• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 63 subscribers
  • Views 684 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ78PL116 cell balancing failure

Steve14663
Prodigy 170 points
Other Parts Discussed in Thread: BQ78PL116

Following production of 100AH  battery packs, we have found a number of latent 'features' that appear to be related to the cell balancing within the BQ78PL116 management chip. The cell balancing seems to terminate on some packs prematurely, often when the cell voltage differential increases beyond 400mV, thought his varies between packs. This condition often leads to a cell over voltage during charge, and always appears to occur on cell 2 or 3 of a four cell stack. At the time of the balancing termination,there are no active warnings or safety conditions. As far as the pack parameters are concerned:

  • The cell imbalance fail voltage is set to 1000mV and disabled (time set to 0).
  • The cell balancing algorithm is set tot the default open circuit balancing.
  • The cell chemistry used is LiFePO4.
  • Typical charge / discharge current: 6 / 15A

We have noticed that even when idle following the cessation of charging and the clearing of safety flags related to the over voltage condition, cell balancing does not resume.

Is there a reason that balancing would stop prematurely?

Is there a reason that cells 2 and 3 appear to always be the cells that are at a higher state of charge?

Which pumping algorithm performs best with the LiFePO4 cell chemistry?

Given the capacity of the cells, what pumping current is recommended?

  • 0
    TI__Guru 73830 points

    The device should not disable cell balancing after charge as long as the cells are out of balance. It will disable balancing on the low end after the cells reach a threshold governed by an equation in the TRM. There should be a reason for Cells 2 and 3 to be higher, although the PL116 internal regulator runs off of Cell 2. This may be loading Cell 2 and causing it to be balanced more. What is your balancing current set to? We have typically seen it set to 1A for high capacity packs. We also have been recommending using the Terminal algorithm for high capacity packs. It seems to work the best. You can send me log files and a pack parameter dump and I will review them.

  • 0 in reply to ThomasCosby
    Prodigy 170 points

    Hi Thomas,

    It appears to stop pumping at the end of charge, though we haven't fully tested at the discharged end of the cycle. unfortunately this came to our attention as the RSOC does not update correctly if the charge termination criteria are not met at the end of charge, whereas the  RSOC is not affected if the cells are out of balance at the end of discharge. I have included a log of an almost charged battery being charged then left connected to the charger. this is evident by the periodic current spikes that occur as the COV flag is cleared and the pack reattempts charging. Current scaling is applied for this pack and as such the current values logged are /10. As you can see, beyond line 77, there is no pumping despite the cells being at significantly different voltages. I have also attached our parameter set for verification. Below is the balancing circuitry in use, from my calculations the average inductor current is in the order of 120mA, this is well below the 1A recommendation. what component value changes would you recommend to achieve the 1A pumping current?

    Test Log Files.rar
  • 0 in reply to Steve14663
    TI__Guru 73830 points

    The problem could be that you have SuperPump mode enabled and the device is detecting current flow. I would try setting all of the Turbo mode bits to zero to disable the SuperPump. I would also set the Inhibit Safety Rules bit low, because I have found that disabling safety rules can disable some functions.

  • 0 in reply to ThomasCosby
    Prodigy 170 points

    Hi Thomas, 

    Sorry, I should have mentioned, I sent you the ppcsv file we use to load new packs, the Inhibit Safety Rules bit is always cleared following calibration. We have enabled super pump on these packs, but from my understanding that should not prevent the pack from pumping, only increase the rate at which it pumps during idle periods. I will however try disabling it and see if t makes any difference. The schematic capture with our pumping configuration obviously did not come through, so i have included it as an attachment.