• State Resolved
  • Locked Locked
  • Replies 15 replies
  • Answers 4 answers
  • Subscribers 63 subscribers
  • Views 465 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:

BQ27542-G1: remaining capacity

BQ27542-G1: BQ27542-G1: remaining capacity

Corey Walker
Prodigy 130 points
Part Number: BQ27542-G1
Other Parts Discussed in Thread: BQSTUDIO

Based on the original question, I now understand that Reserve Cap IT Cfg setting will "reserve capacity setting will show 0% Remaining Capacity at the point the battery has the amount of reserve capacity specified." However, if Android shuts the phone down based on 0% SOC, not reserve capacity, how can this work? I ran a discharge test via GDK board controlled by bqStudio to terminate on 0% SOC. But that termination point did not reserve any capacity.

  • 0
    TI__Guru** 105760 points

    Hello Corey,

    Could you share the discharge log?

    The gauge should've reported 0% SOC when the reserve capacity on the battery was reached. Which termination point are you checking?

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 130 points

    I'm attaching the log. I didn't think it worked (capacity reserved after reaching 0% SOC) because the remaining capacity also went to zero. But if ResCap is an offset for reported RemCap, then that's probably normal, in which case the only way to confirm "it worked" is to run a discharge test AFTER 0% SOC and measure the remaining capacity manually, via discharge rate and time. Right?  Regards, Coreyreserve_test_discharge_1.log

  • 0 in reply to Corey Walker
    TI__Guru** 105760 points

    Hello Corey,

    Could you also share your .gg file, it looks like it is working because your SOC reported 0% when the battery voltage was fairly high. What was your terminate voltage set to?

    You can test it that way if you know the full capacity of your battery. If you continue to drain the battery down after the gauge reports 0% the battery should reach the terminate voltage after the desired reserve capacity as been used.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 130 points

    I believe the vendor has term V set to 3300 mV. I'm attaching 2 gg files, before and after reserving 125mA. GS_baseline.gg.csvGS_125mA_reserved.gg.csv

  • 0 in reply to Corey Walker
    Prodigy 130 points

    "If you continue to drain the battery down after the gauge reports 0% the battery should reach the terminate voltage after the desired reserve capacity as been used."

    My vendor has the [RSOCHOLD1] Pack Config Reg set high, so SOC won't reach 0% until term V is reached. Seems we need to reset this bit low for the above statement to be true, and overall reserve capacity to work?

    ...from bq27542-TRM.pdf doc:

    "RSOCHOLD1 = The fuel gauge will prevent StateofCharge() from reporting 0% until Voltage() is less than or equal to Terminate Voltage. Set to 1 to enable."

    Also in that doc, "A loaded rate or no-load rate of compensation can be selected for Reserve Cap by setting the [RESCAP] bit in the Pack Configuration data flash register." However, after searching the doc and gg file I cannot find the [RESCAP] bit. 

    Another confusion from the doc: pg 34, 6.3 Current/Power Profiles, describes Load Select and Load Mode. "If Load Mode = 0 (constant-current model), then the options presented in Table 6-5." But on the next few pages, and in table 6-5, the doc starts using the term "SHUTDOWN State"; I believe incorrectly. It's very confusing. SHUTDOWN State is unrelated to current or power model, as described in 5.2 SHUTDOWN Mode.

  • 0 in reply to Corey Walker
    Prodigy 130 points

    Correction: that baseline is from a different battery (same type). I'll update with another gg file when I complete my experiments with [RSOCHOLD1] and Load Mode.

  • 0 in reply to Corey Walker
    TI__Guru** 105760 points

    Hello Corey,

    Having RSOCHOLD1 may prevent the reserve capacity from working correctly, I would need to check the FW code to see which condition has priority. Are you performing a test with RSOCHOLD1 off to check this?

    I believe RESCAP was turned into a reserved parameter. Let me double check if the TRM is correct with the usage of shutdown state.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 130 points

    Thanks Wyatt,

    Yes, I performed experiments with [RSOCHOLD1]. It has to be reset to 0 for Reserve Capacity to work as expected. Our vendor was setup for energy mode, so I also had to reset Load Mode to 0. It now appears to be working; however, the accuracy seems to be very dependent on current load after 0% SOC. I think we can live with that, as we're only trying to keep an RTC running. Any other concerns with advising the vendor to change to constant-current Load Mode?

    I now notice that our vendor has 2 different batteries with the same fuel gauge setup a little differently. Can you help us understand why they might have SE_EN (shutdown feature) on for one battery and off for the other? Is there any interdependency with Reserve Capacity, such that I need to advise the vendor to change it?

    Thanks!!

    Corey

  • 0 in reply to Corey Walker
    TI__Guru** 105760 points

    Hello Corey,

    I'm glad you have it operating now. The reserve capacity feature should work in both load modes, were you able to test if the load mode in constant power was operating correctly?

    I'm not sure why they would configure SE_EN different for different packs, you would need to ask them for their reasoning. I am still waiting to see if the TRM is consistent when talking about SHUTDOWN mode vs load mode, once this is verified it will be easier to tell if they are dependent on each other.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 130 points

    I have not yet, but will confirm operation in constant-power mode. I just know that the reserve has to be entered in power terms ("Reserve Energy","0","egy") vs. current terms ("Reserve Cap-mAh","0","mAh") -- separate gg entry points. I know the vendor did not try that. We are considering not using CC Load Mode, because we are concerned that we don't know what that will change, other than the reserve units. Does it also change the whole fuel gauge operation, i.e. impedance tracking algorithm? Would they have to perform another learning cycle too? 

    To recap, I had to do the following to get reserve to work:

    • reset [RSOCHOLD1] low, to allow SOC to reach 0% prior to termination voltage
    • reset Load Mode low, to use the mAh reserve units

  • 0 in reply to Corey Walker
    TI__Guru** 105760 points

    Hello Corey,

    The gauge will use current for the impedance track algorithm in constant current mode. If your vendor ships with constant power mode I would recommend keeping it and following the steps you outlined to have reserve capacity working correctly while in power mode.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 130 points

    Hopefully last question: if we remain in constant-power model (load mode = 1), then we need to enter reserve into the following line of gg file.

    "Gas Gauging","IT Cfg","Reserve Energy","400","egy"

    Should I treat "egy" unit as mWh?

  • 0 in reply to Corey Walker
    TI__Guru** 105760 points

    Hello Corey,

    Yes that is correct, you can also change it to cWh by referring to section 6.1.1.7 Design Energy Scale.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 130 points

    Hi Wyatt,

    I spoke of inaccuracy before, at least at the minimum current load that the GDK board will allow, and now that I have tried constant-power model (Load Mode = 1), it seems even more inaccurate, so I'm confused. I measured the following by applying CC load and measuring time from 0% SOC to term V (3.3).

    CC model, 50mA load, 125mAh reserved = 92mAh measured.

    CC model, 50mA load, 250mAh reserved = 216mAh measured.

    CP model, 50mA load, 482mWh reserved (125mA*3.85V) = 64mAh measured.

    Any idea why CP model is so different? I'm running a full charge of this battery now, to see if the gauge sort of relearns.

    Regards,

    Corey

  • 0 in reply to Corey Walker
    TI__Guru** 105760 points

    Hello Corey,

    Was the test run using a CC discharge or through your normal application? CC mode will be more accurate if you are using a CC discharge because the voltage isn't taken into account. 

    Sincerely,

    Wyatt Keller