BQ34Z651: Clearing SafetyStatus Faults

David,
Safety faults normally clear with their recovery thresholds / conditions. If this does not work, then you can reset the device. The fault should clear unless it is still present in the system.
Tom

David,

Here is an answer to you other posting that got deleted.

QUESTION:

I have been looking at sluc138 to determine which chemistry ID to use for the cells that I am using for my battery pack. and I have a couple of questions.

LGC INR18650MJ1 cells are one cells I am using so sluc138 indicates that I should be using Chemistry ID 2059. Yet that ID is not available in the bqEVSW v0.9.87 that I am using, is there a way to update bqEVSW so that I can select that chemistry?

The battery packs that I am designing are either 2 cell (~8.4V) or 4 cell (~16.8V) (depending on the implementation) but in either case the battery pack itself consists of 8 physical 18650 cells, paralleling either 2 or 4 of them to make the ~4.2V 'cells' of my pack. Are the .chem files such as ID 0259 only applicable to a single physical cell or can I use these files even though the capacity of each of my cells is 2 or 4 times greater than the cell that the Chemistry ID refers to?

Answer:

You can update your chemistry database using the Chemistry Updater.

Use this option.

The online GPCCHEM tool is a more up to date tool that can be used to try to find a ChemID match.

Chemistry files are setup for 1S1P, but the gauge will use them for any cell configuration. You will select the number of series cells in one of the Configuration registers. The gauge treats parallel cells as one parallel cell and will update Qmax and the Ra tables with values that reflect actual capacity and resistances for the combined parallel cells. This is done during the optimization cycle.

Tom

Hi Tom,

I have tried:
Removing the load
Charging the battery
Disconnecting the Cells (power cycling the device)
Sending the reset command 0x0008
Toggling the MRST input
Left it in an idle state for ~12 hours

None of these have cleared the fault, do you have any other ideas?

Thanks,

David

The Reset command is 0x0041. The AOCD should clear with either the reset or by applying a charging current. Is the gauge reporting any current? If you cannot clear the fault, then there may be a defect on your board or the device may be damaged. You can try shorting across pins 36 and 37 and resetting the device. This should force the device to think that there is 0mA current across the sense resistor.

Hi Tom,

Thanks for answering my second question too, I will try that out today and see how I get on, I do also have a quick follow up question:

I see that the bqEVSW has an option to read the .chem file from the gauge. Does the gas gauge modify the chem file that was loaded onto it over time or will this read option just give me the file I loaded onto it in the past?

Thanks,

David

The bqEVSW tool does not export .chem files. You can export a .dfi or .rom file to program into other devices. These files contain the full data flash which does include the updated Ra table data.

Hi Tom,

The corrected RESET command worked perfectly, thanks for pointing that out.

So If I program a prototype battery pack with an appropriate chem file and run it through a series of full charge/discharge cycles and then to allow the Ra table to be updated to match the parallel cells.

When it comes time to commission a second battery pack would I still want to program it with the appropriate chem file or will programming the ROM and DFI files provide the gas gauge with all the relevant information it needs, updated for the parallel cells.

Thanks,

David

The .dfi and .rom files contain the full data flash, including the chemistry data and learned parameters. The difference in these two file formats is that the .rom file contains header information that the bqMtester tools use to prevent loading the .rom file into a different device type.

Thanks for that clarification. Does that then mean that in order to program a new pack, programming just the .rom file would suffice?