Brownout Detection error on LiPo powered ESP-32 PCB

Mojies asked Jan 14,2023
0

Hi all,

For a project I created a custom PCB.


It's a LiPo powered PCB with ESP32-WROOM-32E module, Lipo charging module, 3.3V Voltage regulator and MPU6050 Acc/Gyro unit. Charging via USB-C connector. The ESP-32 module being programmed by using a programming header (and external FTDI232 module, to be attached on position “H1:program” in the schematic below.)


After uploading the code and powering via the FTDI232 module, attached to my USB port in computer (LiPo and USB-C disconnected) everything runs fine.


But when trying to power only from LiPo (2000mah) and even when supported with charging power from USB the script________ is triggering a Brownout Detector error. When uploading a blink________ or ‘bare minimum’ script________ everything is working fine om LiPo-only power. My assumption is that the final script________ (using Wifi, BLE, MPU sensor) is triggering the error while consuming to much current? But I don’t understand why I don’t have this issue when powering via FTDI232 module.

Is my schematic correct (looking at the choice of voltage regulator HT7333-7, placing of capacitors etc)? Any suggestions how I could further investigate and solve the error in a future design are really welcome.

Thanks,

Mojies


Schematic_1PCB_2023-01-14 (1).png


Schematic  in better quality

  • Your Answer(2)
Upload photo
You can only upload 5 files in total. Each file cannot exceed 2MB. Supports JPG, JPEG, GIF, PNG, BMP
0 / 10000
  • 0

    Hi Mojies, I think your problem is the regulator - If I've read part numbers correctly, the ESP32 module has a max current of 0.5A, but the regulator seems to be rated for 0.25A. Might be worth looking at 1117 series regulators. Most are 800mA, but some go all the way to 1A. Other than that, a parametric search at your favourite supplier should do the trick. Other things I've noticed from the schematic is that there's not a lot of decoupling caps. On voltage regulators, I've always placed bulk and decoupling on both input and output, with the decoupling as close as physically possible to the regulator legs. The bulks are right, at about 10µF, but i'd be putting 100nF decouling as well. I'm not sure how the regulator will behave (or be damaged) by having the feed from H1:Program into it's output. I normally bypass LDO and linear regs with a silicon diode (1N4148, 1N4001 or similar (depending what's in the parts box, usually) between the input and output (anode to output) to stop any bulk caps on the output side discharging backwards through the regulator. I've attached a snip of a schematic and PCB... Note that the VIn to the regulator is very close to the VOut of the power switch on this design, so I've only placed one bulk capacitor on the rail.

    Jan 26,20231 Comments
    matt 2023-02-08 15:45:38Reply

    Forgot to mention... If you add a couple more diodes, one connected anode to regulator output and cathode to 3v3 rail, and another one connected anode to regulator ground pin, cathode to real ground, you'll stop any nasty backfeeding into the regulator. The first will stop the backfeeding, and the second lifts the regulator output to compensate for the 0.7v drop.

    Reply
Browse other questions tagged or ask your own question.
List of Questions
pause
  • Weekly Leaderboard
Ranking Name Answers
1 PCBWay Team 9
2 weslley ramos 1
3 Epishko Dmitry 1