• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 631 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.

BQ27542-G1: Making the Gauge more Accurate

e g
Prodigy 190 points
Part Number: BQ27542-G1
Other Parts Discussed in Thread: BQSTUDIO

I have a battery that is 4 Molicel 18650K batteries (Chem ID 2249) in parallel connected to my fuel gauge. I have already calibrated the fuel gauge, performed the learning cycle, created a golden image, and am now at 15 cycle counts of fully discharging and charging it. The Fuel Cell does not show that it is getting more accurate overtime, TimetoEmpty varies from 1% - 4% off ;~15 minutes to 50 minutes difference from the actual time to the TimetoEmpty value. I also always seem to get a Full Charge Capacity value in the range of 9000<x<10000 when ,I assume, it should have a value higher than 10000. I have set the battery in Data Memory as such

Charging Voltage = 4200mV

Taper Current = 500mA

Design Voltage = 3700mV

Design Capacity =  10400mAh

Charge Current = 2000mA

Discharge Current = 500mA

My goal is to have the TimetoEmpty value consistently around 0.4% or less (around 5 minutes). I am wondering if there is a setting I made a mistake on, missed when setting up the gauge, or something wrong with cycle setup on bqstudio. Is it possible for me to improve the accuracy to around 0.4% or less?

  • 0
    TI__Guru* 76230 points
    hi
    When you performed your learning cycle, did update status go to 06? What are your charge and discharge conditions, is it the same CC each time to the same full charge voltage or those ware varying? What is the temperature condition when performing these tests? DO you have log and gg files of your test to show??

    thanks
    Onyx
  • 0 in reply to Onyx Ahiakwo
    Prodigy 190 points

    I did not check the update status value at the end of the learning cycle. I am looking at the learning cycle that I used for the golden image (and have posted as an attachment) and I am not seeing a status with 0x06 on it. I constantly charged the battery at constant 2A, to 4.2V with taper current 50mA. I Discharged the battery at constant 500mA, Termination voltage at 3V. The temperature condition was that the battery and fuel gauge were at constant room temperature as they weren't in any casings. For the first 3 logs I had the wait time to be around an hour of waiting. For the next 4 logs I changed the wait time to 5 hours, I didn't read the instructions explicitly stating that you need 5 hour wait time in order to properly calibrate. Then the last 4 logs I changed the design capacity from 10,000mAh to 10,400mAh; changing from the minimum capacity to typical capacity to see if that would improve my accuracy.

    I did not have any gg files. I do have all the log files I could gather and placed them in a zip file attachment if that helps answer my question.

    LearningCycle18.log

    CycleLogs.zip

  • 0 in reply to Onyx Ahiakwo
    Prodigy 190 points

    I have checked the update status and it showed a 0x06.

  • 0 in reply to e g
    TI__Guru* 76230 points
    Hi

    We can achieve up to 1% SOC error accuracy depending on how closely matched your chem chosen is to your cell. The fact that update status is 06 means that learning successfully completed. To evaluate accuracy of the gauge, ps follow the instructions in the blog below. The excel sheet in the blog has details of the math.
    e2e.ti.com/.../how-accurate-is-your-battery-fuel-gauge-part-2-2

    thanks
    Onyx
  • 0 in reply to Onyx Ahiakwo
    Prodigy 190 points

    I have redone a few more cycles and performed the SOC calculations as you have done in the articles and the fuel gauge came to a 1% error. There is still the question of how to get a smaller difference between the TimetoEmpty value on the Fuel Gauge and the actual time it takes to empty the battery. The difference in time is around 10 minutes but the goal is to get the error in time to be around 5 minutes or less. 

  • 0 in reply to e g
    TI__Guru* 76230 points

    We evaluate for accuracy based on SOC. Such stringent time to empty criteria is next impossible because of the variability in load current and temperature.

    hanks
    Onyx