So you're a Noob? Post your questions here until you graduate! Don't be shy.

User avatar
By Fleming
#67447
rudy wrote:Confirm with a meter that GPIO15 is pulled low. GPIO0 and GPIO2 are at a high level. That Reset and CH_PD are both high.

Confirmed, all the levels are correct, see Album

rudy wrote:Many of my boards are soldered on the same crappy white boards you are using. But I have an on board regulator along with additional capacitors on it. I have never used those boards with an external 3.3 volt supply.

I have had problems with some regulators even when they were soldered to the back of those boards. It would work often but I would get crashing problems. It was only after I brought them to work and scoped the power pin on the module did I find there was a glitch on the power line. The glitch was causing crashing problems.

I am concerned about using large value electrolytic capacitors to provide the surge current. They often have to high an impedance at high frequencies. I use a multi stage approach. 0.1uF in parallel with a 10uF ceramic cap in parallel with a 68uF tantalum cap. All as close as possible to the Vcc and gnd pins on the module.

See viewtopic.php?f=6&t=11068&start=4

With your setup I would take a module and solder a tantalum cap from the Vcc pin and the negative of the cap soldered right onto the metal shield of the module. It is a ground point. I would keep the wire length of the cap as short as possible. That's what I would do, but the -ve wire of the cap soldered to the gnd pad should do.

I rely on the power supply, but is definitely one point to improve... once I manage to upload a single sketch...

rudy wrote:I just use the generic setup. 4M with 1M SPIFFS, QIO, flash freq 40Mhz, CPU freq 80MHz.

Load the blink sketch. Have that run for a while.

Will try