• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 1428 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.

BQ25892: Strange Vsys and/or Bat Fet disable

charles lahure
Prodigy 120 points
Other Parts Discussed in Thread: BQ25892

Hi,

I manage to get a BQ25892 charging a battery @2A while supplying another load through Vsys

but I cannot get it to draw more than 0.7 amp from the 12V Vbus input. I am using  Vsysmin=3.0V.


With more current drawn from Vbus or with a higher Vsysmin, the Vsys output cut off periodically for about 20ms

and throw the host in reset because it is supplied through Vsys. The batfet is also disabled.

I found an easy and strange way to produce a very similar problem:

All registers are set to default, except one bit: REG03[4]=0 to disable charging.

A low charged battery is attached, but no current goes in or out the battery.

When I increased the current drawn from Vsys, almost every time the Vbus input current reach 0.36Amp,

either Vsys drops briefly to zero (20ms) and I loose the host, either the batfet is disabled.

It's strange because without the battery attached (or with batfet off), it works fine (up to Vbus input current = 1.3 Amp).

The battery is not involved in power consumption, but it does create a problem !

I tried different values of Vsysmin without luck.

Please help

Thank you

  • 0
    TI__Guru**** 166876 points
    What are the input current (IINDPM) and VINDPM settings? The INDPM setting should be set to maximum current that your source can provide. If your source can't provide that much current or the connection from the source to the charger is highly resistive, causing a voltage drop at the charger's VBUS pin, then VINDPM will first scale back charge current then scale back input current limit to prevent the charger from crashing the adapter.
  • 0 in reply to Jeff F
    Prodigy 120 points

    Thanks for your reply

    > What are the input current (IINDPM) and VINDPM settings?

    I am testing with REG03[4]=0 and all other bits to default.  So the registers are:

    Reg00 IinLim = 0x7F -> max = 3.25 A

    Reg04 ICHG = 0x20 -> max = 2 A

    Reg0D VinDpm = 0x54 -> 11V

    I think my power supply is good enough, I have the same problem with a 12V car battery

    I doubled the pcb tracks with solder and copper, and the problem remains

    Finally  here are the solutions that I found :

    The best solution is to put a high pass filter between SW and PGND, the very same that is suggested in the evaluation board user's guide (10 nF + 1 ohm). The main problem is that this capacitor heats too much.

    Intermediate solutions are:

    1- Increase the inductance from 1 to 2.2 uH

    2- Put a schottky diode between SW and PGND, the very same that is suggested in the evaluation board user's guide.

    3- Put a schottky diode between BTST and REGN, the very same that is suggested in the evaluation board user's guide.

    My testing procedure is to set REG03[4]=0 and let all other bits to default. Increase or switch the Vsys load and see when the bat fet turns off.

    An empty battery is attached, its current is always zero because of REG03[4]=0.

    Only with the low pass filter, I am able to drive a 6 watts load without seeing the bat fet opening.

    I must admit my pcb design is basic.

    Is this problem related to the negative spike at SW ?

    Thanks again

  • 0 in reply to charles lahure
    TI__Guru**** 166876 points
    Did you disable the ILIM pin resistor (REG00 bit 6), which clamps the maximum input current to its setpoint, regardless of ILIM register?
  • 0 in reply to Jeff F
    Prodigy 120 points

    Hi

    > Did you disable the ILIM pin resistor (REG00 bit 6)
    No I did not disable it. But my Rilim=120 Ohm, so i've been able to draw more than two amps from a 5V Vbus

    As I said my problems arise at a much smaller current

    Again, as I wrote, my best solution is to put a high pass filter between SW and PGND, the very same that is suggested in the TI evaluation board user's guide.

    However, the 10nF C21 capacitor let too much current flows. I calculate that the resistor R17 dissipates 2 Watts. Indeed it's hot. The formula is:  P= fs.C.Vbus^2 (fs = 1.5MHz, Vbus=12V)

    How is it possible that the BOM specify a 0.125 W rating for R17 ?

    As of my testings, 1nF is enough. It gives the same improvement as a 10nF, with only a 0.2W loss.

    Could please check this and elaborate on the use of this filter ?

    Thank you