• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 1349 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ27500-V130: SOC vs OCV estimation inaccurate for battery MP176065xlr (Chemical ID 1576)

boudet nicolas
Prodigy 50 points
Part Number: BQ27500-V130
Other Parts Discussed in Thread: BQ27500, GPCCHEM

Hello,

I'm testing a new battery SAFT MP176065xlr with a gauge BQ27500 v1.3.

The device is loaded with a defaut DFI file and with the chemistry MP176065 xlr id 1576. (from ChemicalUpdater-667-windows-installer )

I have also changed some parameters:

- the full charge capacity from 1000mAH to 6800mAH

- the taper current from 100mA to 120mA

I made measures of the initial SOC estimation versus Open Circuit Voltage.

I found that the SOC versus OCV is inaccurate for high OCV (3.4 to 4.2VDC), the SOC error is about 10%.

From 4.2 to 4.04VDC OCV reading, the gauge estimate the SOC to 100% or datasheet indicate that SOC is about 90% at 4.05VDC OCV.

Below 4.04VDC OCV the gauge start to decrease SOC estimation 

Is there an other/newer chemical file more accurate for this battery?

If not, what can i do to make this SOC estimation more accurate?

Thank you in advance.

Nicolas.

  • 0
    TI__Guru* 76230 points
    hi Boudet
    For starters, the bq27500 is an older device and i would strongly recommend you consider a different device like the bq27520.
    secondly, you need to perform a chem id identification using gpcchem and then perform a learning cycle before the gauge can be accurate
    thanks
    Onyx
  • 0 in reply to Onyx Ahiakwo
    Prodigy 50 points

    Hello Onyx,

    thank you for your response.

    I know that the bq27500 is not recommended for new design but i'm constrained to use it because the system we are producing is not new .

    The only element updated was the saft battery for obsolescence purpose, it have been change from MP176065 to MP176065xlr.

    For this reason  a new chemical id followed by a new learning cycle must be done (as you said).

    But i was surprised to see this SOCvsOCV error with the MP176065xlr because this wasn't the case with the old MP176065.

    This error still exist after one learning cycle but it's seems to be a little less important (2 to 1% less), maybe if i do some others learning cycle this will cancel the error.

    Is this right?

    In parallel, I will also try the gpcchem to find a better id, many thank's for this suggestion, i didn't know this tool. I will keep you inform about the result of it..

    Waiting for your answer about the effectiveness of succssives learning cycles.

    Best regards,

    Nicolas.

  • 0 in reply to boudet nicolas
    Prodigy 50 points
    Hello,
    I've tried to find a new id with gpcchem but my data returns the following error:
    All IDs disqualified because of too high resistance deviation. No suitable ID selection available

    I think this is due to my cycling discharge rate wich was too high: C/5 whereas the guide indicate a rate of C/10
    I will do a new cycle with this rate and see if it's accepted by gpcchem.
  • 0 in reply to boudet nicolas
    TI__Guru* 76230 points
    Yeah you need to use c/10 to perform chem id identification . Also you need to remove any protector circuit that might have come with the battery when doing the chem id identification as that could be the cause for not being able to find an id? carrying out a learning cycle with a wrong id will likely not improve the process.

    thanks
    Onyx
  • 0 in reply to Onyx Ahiakwo
    Prodigy 50 points

    ok, so i focus on the id identification.

    The old ID SAFT MP176065 was very precise, for it, i didn't had to make a chem id identification.

    I believed this was only for unknow battery or for battery without specific ID listed on the chemistry selector.

    Do you know if ever such SOCvsOCV problem have already been detected on chemical 1576 (SAFT MP176065xlr) ?

    May an update of this chemistry available?

    thank's in advance,

    Nicolas.

  • 0 in reply to boudet nicolas
    TI__Guru* 76230 points
    Nicolas,
    If you change chem id, then you must do chem id identification
    thanks
    Onyx
  • 0 in reply to Onyx Ahiakwo
    Prodigy 50 points

    Hello Onyx,

    sorry for this late answer.

    As advised, I'have done the chemId identification and the chemistry selected in the report works well.

    The SOC vs OCV curve is fine and I was able to do a successfull learning cicle with it.

    I'm still wondering why the official MP176065xlr chem was problematic, but the most important is to have found a solution.

    Thank you very much,

    Best regards.

    Nicolas.