Wemos D1 Mini freezes

Moderators: grovkillen, Stuntteam, TD-er

Post Reply
Message
Author
Soundstorm
New user
Posts: 6
Joined: 26 Jan 2019, 17:17

Wemos D1 Mini freezes

#1 Post by Soundstorm » 04 Feb 2019, 13:12

Hi guys

I'm puzzled here. I have 3 Wemos D1 Mini boards, 2 of them have a BMP180 shield, one has a BME280 shield. I'm using all ready made hardware to keep this project simple. All boards have mega-20190202.

All boards function is the same way. The boards send the measured temperatures using MQTT to io.adafruit. There are also some rules implemented, but they don't seem to affect what is happening.
The boards are powered using multiple wall plug USB adapters.

For some reason, the boards freeze after a random time (seconds to days). By freezing, I mean they are not sending MQTT messages anymore and cannot be contacted with the web interface.

I connected all boards using a powered USB hub to my pc, to monitor the serial data and maybe identify what happens when they freeze. Of course, when doing so, they never froze...
I kept using the powered USB hub without being connected to my pc, and they work for much longer periods then when connected to the wall plugs.

Is there anything connected differently when using the USB hub and/or pc compared to the wall plugs? Is the USB-Serial converter pulling some pins up/down and do I need to do the same? I was under the impression the Wemos boards didn't need anything special.

Any tips would be great.

TungstenE2
Normal user
Posts: 123
Joined: 03 Apr 2018, 21:46

Re: Wemos D1 Mini freezes

#2 Post by TungstenE2 » 12 Mar 2019, 14:14


User avatar
ThomasB
Normal user
Posts: 1064
Joined: 17 Jun 2018, 20:41
Location: USA

Re: Wemos D1 Mini freezes

#3 Post by ThomasB » 12 Mar 2019, 16:18

I kept using the powered USB hub without being connected to my pc, and they work for much longer periods then when connected to the wall plugs.
Since your Wemos D1 Mini boards run more reliably with the USB hub I tend to agree with you that it is possibly a power issue.

Unreliable operation on various ESP8266 boards has been caused by "dirty" 5V wall wart supplies. So although you have tried several power supplies, there is a chance they all have poor regulation or high ripple. Sometimes adding bulk capacitance to the ESP8266 board helps resolve it. For example, add 330uF or larger across the board's 5V and GND pins, plus another big cap on the 3.3V pin. If you discover that the caps help you can experiment with smaller values to reduce physical size.

Also, it has been reported that some ESP8266 boards run more reliably with a 10K ohm Pullup on the RST pin. Your board probably already has one on RST, but for the testing try adding an extra 10K to see if it helps.

Be aware that dropped WiFi connections can cause random problems on ESPEasy. This is an ongoing bug. That said, there is a chance that the WiFi reception at the location where the USB hub is located is different than the location you are running on the Wall Wart. So if that is the case, then be sure your testing is all performed EXACTLY at the same location. Even a few inches can make a difference due to the small RF wavelengths.

Lastly, disable Serial Log during your run-time duration testing. When it is enabled it may contribute to unreliable operation due to a full serial buffer's code blocking.

- Thomas

Post Reply

Who is online

Users browsing this forum: No registered users and 28 guests