BQ76952: Lithium Ion Over Discharge

Hello Anil,

In SLEEP mode there will still be current consumption from the IC, which can drain the cells. It seems the device shutdown once it reached a voltage close to ~V_PORA (see Section 7.3 of datasheet), which is probably why you're seeing a ~0.3-V per cell.

I noticed you have cell-based SHUTDOWN disabled. You probably would want to set the cell-voltage based SHUTDOWN to a voltage where the cells are not damaged, usually a value above 1-V per cell. 

You could also set your device to go to DEEPSLEEP mode, for further decrease in power consumption. In this mode, only the LDOs are active.

Best Regards,

Luis Hernandez Salomon

Hello Luis,

The pack was actually in shutdown mode (the "rst_shut" pin was asserted low for few seconds with a button press) during storage.

Would the cell-based shutdown voltage still matter in that case?

Thanks,

Anil

Hello Anil,

Cell-based shutdown wouldn't be necessary in that case. 

Is there anything connected at the load terminals? The current draw in SHUTDOWN should be low enough to avoid the cells draining that much in storage.

To avoid the device from unintentionally WAKE from SHUTDOWN, I would recommend to adjust the Auto Shutdown Time (See Section 13.4.1.8 Power:Shutdown:Auto Shutdown Time in the TRM).

Are you able to detect any leakage currents?

Best Regards,

Luis Hernandez Salomon

Hi Luis,

There was no load connected to the terminals. I will set the auto shutdown time, drain the pack to a low voltage until it goes into shutdown mode, and then let it sit for several weeks and monitor the voltages.

I will update you with the progress.

Regards,

Anil

Hello Anil,

I hope this is able to solve the issue. If it does not let us know. Check to see if there are any current leaks in the board.

Best Regards,

Luis Hernandez Salomon

Hello Luis,

How do you propose checking for current leaks? It does not detect any discharge current when no load is connected.

See attached file. Those are the latest config. Stack shutdown voltage is set to 35 VDC and cell shutdown at 2.4VDC. If I try to put the BQ76952 into config update mode, it shuts down after I load the config file and does a write because it sees that the stack (powered by 12VDC during OTP) and cells are under those thresholds. I am not able to do a OTP for this reason.

Thanks,

Anil

BQ76952_Config_17.gg.csv

Hello Anil,

You could check the current flowing through the resistors in the board to see if there's any flow. It may be a bit difficult to figure out if there's any leakage.

The stack/cell shutdown voltage seems too high. As I mentioned before, we usually recommend a voltage higher than ~1-V. With auto-shutdown, these shouldn't affect much however.

In CONFIG_UPDATE mode, settings should not take place until you exit config mode. Make sure you are in FULLACCESS mode and follow the steps found in BQ769x2 Calibration and OTP Programming Guide. With these steps you probably would still run with some issues as you are asked to exit config mode in the middle of the steps. Once exited, all parameters would be active, which would make the device transition to SHUTDOWN.

If you want to avoid SHUTDOWN after OTP, you can update the SHUTDOWN stack Voltage to 6V (or 0V to disable it temporarily) before exiting from CONFIG_UPDATE mode. This way you are able to still use the device. Once you connect the cells, the device would reset to OTP settings.

Best Regards,

Luis Hernandez Salomon

Hi Luis,

In CONFIG_UPDATE mode, the settings do take effect and cause the BQ76952 to shutdown before I can even exit CONFIG_UPDATE mode. I enter FULLACCESS mode followed by CONFIG_UPDATE mode, then configure the registers ("write_all") and it immediately shuts down without me being able to exit CONFIG_UPDATE.

Am I doing something wrong?

Regards,

Anil

Hello Anil,

That should not happen. There must be something wrong when you try to enter CONFIG_UPDATE. Are you entering CONFIG_UPDATE mode as described in the BQ769x2 Software Development Guide? If you are using CRC, did you send the correct CRC?

Best Regards,

Luis Hernandez Salomon

Hello Luis,

I am pretty sure it is entering CONFIG_UPDATE mode because I can see that the registers are configured correctly. I am using the BQStudio Battery Management App with an EV2400 I2C interface to do this.

See attached PDF.

Thanks,

AnilBQStudio Guide.pdf

Hello Anil, 

That is very odd. Everything seems okay, so it should be working fine. I will look into this more and write a response when I find more about this behavior.

For now, I would recommend you lower the shutdown stack voltage to 6-V and completely avoid this from happening. The stack/cell voltage you have for shutdown are pretty high. The auto-shutdown should be enough to keep the part from waking-up unintentionally. 

Best Regards,

Luis Hernandez Salomon

Hi Luis,

I am using 18650 cells so I do not want them to ever drain as low as 1V. I prefer 2V/cell at the least to go into shutdown mode. I tried 1V cell shutdown voltage and 6V stack shutdown voltage but it still happens.

Thanks,

Anil

I had to set the cell shutdown voltage to 0V in order to write to the OTP.

Please let me know if you find a solution to this because it's urgent. I will move forward and test everything else for now until you get back.

Thanks,

Anil

Hello Anil,

This should not be happening. If you have a stack voltage of 12-V for OTP, the device shouldn't go into shutdown.

Do all setting changes trigger under CONFIG_MODE? Or are you experiencing this only with SHUTDOWN?

Could you share your .gg file so I may review it?

Best Regards,

Luis Hernandez Salomon

Hi Luis,

I am only experiencing it with the "shutdown cell voltage" and "shutdown stack voltage" settings. As a temporary solution I made the shutdown stack voltage 10VDC and the shutdown cell voltage 0 (disabled). I did the OTP successfully with those settings.

Attached is the .gg file.BQ76952_Config_18.gg.csv

Hi Luis,

We will try to use a PIC microcontroller. For now, is there any fix to do this this using the BQstudio? We are crunched for time so it would be faster if it's possible. 

Thank you,

Anil