• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 236 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.

Original question:

BQ27441-G1: No communication over I2C

BQ27411-G1: Does not reset cleanly when voltage glitches

Lawson Fisher
Prodigy 10 points
Part Number: BQ27411-G1

I copied the schematic in Figure 5 from the SLUSBN7D –JANUARY 2014–REVISED APRIL 2017 datasheet.

When I toggle the SW2 power switch on, I occasionally have no I2C communications. It does not acknowledge its address and the VDD pin 5 line is low. I've also seen large glitches, over a volt on the battery (pin 6 BAT) and I'm assuming they are related. Cycling power with SW2 clears it, but I need a robust long term fix. Short term should I  remove some capacitors downstream of the SW2 switch in an attempt to minimize the power glitch? I really though the dual 18650 batteries would not dip in voltage.

I don't think it matters, but I did skip the one time programming step. I only need Voltage, Current, and Temperature  data so I did not bother with the chip programing. At first I thought this was my issue, but I then saw the power glitch and VDD staying low. 

  • 0
    TI__Guru 57747 points

    Hello Lawson,

    Use the EVM schematic for reference. https://www.ti.com/lit/ug/sluuap3a/sluuap3a.pdf

  • 0
    TI__Guru 52996 points

    The gauge may not turn fully if the residual voltage on VDD (pin 5, connected to a capacitor) isn't fully drained before powering up the gauge, for example, if the gauge loses power briefly (it will back-bias the LDO and can lock up the startup circuit. VDD will not power up fully in this case).

  • 0 in reply to Dominik Hartl11
    Prodigy 10 points

    I was going to try a larger capacitor on VDD thinking I could stabilize the internal power supply to stop the lockups, but your explanation indicates  this is the wrong strategy.

    How about a diode between my battery and the BAT (pin 6) would block the glitches and keep the device power all the time? I'm using 18650 batteries, so a Schottky diode with 250mV forward voltage drop will keep me above the 2.45 volt minimum on the BAT input.

    My other thought was allowing the CPU to cycle power off to the chip with a transistor that the firmware could toggle if the BQ27411 does not respond on the I2C bus.Your reply suggest a transistor to pull VDD to ground would also be something to try.

    Would be nice if this chip had a reset input like the I2C expanders I'm using.

    thanks

  • 0 in reply to Lawson Fisher
    TI__Guru 52996 points

    This gauge is a pack side gauge, intended to be permanently attached to a cell. It's not designed to be power cycled, especially not fast enough that the residual charge on VDD is a problem. There's a fundamental system design problem if the gauge gets power cycled like that.

    It really should be attached to the cell one time during production and then be powered up for the lifetime of the product. 

  • 0 in reply to Dominik Hartl11
    Prodigy 10 points

    I have the gauge permanently connected to the twin 18650 batteries. I thought I was following the recommended design.

    When I flip the switch to power my device, I see the battery voltage drop down to around 2 volts. I have a huge capacitor load that I could reduce from 700uF to 350uF in an effort to minimize the battery glitch. I'm assuming if I keep my battery voltage from dropping below 2.5 volts I will keep the gauge happy.

    Another option I have is to place my power switch between the battery and the gauge. Maybe this is the simplest fix.

    thanks for helping.   

  • 0 in reply to Lawson Fisher
    TI__Guru 52996 points

    Yes, if the voltage doesn't drop below the POR threshold, then the gauge doesn't care. Maybe a large capacitor permanently attached to the pack (gauge + battery) which can drive your transient spike when you turn on the load would be a simple fix?

  • 0 in reply to Dominik Hartl11
    Prodigy 10 points

     I can just move my caps upstream of the switch to stop transient spike.

    THANKS!