Closed walkerdanny closed 4 years ago
Probably power issues. Use a different USB port or a separate power supply.
Thanks for the response - I'm powering it from a bench top PSU that's good for 2A so I doubt it's a lack of current. I can try monitoring the supply voltage as the WiFi initialises to see if there's a significant ripple when I have access to a scope.
I missed the custom part of your post. Definitely check the power. I have found that 4.7uF bypass and 100nF decoupling capacitor are enough to stabilize. Do you have a lot of external devices that might be drawing power through the MCU? It would be suspicious that the timing is always as the wifi starts, but perhaps put in a long delay before your WiFi.begin and see if it something else on the rail causing the dip.
I had a poke with a scope - there's a 0.1V drop on the supply line during WiFi init, which I'm not convinced is a huge problem. No amount of capacitance seems to make a difference to how far it'll get through the initialisation process.
I'm thinking it's something clock related, as the reset reason is the RTC watchdog timer kicking in, and the serial output becomes nonsense that can't be decoded. I think next I'll have to have a look at the garbage output it gives me after the crash to see if it's at a different baud rate or something.
It won't hurt to look at other avenues, but I'd say it is 95% likely a power issue. You could also try dropping the cpu speed before you init the WiFi with setCpuFrequencyMhz(80)
which will reduce your power consumption by ~40mA and might give you enough cold cranking amps to fire that solenoid :smile:.
[STALE_SET] This issue has been automatically marked as stale because it has not had recent activity. It will be closed in 14 days if no further activity occurs. Thank you for your contributions.
[STALE_DEL] This stale issue has been automatically closed. Thank you for your contributions.
Hello, I also have this problem. The develop board created by myself. When I use the WiFi.begin()
. My board also reboot....
You solved this problem?
I designed a board for ESP32-PICO-V3-02 and am having a similar issue - constant reboot loop with rst:0x8 (TG1WDT_SYS_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT) as soon as I try to turn on Ethernet (wired, with W5500 chip).
I have the ESP32-PICO-D4 eval board from Espressif and that one does not have the same problem.
But it is strange because I can build code, load it, blink LEDs, send serial messages. But when I enable the Ethernet (via SPI) it enters a reboot loop.
I tried putting a 1K pullup on pin 18 since I saw that in the reference schematic. But no luck.
Your issue is different than the OP. Open a new issue and fill out the necessary information for assistance.
Hardware:
Board: Custom ESP32-PICO-D4 based board Core Installation/update date: 11/10/2019 (today) IDE name: ESP IDF (Stable version updated today) or Arduino 1.8.10 Flash Frequency: 80MHz PSRAM enabled: No Upload Speed: 115200 Computer OS: Windows 10
Description:
I'm developing a custom board which uses the ESP32-PICO-D4. It will happily run simple Arduino sketches which send output over UART, but as soon as I enable the Wifi I get a boot loop, seemingly initiated by the RTC watchdog timer.
I've flashed the sketch using the ESP IDF and changed any settings in the config that I thought might make a difference, but nothing has changed (although it gave me a better level of debug output).
It's not out of the question that it's a hardware issue, but I can't see any obvious cold joints etc. so I'd rather exhaust the software options before soldering a new chip on.
The last line of the debug output occasionally gets as far as
W (334) phy_init: fail
before the debug stream turns into a garbage sequence of characters (as below) and the watchdog timer resets it and I getrst:0x10 (RTCWDT_RTC_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT)
Can anyone suggest a fix? Thanks.
Minimal sketch:
Debug Messages: