logo elektroda
logo elektroda
X
logo elektroda

Strange behaviour of Wemos D1 Pro + BME280: Doesn't switch on after battery discharge

krawietz 681 5
ADVERTISEMENT
Treść została przetłumaczona polish » english Zobacz oryginalną wersję tematu
  • #1 20033382
    krawietz
    Level 16  
    I have made several different modules based on the WEMOS D1 and they generally work....
    However, with a total of two I have a problem and no longer know how to bite it myself.
    The WEMOS is mounted on a slightly larger prototype board using goldpins and a suitable strip so that it can be removed.
    The connectors are soldered to the board so that the power supply, output devices or sensors can be connected easily and without any mistakes.
    And this is where the problem begins, after commissioning the system - specifically the Wemos D1 pro + BME280 - ran for a little over 24 hours on battery power, which as it turns out is irrelevant, and then switched off due to battery drain.
    Unfortunately, it did not switch on again, neither after applying voltage to the battery charging circuit, nor after disconnecting the battery and connecting the USB power supply either from the computer or directly.
    On the console it only reported waiting for host.
    Unplugging it from the prototype board makes it work fine on both battery and USB power. Short-circuiting everything to the goldpins soldered into the WEMOS board makes the circuit work normally.
    I corrected all the connections on the board (just 4 wires going from the respective WEMOS outputs to the connector I'm connecting the BME to) and the chip started properly. Worked a few hours and the BME readings became incorrect.... ot temperature over 100 degrees, no pressure reading and nonsense in humidity. On top of that the WEMOS started to light up with its built-in LED permanently. After a reboot it no longer gets up but spits some "exception". Removing it from the prototype board - it works fine.
    I don't know what's going on anymore... some transients on the other pins? The meter doesn't show short circuits anywhere. And the connections checked are ok.
    On top of that, the other chip that had been working for a few days also went crazy, it stopped reading data from the HC-SR04.... I corrected the soldering of the wires on the prototype board and the HC took up work....

    Somehow I have no idea what is going on.... the only thing I can think of is some problem with the wires I used, or some transients? Generally in other modules I didn't solder, no pull-ups and everything worked just on hanging ESP outputs not connected to anything.

    Krawietz
  • ADVERTISEMENT
  • #2 20033392
    khoam
    Level 42  
    krawietz wrote:
    After reboot it doesn't get up anymore but spits some "exception".
    .
    The "some" is actually quite important and it would be useful to post this message here.
  • ADVERTISEMENT
  • #3 20033434
    krawietz
    Level 16  
    I'll only upload when I get to the module in the evening.

    Krawietz

    Added after 4 [hours] 34 [minutes]:

    Wemos spits out:

    Fatal exception (0):
    epc1=0x40100000, epc2=0x000000, epc3=0x000000, excvaddr=0x000000, depc=0x000000
    But of course only when it is inserted in the board....


    Krawietz
  • ADVERTISEMENT
  • #4 20033823
    khoam
    Level 42  
    If the exception code equals 0, this indicates "IllegalInstructionCause". This type of error often occurs when reading flash during boot when there is a problem with the ESP power supply. Such an error will also occur when D4 (GPIO 2) is shorted to ground (even momentarily) during boot.

    There is also flash verification information before the "Fatal exception (0)" message. Could you please post these?
  • ADVERTISEMENT
  • #5 20033893
    krawietz
    Level 16  
    Well... Looks like moisture in the air and flux residue from the tin or possibly insulation from the wires used to route the connections on the prototype board is to blame.
    I checked the resistance of GPIO2 to GND and it came out about 130 Ohms. All due to flux residue between the solder points. So much so that I checked it all after soldering and there was no resistance, anyway the module worked fine for a few hours.... I guess it must react somehow with moisture from the air and forms a conductive layer somehow though. After scraping, everything works. The question remains: what should I wash it with? Isopropanol?

    Krawietz
  • #6 20033967
    khoam
    Level 42  
    krawietz wrote:
    Isopropanol?
    .
    You can, but do not bathe the whole board due to the quartz in the ESP module.
ADVERTISEMENT