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

How can I calculate the usage of a generator for a system?

  1. Jul 23, 2015 #1
    I've got an RO system and due to power fluctuations I need a generator for it. Can you tell me of what KVA generator should i buy for it and how to go about the calculations. Thank You
  2. jcsd
  3. Jul 23, 2015 #2


    User Avatar

    Staff: Mentor

    Moving to EE section...

    Welcome to PF! Sure, we can help, but you haven't told us anything about the RO unit's power requirements!

    Note: there shouldn't be much, if anything, to calculate here.
  4. Jul 24, 2015 #3
    Thank You Sir.

    This is the quotation i got for 5000gallons/day.

    There are different pumps to be installed and their respective power consumption values are given.

    Kindly help me out in this issue.


    Attached Files:

  5. Jul 24, 2015 #4
    You need to add up all the power needed for the various components. One of these is not listed (your well pump).

    I see:
    • Water feed pump -- 2kW (1.1kW is also listed; use the bigger number)
    • High pressure pump -- 3 kW
    • Pre-dosing pump -- 0.5kW
    • Drip cleaning pump -- 0.55kW
    • Anti-scalant dosing pump -- ???
    • Bore pump (well pump?) -- ???
    In addition you will likely want some extra power for lights, small appliances (your laptop) or other miscellaneous gear. Bore pumps vary quite a bit depending on the depth of the well and the amount of water needed. (Remember you will need lots of extra water for backwash, etc. maybe 50,000 gallons extra for a 5,000 gallon output.)

    As a rough guess, 2kw. More for a deep well. I would guess the Anti-scalant pump is similar to the regular dosing pump at 0.5W.

    This gives a sum of about 10kW. You might have a minor power factor problem with all those motors.

    You should be able to find a qualified installer who will sell you a system which integrates with utility power and comes on automatically when the power fails. They will run the numbers for you, but this should give you a ballpark figure.
  6. Jul 25, 2015 #5


    Staff: Mentor

    Your power utility should also be able to give you historical data of your own power usage. I say that because there may be loads that aren't mentioned in the RO specifications.

    To allow for uncertainty and future expansion, I would add 30% margin to the generator capacity you calculate.

    Good luck
  7. Jul 25, 2015 #6
    I connect a lot of generators and use a very simple method

    Measure the current drawn using a grip-ammeter (you can do this during the times of highest load if you want) at the intake

    Convert to KW, add fudge factor (25%)

  8. Jul 25, 2015 #7
    I like adding 30%.

    I would be cautious about using the power company's data. It is unlikely all the pumps are typically on at the same time, thus the numbers could be low. But all the pumps could be on at the same time, so the generator needs to be able to supply that amount.
  9. Jul 25, 2015 #8


    User Avatar

    Staff: Mentor

    My understanding is that you should NOT convert to kW because the generator's limitation is that amperage through the windings, which means that when you have poor power factor (such as when running multiple small motors at part load), you risk burning it out. Perhaps residential generators aren't rated by kva, but I'm pretty sure commercial ones are.

    But if anything, calculating the kva and using it as kw would provide the extra safety factor needed if the generator is rated by kw and the max amperage isn't specified.
  10. Jul 25, 2015 #9
    Correct me if I'm wrong.

    It is my understanding that a bad power factor is caused by current being out of phase with voltage. Because the current is out of phase, it does register as power (and isn't power for that matter). But it will still register as current.

    William White's measurements should work because he's measuring total current. Where he went wrong is in semantics. He called his measurements kW instead of kVA.
  11. Jul 25, 2015 #10
    When I worked in power distribution I connected generators live to the LV distribution grid a lot

    There was the complicated way,

    and the easy way.

    Add up the all the currents in all the phases, divide by 4* and add a bit - and that's the answer in KW (which is what one would ramp the generator up to before closing its breaker onto the grid)

    This ALWAYS works, you can do it in your head in a second

    One can over-complicate matters. Connecting generators is easy peasy.

    So, if my grip ammeter read, 50A, 120A and 90A thats 260A, divide by 4


    add bit

    ramp generator up to 75-ish KW

    *Assumes 4 amps per KW, which is a good approximation for the UK and allows an easy in your head division (halve you current, then halve it again). In other countries you could have the rule of thumb by dividing by whatever convient number you need.
    Last edited: Jul 25, 2015
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook