• State
  • Locked Locked
  • Replies 9 replies
  • Subscribers 65 subscribers
  • Views 1107 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

BQ34Z100EVM Calibration

Matt Waskett
Prodigy 50 points
Other Parts Discussed in Thread: BQ34Z100EVM, BQEVSW, BQ34Z100

Hi,

I'm using the BQ34Z100EVM with the EV2300 evaluation module to develop a battery monitoring system for a 12V, 40AH LiFePO4 battery. The BQ34Z100EVM board is set up for Multi Cell operation, with external LED's (the 10 LED's on the board), and a maximum input voltage of 16V. I am using an external current sense resistor setup measuring 0.0045 Ohms.

I am running into problems when I try to configure the data flash, mainly the Design Capacity, Design Energy and Qmax Cell 0 values, as it is limiting me to a value of 32767mAh/mWh. I read on the design notes for a different gauge (http://e2e.ti.com/support/power_management/battery_management/w/design_notes/bq27500-max-pack-capacity.aspx) that the limit for Coulomb counting is set to 32767, and would therefore mean that I need to mis-calibrate the sense resistor to factor this in. I have now set the Design Capacity and Qmax Cell 0 to 20,000mAh, and calibrated the EVM with half the current draw (it's actually drawing 6Amps, but I've entered 3A). Is this the correct thing to do when it was mentioned to 'mis-calibrate' the sense resistor?

Another thing I wanted to bring up was how do I run a learning cycle, and export the learnt values for the data flash constants in a way that it could be written to a custom board that we are designing using a PIC processor over I2C? Is this do-able with a DFI / ROM file, and how would I go about obtaining one for my battery system with the BQEVSW?

Thanks

  • 0
    TI__Guru 73830 points

    Matt,

    The device should be able to support 32k capacity, but there is a firmware issue that prevents the passed charge counter from counting above 16k. We are going to fix this in the next firmware update. You will need scale the current and capacity by 3 or by 4 so that the passed charge does not exceed 16384 mAh. The capacity and current reported by the gauge will need to be scaled up to report the actual value. If you were to use a 3x scale, then you would divide all of the capacity and current parameters in the data flash by 3. You will also have to scale the current during calibration. e.g., You would enter -1000mA in the Actual Current field, if you use a 3A discharge calibration current.

    You can use a dfi file with a uC. We have an applications note on the bq34z100 product folder that discusses how to extract a dfi file. It is the Going to Production applications report.

    Regards
    Tom

  • 0 in reply to ThomasCosby
    Intellectual 450 points

    any timeline on when this firmware update will be out?

  • 0 in reply to Matt Palmer
    TI__Guru 73830 points

    I do not have an updated release schedule, but I would not wait for the release to begin the evaluation.

  • 0 in reply to ThomasCosby
    Prodigy 50 points
    Hi,
    We are still trying to get the BQ34Z100 to recognise the learning cycle, but it doesn't seem to be happening. 
    After the battery has been fully charged, the flag OCVTAKEN in the Control_Status register is cleared upon entry to relaxation mode, and then set to 1 when the OCV measurement has been taken whilst in relaxation mode. We have been able to replicate this, so in theory, Update_status should have changed to 0x05H, letting us know that the charge/relaxation portion of the learning cycle had been completed, and was now ready for the discharge/relaxation period to complete the cycle. Unfortunately, it has remained at 0x04H for us, even though OCV measurements have been taken during relaxation, and full charge is detected (FC set). Is there anything else that needs to be changed in order for the BQ34Z100 to complete the first stage of the learning cycle?
    Upon discharging the battery, flags VOK and QEN are set, meaning that the voltages are ok for QMax updates, and that QMax updates are enabled. After discharge termination, the BQ34Z100 should update QMax, and the learning cycle should be complete. This is something we have been unable to replicate, as Update_status remains at 0x04H after the charging stage. Is there any other flag or condition that needs to be set for the battery to complete this stage of the cycle? 
    Thanks
  • 0 in reply to Matt Waskett
    TI__Guru 73830 points

    Matt,

    The device will set a DOD point after the OCVTAKEN flag sets and it will update Qmax, if there was a previous DOD point set and it meets the other QMax update criteria. If you discharge to empty and allow it to rest, then the Qmax should update after the OCVTAKEN flag sets during the rest period. Update Status should change to 05 at that point. You will then need to update the Ra table and it only updates after the discharge cycle. You will need to charge-rest-discharge-rest and the Update Status should change to 06 after that rest period. If you collected a log file, then you can send it to me and I can check to see if other factors may have caused it not to update.

    Tom

  • 0 in reply to Matt Waskett
    Prodigy 105 points

    Hi, I have same problem. Did you resolve this problem?

  • 0 in reply to krzysztof duma
    Prodigy 105 points

    Ok learning cycle works property. Very helpfull is SLUA597.

    But I have discovered some bug in bqEVSW 0.9.90. I can't write manualy CC Gain and CC Delta. Also I can't import .gg file becouse CC Gain and CC Delta become 0 (in gg file is correct 1.092 and 1.072). Only it works when CC Gain is 10. Why?

  • 0 in reply to krzysztof duma
    Prodigy 10 points

    I´ve got the same problem with the calibration data (using the same software version).

    If I "read all" values, than put "write all" (without changing any data) and than put again "read all", the values ("CC gain", "CC Offset" and "board offset") of the first read differ from the values of the second read!

  • 0 in reply to Simon Buhrow
    TI__Guru 73830 points

    Simon and Krzysztof,

    The bqEVSW program does restrict the minimum value of CC Delta to to 4.759, but you can set it to a smaller value using the gg file. I was not able to reproduce the issue that Simon mentioned. The CC Offset may change, because the device will run auto-cal for this parameter. Another problem that people have encountered is that you have to press the <return> key after each parameter is entered. Otherwise, it will not get saved to flash.

    Tom