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.

BQ78350-R1: Is it possible to compensate for different resistances between cells in a pack?

Part Number: BQ78350-R1
Other Parts Discussed in Thread: BQ78350, BQ76940,

Hi all,

We use a BQ78350-R1 in conjunction with BQ76940 AFE in a 14 cell LiFePO4 pack. On the other side this BQ78350 is in communication with our TMS320 MCU. The BMS balancing algorithm running on the BQ78350-R1 is a black box to us.

Due to the layout of cells in our application, they are bunched together in 2 packs of 7 cells, which in turn are separated by 2' of 10 gauge cable (approx resistance of this cable is 2mΩ). I am worried that because of the difference in resistances between cells, and because cell balancing does occur during charging (~10-20A in our application) that this could drive cells adjacent to the joining cable out of balance. Essentially, cell 8 in our pack will at 20A of charging current appear to be 2*20 = 40mV higher than it is in reality. And so I have three questions:

1: Can we set an upper limit on the charge current at which balancing will occur?

2: We have previous experience with an off-the-shelf BMS which allows the user to add what it's manufacturer refers to as "Busbar Compensation". Essentially, we could pre-configure the inter-cell resistances and subtract the product of this times the charge current to reveal the true cell voltage and accurately calculate each cell's internal resistance. Is a comparable feature available on the BQ78350-R1?

3: Am I being overly cautious here, is this small resistance worth being concerned about - will that 40mV be lost in the noise? We have not had the time to put a large number of cycles through our packs, I am more concerned about the longer-term effects of this.

Thanks guys!

Steve

  • Hi Stephen,

    1. The BQ78350-R1 does not have a maximum current for balancing.  The start is at the Cell Balance Threshold. Balancing may be a little odd with LiFePO4 since the voltage of cells rises quickly at the end of charge.

    2. The BQ78350-R1 does not have a busbar compensation feature.

    3. The BQ76940 is not a highly accurate voltage measure as you know from the data sheet, but you are likely to calibrate the offset in the BQ78350-R1 at a voltage point of interest.  With LiFePO4 cells the voltage profile is very flat over the SOC, so a 40 mV dynamic change will be very apparent.  Something which may lessen the concern is that the balance will continue through charge taper, so as the current drops the voltage from the resistance will go away.

    Since you are using 14 cells you may want to consider an option which is not on the data sheet: Use cell 8 of the BQ76940 as the unused cell rather than cell 14 and connect it across the busbar as the unused cell. It would require an extra wire to the pack for the other side of the bus bar.  The AFE cell map of the BQ78350-R1 would be set to show cell 8 unused.  You may want some diodes between VC7 and VC8 to avoid any voltages above abs max during connection.  During operation the diodes would not conduct at 40 mV or much higher (if discharge current is higher). I have not done this but cell 8 is one of the cells which is not used in reduced cell counts.  If you have packs built already it may not be an option since your board and harness designs are already fixed.

  • Thanks for a very thorough reply. Your suggested solution is clever and is definitely thinking outside the box, unfortunately we're at the stage in our design where even changing the 10 gauge cable to something thicker isn't trivial. I was hoping for a firmware fix.

    If I may, I would like to add a follow-up question. I understand that this product only allows balancing during charging. But does the BQ78350-R1 allow balancing with a charge current of 0A? Is there a check to ensure positive charge current, if so what is the minimum charge current allowed to balance? I hope that my thought process is quite clear.

    Failing that, I will resolve this issue by changing the design of the joining cable to something thicker, I think I can squeeze in 8AWG. I will also get our harness manufacturer to move the tap for cell 7 midway along the cable using something like https://www.te.com/usa-en/product-1811027-1.html (I need to find an exact solution appropriate for the cable's gauge). I don't actually need to eliminate the resistance between cells, I just need to ensure that the resistance between points at which cell taps are taken are as closely matched as possible (taking each cell as a direct short).

    Thanks again,

    Stephen