• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 65 subscribers
  • Views 535 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.

BQ34110: EOS learning never reach relaxation

Manuel Haag
Prodigy 40 points
Part Number: BQ34110
Other Parts Discussed in Thread: BQSTUDIO

Hi

I am using the BQ34110 gas gauge for our backup battery system an got some problems with the EOS learning cycle. First some application information.

The Battery:

  • acid lead
  • 1200mAh
  • 12 Cell
  • 24V nominal voltage
  • gg-File:

Steps already done:

  1. Calibrate voltage, current and temperature
  2. Configure the date flash for our application (see gg-file)
  3. Generate a CEDV-profile with the GPC tool
  4. Learned FFC with a learning cycle
  5. Verified SOC with different loads.
  6. Tried to start a EOS learning cycle without success

About the EOS-learning problem:

  1. We are using discharge before charge (LSM = 1), because the charger is usually connected to the battery (floating). 
  2. We disconect the charger. The system is then supplied with the external supply. 
  3. After that we trigger a learning cycle with the EOS_START_LEARN() or EOS_RCELL_RRATE_LEARN() command.
  4. LCMD and LRLX get set 
  5. However, we never reach the relaxation condition. Therefore, the cycle does not start.

Relaxation Condition according to the TM:

If the change in the consecutive 100-s averages of Voltage() is less than 4 μV, then relaxation is deemed achieved.

We never reach this condition. The RXLX never gets 0 and the LDSG never get 1. There is no discharge through the battery.

 

Questions:

  • Is there something wrong with our setup?
  • Is it possible to change the 4uV over 100s?
  • It is clear that in the beginning (when disconnecting the charger) the voltage is decreasing slowly. But how long do I have to wait?
  • Is the 4uV per cell or pack?
  • Are there other issues that I don’t see?

Thank you in advanced.

  • 0
    Prodigy 40 points

    Here is the gg-file if needed:

    bq34110.gg.csv

  • 0 in reply to Manuel Haag
    TI__Guru 52115 points

    We will take a look and get back to you.

    Andy

  • 0 in reply to Andy Liu (BMS)
    TI__Guru 52115 points

    Actually, it is not difficult at all to meet the condition that the change in the consecutive 100-s averages of Voltage() is less than 4 μV.  The REST bit in the Gauging Status can tell you whether the device has reached relaxation or not.

    Let the battery pack relax and use the bqStudio to communicate with the gauge every a few seconds to monitor the gauge. If the battery pack is well relaxed, you should be able to see the REST bit set.

    Andy

     

  • 0 in reply to Andy Liu (BMS)
    Prodigy 40 points

    Hi Andy

     

    Thank you for your response.

     

    We are using the bqstudio and monitor the registers. But, if we disconnect the charger and the battery is not connected to the circuit (no discharge), we have to wait around 45min to get into this state.

    That is a very long time for our application.

     

    However, after the REST Bit is set, we can start the learning cycle. LRLX gets clear and LDSG gets set (as described in the user manual). Then we connect the load (200mA) and discharge the battery for 10min. After that, the LRLX bit gets set again and the device is waiting for relaxation again.
    In this state (post-learn relaxation phase), we get some problems again. When we disconnect the load or disconnect the battery from the load (no discharge), the LUCD error Flag gets set. We do not know why, because the Discharge Detection Threshold (60mA), Charge Detection Threshold (75mA), and Quit Current (40mA) should be high enough. We have found the same issue here but it is not really answered:

    Because if we increase the threshold, the device never change from Relax-state to discharge state. I think that is not the sense and the result of the learning cycle is 1mOhm (should be around 120mOhm).

     

    Is there a solution for this problem? We really have no idea what the problem could be. The learning videos did not help us to solve this issue.
    Thank you for your help.

     

    Best regards,

    Manuel

  • 0 in reply to Manuel Haag
    TI__Guru 61536 points

    The gauge will only set the LUCD bit if the gauge was in discharge or charge state. Please check if both the thresholds (charge/discharge/quit) and relax times (charge/discharge/quit) are set appropriately. Note that the gauge will stay in charge or discharge state for 60 seconds (default) even after the current crossed the quit current threshold.

  • 0 in reply to Dominik Hartl11
    Prodigy 40 points

    Thank you Dominik.

    After a few days, I have solved the proroblem with the LUCD bit. The Problem was, that I had to wait until the Battery Gauge quit the discharge mode, before I fully disconect the Battery from the circuit.

    However, the time until the REST bit get set is still very long (30-45min). So my final question, is this nomal or is there a way to shorten this time?

    Best regards,

    Manuel

  • 0 in reply to Manuel Haag
    TI__Mastermind 24870 points

    Hello Manuel,

    That depends on the battery instead of the gauge. We recommend to keep the default settings there.