• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 2282 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.

BQ25895: BQ25895 as UPS for Raspberry PI 3B+

Gabriele Rizzi
Prodigy 100 points
Part Number: BQ25895
Other Parts Discussed in Thread: REG71050, TPS61256C, TPS61089

Hi,

I am trying to use the BQ25895 as a UPS controller to both charge a LiPo battery pack (2x 18650 in parallel) and supply power to a Raspberry Pi 3 B+ connected to PMID (5V + boost converter).
With Vbus = 0 (no external power supply) and LiPo battery connected, PMID is at 5V and the Raspberry PI boots correctly.

When I connect an external 5V power supply to Vbus, I see a very short glitch (<5ms) on PMID and the Raspberry PI continues to work fine. The battery is charging also correctly.
When I disconnect the 5V external power supply from Vbus, the PMID goes down to ca. Vbat (3.7V) for ca. 38ms and then goes back again to 5V.

Clearly this long voltage drop resets the Raspberry PI.

Is the length of this voltage drop structural? or can it be avoided/minimized? or is it possibly due to a problem with my setup (I use the same circuitry as in the EVM demo board).

Looking forward to getting some hints.... thanks!
 

  • 0
    TI__Guru**** 169886 points

    Gabriele,

    The delay time before the charger changes from charging with power applied at VBUS to boosting from the battery to PMID is fixed and cannot be changed or shortened.  Can you add a simple circuit that senses when VBUS is removed and quickly puts the Raspberry Pi into lowest power mode so that it will not deplete the capacitors at PMID?  Alternatively, you could add a small, low cost 5V charge pump like REG71050 that provides 5V power to PMID from the battery during the switchover.  I suggest using a Schottky diode from the output of the charge pump to pump so that the charge voltage at PMID doesn't overdrive the charge pump.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Prodigy 100 points

    Hi Jeff,

    thanks for getting back to me so quickly.

    Well, this delay in getting PMID output at 5V is a real problem.

    Unfortunately, the Raspberry PI is a very power hungry device and, since I cannot know what the user is doing when the external power is cut, there is no way to put the RPi in a kind of power down mode.

    Reg71050 can only provide a few milliamps, and that will not keep the RPI running in a worst case situation.

    I was considering two options: adding an additional boost stage, powered by Vsys, to get a stable 5V 3A even when switching between battery and external power supply.

    The second option could be to put a supercap, 1 or 2 F connected between PMID and Ground. This should bridge a gap of ca. 40ms even when supplying a current of 1A. Would PMID output stand this?

    Would you have any recommendation? or preference?

    Rgds, Gabriele  

  • 0 in reply to Gabriele Rizzi
    TI__Guru**** 169886 points

    Gabriele,

    The boost converter will not be able to start up into a completely discharged super cap, as that will appear as a short to ground for too long, but it might into a partially charged super cap.  The safest option would be a separate boost converter like TPS61256C or similar.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Prodigy 100 points

    Jeff,

    thanks for your suggestions.

    I will probably go for a completely separate boost converter, powered from Vsys, using a TPS61089.

    Pity that the BQ25895 cannot be fully exploited.

    Best regards, Gabriele