• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 63 subscribers
  • Views 690 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.

BQ20Z655-R1: Initial accuracy of gas gauge

George Mortimer
Expert 1205 points
Part Number: BQ20Z655-R1
Other Parts Discussed in Thread: BQMTESTER

E2E community -

I am having an issue with a customer regarding initial (out-of-the-box) accuracy of the battery gas gauge.

I have created the Golden Image per the bqMTester user's guide, section 4, and have used it to program the pack using the Multi-Tester application. The customer has taken the pack, charged it to full, and then discharged it through their utilizing equipment.  They get about a 5% difference between the 0% SoC point and the designed minimum operating voltage on initial operation. Is there any way to improve this initial number, or does the pack need to be cycled with the production cells in order to improve accuracy?

One thing I have just discovered is a mention in one of your app notes that an inaccurate learning cycle can occur if the pack is allowed to discharge to CUV cutoff instead of stopping the discharge at the design minimum voltage.  My practice has been to charge to taper current cutoff (C/20), rest 2 hours, discharge to CUV, and allow to rest over night.  Would this be causing the inaccuracy?

Also, is the charge and discharge rate critical to learning?  I noted that the app note recommends charge at C/2 and discharge at C/5. Is this a hard-and-fast rule?  The pack I have designed has an overcurrent shutdown that is below C/2.  I have been both charging and discharging at a value between C/4 and C/5. In the initial test at the customer, they were discharging at approximately C/15.  Would this affect the gas gauge accuracy?

Any help would be appreciated.

George Mortimer

  • 0
    TI__Intellectual 1735 points

    •   Hi,
    • When doing the learning cycle, we recommend that to charge at C/2 and discharge between C/5 to C/10. We recommend to discharge at C/7.
    • Please make sure you have done calibration before learning cycle.

  • 0 in reply to answer
    Expert 1205 points

    Can you tell me how much the accuracy will be affected if you charge or discharge at other than the C/2 and C/7 limits?

    As I said in my initial post, I cannot charge at C/2 because the gas gauge programming has the overcurrent Tier 1 trip at approximately 0.4C (battery capacity 6700mAh, 1st level OC in charge 3000mA). Will charging at 0.4C or thereabouts affect the gauge initial accuracy?

    George

  • 0 in reply to George Mortimer
    TI__Intellectual 1735 points

    • You can charge at 0.4C and it has little influence on accuracy.
    • Discharge current should be limited between C/10 and C/5.

  • 0 in reply to answer
    Expert 1205 points

    OK, Thanks!

    One more question: Based on the SLUU238 User's Guide, the creation of the Engineering Perfect battery pack consists of a charge and discharge cycle (repeat a second time for better accuracy), ending up with a pack discharged to the Term Voltage value.  As I am going to be importing this .gg file into packs with the cells charged to 30% capacity per IATA regulations, would this be the correct condition for the gas gauge file, or would I be better off recharging the pack to 30% RSOC after the second discharge and rest cycle in order to have the BMS in the same state as the cells?