I am Marco and got my first EmonTX v3.4 back in 2015, it worked like a charm for a long time, but then somehow the RF module was interfering with our electric gate (I guess they worked on the same 433 Mhz freq), so I switched off the emonTX and forgot about it.
Now 5 years later being in forced quarantine I re-discovered this world, and with much surprise I found out that the emonTX can now work with ESP8266, bypassing my aformentioned RF problem. Huge thanks for this, you guys are really awesome.
So I proceeded with updating the emonTX:
I am using the latest firmware of continuous monitoring, uploaded with Plaftorm.io
I am using EmonESP on a NodeMCU v3 (works like a charm, btw), fw uploaded with Plaftorm.io
I am using a 5V DC PSU able to provide 2A for powering the emonTX + emonESP
I am also using a 9V AC/AC adapter for current sampling
I’ve connected 3 x CT of the type SCT-013-030 on CT1, CT2 and CT3 (fyi I already removed back in 2015 emonTX on-board burden resistors to compensate the fact these CTs aleady have burden resistors)
I then configured emonCMS and MQTT and they work fine.
I immediately noticed some anomalies on the readings I got, so after a few days I re-read the wiki and forums a bit better and noticed I didn’t do one important step.
So yesterday I did the following:
I removed the JP2 so I am using EU plug AC/AC adapter for AC sampling
In that occasion I also did:
Wired a DS18B20 temperature sensor to the emonTX board
Re-powered everything, worked like a charm for a few hours.
Now comes the strange part. I noticed at first that I got all power readings to zero - while temperature reading was fine. My first thought went to a potential lack of power - in fact I also added a new temperture sensor to the setup - so I switched to a more powerful PSU: now the emonTX + emonESP is plugged to a 5V DC 2.4A.
But I am once again receving strange readings - please see attached screenshots: in Screen1, I have normal operation and then the next reading doesn’t include Vrms and power. Also the temperature reading goes from a “normal” 21°C to a hellish 69°C!
Then in Screen2 for some reasons I have a message about “new config saved in EEPROM”
In Screen3 it starts behaving normally.
I have two main canditates for these issues now:
bad WiFi connectivity - I have -85 dBm signal, but I am already working on an upgrade
faulty AC/AC adapter - in particular in regards to the zero readings I got
I think 5V DC 2.4 A should be more than enough to power the setup…or not?
I would really appreciate your thoughts/tips.
It is a problem with your ESP8266. It is sending random (I think) characters back to the emonTx, which the emonTx thinks is you using the ‘on-line’ calibration feature.
I do not have an ESP8266, so I cannot prove this.
There are three ways you can go:
My colleague @borpin prefers to use a Pi Zero W instead of the ESP8266, so he will suggest you use that with a script to send the data, or
You can disconnect the input to the emonTx serial from the ESP8266 - this is very confusingly labelled “Tx” on the emonTx’s UART connector, or
You can delete the lines load_config() and 2 × config( .... ); - you will then not be able to change the r.f configuration nor the calibration ‘on-line’ - you will need to edit the source file, compile and reload. I cannot use platformio, it destroyed part of my system, so if you need more help with this, you will need to look to someone else.
I think all I did was install the emonScripts when it asks if you want to continue you say no (otherwise you get a full emoncms) and then just run the emonhub.sh install script. The only change I made was to switch the branch to the Python3 branch as that was partly what I wanted to test.
You then need to configure emonhub, but it works pretty seamlessly.
as promised here is my updae.
I received my second EmonTX, flashed continuous monitoring.
Then I flashed another NodeMCU with EmonESP FW and started monitoring solar inverter.
After a while I noticed the same issues (0 readings) and EEPROM writing.
I therefore disconnected the TX input and that solved my problem.
It seems that this problem was anticipated with the emonESP module - the Tx pin from the emonESP is not connected to the header socket - but this was never documented nor mentioned to us moderators, so we were never aware of the potential problem.
It seems that it is only a problem when you use an ESP8266 not bought from the OEM Shop.