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.

BQ40Z80: Learning cycle for common cell and different parallel/series product

Part Number: BQ40Z80

Hi team,

I'm going to develop multiple products with the bq40z80.

Even if I use the same cell, if I change the number of series and parallel, do I have to repeat the learning cycle from the beginning each time?

Or is it possible to develop these products based on the same srec file with UpdateStatus=06?

For example, can I create a 5S3P golden image with the steps below?

1. Write the srec file that performed the learning cycle until UpdateStatus=06 on the 5S5P product to the 5S3P product.

2. Change parameters as 5S3P products.

3. Charg the 5S3P device to full, relax it for 2 hours and then discharg it to empty, (at which point it should be UpdateStatus=0E).

Best regards.

Masami

  • Hello Masami,

    If you have done a learning cycle with the same number of parallel count but different series you should be able to re-use the learned values. If you change the number of parallel you are changing the resistance of the cell stages (from the gauges point of view it is lower) and you are increasing the Qmax, so you will not be able to reuse the same learning data, you would need to re-run the learning for the new pack.

    We also recommend doing the learning cycle on all new pack designs for best accuracy, there are some specific circumstances where you could re-use, but accuracy may be impacted.

    Sincerely,

    Wyatt Keller

  • Hello Wyatt,

    Thank you for the answer!

    I understood that I can re-use the learned value with the same number of parallel count, but I can't with different number of parallel count.

    Even if the Qmax, R_a 1~14 and xR_a 1~14 values are changed according to the parallel number ratio, is it still impossible to reuse them?

    I think the accuracy won't be the best, but please let me know just in case if I need to do so.

    Best regards,

    Masami

  • Hello Masami,

    You can for different series count if you use the Ra table from the learned cells in the newer cells Ra table, this is not ideal and it is still recommended to re-run learning but it is possible, for different number of parallel cells it is not recommended at all because everything will be different.

    You can attempt to do this but you should expect very high error in all gauging parameters since Qmax and the Ra table are critical for gauging. It all depends on how much error you are willing to accept in the estimations. I would not recommend this.

    Sincerely,

    Wyatt Keller

  • Hello Wyatt,

    Thank you so much, I learned a lot!

    Masami