Welcome to PhantomPilots.com

Sign up for a weekly email of the latest drone news & information

Send me your defective battery for testing??

Discussion in 'Phantom 2 Vision + Discussion' started by yorlik, May 1, 2015.

  1. yorlik

    Joined:
    Feb 9, 2015
    Messages:
    547
    Likes Received:
    91
    I am going to post this here and dji forum.

    Why do our batteries only last 20-50 cycles when most other lipo batteries last 300-500 cycles?

    No one seems to know....

    Why not? Generally, the reason for premature battery failure is one, or a combination, of the following:

    1) Poor design
    2) Operating above peak pulse current capability (xxC rating wrong or exceeded)
    3) Incorrect software/firmware saying they are worn out when they are not
    4) Overcharging
    5) Discharging too deeply
    6) Cell balance issue

    I think some testing by unbiased, non dji, engineers may gain knowledge why this is so, so I propose to do it....

    But to do this, I NEED defective batteries to test. If you have bad batteries that are of no value to you, consider sending them to me to testing. I have the equipment to do this. I make no guarantees, but in an effort to find the reasons, I offer to test and report back here. Perhaps it is time for us to join the rest of the world who uses lipo batteries and have many many years of reliable life out of them.
     
  2. N017RW

    Joined:
    May 2, 2014
    Messages:
    6,257
    Likes Received:
    1,618
    Location:
    Palm Beach Co.- FL
    Interesting Yorlik.

    I hope you get some volunteers. I'll be watching.

    I don't have any batteries but I must say 50 or 60 cycles is, in my experience with CP-helis, pretty good.
    I always used discharge 'C' values at about 2X the expected drain, never flew below 10.5V discharge under load (3S), and use 'computer controlled' proper charging (@1C) and balancing.
    Then they drop off to almost un-flyable pretty quickly.

    Please investigate or review, if you haven't already, the standard LiPo charge profile of CC/CV and compare that to the DJI chargers.
    My test show my P2NV charger does not use this profile.

    Is this contributing to their seemingly short-lived usefulness?
     
  3. yorlik

    Joined:
    Feb 9, 2015
    Messages:
    547
    Likes Received:
    91
    I have no 'bad' batteries, so have not started an investigation yet. when I do, yes, I will be looking closely at dji charge method and see if it is indeed cc/cv approach.
     
  4. Steve Moore

    Joined:
    Jun 6, 2015
    Messages:
    249
    Likes Received:
    64
    Location:
    Topeka, Ks
    You dont need dead batteries to determine charging method. I dont have any idea what all that means, but Im smart enough to know that ALL DJI batteries will be connected to the same charge, and dead or alive that charger will use whatever method.
     
  5. sdtag

    Joined:
    Apr 5, 2015
    Messages:
    221
    Likes Received:
    23
    I have one bad battery I can send to you. It is out of my P2V+.
    I crashed and the battery ejected and now it won't turn on. I took it apart to see what I could see. Big fat nada. (Nothing obvious)
    Want it?
     
  6. yorlik

    Joined:
    Feb 9, 2015
    Messages:
    547
    Likes Received:
    91

    Sure. Maybe I can fix it for you, since it is not the normal "it wore out" prematurely issue that I want to investigate.
     
  7. yorlik

    Joined:
    Feb 9, 2015
    Messages:
    547
    Likes Received:
    91
    I am confused by your reply, but are you saying you know why they often & typically prematurely die at 30-50 charge cycles? If so, please share! THAT is what I want to investigate.....
     
  8. yorlik

    Joined:
    Feb 9, 2015
    Messages:
    547
    Likes Received:
    91
    I have my first bad battery to test!

    For reference, here is what I found so far:

    I was sent a bad battery - unfortunately it is not a premature failure case but rather one that went for a swim and died.... I sent question to owner who sent it to me...

    All 3 cells in series were only 0.8v! So no chance to turn on.... Is this one that went in water and stayed a while? I vaguely recall it is not a normal 'worn out' due to 20 charge cycles, but??

    So I began with a variable PS across the batteries - all of em, but on the batt side of the MOSFET on/off switches.... I will document as I get enough data to warrant it, but for now:

    - with 1/2amp charge it came upto 2v in 1 minute..
    - then 8v in 3 minutes more at 4amp charge rate
    - then 10v in 5 minutes (impatient mikey upped it to 5amp charge :)
    - 11.0v in 7 min at same 5 amp
    - now she was heating up - hit a tad over 100F so turned down to 3amp...
    - (for reference, a batt can hit 120F w/o harm)
    - still 11.0v@8min@3amp
    - 11.6v@9.5min@3a
    - 11.9v@11min@3a so figured that was all I could do without using the smart charger.

    charge off it dropped to 11.0v
    let it sit for 1min and it dropped to 10.8v
    10.1v@5min sitting

    so since I guess the lipo chemical make up is 3.7v max per cell, I stopped and dropped to constant voltage max of 11.1v with current limit of 0.20amp.... this should be safe to pump em up for a while as I write this and think about it....

    all btw with batt in large ammo can for safety....

    temp back down to 90F now while 0.2a charge toward 11.1v max....

    pushing the on button on pcb does not turn it on, so it still prob has dead sh1t on pcb....

    oh ya, all 3 cells are puffed out about 1/4" on one skinny side! No puff on other 3 sides.... Did you realize this? And THAT puff of gas is what got the hot the fastest....

