P092_DLbus UVR42

[Trainer]

Hello,

i have i UVR42 Solarsystem and it would be nice to control it with my esp.

i found the very nice DL_BUS Plugin https://github.com/letscontrolit/ESPEas ... _DLbus.ino.

Can you help me to edit the code so that i can use my UVR42.
As i found in the code it works only for ESR21,UVR1611,UVR 61-3

[Ath]

Do you have a datasheet available, so the configuration can be set correctly?

The structure of the plugin is quite nice, so it's not very difficult to add a new model, but you will need the correct parameters.
Maybe @uwekaditz will jump in and add these parameters, as he added this plugin to ESPEasy.

[Trainer]

the documentation you can found here https://github.com/martinkropf/UVR31_RF ... %201.6.pdf

i don't know exactly which files i have to edit und what i have to edit that it work.

[Ath]

Thanks for that documentation link, that's very helpful.

I can make the changes, as long as you can do some testing, as I don't own that type of device
What build of ESPEasy do you have installed on your ESP (ESP32 or ESP8266)? (It should be a TEST_C build, as that's the only one that has this plugin included, AFAICS).

[Ath]

It was even easier to add than I originally estimated The source is very well structured, thank you Uwe

I've created a pull request #3977, so the builds can be downloaded from the Github Actions service ('Checks' tab in the PR) once the build is completed.

Please test and report your findings here.

[Trainer]

It was even easier to add than I originally estimated The source is very well structured, thank you Uwe

I've created a pull request #3977, so the builds can be downloaded from the Github Actions service ('Checks' tab in the PR) once the build is completed.

Please test and report your findings here.

you are amazing.
Thank you.

where can i find the new source?
https://github.com/letscontrolit/ESPEas ... _DLbus.ino
this one should be the old one?

i can flash the new FW on a separate ESP8266 to test it next time, when i am in front of the solarfocus UVR42.

[Ath]

The source can be found in the PR, as well as the built binaries, as suggested, in the 'Checks' tab there. On the right there is a drop-down titled 'Artifacts' where you can download Binaries.zip that will include the built configurations (get the TEST_C build, as before).

[Trainer]

The source can be found in the PR, as well as the built binaries, as suggested, in the 'Checks' tab there. On the right there is a drop-down titled 'Artifacts' where you can download Binaries.zip that will include the built configurations (get the TEST_C build, as before).

Maybe I'm stupid but I can't find a way to download the new Test_C build. I can't find the sub-item 'Artifacts' either.
Also in the 'checks' tab i can't find any download link

[Ath]

It may require you to have a github account (it's free) and to be logged in, not sure, I'm always logged in

[Trainer]

It may require you to have a github account (it's free) and to be logged in, not sure, I'm always logged in

thank you. now i got it and i will try it next time, i am in front of the UVR42

[Trainer]

today i tried the UVR42 with my esp.

i use a voltage divider (24V INPUT = DL-BUS (~25V)) 10kohm and a 1,5kohm so i get ~3,3V on my GPIO0.

DL---|
|
10k ohm
|--------------GPIO0
1,5k ohm
|
GND-----------------GND



am i doing something wrong?

[TD-er]

GPIO-0 already has a pull-up resistor to 3V3, so not entirely sure what effect that will have on the signal.
Not sure what you would like to achieve with this though, as using GPIO-0 for an input signal may prevent the ESP to boot if the pin is pulled low when the ESP boots.

But just comparing the mentioned resistors to see what voltage will be over the 1k5 resistor:

(24 / (10k + 1k5)) * 1k5 = 3.13V.
So the resistance values seem to be chosen correctly to get roughly 3.3V from roughly 25V input.

[Trainer]

GPIO-0 already has a pull-up resistor to 3V3, so not entirely sure what effect that will have on the signal.
Not sure what you would like to achieve with this though, as using GPIO-0 for an input signal may prevent the ESP to boot if the pin is pulled low when the ESP boots.

But just comparing the mentioned resistors to see what voltage will be over the 1k5 resistor:

(24 / (10k + 1k5)) * 1k5 = 3.13V.
So the resistance values seem to be chosen correctly to get roughly 3.3V from roughly 25V input.

