BQ20Z655-R1: GPC Error

Hi Andy,

I've tried learning cycle so many times with NCR18650F cells(4S/6P). I used its own Chem ID(0281). But I didn't get any progress to change Update Status.

It stays 0x04 and RDIS set. So i thought maybe Chem Id was wrong. I used GPC tool and I share my results and gg file here.

My learning cycle steps as follows;

1. Send default .senc file.
2. Edit data flash parameters. (Update status is 00)
3. Upload Chem Id which downloaded from e2e.
4. Do calibration. 
5. Send 0x46 to 0006 open mosfets.
6. Discharge until empty.
7. Wait 5 hours.
8. Send IT Enable command
9. Charge Pack until Charge Termination occurs.
10. Wait 2 hours.
11. But nothing happened here.(Update Status Still 0x04).
12. Run learning cycle untill update status become 0E
13. Then export data flash parameters as an gg file
14. Modify that gg file. (Update status 0E to 02, cycle count=>0, lifetime data to default)
15. Program that battery with default senc which downloaded from e2e to clear all datas.
16. Import that modified gg file to battery.Export rom file. This rom file will be my golden file.

NCR18650F.rar

4101.TamGPC_report.zip

Hi Mett,

Let me take a look at your files and see if there was something wrong with your learning cycle.

Andy

Hi Andy,

Thanks for your interest. I' waiting your answer and I hope there is a solution.

Hi Mett,

Do you have any log file that I can take a look at?  I don't see any log file in the files you shared.

Andy

Hi Andy,

Here is my log file for 4S/6P NCR18650F battery pack. My design capacity is 2750mAh*6= 16500mAh as mentioned in datasheet of the cell.

But when I perform capacity test I get capacity around 16800-17000mAh. My learning cycle is correct regarding to sheets of TI . I've done so many learning cycle with other type of 18650 cells. But this one is not updating Qmax and Ra table. I start learning cycle with VOK bit and QEN set. After charge termination or fully discharged RDIS become set. Sometimes it doesn't but still there is no change in Update Status.  

I've tried Chem ID's such as

• 0238 NCR18650F(2700mAh)
• 0281 NCR18650F(2750mAh)
• 0285 NCR18650F(5500mAh)
• 2088 NCA596080SA

Possible errors?

1. Battery capacity >16800mAh but in my data flash the value is 16500mAh. (There is minimum 300mAh difference) ,
2. Chem ID is not suitable for my application ,
3. Data Flash is incorrect ,
4. Cycle currents are too low for detecting internal resistance,
5. My learning cycle is wrong,

I'm waiting your urgent solution.

Thank You.

gpcNCR18650F.log

Hi Met,

I see that only a few items are logged in your log file. It is more like a IVT file and therefore it is a bit difficult to find out the cause. 

Do you have a log file that basically logs everything? We want to see more info on the gauge side.

Andy

Hi Andy,

I'm waiting 5 hours of relaxation .After that I'm going to send my log file which consist all data from the bqStudio.

Thanks.

Hi Andy, 

Here is my log file of Panasonic NCR18650F type cell. I used Chem ID 2076 because GPC Tool recommended. 

Log file profile is C/2 Charge, Relax more than 2 hours, C/5 Discharge, relax more than 5 hours. 

I need urgent solution. 

And do you have any online chat/ support service that we can solve problems much faster ? 

Thanks.NCR18650F.zip

Hi Andy,

I'm waiting your reply.

Thank you.

Hi Mett,

Sorry for my late response.

I'm looking into your log file now. Will provide my feedback ASAP.

Andy

Hi Andy,

 I hope there is a solution. Waiting your suggestions.

Regards.

Hi Andy,

I’ve changed design capacity 17400 (max cap of cell)  because when I discharge the battery I can calculate more than 17000 mah capacity. Have you calculated capacity from my log file that goes below 90%?

Is that any problem at RDIS,VOK flags in my log file?

If there is no change at Update status after charge should I still continue discharging? or send default .senc and start all over?

Does Qmax update during discharge?

Please answer each question it will be very helpfull .

Thanks.

Hi Andy,

I've realized that I did big mistake on schematic. I was trying to perform learning cycle for weeks. Finally I can achive update status 0E. 

My mistake was at SRP-SRN pins of the IC. They are connected wrong side of sens resistor as shown in the picture. Therefore when I discharge the battery in calibration mode it shows positive current reading. I just thought that is normal and miscalibrated, I continued to make calibration. After current calibration I can read correct measurement. 

But coulomb counting circuitry in gauge reads current the exact opposite I guess. I've made some modifications on BMS circuit and I tied SPR/SRN pins correctly. After that implementation I could achive correct learning cycle.

I have problem with over 200 Bms PCBs which waiting to deliver customer. PCB's produced and components assembled.

My questions are;

Can I use those BMS's with my Golden File(.rom) and good calibration?

What can go wrong after years of usage?

Maybe Ra table updates or Qmax disabled?

Capacity estimation can be wrong?

Thanks for your help.  

Hi Andy,

Thanks for your helpful reply. But I didn’t mean that I’ve corrected connections of 200 pieces of pcb. They are already assambled and coated with conformal. I cannot change Srp/Srn pins. I’ve just modified tracks of two pcb’s for learning purposes.

My question is that Can I use that Pcb’s with wrong connection?

Just doing calibration and sending golden file. What can go wrong after usage?

Thanks.