bq27742-g1: Large change in SOC during relax

Andrew Dueckman96

Prodigy 90 points

Part Number: BQ27742-G1
Other Parts Discussed in Thread: GPCRA0, GPCRB, BQSTUDIO

Hi,

Chem id 414 LIFEPO4 Battery

Ran learning cycle, no errors tested SOC accuracy 1-2% which is good for my application

But....

after 4-24 hours with just the gauge current draw (Relax)

Full Charge capacity, Filtered FCC, Unfiltered FCC changes from ~2500 to 1400

and the SOC jumps to 1 % even if the cell was at 85% charge

I have been looking into this for 2 weeks now and need some help

It almost appears like there is some kind of negative value calculated.

Thanks

over 7 years ago

0 Damian Lewis over 7 years ago

TI__Mastermind 34775 points

Hi Andrew,

Can you provide us a gg.csv file of your settings as well as gauge logs of current, voltage, temp, flags, FCC, Qmax, RemCap, SOC, etc. of the behavior occurring? We will try to get you some answers.

0 Andrew Dueckman96 over 7 years ago in reply to Damian Lewis

Prodigy 90 points

Hi Damian,

This file goes from 31 percent to 1 percent

Thanks

31 percent to 1.zip

0 Damian Lewis over 7 years ago in reply to Andrew Dueckman96

TI__Mastermind 34775 points

Hi Andrew,

I'm out of office, so give me a few days to review the files and get back to you.

0 Damian Lewis over 7 years ago in reply to Damian Lewis

TI__Mastermind 34775 points

Hi Andrew,

Looking at the log file, it appears that a cell temperature drop > 5C during relaxation what's triggering an IT simulation that results in the SOC/RemCap drop. Your DODatEOC Delta T is currently set to default 5C. You can try increasing value to see if it helps.

My recommendation is setting the correct value for T rise and T Time Constant. These correct values can be obtained by using our GPCRA0 Tool. The instructions are available at the website.

http://www.ti.com/tool/GPCRA0

0 Andrew Dueckman96 over 7 years ago in reply to Damian Lewis

Prodigy 90 points

Hi Damian,

I have set DODatEOC Delta T to 100 on several units and will check them again tomorrow

Thanks

0 Andrew Dueckman96 over 7 years ago in reply to Andrew Dueckman96

Prodigy 90 points

Hi Damian,

Units are still falling to 1% even with DODatEOC Delta T set to 100

I will proceed in getting the file ready for GPCRA0 tool.

Thanks

0 Onyx Ahiakwo over 7 years ago in reply to Andrew Dueckman96

TI__Guru* 76230 points

Hello Andrew,
After you make changes to Dataflash, you always need to issue a reset command (0x41) so the parameter gets updated in RAM so the effects of making that change can be seen.

100 is too large for your actual application should that happen to help solve the problem. The best approach would be to use the GPCRA0 tool.
thanks
Onyx

0 Andrew Dueckman96 over 7 years ago in reply to Onyx Ahiakwo

Prodigy 90 points

Hi,

I ran the GPCRA0 tool (files included) and am still finding that the SOC drops to 1%

Sysrate2-report.zip

I programmed 8 units with the new golden image file produced with the GPCRA0 files.

1) Each of the batteries was charged to a different charge level,

	START SOC	After 1+1/2 days at room temp	after 12 hours at -2C	after reset	aftre 1/2 day room temp	after 1/2 day at -14 C
1	73	73	70	73	73	1
2	17	17	16	17	17	14
3	100	100	99	99	99	99
4	17	17	1%	17	17	1
5	34	34	1%	34	34	1
6	99	99	99	99	99	99
7	99	99	99	99	99	99
8	34	34	28	34	34	1

Number 2 is interesting because it did not fail

This is starting to look like a major issue.

Any help would be appreciated.

Thanks

0 Onyx Ahiakwo over 7 years ago in reply to Andrew Dueckman96

TI__Guru* 76230 points

Hi Andrew,
Can you the GPCRb tool as well? Ps follow the instructions there to use the tol. After you program the obtained file, pls make sure to issue a reset command, run acharge discharge cycle before repeating your test above to see if it helps

thanks
Onyx

0 Andrew Dueckman96 over 7 years ago in reply to Onyx Ahiakwo

Prodigy 90 points

Hi Onyx,

I will run the GPCRb tool in a few days after I gather the data.

But I am not sure why a discharge cycle would help?
I have tried both with and without reset and the results are the same.
I have also tried IT enable
Thanks

0 Onyx Ahiakwo over 7 years ago in reply to Andrew Dueckman96

TI__Guru* 76230 points

The GPCRb calculates new Ra values. you need to perform a discharge so as to make sure the Ra values are updated to the actual values. Updates to ra tables only occurs during discharge.
thanks
Onyx

0 Andrew Dueckman96 over 7 years ago in reply to Andrew Dueckman96

Prodigy 90 points

Hi Onyx,

I ran the GPCRb tool,

I am concerned with the value of T Rise calculated as it is negative.

Any thoughts?

GPCRB1-report.zip

Thanks

0 Onyx Ahiakwo over 7 years ago in reply to Andrew Dueckman96

TI__Guru* 76230 points

Hi Andrew,
Yes a negative Trise doesn't seem correct. It is an unsigned integer . Did you follow the instructions in the user guide on carry out the test, i.e ensuring that the thermistor is measuring the battery temperature by placing and attaching the thermistor to the cell so it is actually measuring the heating and cooling of the cell? If no, then i am afraid you will have to re-perform the test

thanks
Onyx

0 Andrew Dueckman96 over 7 years ago in reply to Onyx Ahiakwo

Prodigy 90 points

Hi Onyx,

I re-did the test and did not get a negative number as I noticed that my low temp file was incomplete.

gpcrb2-report.zip

But even after cycling the battery and a reset making sure the Ra table was updated with the new files the SOC dropped to 1%

roomtemplowtempgraph.docx

I also tested an older file that I created before I used the Chem_Id selector using Chem ID 451 instead of the 414 recommended by the CHEM id tool.

The Chem ID 451 file did not drop to 1% but we wanted to re-do it because after 3 days it would read 70% on a full battery.

I will be away for two weeks starting tomorrow.

Thanks

0 Onyx Ahiakwo over 7 years ago in reply to Andrew Dueckman96

TI__Guru* 76230 points

Hi Andrew,

pls do the following:
-On the unit that you programmed the file from gpc, extract a gg file and srec file.
-Start bqstudio logging
-Charge the battery to full, rest 30 mins
-Discharge to empty.
-Charge to desired voltage/SOC
-Drop the temperature to the cold temp you are using
-see if the behaviour occurs.

If it does, send me the gg and srec file from before the test, the log file and then extract a new gg and srec and send over as well.

This will allow me analyze the files to see what is going on and offer further recommendations.

thanks
Onyx

-

0 Andrew Dueckman96 over 7 years ago in reply to Onyx Ahiakwo

Prodigy 90 points

Hi Onyx,

I will try this when I get back in two weeks.

Thanks

Power management

Power management forum

bq27742-g1: Large change in SOC during relax