BQ25672: charging issue

Cherry Zhou

Part Number: BQ25672

Hi team,

Here's an issue from the customer may need your help:

When the battery goes into battery protection, output is 0V. At this time connecting the charger, the 25672 gives the battery more than 2V, but cannot activate the battery and never charge.

The customer would like to know how to set the parameters to activate the battery?

Could you help check this case? Thanks.

Best Regards,

Cherry

over 3 years ago

0 Jeff F over 3 years ago

TI__Guru**** 166996 points

Hi Cherry,

can you reply with IC top marking?

Regards,

Jeff

0 bao huang over 3 years ago in reply to Jeff F

Prodigy 160 points

Hi Jeff,

It is BQ 25672 14A9028. Not all of them, parts of them can not charge the depleted battery. According to the specification of the battery, the release voltage for the protection board of the battery is 6.1V, it is two cells.

0 bao huang over 3 years ago in reply to bao huang

Prodigy 160 points

We tested the Vbat from BQ25672, it is about 1.9V form the devices which could not charge batteries, while it is about 8.2V for the good ones.

0 Jeff F over 3 years ago in reply to bao huang

TI__Guru**** 166996 points

Hi Bao,

based on the CELLS register bits=2S, The charger sets a 1.5s intermediate reg voltage after trickle charge (2.5V) of about 6.1V. If your customer changes CELLs to 3S, this increases that intermediate voltage. Does that work? If so make sure that the host software lowers SYSMIN and VBATREG register settings from the CELLs-3S defaults.

Regards,

Jeff

0 bao huang over 3 years ago in reply to Jeff F

Prodigy 160 points

Hi Jeff,

Thanks for you quick reply. It is always 2S, the battery is inside our devices. The problem is we found about 20% of the devices could not charge if the batteries are all depleted. The battery voltage always reads about 1.9V, and never change even i charge it for a long time, looks like the PCM of the battery is not released. We measured the Vbat from BQ25672 without batteries, it is about 1.9V form the BQ25672 which could not charge, while it is about 8.2V for the good ones which could charge depleted batteries.

Best Regards

0 bao huang over 3 years ago in reply to bao huang

Prodigy 160 points

And the good devices always could charge all of the depleted batteries.

0 Jeff F over 3 years ago in reply to bao huang

TI__Guru**** 166996 points

Hi Bao,

I have confirmed that the ICs from your lot used an incorrect test program during production. The ICs that currently function correctly will continue to function correctly but those that are failing to charge will not. I recommend starting an RMA for those failing units.

Regards,

Jeff

0 bao huang over 3 years ago in reply to Jeff F

Prodigy 160 points

Hi Jeff,

Thanks.

0 Jeff F over 3 years ago in reply to bao huang

TI__Guru**** 166996 points

Bao,

The PFM mode ripple is a trigger for this issue. Some customers have been able to work around this issue by disabling PFM.

Regards,

Jeff

0 bao huang over 3 years ago in reply to Jeff F

Prodigy 160 points

Hi Jeff,

The default value for PFM_FWD_DIS is 0. We tested by changing PFM_FWD_DIS to 1 to disable the PFM, while still could not charge the depleted battery. Is there any other registers need to be changed accordingly?

0 Jeff F over 3 years ago in reply to bao huang

TI__Guru**** 166996 points

Hi Bao,

Ok. Then I recommend requesting replacement ICs through the vendor (or TI) from whom purchased the ICs.

Regards,

Jeff

0 bao huang over 3 years ago in reply to Jeff F

Prodigy 160 points

Hi Jeff，

Is there any other ways to solve this problem, we already made 1000 devices with this IC, it is pretty difficult to change the IC.

Best Regards

0 Jeff F over 3 years ago in reply to bao huang

TI__Guru**** 166996 points

Hi Bao,

The boards that function normally will continue to function normally. Those that cannot exit trickle charge will function normally as long as partially charged battery is attached.

Some customers were able to exit trickle charge by turning off PFM and increasing the cells bits (REG0x0Ab7:6) to the next configuration. Changing cells bits auto changes VBATREG, SYSMIN and ICHRG so the hosts needs to rewrite those down to prevent OVP.

Regards,

Jeff

0 Jeff F over 3 years ago in reply to Jeff F

TI__Guru**** 166996 points

Hi Bao,

I have found that on many devices with this issue, if I reset the SYSMIN (REG0x00b5:0) to 2.75V, the charger exits trickle charge. Can you try that? Then you will want to reset SYSMIN to your desired value once the protector closes.

Regards,

Jeff

0 bao huang over 3 years ago in reply to Jeff F

Prodigy 160 points

Hi Jeff,

I didn't found the register REG0x00b5:0, could you please show me where it is in the datasheet? Thanks.

Regards.

0 bao huang over 3 years ago in reply to bao huang

Prodigy 160 points

Do you mean the Vsystem? For 2s battery, i think the default value is 7V as shown below.

0 Jeff F over 3 years ago in reply to bao huang

TI__Guru**** 166996 points

Hi Bao,

Yes, that's it. There is an internal mismatch that gets worse with higher SYS voltage and with PFM switching noise. Setting SYSMIN=2.75V and disabling PFM mode until the protector closes has worked for some customers.

Regards,

Jeff

Power management

Power management forum

BQ25672: charging issue