• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 1454 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.

BQ76930EVM / bqStudio communication problem

Michal Radziwon0
Prodigy 145 points
Other Parts Discussed in Thread: BQ76930, BQSTUDIO

Dear fellows,


I recently started to develop a BMS for 8 LPH cell system.  For start I decided to evaluate BQ76930 with TI evaluation kit and EV2300 interface module.

There is no problem, and everything works as described in manual, when I connect directly to device via I2C from "bq769x0 Evaluation Software". Problems start with bqStudio.

First, current target selector is different then one described in manual -  first it asks to choose an application as "gauge", "charger and gauge" or "charger". Whichever I will choose, the next window is similar to one described in manual, where I can select 1E9B_0_06-bq78350.bqz  (as described in manual "the latest one"). However, then I am getting a message"

The detected device is not compatible with selected application!
Some actions could cause the device to loose communication permanently and enter an unrecoverable state.
Proceed at your own risk....

If I will choose to continue, the communication will work for ~15sec, and then drop. The error seems to be caused by disappearing voltage from regout and C1 pins ~15 sec after releasing boot button. Which obviously turns companion controller off.

Could you help me to address this issue? I suspect that I am making some very simple error here.

(in case that may matter, I run whole toolchain on 32bit WindowsXP from virtual machine; I am using SMBus communication, I have set jumpers accordingly to manual, and there is nothing dangling on I2C line)

Best regards,

Mike

  • 0
    Prodigy 145 points
    I think I have spotted the problem.
    There were more batteries set in the system, then actually present - therefore reading on some was~0mV - this most likely (but not certainly ) was triggering detection of shortened battery and shutting down the system.
  • 0 in reply to Michal Radziwon0
    TI__Guru 54715 points

    bqStudio changes with updates/added devices, unfortunately the software change does not change the prior instruction manuals.  The extra screen and the warnings sound correct for the recent version.

    You seem to have found the most likely cause.  If the gauge firmware is loaded and the board is booted, the firmware will check cell voltages, if it can't find them it will turn off the AFE as you observed.  The method to avoid this to correct the configuration is to simulate a charge current as described at the end of the "Reducing the cell count" EVM user guide section.  A few mV on the SRN test point should be sufficient to keep the gauge awake to allow correction of the configuration.

  • 0 in reply to WM5295
    Prodigy 145 points
    Thanks a lot for your input!
    However, I still have some conceptual problem. With cell simulator turned on, the SOC was ~6-7% and was not changing with changing voltage - is it correct? Or simulated cell stack will change its SOC only when some current is drawn in-an out from system?

    As for current tests I have moved to with small Li-Poly cells of lowest count (6pcs for bq76930) and real, load / charge, I expect SOC to behave more predictable. Bot how/when health of cells is evaluated? Is this information I will get for each cell individually, or it's health of whole battery stack?
  • 0 in reply to Michal Radziwon0
    TI__Guru 54715 points

    This seems normal.  At reset the battery capacity is estimated from the voltage (see TRM section 8.1.5).  The default chemistry is for a 4.3V cell (chemistry 1210). Reset at ~ 3.6V is typically ~ 6%.   This gauge then counts coloumbs to change the SOC, so current flow will be required to change SOC.  To get a better estimate of the capacity at reset, use a chemistry which more closely matches your cells, one of the 04xx series chemistries should provide a better result.

    The state of health is a single number for the pack based on FCC and the design values (see 16.43).  FCC can be determined from gauging based on the low cell or the external average voltage (see section 8 and 8.3 Gauging Configuration Options) 

  • 0 in reply to WM5295
    Prodigy 145 points
    Thanks a lot for your input!