• State Resolved
  • Locked Locked
  • Replies 15 replies
  • Answers 6 answers
  • Subscribers 64 subscribers
  • Views 790 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:

bq34z110 update status

BQ34Z110: BQ34Z110 optimization/learning cycle procedure for PbA chemistry with high voltage rebound

Rom Caroselli
Prodigy 130 points
Part Number: BQ34Z110
Other Parts Discussed in Thread: BQ34Z100-G1, BQSTUDIO, BQ34Z100, BQEVSW

Hello,

My target battery is 12V 35Ah PbA AGM Valve Regulated type and I'm using the BQ34Z110 most commonly used for PbA.

The 2013 data sheet and posts for PbA applications call for starting with a fully charged relaxed battery before IT Enable.
However, older and newer application reports, SLUA903-July 2018, and SLUA597-MAY2011 call for starting with a fully discharged relaxed battery before IT Enable.
Examples in these reports are for lithium chemistries, so I'm not clear if they apply to PbA.

(1) Can someone explain why for the PbA chemistry, status 5 is to be achieved on discharge, and for lithium chemistry it's on charging?
(2) Which learning cycle procedure is currently recommended for my battery?
(3) Is the modification of the senc file usually required for batteries with high voltage rebound?

(4) After IT Enable 0x0021 is set prior to the first learning cycle attempt, does it need to be reset before starting subsequent attempts?
(5) When is RESET 0x0041 to be used?

(6) Will the learning process accept interference with the voltage rebound?
For instance, once the discharge load reaches the desired termination voltage and is removed,
can I reapply the load in bursts as needed to keep the rebound below the intended termination voltage to achieve the update status 5?

(7) One option to get past the voltage rebound issue is to increase the cell termination voltage and thereby increase the capacity reserve.
After learning is complete, could the BATLOW Alert thresholds be changed to indicate a SOC level below zero but within the reserve range?
This won't create a conflict with the gauging algorithm?

(8) Are these settings still recommended for PbA where voltage rebound is an issue?

Discharge =>C/8 to 1667mv/cell (10V for 6S)
Cell Termination Voltage=1800mV/cell (10.8V)
Cell BL Set Volt Threshold=1600mV/cell (9.6V)
FConvEn=0

I'm getting a voltage rebound of 1.35V which overshoots the 10.8V target by 0.55V.

Thank you.

  • 0
    TI__Guru** 105600 points

    Hello Rom,

    I would not recommend implementing the BQ34Z110 gauge for a new design, you should use the BQ34Z100-G1 which is newer and also supports lead acid battery type. The BQ34Z100-G1 TRM has a completed guide for the learning cycle you should follow. The new guides do not 100% correspond to all gauges, but they do overview a lot of the common practices that will get a good learning cycle.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 130 points

    Hello Wyatt,

    Yes, I'm aware that the Z100-G1 supports PbA but I'm already committed to the BQ34Z110 with ChemID selection completed and set-up is now ready for learning.

    I also like that the Z110 is targeted to PbA, as most info on Z100 is lithium.

    Sincerely,

    Rom

  • 0 in reply to Rom Caroselli
    TI__Guru** 105600 points

    Hello Rom,

    You can use the BQ34Z100-G1 with the chem ID you have found, both use impedance track. It is also pin-to-pin compatible and has the same features to support lead acid batteries, It just has additional bug fixes and updated firmware to work with the BQStudio platform. The BQ34Z100-G1 has more documentation and we are able to assist more effectively since it uses our newer platform and firmware.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 130 points

    Hello Wyatt,

    Yes, I understand that as the device architecture and usage is essentially the same ,I would have the same questions for BQ34Z100-G1.
    I'm already set-up and ready to start the learning cycle using BQ34Z110 EVM so I'd like to continue.
    I would expect my questions relevant to both device versions.

    Sincerely,

    Rom

  • 0 in reply to Rom Caroselli
    TI__Guru** 105600 points

    Hello Rom,

    I'm not sure what you mean by voltage rebound, when a load is removed from any battery chemistry under high load there will be a voltage "rebound" back to the OCV. You should not provide bursts of current during learning, it should be a constant current discharge down to the terminate voltage. The voltage may rebound but under any load it will immediately drop below terminate due to internal resistance.

    I would follow the guide that is in the BQ34Z100-G1 TRM for reference during the learning cycle process.

    The terminate voltage is where the gauge will force SOC to be 0%. You can set it anywhere you want after learning. During the learning process we need at least a 90% change in DOD between the two relaxation points. 

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 130 points

    Hello Wyatt,

    The BQ34Z110 datasheet SLUSB55B pg30, provides a CAUTION statement regarding voltage rebound. It reads:

    It may be necessary to stop the load, check the rebound, then apply the load again to ensure the rebound voltage will not be significantly higher than the Cell Termination Voltage.

    In the example given for a 12V Lead Acid Battery, like mine, the solution is a DOD to 10V, 0.8V deeper  than the manufacturer's recommended termination voltage of 10.8V.   Then a voltage rebound of 1V would bring the relaxed voltage to 11V, apparently close enough to the intended 10.8V termination voltage. 

    SLUA903 pg4 reports that for learning purposes (no specific chemistry), the Termination Voltage should be set to the minimum voltage of the battery as specified in the manufacturers data sheet.   This suggests that the only way to widen the rebound allowable is the DOD, i.e. discharge deeper than the manufacturers spec by an amount equal to the voltage rebound. However BQ34Z100-G1 TRM SLUUBW5A  pg61 warns against this as it may damage the battery.

