• State Resolved
  • Locked Locked
  • Replies 15 replies
  • Answers 6 answers
  • Subscribers 63 subscribers
  • Views 3478 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.

Original question:

"Update Status" issue of Bq34z100-G1

BQ34Z100-G1: Learned Status and Update Status issues for Bq34z100-G1

Ramon Santos
Prodigy 90 points
Part Number: BQ34Z100-G1
Other Parts Discussed in Thread: BQSTUDIO, GPCCHEM

Hello,

After properly following the first few steps of an optimization cycle for a 12V 100Ah Lead acid battery, the Update status and Learned Status does not update beyond 4.

I followed these steps in the optimization cycle

1. Discharge the battery to empty as much as possible, then detach the power supply and electronic load completely from the battery

2. Send IT_enable and Reset commands. Rup_dis and QEN should be 1 at this point.

3. Relax until RUP_DIS and VOK are clear, and OCVTaken is set.

4. After OCVtaken is set, reattach the power supply and charge at C/2 (in my case I set it to 50 A). VOK should set at start of charge

5. Wait until the gauge displays the set taper current (in my case I wait until "150" is displayed on the left hand side display of BQstudio), and make sure that FC is set to 1.

6. Quickly detach the power supply and eload again. 

7. Relax the battery again unitl RUP_DIS and VOK are clear and OCVtaken is set. At this point Update status should become 0x05.

8. Reattach the eload and discharge at C/10 (in my case I have tried both 20 A and 10 A) until termination voltage is reached (in my case I wait until 10800 mV is displayed on BQstudio). The Grid Number should update.

9. Relax the battery again.

In my case, I have tried the cycling twice, the first time with 20A discharge and the second time I stopped before discharging at 10 A since Update Status did not change.

Basically what baffles me is that the important bits RUP_DIS, VOK, OCVTaken, FC, and QEN behave as expected and at the proper times. However both Update status and Learned Status both only reach a maximum value of 4. From the datasheet a value of 4 for Learned Status just means that IT_enabled, and 4 for Update Status means IT is enabled but Qmax and Ra data are not learned. From what I understand they should both reach a value of 6.

I did notice that the current readings of BQstudio tend to drift away at extreme values from the designated Design Energy Scale of 7. Also it does not go to zero when the battery is disconnected. Is it possible that that is the reason? At first I thought that maybe the battery chemistry I chose, 809 from tool TI provided, is wrong, but RUP_DIS did not set during discharge; I read somewhere that RUP_DIS settting during discharge means that the chemistry chosen is wrong.

I also read from a previous thread here that Update Status for Lead Acid batteries does not update until after discharge? In the above list of steps I expected Update Status to change during step 7, so should I expect it to change only after or during step 9? In the files I am attaching here I stopped basically at step 7 because Update Status did not change. The discharge in the log attached is just me discharging the battery to empty it again so I can restart the optimization cycle.

I have attached the log file (sampled every 5 seconds) and gg files (autoexported every 10 minutes). Thanks!

  • 0 in reply to Ramon Santos
    TI__Mastermind 25955 points
    Hi Ramon,

    Thank you for the detailed post and logs. I will update this thread by COB Tuesday.

    Sincerely,
    Bryan Kahler
  • 0 in reply to Bryan Kahler
    TI__Mastermind 25955 points
    Hi Ramon,

    With lead acid, please modify your steps to be based around discharge cycles. Allow the battery to rest after a full charge, then fully discharge and allow the battery to rest to update to 0x05. Afterwards, again charge until fully charged, allow the device to fully reset, then fully discharge and allow the battery to fully rest to update to 0x06.

    Some other changes to make:

    Please reduce Flash Update OK Voltage to 1000 mV.

    Please scale T Rise by a factor of 7 as well.

    Sincerely,
    Bryan Kahler
  • 0 in reply to Bryan Kahler
    Prodigy 90 points

    To be clear, what is your definition of "rest"? Is it when the voltage stays at a single value for a long time?

    So from what I understand, I am going to:

    1. Discharge battery to empty. Completely disconnect the e-load and power supply.

    2. Send IT_enable and then Reset. RUP_DIS and QEN should set.

    3. Let the battery relax/rest until RUP_DIS and VOK are clear, and OCVTaken is set, then relax some more(?)

    4. Reconnect the battery to the supply and charge to full, i.e. until the Taper Current previously set is reached and FC bit is set.