Do you think the problem is with the gpio port?
which one would you recommend?

[TD-er]

Not sure what you want to do with it.
What do you expect to read from GPIO-0?

[Trainer]

Not sure what you want to do with it.
What do you expect to read from GPIO-0?

I would like to read out the sensor data.
I have only taken one remaining GPIO pin.
If another GPIO pin is recommended, I am happy to change it.

[TD-er]

OK, what signals do you need to connect between your sensor and the ESP?
A 24V signal is not really common, so I wonder what kind of signal it is.
Is it an analog voltage you need to measure?

[Trainer]

OK, what signals do you need to connect between your sensor and the ESP?
A 24V signal is not really common, so I wonder what kind of signal it is.
Is it an analog voltage you need to measure?

it should be a 24V digital signal if I understand correctly.
https://github.com/ekuiter/uvr2web/blob ... %201.7.pdf

[TD-er]

Hmm, I see it on the last page of the PDF you linked.
To me it seems a bit strange to see that a serial data link needs such high voltages.

[Trainer]

what else could i try?
Is it possibly due to the 25V I measured instead of 24V ?
Is only 3.3V recognised by the ESP or also when only 3V are present?

[TD-er]

The ESP is rather tolerant, so that's not an issue.
But what might be an issue is that it isn't an UART signal.
It seems like it is a 50 Hz clock and not sure if this low baudrate can be read using a UART.

Can you simply test using the pulse-counter plugin to see if you do receive pulses?
If that's happening, then you know it is electrically connected how it should be.

[Trainer]

The ESP is rather tolerant, so that's not an issue.
But what might be an issue is that it isn't an UART signal.
It seems like it is a 50 Hz clock and not sure if this low baudrate can be read using a UART.

Can you simply test using the pulse-counter plugin to see if you do receive pulses?
If that's happening, then you know it is electrically connected how it should be.

i try pulse count and get some values

Bildschirmfoto 2022-07-03 um 17.10.52.png (99.14 KiB) Viewed 15887 times

Bildschirmfoto 2022-07-03 um 17.09.28.png (43.55 KiB) Viewed 15887 times

[Trainer]

How can i fix it?
For me, the temperature data would be very important, as I have to use up the excess energy.

I can still test the next two days if you have a suggestion.

Thank you in advance.

[Trainer]

@TD-er or Ath can you please help me to solve my problem.
i get some pulses but no temp values

[Ath]

You haven't yet confirmed if your device is actually sending an UART/serial/DLBus compatible signal. If it is using a different, proprietary protocol, not supported by the DLBus/P092 driver, then there is no way of decoding any sensible values from it.

The pulse-counter just detects any signal that changes from high to low, and reverse, so there is no way of extracting any temperature information from that, it merely counts the irregular pulses, that's why it seems to work .

[Trainer]

You haven't yet confirmed if your device is actually sending an UART/serial/DLBus compatible signal. If it is using a different, proprietary protocol, not supported by the DLBus/P092 driver, then there is no way of decoding any sensible values from it.

The pulse-counter just detects any signal that changes from high to low, and reverse, so there is no way of extracting any temperature information from that, it merely counts the irregular pulses, that's why it seems to work .

how can i check if it is a UART /serial/ DLBus Signal without a Oscilloscope?
Since I use a UVR42 and the circuit board says DL, I assumed that this was also a DL signal.

[Ath]

Well, the last image you shared shows a Pulsecounter task, not a DLBus task. I don't think there is a relation between P092 and the Pulsecounter. That's why I asked.

And to have UVR42 support in P092 you will need a build (with TEST_C in the name) from the PR I created for that, as linked above (Just updated it a few seconds ago).
When connecting the UVR42 to the ESP, proper level conversion has to be done, but I think you've got that part covered.

[Trainer]

Well, the last image you shared shows a Pulsecounter task, not a DLBus task. I don't think there is a relation between P092 and the Pulsecounter. That's why I asked.

And to have UVR42 support in P092 you will need a build (with TEST_C in the name) from the PR I created for that, as linked above (Just updated it a few seconds ago).
When connecting the UVR42 to the ESP, proper level conversion has to be done, but I think you've got that part covered.

