Review Request: 3.3V Power Supply & Li-ion BMS Schematic for ESP32

Thread Starter

Aman G

Joined Jul 28, 2026
2
Hi everyone,

I would really appreciate it if someone could review my schematic and provide feedback.

I'm currently designing the 3.3V power section for an ESP32-based project. The schematic is attached.

The power section includes:

  • USB charging circuit (BQ24074)
  • Single-cell Li-ion battery
  • Battery protection circuit (BMS) (BQ29700)
  • 3.3V power rail for the ESP32 and other peripherals (TPS62840)
  • Load switch to control power to the NFC module for power saving (TPS22917)
The main goal of this circuit is to provide a reliable and efficient 3.3V power rail for the ESP32 and any other modules that require 3.3V while ensuring proper battery charging and protection.

I would appreciate feedback on the following:

  1. Is the overall power architecture correct?
  2. Are there any design mistakes or reliability concerns?
  3. Are there any missing components, protection features, or recommended improvements for a design?
If you notice anything else that could be improved, please let me know. I'm trying to make this design as robust as possible, so I'd really appreciate any suggestions or constructive criticism.

Thank you very much for taking the time to review my schematic!
 

Attachments

Thread Starter

Aman G

Joined Jul 28, 2026
2
Hi everyone,

I would really appreciate it if someone could review my schematic and provide feedback.

I'm currently designing the 3.3V power section for an ESP32-based project. The schematic is attached.

The power section includes:

  • USB charging circuit (BQ24074)
  • Single-cell Li-ion battery
  • Battery protection circuit (BMS) (BQ29700)
  • 3.3V power rail for the ESP32 and other peripherals (TPS62840)
  • Load switch to control power to the NFC module for power saving (TPS22917)
The main goal of this circuit is to provide a reliable and efficient 3.3V power rail for the ESP32 and any other modules that require 3.3V while ensuring proper battery charging and protection.

I would appreciate feedback on the following:

  1. Is the overall power architecture correct?
  2. Are there any design mistakes or reliability concerns?
  3. Are there any missing components, protection features, or recommended improvements for a design?
If you notice anything else that could be improved, please let me know. I'm trying to make this design as robust as possible, so I'd really appreciate any suggestions or constructive criticism.

Thank you very much for taking the time to review my schematic!
i'm attaching schematic as block representation as well as with wire connections please refer these files and provide me your observations
 

Attachments

MisterBill2

Joined Jan 23, 2018
28,290
There appear to be quite a few diodes, with forward voltage drops, in series with the power feed. While the voltage will still be correct, the efficiency would be reduced some amount.
That may matter in your application. OR possibly not.
 
Your architecture is sound, the charger plus a separate cell protector plus a load switch on the NFC rail is the right shape for this. BQ24074 gives you power path and charge termination, BQ29700 covers over/under-voltage and overcurrent as a second layer, and gating the NFC module off a TPS22917 is a clean way to drop that drain. the part I would rethink is the TPS62840. It is a buck, so it can only hold 3.3 V while the cell sits above 3.3 V. Once the battery sags below about 3.4 V the rail starts to droop even though the BQ29700 will not open until 2.8 V, so the ESP32 browns out long before the protector trips and you leave the bottom of the battery unused.

Swap it for a buck-boost and that gap disappears. The TPS63802 holds 3.3 V from 4.2 V down to the 3.0 V cutoff and idles around 11 uA, which is about what the ESP32 draws in deep sleep anyway, so you give up a few microamps of sleep current versus the 60 nA buck and get the full cell in return. also make sure the converter input hangs off BQ24074 OUT rather than the cell directly, that is the whole point of the power path. If you would rather not spin the buck-boost yourself there are small boards that already pair a 1S charger with that converter so the rail never sags at the low end, MicroPowerCore is one of them.
 
Top