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

Battery life performance

  1. Mar 2, 2009 #1

    I have 2 24V lead acid batteries in parallel to make a power supply for my circuit. I want to do some drain tests, and plot a curve for % energy remaining vs. time. The load on the battery isnt that large, and I dont want to have to supervise my experiment for 12 hours solid.

    Im presuming the curve will be the same if I take one of the batteries out (to shorten the drain time)? If I multiply my values by a factor of 2, will that be a fair representation of the system with both batteries.

    I will be working out the energy remaining using V=E/Q and Q=It, the Wh/kg value I have for my batteries, and the conversion rate of 1Wh = 3.6kJ

    I hope this makes sense, thanks in advance
  2. jcsd
  3. Mar 2, 2009 #2


    User Avatar

    Staff: Mentor

    Are these batteries designed to be used in parallel? Parallel connection of batteries is not usually done, at least not connected straight together. For sure, if you drain on battery in a 1-up experiment, and then hook them back in parallel, there will be problems.
  4. Mar 2, 2009 #3
    Connecting batteries in parallel can be very dangerous if you don't know what your doing. The reason for this is large currents going through the batteries due to differences in potential. Batteries have a relatively low internal resistance, so if you have one at 11.5v and one at 13v and connect them in parallel you are essentially creating a current unlimited short circuit that could cause one of the batteries to explode or catch fire. For batteries in parallel they must always be charged and discharged together. If you discharge one, but not the other, don't put then back in parallel again.

    Back to your question, ideally no it wouldn't make a difference. But in reality, yes it will. Not all batteries are manufactured to the exact same specification, even if they are the same model of the same assembly line. The easiest and safest thing to do would be to discharge them both separately and average the results.
  5. Mar 2, 2009 #4
    You need to study Peukerts Law before embarking on your experiment. Battery capacity is NOT a linear function of load.

    There is no such thing.

    Equally nonsensical as the prior quote.

    Nothing wrong with conducting an experiment, but you can find many such results by looking online at manufacturer specifications
  6. Mar 2, 2009 #5


    User Avatar

    Staff: Mentor

    I suppose not, but I meant more along the lines of same type, same size/capacity, same state (age and chemistry changes due to history of use) same initial charge, etc. All to minimize differences in the voltages when connected together, and during cycling.

    Might be safer to just Schottky diode-or them...
  7. Mar 2, 2009 #6


    User Avatar
    Science Advisor
    Gold Member

    Well, not actually. I have seen UPSs with parallel battery banks. It is dangerous if the batteries aren't at the same voltage however. And I would not recommend doing the experiment.
    Last edited by a moderator: Apr 24, 2017
  8. Mar 2, 2009 #7


    User Avatar

    Staff: Mentor

    I think Naty was referring to my comment about batteries "designed" to be put in parallel, not that parallel battery banks don't exist. I think we agree that they do, but you have to be very smart and careful and know the history of the batteries before hooking up that last wire...
  9. Mar 2, 2009 #8
    He's not trying to determine the capacity based on load, he said he wants to generate a discharge curve based on a constant load.

    Correct. Batteries, or rather battery packs can be designed to be use in parallel configurations by designing cells that "match" each other. "Matched" battery packs or arrays are built using cells that are manufactured to have small differences in specifications such as internal resistance and open circuit voltage so they have very little stray current between the cells. Although, this can't be done with all chemistries. For example, LiPo cells in my RC heli become unmatched after about 20 cycles.
  10. Mar 3, 2009 #9
    Very good to know Topher. What tells you that the batteries are becoming mismatched?

    As Berkeman implies, all things being equal, one should expect each battery should preform as the other, though process variations are a factor that changes things, and there are other unexpected variations such as thermal resistance to ambient during charging and discharging. Apparently Lithium Poly are very touchy?
    Last edited: Mar 3, 2009
  11. Mar 3, 2009 #10
    [...as batteries designed to be used in parallel.]

    I don't know if you're talking about lead acid batteries in particular or batteries in general.

    Do you think you might be able to come up with a reference, or a quote that says something about this?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook