BQ40Z80: Open balance connection behavior

Hello Daniel,

Daniel Librio said:

I'm wondering, what is the default behavior, should a single balance connection be broken during pack discharge, or a single cell fail high impedance?

-What type of load are you seeing with your cells? It seems like you are pushing really high current and causing the cell to burn up or a fuse to burn up 

Daniel Librio said:

In our application, it is critical that output power remains on.  We can tolerate a failed cell in terms of a reduced pack voltage, but not the pack output switching off .

-Assuming that the cells are wired in series, if one of the cell goes out that the either pack would be a open circuit. The eaiest solution could be to switch out your current with a fresh set of cell with more cells in parallel to the point you reach the max capacity of the battery gauge to better handle heavier loads.

Thank you,
Alan

Hi Alan, thank you for your reply.

Alan Frias said:

-What type of load are you seeing with your cells? It seems like you are pushing really high current and causing the cell to burn up or a fuse to burn up 

It's not so much that we are experiencing cell failures as a common occurrence, it is more a case of mitigating potential failures in the field, as it a critical application which must be shut down by the operator in a safe manner to protect the device.  Unscheduled power loss would almost certainly damage the device.  In development, we have only seen one failure due to a physically impacted cell, but it sparked a discussion.

Alan Frias said:

-Assuming that the cells are wired in series, if one of the cell goes out that the either pack would be a open circuit. The eaiest solution could be to switch out your current with a fresh set of cell with more cells in parallel to the point you reach the max capacity of the battery gauge to better handle heavier loads.

Once the device is safely shut down, we will definitely replace the entire pack.  But we need to maintain some form of power to the pack while the shut down is occurring.  In the case of the damaged cell I cited above, the damaged cells voltage dropped to nearly zero, but the overall pack was still supplying current to the load right up to the point the BMS shutdown, despite the failed cell being in series with the other cells.  If it had have been a basic pack without a BMS, we'd have had enough power available to safely shut down the device.  However, the BMS shutdown, turned off the CHG/DSG FETs, and cut all power to the device.
If there is a way we can configure the BQ40Z80 to not shutdown in this scenario, it might buy the operator enough time to safely power down the device.

Hello Daniel

Daniel Librio said:

It's not so much that we are experiencing cell failures as a common occurrence, it is more a case of mitigating potential failures in the field, as it a critical application which must be shut down by the operator in a safe manner to protect the device.  Unscheduled power loss would almost certainly damage the device.  In development, we have only seen one failure due to a physically impacted cell, but it sparked a discussion.

-Do you have a exported .log from bqstuido or able to recreate this to better see the cells interaction since we can view the temperature, individual cell voltage/current and etc during the event to better recommend a solution for event that don't happen to be due to physical damage.

Daniel Librio said:

If there is a way we can configure the BQ40Z80 to not shutdown in this scenario, it might buy the operator enough time to safely power down the device.

-This is possible but you would essentially would need to bypass all the protection and permanent failure threshold. This would still cause the gauge to work in the extreme situation but will open for the possibility of father damaging the cell to the point where it could lead to fire/explosion. If the cell gets physically impacted to the point it drops to 0V than this should warrant the fets opening but if you update the threshold so that it would still leave the fets closed than it would most likely lead to farther damaging the neighboring cells and to a thermal runaway. To summarize this it is possible but heavily not recommend due to safety concerns.

Thank you,
Alan