I have flashed the P092 firmware Test_C on my ESP8266 and set up my UVR42 for sensor 1. Unfortunately, only NaN is displayed.
Now that I am not on site again, I can unfortunately only provide a screenshot next time.

If you have not changed anything in the firmware, I should have installed the latest firmware that you have adapted.
Can the 50Hz signal be a problem for the ESP8266?

[Ath]

If the device is using the DLBus protocol/signal, then the plugin and/or the ESP8266 won't have trouble reading it.
The device configuration just has to be set correctly, and that should take care of things. For UVR42 an extra sensor option had to be added in that configuration, as that code is not yet included in the regular releases of ESPEasy, you should use the .bin file from the pull request, that has the new code. The regular releases don't yet have the option to select the UVR42 device.

[Trainer]

If the device is using the DLBus protocol/signal, then the plugin and/or the ESP8266 won't have trouble reading it.
The device configuration just has to be set correctly, and that should take care of things. For UVR42 an extra sensor option had to be added in that configuration, as that code is not yet included in the regular releases of ESPEasy, you should use the .bin file from the pull request, that has the new code. The regular releases don't yet have the option to select the UVR42 device.

yes, i choose the DLBus Plugin and in the Plugin i choose UVR42 from your Test_C build.
at least i choose the GPIO0 Pin and Sensor 1.

But the values are NaN.

Next time i am there i can send you screenshots.
If you have any further tips, I would ask for help.

It would be really great if I could read out the values.

[Trainer]

Hello,

This ist what i can measuring with my old oscilloscope (50ms/ 1V per devision) bereden GND and DL

[espeasyUser]

Hi all,
does the ESPEasy DL-Bus plugin also support the master device functionality for the Cora DL-bus, i.e. is it possible to directly connect a DL-bus sensor device like the TA Fristar 3 WP to ESPEasy. With "direct connect" I mean without using a TA UVRxy and/or TA CMI.

I want to connect my TA Fristar 3 WP to my smart home system using espeasy in order to read/wrote cora-dl bus variables.

[arducoder2024]

Hello everyone,
I'm interested in this too! I would also like to read out the Fristar 3 WP directly. I would like to do this with an Arduino or ESP so that I can process the data via MQTT.
@espeasyuser: Have you already found a solution?

[Trainer]

If the device is using the DLBus protocol/signal, then the plugin and/or the ESP8266 won't have trouble reading it.
The device configuration just has to be set correctly, and that should take care of things. For UVR42 an extra sensor option had to be added in that configuration, as that code is not yet included in the regular releases of ESPEasy, you should use the .bin file from the pull request, that has the new code. The regular releases don't yet have the option to select the UVR42 device.

i tried it again but no luck. i use a optocoppler to transfer the UVR42 voltage to a 3,3V GPIO signal.
I count Signals an the input but no value shown for my Sensors.
Do you have any hints why i can't see any values?

[Ath]

Again you also have a pulse counter task on the GPIO that's to be used by the DL-bus reader. Can you disable all but 1 task that is trying to use that GPIO, (one of the DL-bus tasks should be the only one enabled), to exclude that as a possible cause of interference?
There are, AFAICS, no other reports of this plugin not working, so I expect something in your configuration isn't correct.

[Trainer]

Again you also have a pulse counter task on the GPIO that's to be used by the DL-bus reader. Can you disable all but 1 task that is trying to use that GPIO, (one of the DL-bus tasks should be the only one enabled), to exclude that as a possible cause of interference?
There are, AFAICS, no other reports of this plugin not working, so I expect something in your configuration isn't correct.

thank you for the hint.
i only add now one device but still the same problem.

Do you see maybe a problem in my schematic?

[TD-er]

100 Ohm as pull down does seem a bit low.
Maybe 1k or 4k7 is better?

Also I don't know if the base of the transistor in the optocoupler should be left floating?

[Ath]

We even have a wiring suggestion for the UVR42 in the documentation https://espeasy.readthedocs.io/en/lates ... uvr42-page
This doesn't have a galvanic separation, but would be the first thing to try 'if all else fails'.