    I have been on .2a charge toward 11.1v max now for about 60 minutes and it is sitting at 10.8v.....

    I was hoping the 100F caused the batt to not turn on (there is a temp sensor) but alas, no. still dead at 90F for smart batt turn on......

    10.8/3=3.600v per cell set.... @ .2a charge
    10.7v/3=3.567 no charge....

    will be interesting to see how they come out on lipo charger and what mah they have... I am afraid your leaving them to drop to 0 volts each cell may have killed any chance of them being of value to use in the future in a quad, or even to test capacity balance issue. Shoot!

    More as I figure it. I need batts that act bad without taking a swim that just seem to die prematurely at 20-50 cycles to prove my theory and allow me to show the various dji repair houses how to actually revive them for folks! It has to be cheaper than buying new when these batts have potential to actually run 300 reliable cycles if they do not become unbalanced!

    Meanwhile, if YOU - *YOU* - have a battery that still works, has fewer than the speced 300 cycles on it, but feel (rightly so!) it is not safe to fly anymore, and want to donate it to the cause, contact me! There HAVE to be a LOT of these kind of batteries out there that folks are not willing to use anymore but still 'SEEM' ok & report ok in dji Assistant.... I am not charging for my study; I am doing it for our sport.
     
    #8 yorlik, Jul 14, 2015
    Last edited: Jul 14, 2015
  9. yorlik

    Joined:
    Feb 9, 2015
    Messages:
    547
    Likes Received:
    91
    SDtag, played more today with your batt that would not turn on after crash. Recall at least 1 cell did not charge well.... took it apart to the 6 individual cells. It has been sitting now for 2 weeks.

    2 of the 6 cells are back at 0v - of course the sets were in parallel. These get tossed.

    2 of the 6 cells are at 1.4v - of course the sets were in parallel. These get tossed.

    One cell set is good still; 3.6790 & 3.6791. I will charge/discharge these individually and check how they do.

    Each cell is marked -ATL 9.62Wh 3.7v
    673497 then a serial number
    A4520FCJV1

    In case anyone in interested in trying to buy replacement cells!
     
  10. yorlik

    Joined:
    Feb 9, 2015
    Messages:
    547
    Likes Received:
    91
    Thought I would post that this is a wrong comment. After taking these apart, I see the two wide flat sides (that can show puffiness) are only the outside two cells (of 6 total). They are packed together so tightly that only the wo outside ones are capable of puffing for us to see; the inard cells are just packaged too tightly to puff at all. The short sides of the smart battery are the skinny sides and as such cannot puff.

    Thought I would toss that out.
     
  11. yorlik

    Joined:
    Feb 9, 2015
    Messages:
    547
    Likes Received:
    91
    SD, FYI, I have charged/discharged one of the two parallel 'good' cells. Guess what? It is like new.

    It took 2250mah on CC session of the charge, then finished on CV session at 4.2v, only heated to 77F, and finished with 2622mah going in.

    It gave up 2711mah at 86F with 1C discharge down to 3.2v. Showed it was a good 36milliohm internal resistance too. Remember there are two of these in parallel for each of the 3 sets - so discharge >2600mah is like new.

    I call this basically a brand new cell.

    Will get to the other cell sometime next. It is a shame this battery was allowed to discharge to 0v and sit or I bet all the other cells would respond as new too....
     
  12. sdtag

    Joined:
    Apr 5, 2015
    Messages:
    221
    Likes Received:
    23
    Yeah I think what happened in my crash was the battery appeared to be dead but I bet it was still still on and discharged down to infinity. with the button not working there was no way to turn it off.
    Stinkin trees!
     
  13. ahmida

    Joined:
    Jul 25, 2014
    Messages:
    43
    Likes Received:
    0
    thanks for your work man :)
    anyone know where I can buy replacement cells for the P2V+ battery?
     
  14. yorlik

    Joined:
    Feb 9, 2015
    Messages:
    547
    Likes Received:
    91
    SD, just finished charge/discharge cycles on your other cell from defective battery. It too is like new. charged to 2446mah (instead of 2500) but I then pulled 2646mah out of it down to 3.2v. I recharged it 1/2 way and will store both good cells in the beer fridge (25.5F) for the forseeable future (until needed to fix some other battery). Thanks for your donation and it was good to see that 2 of the 6 cells that discharged to 0v and sat there for some time recovered. I must remember this was NOT one of the 20x charge batts that showed bad cell or just drops voltage when one gets to 40%.

    I am off to test Richwest complete battery that dji DOES report bad cell on next! It already went thru first round testing that showed dji assistance data was right on calling it 3800mah capacity left. Now I am going to work with each cell group and see if my theory of poor balancing brings it back.... Will report that in his thread here:

    http://www.phantompilots.com/threads/reconditioning-batteries.45401/page-3
     
  15. battman2036

    Joined:
    Aug 12, 2014
    Messages:
    246
    Likes Received:
    67
    Location:
    Cleveland, OH
    Still wanting P2+ batteries? I have one that drops mid flight to low voltage after only about 8 minutes.
     
  16. yorlik

    Joined:
    Feb 9, 2015
    Messages:
    547
    Likes Received:
    91
    absolutely! THAT is exactly the type issue I am investigating!

    I will pm my address. thank you.