Espruino JavaScript
Flash map error on NodeMCU with 12E #5486
Replies: 1 comment
-
|
Posted at 2016-05-26 by MrTimcakes Hey, I'm getting this error on one on my ESPs too, I was playing around with some WS2812b LEDs and I saved the code on the ESP, it became non responsive so I disconnected and couldn't reconnect. I tried to flash Espruino back on but when connecting from the WebIDE I just get this printed every few seconds. I don't mind one of my NodeMCU's being out of commission, I just hope the saving thing will allow someone to recreate this issue for debugging. Posted at 2016-05-26 by Wilberforce Here is the reason for the reset:
That Did you have Posted at 2016-05-26 by dwallersv Do not understand what you mean about the I2C code and saving. I'm a noob :-) However, it doesn't matter -- I get this problem with a virgin, freshly flashed NodeMCU esp8266. I had a problem at first that I'd flashed for the 512kb 01 board, but fixed that and am still getting this watchdog timer issue with nothing loaded and running at all. Oh, and the strangest part of this -- it acts like noise or some capacitive coupling issue because if I rest my finger on the board in the right place (among other things, the CP2102 chip is there -- USB->serial) the problem doesn't happen. Posted at 2016-05-27 by Ollie @dwallersrv @ducky first thing to do is use 'erase_flash' with esptool.py. Then reflash it. This will clear any saved code. Posted at 2016-05-27 by dwallersv @ollie thanks for the suggestion -- unfortunately, that's what I did every time I flashed it, so that's not the problem. I'll give it another go, though, and see what happens. Posted at 2016-05-27 by Ollie @dwallersv Odd. With nothing running, on a totally wiped board and to be stuck in the reset loop.. I've read elsewhere that this can be power, or pull up related but if you aren't running anything at all that's unlikely. Potentially a defective NodeMCU? Posted at 2016-05-27 by Wilberforce Paste the erase and flashing scripts you are using here.... Could be something wrong with those.... Posted at 2016-05-27 by MrTimcakes Thanks, that did it. Posted at 2016-05-28 by dwallersv I've determined with certainty it's a flaky board (what do you expect for $7?). If I have contact with the area of the board where the CP2102 USB/serial chip is (but not just contact with that IC), it runs fine. As soon as I take my hand away, it get's stuck somehow and the WDT fires. I can even carefully take the board on the end of the USB cable and carefully put it under my shirt against my stomach, carefully remove my thumb from the "magic" area, and then use both hands to interact with it in the IDE. Pull it away from my body, same crash. So, it's capacitive coupling of some sort due to seven dollar manufacturing. Got a few more ESP12E variants on the way, as well as a dozen raw 12E SMD boards (couldn't pass up $1.50 each w/free shipping on eBay), so I'll confirm shortly that it's what I think it is. Until then, I have my Pico to work with, solid as a rock. |
Beta Was this translation helpful? Give feedback.
-
Posted at 2016-05-26 by dwallersv
With the help of fellow travelers here, I got my WS2811 50LED test strip working using the neoPixel call. I was unsuccessful using SPI directly through GPIO13 (NodeMCU D7) with the exact same code that works like a champ on the Pico. I followed the example in the tutorial using the send4bit call to shape the NZR waveform the 2811B modules are expecting.
Got some light, but weird colors, not what I sent. I'm guessing the baud rate setting is not precise, or, as Gordon mentioned in another thread, the H/W SPI still does some waiting/checking/something for each byte, so the continuous 800khz stream isn't actually happening. (Aside, where can I find the source for the neopixelWrite() call? I'd like to see how its implemented).
Anyway, my main problem's solved: I can drive these strips effectively with an 8266, which was the goal.
Here's the problem: I get these random crashes:
What's going on? Looks like a reset -- is that a flaky reset button on the NodeMCU board? I note that if I rest my finger over the button area, contacting the USB connector, and over some other components on the board, the problem goes away. So it looks like noise. However, I don't know how to interpret "rst cause:4" and the other stuff sent to the console.
Beta Was this translation helpful? Give feedback.
All reactions