• State TI Thinks Resolved
  • Locked Locked
  • Replies 15 replies
  • Answers 6 answers
  • Subscribers 63 subscribers
  • Views 836 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.

BQ34Z110: Utilizing lead acid gauge without current sensing

Michael Emery
Prodigy 30 points
Part Number: BQ34Z110
Other Parts Discussed in Thread: BQ34110, BQ34Z100, BQ34Z100EVM, EV2400, BQ34Z100-G1, GPCCHEM

Hello,

I’m looking to implement a battery gauge for 6v and 12v deep cycle lead acid batteries without putting a current sense resistor inline with load.

I’m creating a battery monitoring system not a battery management system and it isn’t practical for the device to be able to handle the full load or to install bus bar current sensors.

Looking at the documentation for the bq34z110 I see that there does exist a configuration to use design capacity/5 or a user configured rate to calculate load.

Is it possible to use this device as I want?

  • 0
    TI__Guru** 105760 points

    Hello Michael,

    The gauge would not know when you apply the user rate or scaled design capacity current values for predicting SOC because it wouldn't be able to detect a discharge since current would always be 0. I think you would need to use a monitor for this application.

    The only gauge we have that is designed for voltage correlation gauging is for 1s applications: https://www.ti.com/product/BQ27621-G1

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 30 points

    How about employing current sense solely for that purpose, Connecting a nice little 18 gauge wire in parallel to 00 wire providing the pack - connection to the motor controller and employing a much higher resistance resistor to limit the current flowing through the device to the minimum required to dependably sense that the motor is running?

  • 0 in reply to Michael Emery
    TI__Guru** 105760 points

    Hello Michael,

    This will still not work, the gauge must have the coulomb counting measurement in order to calculate the passed charge used for it's RemCap estimations. The User Rate tells the gauge what to use for the FCC calculation, we still need the current measurement for the RemCap information.

    Please check SLUA450 for information on impedance track gauges, you can see the RemCap (RM() in SLUA450) uses the accumulation of passed charge.

    https://e2echina.ti.com/cfs-file/__key/telligent-evolution-components-attachments/00-24-00-00-00-00-02-91/Theory-and-Implementation-of-Impedance-Track-Battery-Fuel_2D00_Gauging-Algorithm.pdf

    It is a requirement for this gauge to also measure current. The only work around would be to scale down the current and use an amplifier on a very small sense resistor. The reason we don't recommend this is because it can lead to high error from noise since the voltage on the sense resistor would be very small.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 30 points

    Thank you Wyatt,

    In your first post you suggested I might have to use a monitor.  Do you have one that doesn’t require current monitoring you might suggest?

  • 0 in reply to Michael Emery
    TI__Guru** 105760 points

    Hello Michael,

    I would recommend making another post linking a monitor you think may work and our monitor team will give you some suggestions. They won't be able to gauge but you can monitor the cell voltages.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 30 points

    I didn’t actually see one I thought might work any better than the gauge we are discussing.

    I guess I’ll look again since you seem to think it exists.

  • 0 in reply to Wyatt Keller
    Prodigy 30 points

    To flush out the idea of the work around:

    I think you’re suggesting upping the value of the current sense resistor and using a current sense monitor?

    I think there’s something I’m “not getting.” The problem I’m trying to avoid is having too much current.  If I add a resistor in between my pack- terminal and the current sense resistors to bring the current on the board down to something more reasonable then the voltage differential measured would be in operating ranges of the gauge.  Wouldn’t the problem just be configuring the gauge to scale that signal to represent the current on the main load branch not flowing through the device?  I’d be loosing resolution on the low end but operating ok under larger loads.

  • 0 in reply to Michael Emery
    TI__Guru** 105760 points

    Hello Michael,

    I suggested the monitor because the monitors do not need the current measurement for algorithm purposes. Looking at the post again since this is a lead acid application it may be better to use a top of stack method.

    If you're planning to measure the current then the BQ34Z110 will be able to gauge the battery, if it's scaled the accuracy will be reduced but you should be able to still get SOC information. As you say you can use a smaller sense resistor but you will lose accuracy at lower current values.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 30 points

    I need to get an evaluation module but don’t see the bq34z110evm.  Is it discontinued or am I just not seeing it?

    I see the bq34110evm (Should I consider that part?  I had read that IT is better for rechargeable cells and CEDV for non-rechargeable cells but the bq34110 seems like it is for rechargeable lead acid cells.)

    I also see the bq34z100evm (The documentation is a little unclear whether the bq34z100 is for lead acid or not.  Most of the documentation only says li-ion.)

  • 0 in reply to Michael Emery
    Prodigy 30 points

    it looks like the gauge development kit is discontinued as well.  Is there a replacement for it (something with the ev2400 built in).

  • 0 in reply to Michael Emery
    TI__Guru** 105760 points

    Hello Michael,

    You should use the BQ34z100-G1 for lead acid batteries, the BQ34z110 uses an older firmware which only supported the lead acid batteries.

    The process to do a learning cycle for lead acid battery can be found here:

    https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/927162/bq34z100-g1-lead-acid-battery-learning-process

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 30 points

    Thanks Wyatt,

    Does the BQ34z100evm ordered through TI online have the bq34z100-g1 on it?

    Is the ev2400 the only TI supported option available to interface with it?

    Just for the sake of education can you compare and contrast the bq34110 vs the bq34z100-g1?

  • 0 in reply to Michael Emery
    TI__Guru** 105760 points

    Hello Michael,

    I believe it now has the G1 loaded, if not you can download the srec on the product page and update it for testing.

    The EV2400 is the device we use to generate our programming files, calibrate, program the chem ID, and do most of the testing with. You will need one in order to upload the chem ID.

    The BQ34110 and the BQ34z100-G1 use different gauging algorithms. The BQ34z100-G1 uses impedance track, which takes into account the OCV of the battery, the battery resistance, temperature and current to find gauge parameters. The CEDV is a more simple algorithm that relies more heavily on coulomb counting. It is good for simple loads that have a constant discharge current.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Prodigy 30 points

    Thank Wyatt,

    I’m going to go forward with the BQ34Z100-G1 and drop the idea of using a parallel circuit to measure the current.

    I got the EV2400 and the bq34z100evm but don’t think I’ll be able to use the bq34z100evm for anything since it can’t handle the current I’m working with.

    I’ll have to go ahead and prototype.

    On the same note, I’d like to have TI profile a few of the batteries I’ll be working with.  Who do I talk to about that?

  • 0 in reply to Michael Emery
    TI__Guru** 105760 points

    Hello Michael,

    You may be able to use the EVM to collect the data for the GPCCHEM tool, we do not need to characterize all batteries, for most applications we already have a chemistry ID that is a close match. You can submit a Relax Discharge Relax log (RDR) to the GPCCHEM tool to get the chemistry ID with the closest match.

    Sincerely,

    Wyatt Keller