btu.

    Your assumptions and end-result calculations look pretty good to me.
  7. Oct 29, 2008 #6
    This information is fantastic.
    So Its going to take along time to heat up, but if it works this isn’t a problem.

    I am having some reservation about my original manifold design, its very simple and I think might not work.

    Sorry i don’t have a photo of it but i looks like this.

    So it coils in. see attchment

    I know that is an horrendous drawing but stupid computer doesn’t even have paint.

    I have already cut allot off copper for the first it, so I might carry on with that and test it.

    AcidBathSDMF = Yes they do make solar ones and they do work well, but i live in sunny UK./…..NOT. So solar is kind of out of the question.
    And your right having a faster flow is better, the last thing i want is for the water to boil.

    Danger= Thanks for the suggestion, if it works i will do this.

    Naty1= The information you have given me is great, if a little demoralizing lol. It looks like its going to take a age to warm up the water, but i have learn a lot from your post.

    Russ_watters = I am going to keep the 15mm tubing (carnt afford to buy any more). Thank you for your comments as well, they have been very helpful.

    I have one more question if any minds answering it.

    Your say that 4lb of coal should/could produce 25000 BTU.

    Would doubling the amount of coal burnt double that to 50000 BTU, (or doesn’t it work like that)

    If i filled the pool with water and then went about heating it, if i could get this to this magic
    25000 BTU then i think it would take about 40 hours to heat to a nice temperature.
    I have worked that out from tap water being about 4 degrees c (40 F) and wanted it at least 27 degrees c (80 F)
    So this is 40 hours of heating.

    That is really out of the question.

    So if i put the water out the tap, through the heater first. This should get my pools starting temperature at my output temperature, although the pool temperature would drop due to the very cold air temperature (I will put a cover over the pool while heating to try and keep as much heat in as possible..

    Ideally i could have two manifolds, one to heat the tap water, and another to circulate the rising pool water.

    This way i think i could reduce the 40 hours substantial.

    I could also start the heating on gas BBQ, use 1 bottle (13hours) leave it on all night, then in the morning i could change over to the coal, because i can then keep an eye on it.

    Do you think this would be a better way of heating my pool.

    P.s thankyou again for your help

    Attached Files:

    Last edited: Oct 29, 2008
  8. Oct 29, 2008 #7
    Your best single approach would likely be to use hot water from a home type house heater to provide warm/hot water initially; it's been designed to work with the right match of input heat, transfer to water, and then you can fill your hottub slowly..... and then it's easy to keep it that temperature for a while with modest additional heat.

    Next best heating is, as you suggest, likely a propane bar b q for ease of heating...But likely that heat transfer will be much lower than a hot coal fire....place your tubing with water CLOSE to the flames, remove grate and flame deflection shields, etc...

    for the record, 15mm tubing is too small..but I see your length is 8 meters, I was thinking 8 ft...so it might be barely enough surface area......but it will take longer than I crudely estimated to transfer enough heat....

    Glad you planned this before the beer began to flow!!!! Don't forget life jackets!!!
  9. Oct 29, 2008 #8
    As long is its possible i'll give it a go.
    And even if it isnt i'll still try.

    As for filling it with hot water from the house, i'd like to but i think my parents would kills me.

    I will post some pic's of the copper piping when i have made that. and test results.
    I have a couple of aquarium Thermometers.

    I might fill a standard bin with cold water, measure the temp. Then pump that water through the fire into another bin (not letting the pump run dry) then measure the temp of the "hot" bin.

    This should give me an indication of the temperature raising ability of my set up.
  10. Nov 26, 2008 #9
    Just an update.

    I have built it, and it passed the tests.

    I heated a tank of water 50 litres up to 140F in about 10min.

    Now for a bigger tank.

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