Problems with ESP-WROOM-32 and <WiFi.h> Library

[daniel_hammer]

I have a problem trying to use standalone ESP-WROOM-32 module with a LiPo battery and using TP4056 module, when I do a upload of a simple test sketch without libraries, my ESP works. But when I try to use WiFi or Bluetooth, the brownout happens. Someone can help me to see what is wrong in my schematic and show me what I have to do?

[ESP_Sprite]

HT7B33 is not a chip Google recognizes. Are you sure it can deliver >=500mA to the ESP32? Maybe also add some decoupling caps on the line between U1 and U2?

[daniel_hammer]

The resolution of printscreen has transformed the 8 in a letter B, the chip is a HT7833 and this LDO haves 500mA to deliver to esp32. But i will follow your tip.

[ESP_Sprite]

Could you also show your PCB design, by the way? Possibly something can be glanced from that.

[daniel_hammer]

I put my PCB and the 3D model too. But I don't make my pcb yet, I'm using my protoboard to test my project.

[ESP_Sprite]

Could be that the protoboard is the issue... those things are known to be somewhat shitty, and it's easy to get cheap Dupont wires that are thin as a hair inside.

[daniel_hammer]

I will try to swap the LDO HT7833 for other LDO and if doesn't work I will make a PCB to eliminate the problems of protoboard.

[ESP_Sprite]

Your PCB design is... interesting. I'd turn the ESP32 around so the antenna is near (or sticks over) the edge of the PCB, that way you don't have groundplanes getting 'in the way' of a signal. You also have no way to program the ESP32; I'd add a header with at least Tx, Rx, GPIO0, EN and GND on it so you can add an external programmer and also debug the board that way. Also, I didn't spot the 220uF cap on your 3.3V line; with that, you may have issues that the ESP32 is taken out of reset while the power hasn't completely come up yet (will look like the ESP32 does not start up after you applied power). If that happens, you may want to increase C3 to 10uF or maybe even more.

Ah, just spotted your board does not even have a ground plane. Suggest you add that. (In Kicad, remove all traces that carry ground, create a zone on the back of the board, select 'GND' as the net, and make a square covering the entire PCB.) Also, I'd fatten up the traces carrying power a bit; you have the space for it anyway.

[daniel_hammer]

Your PCB design is... interesting.

I know that my PCB design is bad, but I'm doing a proof of concept for this moment.

I'd turn the ESP32 around so the antenna is near (or sticks over) the edge of the PCB, that way you don't have groundplanes getting 'in the way' of a signal. You also have no way to program the ESP32;

Yes, it is the correct thing to do.

I'd add a header with at least Tx, Rx, GPIO0, EN and GND on it so you can add an external programmer and also debug the board that way.

I'm using a ESP32 burning board while I test in my protoboard (I don't use the energy provided of burning, only the USB programmer of the board), but, now I put 1 pin header with 3 pin when I want to use a usb tool for debug.
*The ESP32 burning: https://pt.aliexpress.com/item/32964740240.html

Also, I didn't spot the 220uF cap on your 3.3V line;

In my tests using my protoboard, my ESP doesn't work without electrolytic capacitor. However the capacitor seems connected on my PCB project.

Ah, just spotted your board does not even have a ground plane. Suggest you add that. (In Kicad, remove all traces that carry ground, create a zone on the back of the board, select 'GND' as the net, and make a square covering the entire PCB.)

I didn't it because I'm not going to make in a manufacturer yet.

But I believe that the HT7833 doesn't provide the enough quantity of current for the WiFi operation and I will test a new LDO.