I have modified a second MP1584 buck converter module so that it would not activate until input voltage surpasses 13V, comfortably above the output voltage of 3.3V. I want to connect input pins to a solar panel so the associated components (ESP8266 WiFi microcontroller and INA219 voltage/current sensor) would be powered exclusively by the sun.

First is a test run with my bench power supply. Gradually increasing supply voltage starting from zero volts. Thanks to the modification, there were no odd behavior or sounds of a MP1584 trying to work with too low of an input voltage, which is great. As I increased voltage past the ~13V threshold, I saw the blue LED of my ESP8266 blink and it booted up as planned. This time, it was able to find INA219 on I2C bus, which is further than I got before.

Feeling optimistic, I connected this circuit to my solar panel at night and hoped I would wake up to find the system running. Sadly I woke up to disappointment, as there were no logged messages from the ESP8266. Probing the circuit with my volt meter, I confirmed a 3.3V supply voltage was present, but for whatever reason the ESP8266 failed to boot that morning. I manually disconnected and reconnected the circuit board, and this time ESP8266 booted up fine (now it has full daylight power) and started reporting values measured by INA219.

I don't know what happened at sunrise. I hypothesize that when the solar panel output voltage rose past 13V, it has still yet to produce enough power to successfully start an ESP8266. So when MP1584 activated, it could supply 3.3V but not enough amperage to supply an ESP8266 through its boot process, putting it in a glitched state that was neither on nor off and stuck there until I power cycled the system. [UPDATE: Further experimentation found this hypothesis was correct, the panel would reach operating voltage well before generating appreciable power.]

I didn't have my oscilloscope set up to capture the startup waveform to confirm or disprove this hypothesis. It's clear there are additional subtleties I don't know about starting up a circuit on solar power. Do I want to invest the time to learn and experiment with this problem domain? After thinking it over for a bit I decided "nah" and abandoned the idea of running everything exclusively on solar power. I'll retreat to what I know and incorporate batteries into the system instead. Starting simple with household alkaline AA batteries.