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.

BQ76920EVM: Unable to find Golden image/way to upload on actual system

Part Number: BQ76920EVM
Other Parts Discussed in Thread: EV2400, BQSTUDIO, BQ78350, BQ78350-R1A, BQ78350-R1, BQ76940, BQ76920

Hello,

I have been using the development board to see how the system works and setup  the "golden image" to our specification.   However, I do not entirely understand how I can save the file and upload it to the actual system?  We plan on using the EV2400 interface between our system and the software.

I have been through all the available datasheets but I am still confused as to actually getting started with a real system.  Any further reading suggestions are greatly appreciated.  We have already defined the specs we require, such as OCD proection etc.  We are trying to understand how these are actually transferred on the system.

Thank you,

Chris

  • Hi Chris,

    Have you taken a look at the EVM user's guide? It provides a detailed step-by-step process. If so, which step are you having trouble with?

  • Hello,

    Yes I should have stated that I did try the EVM user guide.  Firstly, I tried the AFE on its own, I was able to change the CHG_ON and DSG_ON bits and I could monitor my load resistor with a DVM to see the voltage across it.  The values I read were correct.  Then I setup the system to check the Gauge and I was able to reach up to step 17.

    For step 18, sending FET_EN doesn't seem to make a difference.  Initially, bit FEN_EN is high and bit DSG and CHG are low.  Sending the command FET_EN still keeps bit FET_EN high and DSG and CHG low.  I assume that that's the state my system should be in so I move to step 19.

    For step 19, I write in values for temperature but I get a red cross instead of a check mark.  Same thing happens when I input a value for the Ext Average Cell Voltage.  This is where I stopped and decided to try something else.

    I scrolled down to Section 5 and decided to update the firmware, in case something was off.  Importing the file and pressing Program, gives me an error that "The device must be in ROM mode".  Thus I tried searching for a way to actually update my firmware, even though the EVM user guide didn't state anything about the ROM mode.  I found another question on the forum which stated that, in order to enter ROM mode I had to send command 00, however the software informs me that that's not a valid word to write.

    Thus I wanted to understand why I couldn't use the Device Control.  I found application report SLUA742 and tried the example in section 2.  I was able to write 00 40 onto register 0x44, then read back from 0x44 the following:

    00 40 84 07 02 75 96 19
    A5 40 01 00 FF FF 00 00
    00 00 00 00 00 00 00 00
    00 00 00 00 00 40 00 00
    00 00

    This looked correct, in accordance to what is given, which makes me wonder why I couldn't enter ROM mode.  Then I made the post to understand if this was the correct approach for the final system.

    Thank you for your time,

    Chris

     

  • Hi Chris,

    After you click the FET_EN button in the Command Window, did you click the Refresh button? Can you check to see if any protection faults are being reported in the registers that may be preventing the FETs from turning on?

    If you are trying to update the firmware, you may want to try turning off the Dashboard (click on the green banner in the upper left corner of BQStudio).

    Thanks,

    Matt

  • Hello Matt,

    I restarted the whole setup.  FET_EN bit initially is low.  After pressing FET_EN, I refresh the list.  Bit FET_EN becomes white background momentarily before changing to high.  However, bit 6 of Battery Status (DSG) remains high throughout, bit 2 of Operation Status A (DSG) remains low throughout, bit 6 of Gauging status (DSG) remains high throughout and bit 1 of Operation Status A (CHG) remains low throughout.  None of them seem to be affected.

    I checked for errors AFER and AFEC bits stay low throughout and also no errors on the Error window within bqStudio.  Any other registers I should be checking?

    I also tried updating the firmware by turning off the dashboard but I still get the same ROM error and I'm still not able to enter ROM mode.  However, my bq78350_R1 version is 1e9b_1_03 which is the version shown in the user manual as well.

    I think the golden image consists of the Data memory configuration, Chemistry View and Calibration.  Assuming I am able to prepare the data memory config and the chemistry view using the bqStudio GUI, I am still running into problems with the Calibration tab,

    Thank you,

    Chris

  • Hi Chris,

    The version of firmware you have should be fine. This is the version that comes shipped on the BQ78350-R1 device. There is a newer version, but it only has one change which is an improved Shutdown routine. This would not affect anything you are seeing now since it is functionally the same otherwise. Version 1.04 firmware comes loaded on the BQ78350-R1A device.

