• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 492 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.

BQ20Z655-R1: at low temperatures

Tobias Jung
Prodigy 80 points
Part Number: BQ20Z655-R1
Other Parts Discussed in Thread: BQ40Z50, , GPCRB

I got two problems with the BQ20z655 at low temperatures, first I don’t understand why at temperatures below JT1 (-10°C in this case) the Charge-FET isn’t switched off and second when I start charging with a normal laboratory power supply at room temperature, then during the charging I cool the battery down to e.g. -20°C and during this process a SOC of 48% is reached, directly after pulling the battery out of the charge or disconnecting from the power supply and inserting in to the application the SOC jumps down to e.g. 4% although the max current peaks form the application are around 150mA. Which shouldn’t have a big impact on the SOC of a battery with 3300mAh after such a short time.

After charging:

After about 2 min in the application: 

  

I also attach a log-file where I charged the battery with the power supply (the switching just occurred due to COV because the supply was set a little high, but the charging at temperatures below JT1 wasn’t stopped).

In this project we are using the bq20z655 and not the bq40z50 because also an LCD is used.

 

 20nachminus20chgfw16.csv

  • 0
    TI__Guru 57697 points

    Hello Tobias,

    The bq40z50 is more accurate at low temperature compared to bq20z655-R1. There are very few adjustments available on bq20z655-R1 for cold temperature.

    I would recommend that you use the GPCRB tool to improve low temperature performance. Another factor that is applicable is that bq20z655-R1 expects charging temperatures to be more stable. There is no compensation for drastic changes in temperature during charging. Also note that the battery capacity at -20C is very low in terms of maH even though percentages may be the same. That may be the cause of the drop.

  • 0 in reply to Shirish
    Prodigy 80 points

    Hello Shirish,

     

    thank you for your reply, I’ll go on and make the log files at 25°C and at -20°C for the GPCRB Tool and try if this leads to more stable SOC values.

     

    But what about the Discharge FET not turning off below JT1, is the datasheet wrong at this point and there is not function that turns the FETs off or must I set a certain bit for this?

     

    Yours Tobias

  • 0 in reply to Tobias Jung
    TI__Guru 57697 points

    Hello Tobias,

    Is [CHGIN] or [CHGSUSP] bit in Operation Cfg B set?

  • 0 in reply to Shirish
    Prodigy 80 points

    Hello Shirish,

    it took us a little while but now we have set the [CHGSUSP]  and [CHGIN] and done the cycle for the GPCRB Tool. But unfortunately, we still see the SOC jumping in low temperature ranges. 

    First, we charged four batteries at room temperature, then the batteries were removed from the charger / power supply. 

    One battery which was also logged was the discharged with an electronic load and as visual in the log file the SOC jumped from 15% to 0%

    I also attach the log file.

    Then two batteries where unconnected in the climate chamber (which was set to -20°C) and also the SOC jumped from 100% to 0% over night.

    The last battery was in the application in the climate chamber overnight and at 81% SOC in the next morning. Then this battery was disconnected from the application and then connected to the EV module to log the future behavior. Also, this battery jumped after some hours to 0% SOC although it wasn't charged or discharged in this time. 

     

    I also attach this log file. 

    What do you propose us to do? Redo the cycle with the GPCRB Tool or change any setting?

    LOG_1.xlsxLOG_2.xlsx

  • 0 in reply to Tobias Jung
    TI__Guru 57697 points

    Hello Tobias,

    Is this observation during the logging to collect data for submission to GPCRB? If yes, then please ignore SOC. We only need Voltage, Temperature and current for submission to GPCRB.

  • 0 in reply to Shirish
    Prodigy 80 points

    Hello Shirish,

     

    the logging was done after the gg_out.gg from the GPCRB Tool was programmed to the BQ20z655-R1. Because of this we where expecting to get consistent values for the state of charge.

  • 0 in reply to Tobias Jung
    TI__Guru 57697 points

    Hello Tobias,

    Thanks. I will analyze the log and reply

  • 0 in reply to Shirish
    TI__Guru 57697 points

    Hello Tobias,

    For LOG1, the error is not very large for such low temperature. The Ra tables may need to be updated to optimal values. Try running another discharge/charge cycle which will update the Ra tables.

    For LOG2, the change in RSOC can be explained. The voltage rose continuously and reached a state of rest. At this point the gauge recalculated RSOC based on OCV readings. The recalculated RSOC was 0.

    Low temperature error is normally higher because of self heating of the battery. When the battery is at rest, there is no self heating and this reduces the power it can supply.

    Check that Load mode matches your application. I would recommend using Load Select = 7

    Another frequently used setting is Reserve Battery Capacity. Set this to a higher value to be more conservative and prevent unexpected system shutdown.