BQ4050: Repetitive Overcharge Problem Appearing on a lot of gauges

Ahmed Gebril

Part Number: BQ4050
Other Parts Discussed in Thread: BQSTUDIO

Dear BMS Experts ,

We have been used BQ4050 for 2s Li-ion battery (18650 - 3350 mAH) as BMS gauge in our products .

The issue we are detecting a lot of gauges reporting Overcharge Flag (Safety Status 4194304 = Overcharge ) . This flag occurs randomly during charge that let the product stuck in charging suddenly .

We have no clear explanation on how this gauges get the flag while either conditions not met ( Remaining Capacity smaller than Full charge capacity ) and we tried OC threshold with 300 mAH / 600 mAH . This issue is not reproducible and can be appeared on any number of cycles 3,4,5 ,etc . So we didn't get to a root cause to this flag or how to take preventive action. This issue affecting our productivity in a critical manner .

I am attaching logs of failed products from our software in addition to our default gg.csv and .SREC file.

Any advice on how to resolve this issue or finding the root cause .

defaultdatamemoryfromscanner.gg.csv oldgaugedatamemory.rar

over 3 years ago

0 Wyatt Keller over 3 years ago

TI__Guru** 105760 points

Hello Ahmed,

I think these two threads may be the same: https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/1049774/bq4050-safety-status-isn-t-zero

Can you share a bqStudio log of this happening? It's shouldn't happen randomly throughout a cycle, but I would need to see the other registers to debug further.

Sincerely,

Wyatt Keller

0 Ahmed Gebril over 3 years ago in reply to Wyatt Keller

Prodigy 60 points

Hello Wyatt,

It appeared again today on two of our products . please check the following log , gg.csv , srec during the OC issue from BQ studio . Please advice what can be be done else to resolve . It seems that neither of two conditions has been met.

oc.gg.csv srecwithoc.rar

0 Ahmed Gebril over 3 years ago in reply to Wyatt Keller

Prodigy 60 points

Hello Wyat ,

Did you have a chance to check my reply below ?

It seems that gg.csv safety status is "0" while OC flag is high on the register log . What do you think ?

0 Wyatt Keller over 3 years ago in reply to Ahmed Gebril

TI__Guru** 105760 points

Hello Ahmed,

I don't think the log was attached, can you try attaching it again? I only see the sec in the .rar file.

From the dashboard the conditions don't look like they have been met but we need the log to confirm when exactly it gets set.

Sincerely,

Wyatt Keller

0 Ahmed Gebril over 3 years ago in reply to Wyatt Keller

Prodigy 60 points

I uploaded fild " oc.gg.csv " after the issue has been appeared and I deassembled the product getting the flag to connect bqstudio as I don't know which one will trip the OC . Do you find it?

0 Ahmed Gebril over 3 years ago in reply to Wyatt Keller

Prodigy 60 points

If you mean the continous log from bq every 30min I have no mean to expect unit that this issue will appear and then solder I2c pins because after recovery It may pass tens of cycles without any flags and I still don't know how to recreate the issue to log before and after.

0 Wyatt Keller over 3 years ago in reply to Ahmed Gebril

TI__Guru** 105760 points

Hello Ahmed,

I reviewed your .gg file further, it looks like you have the NR bit set to 0, this means you have to transition the PRES pin to clear OC. So the alert could've happened many cycles ago when the remcap was above FCC. You can also increase the threshold to reduce the possibility of variations in charging causing an OC event unintentially.

Sincerely,

Wyatt Keller

0 Ahmed Gebril over 3 years ago in reply to Wyatt Keller

Prodigy 60 points

Hello Wyat ,

-According to my understanding that both conditions should be met simultenasouly isn't it?

How the second condition was met at this high SOC ?

Which value of OC threshold do you think is possible and safe to limit such flag as present is not available for MCU in this batch ? It is 600 mAH can we change it to 1200 mAH ? Or higher?

0 Ahmed Gebril over 3 years ago in reply to Wyatt Keller

Prodigy 60 points

Hello Wyat ,

Related to this issue , I have questions from my collogues working on it

-Is there any correlation between reducing charging current and limited OC trip ? Because when we tried to decrease current the issue almost disappear?
- Is there any recommended equation to define OC threshold ? Is there recommended OC Threshold according to specific charging current?

0 Wyatt Keller over 3 years ago in reply to Ahmed Gebril

TI__Guru** 105760 points

Hello Ahmed,

The OC will be triggered when the OC threshold is met and surpassed. OC will stay set until one of the recovery options are met.

The OC threshold depends on your system for what would be considered a safety issue.

The charge current should not matter for the OC protection. Can you share a log cycle of your tests? This is the fastest way for us to help debug.

Sincerely,

Wyatt Keller

0 Ahmed Gebril over 3 years ago in reply to Wyatt Keller

Prodigy 60 points

Hello Wyatt,

Would you please check the attached log for the unit that have detected OC last week I logged DSG->CHG->DSG but OC haven't be detected through the cycle?

Kindly I am waiting you advice whether if this log have any explanation to this repeated flag .

LogCycle_DIS_CHG_DSG_2.xlsx

0 Ahmed Gebril over 3 years ago in reply to Ahmed Gebril

Prodigy 60 points

BTW , SafetyAlert[OC]=1 @ end of charge but when I remove the charger it cleared again ? so how it is possible that alert may be happened several cycles ago as you suggested in our case ?

0 Wyatt Keller over 3 years ago in reply to Ahmed Gebril

TI__Guru** 105760 points

Hello Ahmed,

The SafetyAlert does not mean the OC is triggered, it means it is in the process of accumulating charge for the OC:Threshold. If the accumulated charge past FCC = RemCap is over the threshold then the SafetyStatus is set and would need to be cleared as I mentioned.

Sincerely,

Wyatt Keller

Power management

Power management forum

BQ4050: Repetitive Overcharge Problem Appearing on a lot of gauges