    Can you send a screenshot of your BQStudio registers after refreshing (after FET_EN)? You should see the XCHG and XDSG bits turn red (off) and the CHG and DSG bits turn green (on) in OperationStatus. These bits have a different meaning in the BatteryStatus and GaugingStatus registers since these depend on the current flow as well.

    Also, for your calibration issues, make sure you have the device configured properly for the temperature measurements and external average cell voltage. By default, only TS1 is enabled. For external average, you need to make sure your AFE Cell Map is configured and that you are entering the average cell voltage value. More images will help us debug this.

    Best regards,

    Matt

  • Good afternoon,

    Started fresh again today with images. Here is my bqStudio before I press the FET_ON command:

    As you see, the XCHG and XDSG are already high and CHG and DSG are low. Bits FET_EN and SEC1 are also high. Then I press the FET_ON command and get this view:

    XCHG, XDSG, CHG and DSG stay at their previous state. The only bits which change are FET_EN which goes low and the SEC0 bit which goes low again. I was also monitoring the voltage output from the PACK+/- pins and, for 14.96V on the BATT+/- (from a power supply), I got a reading of 11.54V on the DVM. When I put a 7.5Ohm resistor across, I got 0V across it and the current consumption from the power supply did not increase (stayed at ~0.019A)

    I also retried the calibration procedure and here is what I got:

    The only one which gives me a checkmark is the CC offset. I also tried the current calibration using the 7.5Ohm resistor, but as there was no current flow through it the calibration failed as well. Then I decided to just check the AFE, using just the bq76940 evaluation software.  I tried the four different combinations of DSG/CHG LOW/HIGH, while having 14.96V on BATT+/- and monitor PACK+/- with a DVM and got the following:

    CHG LOW CHG HIGH
    DSG LOW 11.54 V 1.403 V
    DSG HIGH 14.66 V 14.96 V

    Which confirms that the DSG and CHG stay low either way.  Let me know your thoughts, I do hope I am doing something wrong!

    Thank you,

    Chris

  • Hi Chris,

    Thanks for sending pictures - this helps reveal some possible sources of the problem you are seeing. I notice a couple of things. The PF_EN bit and FET_EN bits are high at power-up which means that these bits are set in the Mfg Status Init data flash register. If the FET_EN bit is set in this data flash register, it will enable the FETs by default. So you should not need to execute the FET_EN command (clicking this button at this point should turn the FETs off).

    In additional to the PF_EN bit being set, I notice the PF bit and the SOV and SUV bits are set. That means that a permanent failure has been triggered - both an AFE over-voltage and an AFE-undervoltage. Permanent failures will disable both FETs. You can clear the permanent by clicking the PF_CLEAR button in the command window. If the failures are still observed, it likely means that the AFE Cell Map data flash register is not set up correctly. Maybe the correct voltages were not present at some point which triggered these PFs? It might be best to disable permanent fails until everything else is set up and working correctly.

    For calibration, I assume you are using the BQ76920 EVM? Let me know if anything changes once permanent failures are disabled.

    Best regards,

    Matt

  • Hello Matt,

    Great news, sending the PF_CLEAR command works and I can see the FET bits change.  I was also able to run full calibration on the system., thank you for your help!  I assume that I will not run into these problems after I set the proper OV, UV etc. parameters.

    Am I correct in understanding that if I connect our custom PCB to the EV2400 using SMBus connector, upload the Data memory configuration, Chemistry View and run a Calibration through the bqStudio, then I will have set up the "Golden image" onto the BMS?  I am wondering if this is the correct approach for production. 

    Thank you,

    Chris

  • Hi Chris,

    Yes, this is correct. You may want to write some of the bits in the Mfg Status Init register as a last step (like PF_EN, BBR_EN, LF_EN). 

    In the Data Memory, there are also some parameters for the CEDV Gauging that are unique for you battery and are important for accurate gauging (if you are using comensated EDV). The steps for optimizing the CEDV parameters are described in the Using the BQ78350-R1 app note: http://www.ti.com/lit/pdf/SLUA924

    I also recommend using v1.04 firmware for production if your application uses the Shutdown routine. This firmware comes loaded on the BQ78350-R1A, or you can load it onto a BQ78350-R1 device (which comes loaded with v1.03 firmware). The only difference is improved robustness in the Shutdown routine.

    Best regards,

    Matt