Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

  • State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 2 answers
  • Subscribers 64 subscribers
  • Views 260 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

BQ34Z100-G1: Strange Full Charge Capacity Increase

Kevin Knicker
Intellectual 951 points
Part Number: BQ34Z100-G1
Other Parts Discussed in Thread: BQSTUDIO

Dear TI,

We have a LiFePO4 battery pack assembly that uses the bq34z100-G1 fuel gauge.

The packs are in the field so I don't have any bqStudio log data, but I have attached some data that the customer was able to provide.

Note the pack is near full charge, but scrolling down in the data, you'll see the Full Charge Capacity (FCC) suddenly start to increase.

Eventually the Remaining Capacity increases along with it.  However, they are increasing to values that don't make any sense and reach values that in the 30,000 range.

I would think the max FCC could possibly be is 32767, so are these values legitimate?  They eventually return to a normal values, but this increase is creating problems and throws off the Fuel Gauge % for some time.

Here is some background on the data flash setttings:

SMOOTH is on

RELAX_SMOOTH_OK is on

RELAX_JUMP_OK is off

LFPRelax is on

DoDWT is on

FConvEn is on

Load SELECT is 06 and Load Mode is 0 and the User Rate is 2500mA.

I have the data flash attached as well.

I've become aware of the bq34z100-G1's shortcomings with LiFePO4, and I've gone through the forum as much as possible to get information, but I can't find anything in regards to this.
The pack has completed a successful learning cycle and we haven't had much issue until this came up.

Is there something that could be causing the FCC to increase with the pack being left on a charger? You may notice the charge kicks in and then there is a slight discharge, and this repeats when left connected.

It's just strange that it happens so far along into being connected, and then corrects itself.

Are there any settings I should look into and perhaps turn off in that I have listed above?

Thanks much,

Kevin

FCC_Increase.xlsx7S_LFP.gg.csv

  • 0
    TI__Guru** 104695 points

    Hello Kevin,

    That is abnormal behavior, is your packs nominal Qmax (capacity) around 5200mAh?

    Is the .gg file shared before or after the test was done? Why was the user-rate selected for the load select, is your discharge fixed rate at 2500mA? Usually the default is okay, or sometimes 0 and 2 can work with a variable system.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Intellectual 951 points

    Hi Wyatt,

    Correct, the nominal capacity is around 5200mAhr.

    The .gg file attached is what is programmed to the pack when this test was done.We've learned that a charger was connected to the system when this happened.  So the system draws some current from the pack, then recharges.  This behavior repeated, and then the sudden jump happened.  It did return to normal, but I'm wondering if that charge/discharge pattern may have caused this because the pack didn't see a good RELAX state or something of that nature?

    User-rate was selected because I read it can help to prevent sudden jumps on discharge, which we had been experiencing.  Setting it to user-rate helped to prevent that for us.  The normal current rate is around 2500mA.

    We can certainly change the load select to 0 or 2.  In fact, I do believe I have seen that mentioned in a similar forum post that was in regards to capacity jumping.  I will try that, but in the meantime, if there is anything else you can think of that I can look into, please let me know.

    Your help is much appreciated.

    Kevin

  • 0 in reply to Kevin Knicker
    TI__Guru** 104695 points

    Hello Kevin,

    I would be curious if this still occurs with the different load select, but the behavior is strange, usually we see jumps in FCC or remcap but you're seeing a slow increase then a drop. Not many applications use the user-rate load select so that's why I'm wondering if it may be the cause to increase FCC when it runs a simulation.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Intellectual 951 points

    Hi Wyatt,

    I haven't been able to personally replicate this strange behavior.  This was reported to us from in the field.

    I can try changing the load select to see if that helps, but I'm still curious as to why is might be happening at the end charge if the user rate has to do with discharge. 

    Would not using CHCDodEoC have any impact on this? It is presently turned off.

    Thanks,

    Kevin

  • 0 in reply to Kevin Knicker
    TI__Guru** 104695 points

    Hello Kevin,

    Yes I didn't notice that this bit was disabled. I would leave the CHGdodEOC enabled. It may cause the DOD to update incorrectly and cause the FCC variation.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Intellectual 951 points

    Hello Wyatt,

    I've enabled the CHGDoDEoC bit and I changed the Load Select to 0.

    I am suspecting that because the battery pack was oscillating between a small charge and small discharge near EoC, that the Fuel Gauge was hitting the EoC threshold or FC repeatedly and then somehow got into a runaway state.

    I will let you know how this works

    Thanks,

    Kevin

  • 0 in reply to Kevin Knicker
    Intellectual 951 points

    Hello Wyatt,

    Changing the load select to 0 or 2 now causes the gauge to adjust the Full Charge Capacity to extreme values if left to rest.

    The pack I have is rated at 5200mAhr.  The FCC had been reporting 5189, but after a discharge down to 50%, the pack rested and then updated to 2880mAr and 41% capacity.

    There is no way the gauge should adjust to almost half it's rated capacity simply by resting.  I have a good Chem ID and I've performed a successful learning cycle.  I know I'm not the only one to have issues with this part as I see the forum has numerous inquires along the same lines.

    I'm tempted to go back to the load select of 6 and a user rate mA discharge rate. That at least prevented the pack from make sudden large jumps in FCC.  I have RELAX JUMP OK off.  I'm not sure why the load select being set at 6 would cause the FCC run-away that lead to my inquiry.

    I appreciate your support and will let you know if I see any more issues.

    Thank you,

    Kevin

  • 0 in reply to Kevin Knicker
    TI__Guru** 104695 points

    Hello Kevin,

    I'm not sure what caused it exactly, but the parameters that can cause large fluctuations are the Ra table (bad Ra update) or extreme temperature/load changes. Nothing else should cause large jumps like you're seeing in the logs.

    It's strange that when the load select is switched the FCC changes that much, that also indicates to me that the thermal modeling is not good or the Ra table did not update properly. It is also possible since this is an LFP battery the OCV is flat enough to lead to DOD estimation errors.

    Sincerely,

    Wyatt Keller