    From forum posts using BQ34Z110, it appears that voltage rebound has been an issue for PbA chemistries and often resolved when support makes an adjustment to the senc file.

    2021 SEP BQ34Z100-G1 TRM SLUUBW5A (no specific chemistry) suggests discharge to the termination voltage without adjustment, and does not address the voltage rebound problem.

    2018 NOV SLUA925 pg9 has learning cycle instructions for PbA using BQ34Z100,  It suggests to discharge at least 10mv/cell below the cell termination voltage.  Voltage rebound is not addressed.

    The learning cycle steps in this 2018 OCT FAQ post for using BQ34Z100-G with PbA says to discharge to the termination voltage - no adjustments indicated.  Voltage rebound is not addressed.

    In summary, it seems that the voltage rebound issue using BQ34Z110 may have been resolved with the BQ34Z100 & BQ34Z100-G.

    Is this your understanding?
    Is voltage rebound issue no longer an issue in achieving a successful learning cycle as far as you know?

    Sincerely,

    Rom

  • 0 in reply to Rom Caroselli
    TI__Expert 8315 points

    Hello Rom,

    Wyatt is currently Out of Office. He will get back to you early next week when he returns.

    Thank you for your understanding,

    Jackson

  • 0 in reply to Jackson DiGiovanna
    Prodigy 130 points

    Thank you Jackson for the notice.
    I will standby.

    Rom

  • 0 in reply to Rom Caroselli
    Prodigy 130 points

    Hello Jackson,

    This issue remains unresolved and It seems Wyatt remains unavailable.
    Is anyone able to answer my questions?

    Thanks,
    Rom

  • 0 in reply to Rom Caroselli
    TI__Guru** 105600 points

    Hello Rom,

    I would still recommend trying the learning cycle with the termination at the cell manufacturer limit. If you are not able to get the 90% change in DOD with this method, you can decrease the voltage so when relaxed the cells OCV will be low enough compared to the relaxation at full charge to get the 90% change in DOD. Not all of our documents directly relate to gauges which support PbA chemistry, like SLUA903. What it mostly comes down to is getting enough change of DOD during the learning cycle, you can decrease the load to the minimum C/10, increase charge slightly, or decrease term v slightly to get the same result.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 130 points

    Welcome back Wyatt,

    If I understand your reply, the critical metric for the learning algorithm is the difference between the OCV measurement after fully charged and the OCV measurement following discharge, which you refer to as DOD.  Is this the case? 

    This is important to clarify, as I understood  DOD to refer to the voltage drop while under load, i.e. where the passed charge occurs.  It doesn't make sense to me that the 90% passed charge requirement need involve the voltage rebound.  Please explain.

    This thread concerns how the voltage rebound after the discharge affects the learning cycle.
    As mentioned earlier, voltage rebound with regard to a successful learning cycle, was of such great consequence that the BQ34Z110 data sheet included a cautionary statement giving an example of a 12V PbA battery like mine.

    Earlier, I sited other references addressing the issue, specifically for PbA chemistries, but all were dated before the introduction of the newer device BQ34Z100-G1.  I didn't find any after it's introduction.

    So I like to know if the voltage rebound issue using BQ34Z110 was resolved or mitigated somewhat with the BQ34Z100-G?
    If  this is your understanding,  I would be inclined to make the transition now.

    Sincerely,

    Rom

  • 0 in reply to Rom Caroselli
    TI__Guru** 105600 points

    Hello Rom,

    Yes that's correct, the DOD value is based on the absolute voltage range which is the chem ID you have programmed.

    The DOD is not connected to the voltage drop or rebound, it is related to the OCV vs DOD table, so it only matters what the relaxed voltage is after a charge or discharge to find the DOD.

    I am not aware of a new feature in the BQ34Z100-G1 that will mitigate the issue, both the BQ34Z110 and BQ34Z100-G1 will have similar logic engines for computing the DOD from the OCV. The main reason for the switch is the updated toolchain, which will help us support you quicker for debug.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 130 points

    Hello Wyatt,

    Thank you for clarification of DOD for gauging.  It's not what I thought.

    But with regard to the voltage bounce back, it seems something changed given he BQ34Z110 datasheet cautionary statement.

    It recommends dropping discharge termination below normal to ensure the resulting voltage bounce back will be close to the intended termination voltage.  I don't see this recommendation, or even a mention of this issue in documentation for the newer generations.

    Was this recommendation faulty or are the learning algorithms with BQ34Z100-G1/STUDIO more tolerant of voltage rebound than with BQ34Z110/bqEVSW?

    Sincerely,

    Rom

  • 0 in reply to Rom Caroselli
    TI__Guru** 105600 points

    Hello Rom,

    I believe the caution is just a suggestion for if you do run into an issue. Most of the time it will probably be okay not to lower the threshold. I'm not sure exactly why it was removed from the other documents. There is no significant difference in the learning algorithm between the two gauges. If you want to be extra cautious you can decrease the termination slightly lower to ensure the 90% DOD change.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 130 points

    Hello Wyatt,

    Okay, I get it now, thank you.
    Which parameter should I be watching for measuring the DOD change?

    Sincerely,

    Rom