    5. Quickly disconnect the battery from the supply and let it relax until RUP_DIS and VOK are clear and OCVTaken is set.

    6. Discharge the battery down to 10.8V

    7. Let it relax until VOK and RUP_DIS are clear and OCVTaken is set. Just wait until Update Status becomes 0x05.

    8. Charge the battery to full, and let it rest afterwards

    9.  Discharge the battery to empty. At this point Update Status should become 0x06. 

    From what I understand I have to be more patient and let the battery relax more until the desire change (Update Status) happens? I am unclear by what you mean with "modify [my] steps to be based around discharge cycles".

    I am gonna apply the changes to the config files you suggested as well. 

    Thanks for the reply! I am gonna try again as soon as I am clear with your suggestions. It may take me a few days to update this thread then because frankly the cycles take way too long for a 100 Ah battery.

  • 0 in reply to Ramon Santos
    TI__Mastermind 25955 points
    Hi Ramon,

    Yes, resting until relaxation is when the battery voltage stays at a single value for a long time. This can be up to 5 hours after a full charge, or longer after a partial charge. Please ensure that the pack is fully charged and the FC bit has set.

    A quick modification to step 1: Start with fully charging, instead of fully discharging. The updates will occur on the discharge so this will save you some time.

    Step 8: Charge the battery to full, FC set, let it rest until relaxed.

    Step 9: Discharge the battery to empty and let it rest until relaxed. Update Status should then become 0x06 when a qualified OCV value has been taken.

    Yes, relaxation is so the device takes a qualified OCV value. This will not occur if the battery is not relaxed.

    If the device does not update following your above method, we will need to continue investigation with the chemID to ensure its a match and also device settings.

    Sincerely,
    Bryan Kahler
  • 0 in reply to Bryan Kahler
    Prodigy 90 points
    Thanks, I will update once I'm done with cycling.
  • 0 in reply to Bryan Kahler
    Prodigy 90 points
    Quick question, when you say scale T Rise by 7, do you mean divide the default (20) by 7, or multiply it by 7?
    Thanks.
  • 0 in reply to Ramon Santos
    TI__Mastermind 25955 points
    Hi Ramon,

    Please divide by the scale factor.

    Sincerely,
    Bryan Kahler
  • 0 in reply to Bryan Kahler
    Prodigy 90 points

    Interesting.

    The Update Status now has become 0x05, and Max error is at 3%. I just finished charging for the second time (charge - relax - discharge - relax - CHARGE - relax - discharge). The interesting part is that OCVtaken did clear after charging but VOK stayed set. I let it rest for about 3 hours and now that OCVtaken has set, yet VOK is still set. Basically both are set to 1. Is this to be expected? I thought once OCVtaken gets set VOK should clear, but this time it did not...

    I did just finish charging 3 hours ago, I will let it rest for 24 hours more. Any thoughts Mr. Kahler? 

    Attached is the screenshot of the registers from BQstudio for clarity. Again, from what I understand VOK should be cleared by now. Thanks!

  • 0 in reply to Ramon Santos
    TI__Mastermind 25955 points
    Hi Ramon,

    Yes, please let it relax.

    If it still has not updated to 0x06, please review the log to see if there were any current spikes, noise, and/or jumps in the log that would cause Ra not to update.

    This can occur because of negative calculated temp values or because Rdis not updating due to accumulated error.

    Please charge to full again, allow FC to set, relax fully, discharge at a constant current, and then relax fully once more. Update Status should then update to 0x06. If problems persist, please attach your gg.csv file, SREC file, and a log of the event to this thread for analysis.

    Sincerely,
    Bryan Kahler
  • 0 in reply to Bryan Kahler
    Prodigy 90 points

    Hello,

    So do I just wait it out, I mean wait until the VOK bit is cleared even when OCVtaken is set already before going to the next part (discharging at C/10 then relax until 0x06)? It still not cleared after more than 24 hours of relaxing... Or can I go to discharging now? Also just to be clear the last thing I did was to charge the battery a second time (you suggested relax-charge-relax-discharge-relax-CHARGE-relax-discharge-relax), which is why I am relaxing it right now. Update status changed to 0x05 BEFORE that charging part,i.e. during the relax part before charging.

    I am still not done with the learning cycle you suggested since I am still waiting for the voltage to stabilize at a single value (honestly I am also confused why it's taking so long). The next step would be the last discharge and last relaxation.

  • 0 in reply to Ramon Santos
    TI__Mastermind 25955 points

    Hi Ramon,

    No, waiting it out won't be an option here (unless it hasn't actually relaxed). Most likely what has occurred (based on the reported flags) is that the value was not qualified even though an OCV value has been taken.

