• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 63 subscribers
  • Views 1153 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.

BQ34Z100EVM: cannot calibrate my bq34z100g1/evm

Mark Watson
Intellectual 530 points
Part Number: BQ34Z100EVM
Other Parts Discussed in Thread: BQSTUDIO, , BQ34Z100-G1, BQ34110

Hello,

I am running bqStudio (version 1.3.54) on a 64-bit Windows 7 PC. I have the PC connected to an EV2300 (version 3.1r) which is connected through I2C to a bq34z100evm (firmware 0100_0_16). I am trying to get the gauge to read an accurate voltage as I have the BAT+/BAT- pins connected to the +/- terminals of a 12 V lead acid battery. 

I configured the device to read voltages above 5 V by setting the appropriate jumpers on J5 and J2; I set the VOLTSEL bit in the Pack Cfg register; I entered the Voltage Divider parameter to 13000 mV (the max expected voltage), and I changed the Number of Series Cells to 6. Everything else is at the default settings.

I have 12.4 V applied across BAT+/BAT- but the gauge is reading 8.4 V. I am trying to calibrate the device but when I enter 12440 mV into the Applied Voltage box I get the error "calibration cell voltage is out of range. This value is in millivolts. Please check value and retry."  How do I fix this?

If it helps, the Control Status Register is 0x0014, and the Pack Cfg register is 0x49D9. Flags is 0x1116, and Flags B is 0x2400 (see pic below). Note that the CALEN bit in the Control Status register is cleared: am I supposed to be in some kind of calibration mode? The bq34z100g1 data sheet mentions a calibration mode but there are no clear instructions in neither this data sheet nor the EVM user manual. Can somebody help?

  • 0
    TI__Guru 73830 points

    Mark,

    Did you place the .bqz file that I sent earlier into your bqStudio directory? There was a issue with a older .bqz. You could also try updating to the test version of bqStudio. The comment on the bqStudio product folder says that it is v42, but v59 gets downloaded.

    Try using this sequence and do not change the Voltage Divider parameter manually.

    •    Change the VOLTSEL bit.
    •    Change the number is series cells to 6.
    •    Calibrate the voltage.

    The register settings that you show look normal.

    Tom

  • 0 in reply to ThomasCosby
    Intellectual 530 points

    Hello Tom,

    Thanks again for your reply. Yes, I placed 2134.0100_0_16-bq34z100G1.bqz in the folder C:\ti\BatteryManagementStudio\config, as you instructed yesterday. I did not delete the older .bqz file and I did not change the name of the new one: should I do either/both of these two actions?

    Why shouldn't I enter the Voltage Divider manually? The user guide/data sheet says to change it manually and then it will be updated to a different value after the calibration step.

    About updating to the test version of bqStudio: why would I need to do this? Are you not recommending the use of bqStudio version 1.3.54? I am cycling some lead acid batteries in a lab for a client for the purposes of comparing the output of the bq34z100evm against some highly sensitive lab equipment we have here because our client is considering using the bq34z100evm in a field application, so should I be using the test version for his type of application? If I do use the test version, will I have any of the driver issues I experienced while getting this version of bqStudio up and running on my 64-bit Windows 7 machine? Will there be any issues with old files for bqStudio left behind on my computer (because I know the uninstall application does not actually delete everything)? Do I even need to uninstall the current version of bqStudio?

    Thanks Tom.

    Mark

  • 0 in reply to Mark Watson
    TI__Guru 73830 points
    Mark,
    There should only be one 0100_0_16-bq34z100G1.bqz in the directory. You should delete the original one or copy over it. I do not recall whether bqStudio v54 had issues with the bq34z100-G1. I do not currently have that version installed, so I cannot test it. The 64-bit drivers are associated with the EV2300, so upgrading bqStudio should not cause a problem.

    Changing the Voltage Divider parameter is not necessary, if you calibrate the device without IT enabled.

    if you are working on a new design, then you may want to consider the bq34110 device that we released late last year. It uses the CEDV gauging algorithm. IT will also work, but it does have a tendency to upgrade the capacity estimate during the rest period. The issue is that PbA cells rest to a much higher voltage after discharge and the gauge runs periodic capacity simulations during rest and can upgrade the capacity based on the OCV. The CEDV gauging algorithm does not have this issue.

    Tom
  • 0 in reply to ThomasCosby
    Intellectual 530 points
    Hello Tom,

    OK, I am now able to calibrate the IC after replacing the .bqz file. Thank you very much.

    I will mention your comment (about switching to the bq34110) to our client (we are a research university and are not involved in the actual production design process).

    Do you still recommend using the test version of bqStudio?

    Thanks again.
  • 0 in reply to Mark Watson
    TI__Guru 73830 points
    You do not need to update bqStudio, if it is working okay.
  • 0 in reply to ThomasCosby
    Intellectual 530 points

    Tom,

    What are the differences between the CEDV algorithm used by the bq34110 and the IT algorithm used by the bq34z100g1?

    After a quick look at the technical guide for the bq34110 is appears the 110 is just doing simple Coulomb counting (except TI appears to call it "Accumulated Charge calculation").

    The 100g1 has a sophisticated algorithm that includes correlating OCV and SOC from a table in data memory (that depends on battery chemistry) to the measured voltage during charge/discharge in order to calculate how the internal resistance of the battery changes as the battery ages. Furthermore, the 100g1 uses SOC values with Coulomb counting to update the Qmax.

    Also, where can I find more technical info on the CEDV algorithm?