• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 241 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.

BQ76942: REG18 only 0.975 V

Dejan Hrovatin1
Intellectual 260 points
Part Number: BQ76942
Other Parts Discussed in Thread: BQ76952

Hello,

I have a problem with my 8 cell battery monitor circuit which was worked perfectly till some point.
Schematic is here:8cellBQ76942monitor.pdf

In compare to default settings I wrote to OTP some changes, but after that everything was working fine for more than 2 weeks, so I'm sure that this is not source of the problem. However, here is the list of settings I wrote to OTP::
REG12 Config :: 00 (default) -> 0D (new value)
REG0 Config :: 00 (default) -> 01 (new value)
I2C Address :: 00 (default) -> 60 (new value)
Alert Pin Config :: 00 (default) -> 29 (new value)
TS1 Config :: 07 (default) -> 0B (new value)

The problem is that device is not detected on I2C bus. Additionally there is no 3.3 V on the regulator's (REG1) output and I can measure that on REG18 pin is only 0.975 V instead of 1.8 V. I also tried to wake-up an IC with switch on TS2 but It doesn't help.

What are possible reasons for this issue?

Many thanks,
Kind regards,
Dejan.

  • 0
    TI__Expert 8461 points

    Hi Dejan,

    It is very unusual for the REG18 pin to have a voltage other than 1.8V or 0V.  Your schematic looks good, and the new config values should not have caused an issue.  In this thread ((+) BQ76952: I2C Communication Failure - Power management forum - Power management - TI E2E support forums), a customer reported a similar issue. 

    I recommend ordering a new IC.  If you can remember any abnormal behavior leading up to the failure, that would be helpful as well to potentially diagnose the issue.

    Best,

    Andria

  • 0 in reply to Andria McIntyre
    Intellectual 260 points

    Hello Andria,

    IC was replaced with a new one and everything looks okay for now.

    I have two ideas/questions...on schematic is shown that a voltage of 8S battery is measured and it looks like that 8S battery is in one package (balancer comnector has 9 pins). In reality I'm measuring a series connection of two 4S batteries, and I can not guarantee that balancer of "lower" 4 cells is connected firstly and than balancer of "higher" 4 cells. It could happen that also "higher" 4 cells are connected firstly and than lower 4 cells. Could be this a problem?

    Another idea is - I2C communication line is exposed to interferences/unwanted noise from electric motor which is not far away. Could such noise be a reason for such problem?

    Many thanks, Dejan.

  • 0 in reply to Dejan Hrovatin1
    TI__Expert 8461 points

    Hi Dejan,

    For the first question, I recommend taking a look at this FAQ, which walks through cell connection guidelines: (+) [FAQ] BQ76952: Connecting battery cells to the BQ769x2 family of monitors - Power management forum - Power management - TI E2E support forums.

    For the second, since you are using the REG1 to pull-up the i2c lines, I would make sure any noise present is not able violate the abs max specs of the REG1 pin (or other device pins).  I would recommend an isolator for the comm lines as well, but it looks like you are already using one. 

    If you are concerned about voltage transients, you can probe the i2c lines and the REG1 pin to see if you can record any voltage spikes that could be causing stress to the IC.

    Best,

    Andria

  • 0 in reply to Andria McIntyre
    Intellectual 260 points

    Hello Andria,

    As shown on schematic - from hardware point of view, I connected some VC inputs together (VC7, VC8 and VC9) what is needed when measuring voltages of 8 cell battery stack. However, I didn't make any changes in Settings:Configuration:Vcell mode register. Is this mandatory?

    Many thanks, Kind regards, Dejan.

     

  • 0 in reply to Dejan Hrovatin1
    TI__Expert 8461 points

    Hi Dejan,

    Yes, this is mandatory.  You shorted the cells correctly, as recommended, but you must reflect these changes in the Vcell mode register.  Please see Section 13.3.2.19 in the TRM (BQ76942 Technical Reference Manual (Rev. B) (ti.com))

    Best,

    Andria

  • 0 in reply to Andria McIntyre
    Intellectual 260 points

    Hello,

    so, according to schematic, value of register in my example should be b1001111111 or 0x27F. 

    However, I see that configure bits in this register to 1 also enables protection on coresponding cells. I don't like to enable any protection at all, I really would like just to measure voltages as they are. How can I prevent that no one protection is triggered if bits in this registers are configured to 1?

    Many thanks, Dejan.

  • 0 in reply to Dejan Hrovatin1
    TI__Expert 8461 points

    Hi Dejan,

    I recommend reading through the TRM to learn about different protections and configurations for cell voltage.  The TRM should walk through what the device is and is not able to achieve through register settings.  Protections can be edited/enabled/disabled.

    Setting VCELL mode to 0 to avoid cell protections, will signal to the device that all inputs are considered as cell connections.

    Best,

    Andria

  • 0 in reply to Andria McIntyre
    Intellectual 260 points

    Hello Andria,

    I read through the TRM and regarding cell protections (which are enabled by configuring Vcell mode register). I see there is a CUV protection (2.8 V by default) and COV protection (4.35 V by default). These default settings are appropriate for my application.
    However, I would like to understand one thing. If I don't change Vcell mode register in any way (I leave zeroes there), CUV faults can trigger for the shorted cell inputs. Is this all that can happen? I'm asking this because I'm not using any FETs on the board, because I don't need any protection, so for my application is not important if CUV is triggered.
    With BQ76942 I would like just to read voltage of cells and that the device doesn't go in shutdown mode for any reason, any other protections are not needed.

    Many many thanks,
    Dejan.

  • 0 in reply to Dejan Hrovatin1
    TI__Expert 8461 points

    Hi Dejan,

    Setting Vcell mode to all 0s will tell the BQ76952 that 10 cells are connected to the device, and protections are not wanted for any of the cells.  The issue with this is that if you want to perform cell balancing, the device will think that all 10 cells are connected and will attempt to balance all 10 cells.

    Additionally, if Vcell mode is set to 0s, but Enabled protections are manually changed by the user, then the protections specified by the user will take effect.  Since you don't have FETs or a fuse controlled by the BQ76952, faults triggering is not really a concern. 

    What you should be mindful of is SHUTDOWN mode.  I recommend going through the TRM to determine what conditions and register settings can cause the BQ76952 to enter SHUTDOWN mode and adjusting them accordingly. 

    For example, Section 13.4.1.1 Power:Shutdown:Shutdown Cell Voltage discusses how setting the Shutdown Cell Voltage register to a value >0 will cause the device to SHUTDOWN based on low cell voltage.

    If you'd like, you can configure the BQ76952 to output a signal on the ALERT pin to notify the MCU of any faults.  This way, even if FETs are not controlled by the device, the MCU can act quickly based on interrupts from the BQ76952 if a fault occurs.

    Best,

    Andria