    Please fully charge, rest until relaxation, discharge, then rest until relaxation to see the device update from 0x05 to 0x06. If issues persist after completing this new cycle, please upload the full log and the gg.csv file from the device for further analysis.

    Sincerely,
    Bryan Kahler

  • 0 in reply to Bryan Kahler
    Prodigy 90 points

    So I am going to charge it again? Because technically it's still newly charged(its only been resting since last I charged it) though the voltage is still dropping (still not relaxed?).

  • 0 in reply to Ramon Santos
    TI__Mastermind 25955 points
    Hi Ramon,

    In the learning cycle, we are looking for OCV points. These open circuit voltage (OCV) points are seen when the battery is not under load. Simply put. batteries take time to recover and fully stabilize after being loaded - we refer to this as relaxation. We are interested in capturing an OCV reading at the top of the usable battery voltage (after fully charged and relaxed), followed by at the bottom of the usable battery voltage (after fully discharged and relaxed).

    Please fully charge, rest until relaxation, discharge, then rest until relaxation to see the device update from 0x05 to 0x06. If issues persist after completing this new cycle, please upload the full log and the gg.csv file from the device for further analysis.

    Sincerely,
    Bryan Kahler
  • 0 in reply to Bryan Kahler
    Prodigy 90 points

    Hello!

    Update status and Learned status both became 0x06, finally.

    For reference, here's what I did, detailed, for the 100 Ah 12Vnom Lead Acid battery with chem ID 809 according to GPCchem. Note that all the values that I mentioned here are read from BQStudio interface, not at the power supply or eload (e.g. I waited until 10800 mV is displayed on BQstudio):

    1. Discharge the battery to empty as much as possible, with the OCVoltage as low as possible too. Completely detach the battery from the eload-supply setup.

    2. Send IT_enable, and then send the Reset Command. At this point, for me, RUP_DIS and QEN should set (become 1) at this point. Also Update Status becomes 0x04.

    3. Let it relax until RUP_DIS and VOK are cleared, and OCVTaken sets. For me I think I waited some more (<18 hours) even after OCVTaken set.

    4.  Reattach the battery to the power supply and charge at C/2 rate (in my case 50 A). Right at the start of charging VOK should set.

    5. Charge until the Taper Current (that you set, in my case it is 150) is reached AND FC bit is set. I waited until both conditions are met. Reading the datasheet, they should happen at approximately the same time anyway.

    6. QUICKLY detach the supply from the battery after the battery is fully charged (previous step). Honestly I don't know if it's necessary to do this quickly.  I waited for about 80 seconds after the conditions (FC bit is set, Taper Current is reached) since I read from the datasheet that FC needs two Taper Current Window periods to set. By default the value is 40 seconds.

    7. Let the battery relax again until RUP_DIS and VOK are cleared (RUP_DIS actually never set again after step 3) and OCVTaken is set. For me I this happened after about 2 hours of relaxation, but I still waited for about 18 hours just to be sure.

    8. Reattach the eload and discharge at C/10 (in my case, check for your own setup). Wait until the minimum voltage you set it reached and BATLOW is set (not sure if this is necessary), then quickly disconnect the battery. Grid Number increases as I discharge the battery.

    9.  Let it relax until VOK is cleared and OCVTaken is set. Some time during this relaxation, Update status and Learned Status should become 0x05 and MaxError went to 3%. Also, the parameter QMaxCell0 increased for me from 14286 to some value around ~15k

    10. After resting, charge the battery again the same way before.

    11. Relax until OCVTaken is set. For me, however, VOK did not clear anymore as in previous steps. This is referenced in my earlier post to Mr. Kahler. Given that in the end Update Status become 0x06 this might be actually expected and correct.

    12. After Relaxing, discharge the same way as before. For me, during this second discharge, MaxError actually increased to 4%, I dont know why. 

    13. Wait again until OCVTaken is set, and this time, VOK should clear. At this point Update Status and Learned Status became 0x06 and MaxError became 1%. QMaxCell0 also increased to a value of 17182.

    What are your thoughts on this Mr. Kahler? It seems that I had to do two relax-charge-relax-discharge-relax cycles, which is different from the original appnote I read which prescribed only 1 such cycle (relax-charge-relax-discharge-relax). Why is that? Is the learning cycle different for different chemistries? 

    Also I am still confused with the previous VOK issue. Actually I re-charged the battery after step 10 as you suggested, but again during that second charge VOK still did not clear even if OCVTaken set.

    And final question, if I may, what's my next step? I still don't know how to generate the so-called "Golden Image".

    Thanks for the help by the way! I am gonna upload the logs for those interested, but those logs span a total of about 10-14 days..