Welcome to InspirePilots.com

Join the leading DJI Inspire community for free!

Inspire 1 battery charging profiles

Discussion in 'Inspire 1 Discussion' started by Vincenzo, Dec 6, 2015.

  1. Vincenzo

    Joined:
    Aug 5, 2015
    Messages:
    14
    Likes Received:
    5
    Hello folks,

    I'm building an automated charger that will charge up to 8 (inspire batteries and/or remotes).
    It is based on an arduino that controls relays and monitors the voltage, current and power taken by the battery been charged.
    Note this is a sequential charger and uses the standard 100W brick charger. The idea is that the arduino can detect when the battery is fully charged (since the battery turns itself off) and move to next battery untill they are all charged.
    So I wanted to check what voltage, current and power changes are detected when the battery being charged stops drawing power. I added an SD card I can log V,A and W for the whole charging time and here are the results:

    With the regular 100W brick PSU:
    The TB48 is charging at constant current ~4.3A for about 48min (I guess it depends on the starting charge\battery health). My PSU puts out up to 110W apparently.
    The remaining time the current decreases gradually for anoter 50min or so and then the battery reaches full charge (by its internal logic) and turns itself off.
    At turn off the PSU voltage jumps up by 40mV (no more load), the current drops to 0 from ~250mA, and the power drops from ~6.5W to zero. So it is extremely easy to detect when the battery turning off just from the current or power measures.

    Anyway I'm attaching the graphs, and maybe I'll let you guys know when the whole thing is finished.
    BTW, the way I have it running it can log ~8.5 measures per second (Way overkill :) I know). The csv file I generate is about 2MB per battery onto a 1GB SD card :).

    Ciao!

    Vincenzo
     
    Eric, Mike Walker and DroneX like this.
  2. DroneX

    Joined:
    Nov 13, 2015
    Messages:
    13
    Likes Received:
    2
    Cool. Would love to see the graphs. Cannot open.
     
    Mike Walker likes this.
  3. Vincenzo

    Joined:
    Aug 5, 2015
    Messages:
    14
    Likes Received:
    5
    I changed the album's privavcy settings... it should be public now.

    V.
     
  4. Vincenzo

    Joined:
    Aug 5, 2015
    Messages:
    14
    Likes Received:
    5
    One more thing,

    having current and time allows me to calculate the mAh being pushed to the battery and compare to the capacity.

    For example, for the TB48 took in 4675mAh 20% less than nominal capacity.
    For the TB47, it took in 3848mAh 15% less than full capacity.

    Unfortunately I didn't record the capacity left in them before charge so I can't tell how efficient is the charging process. I will next time.

    V.
     
    Mike Walker likes this.
  5. Mike Walker

    Joined:
    Jan 31, 2015
    Messages:
    40
    Likes Received:
    11
    Location:
    Humberside, England.
    Brilliant test, thanks for sharing.

    Regards
     
  6. DroneX

    Joined:
    Nov 13, 2015
    Messages:
    13
    Likes Received:
    2
    Great job. Looks like the 100W PS is limiting charge rate. I wish you had a 180W PS for comparison. Does the battery circuitry have the ability to limit charge current? Or do they depend on the PS voltage dropping at high current? I think the latter and the 180W PS is the limit. What think you? What would happen with a 500W PS charging 1-4 batteries simultaniously. I have a 10A <= 30V bench PS. I may do some test also. This kind of stuff is what makes owning a drone a hobby. Thanks
     
  7. DroneX

    Joined:
    Nov 13, 2015
    Messages:
    13
    Likes Received:
    2
    Really nice! What data logger do uoy use?
     
  8. Vincenzo

    Joined:
    Aug 5, 2015
    Messages:
    14
    Likes Received:
    5
    I only have the 100W DJI charger so I can't say for sure, but since DJI is selling a bigger one I would think that would increase the current. However I believe the battery electronics is what determines the current based on how much the power supply voltage drops. For the TB48, the voltage drops at ~23.2V and I think that's what the circuitry checks.

    V.
     
  9. Vincenzo

    Joined:
    Aug 5, 2015
    Messages:
    14
    Likes Received:
    5
    I made one myself based on an arduino and a INA226 :)
     
  10. Vincenzo

    Joined:
    Aug 5, 2015
    Messages:
    14
    Likes Received:
    5
    DroneX, yes the arduino logs the data to an SD shield. The current, voltage and power is measured with an INA226. The INA226 is a chip that measure voltage across a shunt resistor. I'm using a 0.015Ohm resistor in series with the load (the battery) as shunt. The INA226 also monitors the bus voltage and does the multiplication to get the Power. Hope this makes sense.

    V.
     
  11. DroneX

    Joined:
    Nov 13, 2015
    Messages:
    13
    Likes Received:
    2
    Thanks V, I have a TB47 that has two weak cells after only 5 charges. Still works but not safe to fly. I will do some test with my bench power supply to see if I can kick it in the butt and save it. I have a Adafruit Adalogger (yep, that's it's name) and just ordered a INA226 and .002 shunt. Thanks for the work you did. I love this stuff.
     
  12. Vincenzo

    Joined:
    Aug 5, 2015
    Messages:
    14
    Likes Received:
    5
    No Prob DroneX. Note that the INA226 reads up to ~80mV across the shunt, so with a 0.002 Ohm shunt your full scale current is 0.08/0.002=40A. So you may want to order a few different shunts to match your expected max current. In my case with 0.015Ohm I can read up to ~5.3A, and that's ok with my 100W PSU. Check the Arduino INA226 library ;-)

    V.
     
  13. DroneX

    Joined:
    Nov 13, 2015
    Messages:
    13
    Likes Received:
    2
    Thanks V. I am working on a monitor for the solar system in my RV and expect the charge current could be greater them 20A. I will have to order a larger shunt for the Inspire battery test. Your .015 looks like the perfect size.