BQ40Z80: BQ40Z80 learning issue

Dear Support.

I'm Sorry.
I forgot to upload the log.

Thank you in advance.

0247.BQ40Z80_log.zip

Hello,

Looking at the log file, ITStatus has VOK high for the entire log file, along with the RDIS not being set from the initial discharge. This could be causing Update Status to not change. Can you please confirm that the RESET below in the beginning of the cycle is being done for RDIS to be set?

Regards,

Anthony Baldino 

Thank you for replay.

I checked ITStatus in the log file again.

DSG Start  2024/1/15  10:40:39 -> IT Status:0x0814 : Vok=0, RDIS=1

DSG Rest 2024/1/15  17:40:13 -> IT Status:0x0811 : Vok=0, RDIS=0

CHG Start 2024/1/16  8:43:01 -> IT Status:0x0919 : Vok =1, RDIS=0

CHG Rest 2024/1/16  12:15:34 -> IT Status:0x0911:Vok=0, RDIS=0

It should be reset at the start of discharge.

Could you please check the file again?

Best regards.

Hello,

I apologize, I was referring to the wrong bit map for these registers. After looking at it again, you are correct regarding VOK and RDIS.

Looking further into the log, I believe the reason this is occurring is that the [REST] bit is not being set after the charge, determining if a valid OCV measurement is being made even though VOK and RDIS are being cleared.

I believe that when a valid OCV measurement is not being taken, one of these aspects is out of line:

Can you please also tell me how the ChemID was chosen for this learning cycle? Was the GPCCHEM tool used?

Regards,

Anthony Baldino

Thank you for replay.

I got ChemID reading SLVA725A.

I used BqStudio to log the data.

This is the uploaded data.

3718.GPCPackaged.zip

Am I doing it wrong?

Best regards.

Hi Shuichi,

Can you please send the report that the GPCCHEM tool produces so I can check the results?

Regards,

Anthony Baldino

Thank you for reply.

This is the results data.

7215.GPCPackaged-report.zip

Could you please check?

Best regards.

Hi Shuichi,

Thank you for sending, the ChemID chosen seems to be correct, seeing that ChemID 1892 has a DOD error of 1.21%.

Looking again at the .log file, I believe this could still be occurring due to the voltage after the last charge being unstable while it should be in relax. 

Regards,

Anthony Baldino

Hi Anthony,

Thank you for reply.

Do you mean that the voltage remains unstable two hours after the last charge?

This could be a battery specification.

Is there any way to extend the time to Rest?

Best regards.

Hi Shuichi,

Yes, however I have found something else that could be causing this issue.

When the initial discharge begins, the VOK bit is never set. This could be effecting the process of the learning cycle. Please ensure that when the initial discharge begins, the VOK bit is set:

Regards,

Anthony Baldino

Hi Anthony,

Thank you for reply.

I began the initial discharge with the following steps.

1. Commands : FET_EN

2. Commands : GAUGE_EN

 QEN=1, VOK=1

3. Commands : RESET

 VOK=0, RDIS=1

4. Discharge start

Do I need to wait until VOK=1 after RESET?

Best regards.

Hi Shuichi,

Yes, after the reset is done, please ensure that the charge and discharge FETs are still enabled from the FET_EN, since VOK is set when a charge or discharge is observed. Below has the process of the first discharge and when these bits are supposed to be set and cleared.

Regards,

Anthony Baldino

Hi Anthony,

Thank you for reply.

I read SLUA848 again.

In Figure 3, Learning Cycle Starts is after DSG+REST (5H).

I will try again to do the learning in this way which is different from the first description in 4.2.1.

Thank you very much.

Hi Shuichi,

Please send the .log and .gg file when this is completed so we can take a closer look.

Regards,

Anthony Baldino

Hi Anthony,

I tried to train them by referring to Figure 3, but it did not work.

Please let me check the learning cycle for the BQ40Z80.

And, please answer my questions.

1. Discharge the cells to empty and let them relax for at least 5 hours.

2. Enable Impedance Track (0x21), issue a reset command (0x41). LStatus() changes from 00 to 04

[Q1. If I perform RESET at this timing, VOK=0, RDIS=1. It is different from Figure 3, is that correct?]

3. Charge the cells to full ensuring that the [FC] bit gets set and let it relax for at least two hours.

[Q2. Does external balanced charging with a dedicated charger affect the learning?]

4. Discharge the cells to empty using the typical discharge rate of your application. 

[Q3. Is it must be between C/5(1000mAh -> 200mA) to C/10(1000mAh -> 100mA) rate?]

[Q4. RDIS=1 during discharge. Is this a problem?]

5. Let the cells relax for 5 hours.

[Q5. If REST=1, RDIS=0, VOK=1,Lstatus=0x05 how can I relearn?]

Please let me know if I am learning something incorrectly.

Thank you in advance for your cooperation.

Hi Shuichi,

Shuichi Hashimoto said:

1. Discharge the cells to empty and let them relax for at least 5 hours.

2. Enable Impedance Track (0x21), issue a reset command (0x41). LStatus() changes from 00 to 04

[Q1. If I perform RESET at this timing, VOK=0, RDIS=1. It is different from Figure 3, is that correct?]

Item #1 and #2 need to be done in the opposite order. Sending IT Enable then the reset needs to be completed before the discharge begins. At the beginning of discharge after this has been done, RDIS and VOK should both be set to 1. RDIS is set to 1 from the reset and VOK is set to 1 when either charge or discharge is observed.

Shuichi Hashimoto said:

3. Charge the cells to full ensuring that the [FC] bit gets set and let it relax for at least two hours.

[Q2. Does external balanced charging with a dedicated charger affect the learning?]

Item #3 is correct. Can you please go more in depth about how you have this external balancing set up?

Shuichi Hashimoto said:

4. Discharge the cells to empty using the typical discharge rate of your application. 

[Q3. Is it must be between C/5(1000mAh -> 200mA) to C/10(1000mAh -> 100mA) rate?]

[Q4. RDIS=1 during discharge. Is this a problem?]

Item #4 is correct. 

Q3: Correct.

Q4: Yes, this could cause problems. Please let me know if this issue is seen more than once.

Shuichi Hashimoto said:

5. Let the cells relax for 5 hours.

[Q5. If REST=1, RDIS=0, VOK=1,Lstatus=0x05 how can I relearn?]

Q5: VOK should be 0 at this point. Please let me know if this issue is seen again. To relearn, please reset each of the Cell # R_A Flag back to 0xFF55 in the data flash, along with resetting the LStatus back to 0x00.

If another learning cycle is attempted, please start a log to track the measurements. This will help us debug any issues if they occur.

Regards,

Anthony Baldino