BQ76952: Abnormal data will be read when using analog IIC to read data

Hello Annie,

How often does this abnormal data show up for the customer, is the system functioning most of the time? Have your customer calibrated the Stack, PACK and LD Voltages? Although cell voltages are pretty accurate without calibration, the latter may require it. 380-V does seem pretty high, so it may not be that, but it may be worth double checking.

Best Regards,

Luis Hernandez Salomon

Hi,

Thanks for your reply and I will check the follow up questions with you.

Luis H. S. said:

How often does this abnormal data show up for the customer, is the system functioning most of the time?

After the system closes the charge and discharge FET, some of the data may be abnormal for about 2 minutes. The data abnormality just occur occasionally.

Thanks and expecting your updates.

Best Regards,

Cherry

Hi Luis,

May I know is there any updates?

Thanks!

Best Regards,

Cherry

Hello Cherry,

Could your customer please share a logic analyzer picture of the communication line when an error in the data is detected? 

Also, could they provide their schematic so we may review it and make sure everything is okay? It is quite odd that they find abnormal data after the charge/discharge FETs close.

Best Regards,

Luis Hernandez Salomon

Hi Luis,

Thank you so much for your reply.

The following figure shows the waveform returned by reading the data with CRC mode, according to which the data is parsed in accordance with the data received by the MCU：

And still waiting for the log, if there is anything new, I will update to you asap.

Thanks.

Best Regards,

Cherry

Hello Cherry,

Looking forward to the log.

Could your customer upload the I2C data from a logic analyzer? It is very difficult to know what's going on from the oscilloscope wave.

Best Regards,

Luis Hernandez Salomon

Hi Luis,

Thanks for your patient,  I have updated your question to the customer, and the following are the schematics of  the IIC on the AFE side and the MCU side:

Looking forward to your reply. Thanks!

Best Regards,

Cherry

Hello Cherry,

Thank you for the schematic. I hope they are able to provide soon the logic analyzer picture of the data. 

I will review the schematic and respond on that by the end of the day tomorrow.

Best Regards,

Luis Hernandez Salomon

Hi Luis,

Thanks for your reply, and the customer has no logic analyzer device available for use now. Feel sorry.

Looking forward to you updates!

Best Regards,

Cherry

Hi Cherry,

I think we will at least need their data log to help debug this. I do not see an issue in the schematic. The scope capture is not very helpful unfortunately.

Best regards,

Matt

Hi Matt,

Thanks for your reply, and the following is the info the customer provided:

And the the data in the box is abnormal data.

Thanks!

Best Regards,

Cherry

Hi Cherry,

I agree this data looks very strange. Can you send a better image of the schematic? Maybe in PDF form?

Thanks,

Matt

Hi Matt,

Thanks for your review and the customer would like to know what else part of the schematic be needed. Because of the confidentiality principles for the customer company, the customer may cannot provide more info on public forum. Thanks for your comprehension!

Best Regards,

Cherry

Hi Matt,

May I know is there any updates?

Thanks and Best Regards,

Cherry

Hello Cherry,

Can you upload a higher resolution image of the schematic you sent before? It doesn't need to have anything confidential, just the schematic of the BQ76952. Preferably in PDF format.

A couple of more questions:

1. Are they activating cell balancing during operation?
2. What are their register settings? Could you upload those? So we can make sure everything is okay.

Best Regards,

Luis Hernandez Salomon

Luis H. S. said:

Can you upload a higher resolution image of the schematic you sent before?

I  am sorry for this, the customer has not upload this yet.

Luis H. S. said:

Are they activating cell balancing during operation?

No, the customer never activate cell balancing during operation.

Luis H. S. said:

What are their register settings? Could you upload those? So we can make sure everything is okay.

The following is the configuration and the other unconfigured is the default parameters:

Thanks and Best Regards,

Cherry

Hi Luis,

May I know is there any updates?

Thanks and Best Regards,

Cherry

Hello Cherry, 

I will check these settings and respond by tomorrow at the latest.

Best Regards,

Luis Hernandez Salomon

Hi Cherry, 

One thing I noticed in the settings is the LOOP_SLOW bits in the Power Config register are set to the slowest setting. I wonder if it is possible that they are reading registers before the measurement has occurred. With their settings, the measurement loop is around 500 ms. Can they try to set the Power Config register to 0x0000 to see if the problem remains?

Thanks,

Matt

Hi Matt,

Matt Sunna said:

Can they try to set the Power Config register to 0x0000 to see if the problem remains?

The customer has set the Power Config register to 0x0000, but the problem remains.

Thanks and Best Regards,

Cherry

Hi Cherry, 

I also ran an experiment with the settings you sent and I cannot observe this issue. I have not seen this behavior reported before by any users and the device has been released for more than 1 year with a very large number of users. There must be something different about their system that we are not understanding.

Can you send me their full schematic by a private message? Early in this thread you mentioned the issue only happens when the FETs are first closed.

Matt

Hi Matt,

Thanks for your reply and I have sent the email for the full schematic. I will forward to you the email from customer, please be noted there will be some Chinese, and I will translate the useful part for you.

Thanks and Best Regards,

Cherry