Have you set the button type to momentary?
Apologies my bad based on the 100 mA output
-
Check your button to see if it has an issue (i.e. sticky)
-
Change your shelly input to be active high, with a pull-down resistor. Although to be honest I donât know how it works with it being AC. âactive highâ only works on DC. With AC, it it needs to be in relation to the L / N. You can test it by disconnecting the input from your circuit and directly connecting it to either L or N and see which one it detects.
1 Like
You can do it by placing any electronic magnet in line with a wire going to the bell or switch located where your main power transformer is. When the bell is pushed it creates a magnetic field, then use any off the shelf door contact sensor that looks to see a magnet. You will need to find a coil that does not limit the current too much and creates a strong enough field to change the sensor.
You can also do it with a device that measures current used by the power supply. Some where someone did a tutorial on how to detect a washing machine was finished using this method.
Since the Shellyâs run on 12v you may also be able to install it at the switch location inside the wall cavity.
No, right now I have set it to Detached, since I donât have the output connected.
I cannot find any input configuration for the Uni in the Shelly app. It has some for the ADC input but none for the binary switch inputs. Iâll try taking out the switch and directly connecting to L/N today.
Did some further tests now and it really seems the bell is the problem. When I disconnect the bell from the power source (blue wire), then the input on the Shelly goes to OFF, and when I then press the button it goes ON when the button is pressed and off again once it is released.
So it does seem there is some current flowing through the bell to the Shellyâs input, enough for the Shelly to believe the input is ON. Again, my very limited understanding of electrics fails me here, not sure why this happens, but it does.
So, I guess Iâll have to wire the bell into the output of the Shelly, will try this a bit later today.
Rewired it now, but the bell is not working if activated through the Shelly. If I shortcut the wires going in and out of the Shelly, the bell rings, but if I connect them to the Shelly and activate the switch via the Shelly app it does not ring. Currently I have set the switch in the Shellyâs configuration to âToggleâ and set an auto-off timer of 1 second. But I also tried other settings and without the timer, none works.
Any ideas of what I might be doing wrong?
Here is my current wiring:
maybe switch l with n at input or purple at 3 instead of 1
With âswitch Iâ you mean âIN1â? The input side is working fine, the Shelly is powered and detects button presses. Only the output does not work as expected, regardless if triggered by a button press or via the Shelly app, the bell does not ring. But if I directly connect the brown and black wires going into the output side of the Shelly, the bell rings, so the wiring seems to be correct.
So, either it is a software issue and the Shelly actually does not connect the OUT1-L with the OUT1, or there is another electrical issue at play, maybe whatever is needed to drive the bell does not make it through the Shelly?
just switch blue wire from 3 to out 1 L
Did you measure if the output is switched correct ?
You mean like this? Does this make a difference with AC?
To be honest, I would not know how to measure it. I do have a simple multimeter, but it does not seem to have a setting for low-voltage AC.
The bell is a Friedland Type 4, it says max 15W on it. According to the spec, the Uni can handle 0.3W max on those outputs. Could it be that the bell simply needs more power than can flow through the Uni?
ok so output of the uni is just n and l so basically the ding dong has to come straight to that so both wires from it straight in the output. now with the new picture you send itâs possible that you burnt the MOSFET because the Shelly uni is only capable of switching 0.3 w and the ding ding takes 15 watts.
I should of checked that as well.
now going back to your other posts why not use the bticino modules you have in the panel to switch this load instead of the Shelly.
1 Like
Yes, I guess that would be possible. But I would have liked the solution to be as robust as possible.
Hence my first thought of keeping the bell within the conventional circuit, which did not work. With having the bell wired to the Shelly at least it would be only the Shelly that has to work. If I switch it to BTicino, then suddenly the Wifi, openHAB and BTicino become involved, all of which can fail.
Probably I could connect the switch to BTicino as well, but that would again require additional components.
What I wanted to achieve is get the bell signal into openHAB and play a chime on our Sonos, but also that does not seem as trivial as I thought. It can be done via openHAB, but it stops any music being played from Spotify, which is inconvenient. Maybe a workaround via additional Spotify binding is possible but complicated.
My ideal solution still would have been this one, with the bell remaining independent part of the conventional circuit and continuing to work if the automation fails. But with that wiring, the Shelly input always remain ON, it worked only when I disconnected the bell. So for whatever reason there seems to be enough current flowing through the bell to trigger the input of the Shelly.
So while I donât like giving up, at this point I am thinking about simply forgetting about this project and moving on to more worthwhile undertakings. It does not seem to be worthwhile to put in all this effort that end in an over-complicated solution, for relatively little gains.
dont give up remove the ding dong from first out of Shelly and then just set your multimeter to 200v ac ~ and and measure between out1 and out1 l and if itâs 0 something ok then switch it on in the Shelly app and check if you get 12 v~ if yes the output is not broken so what remains is to get something strong enough to switch your dingdong and because you can put all in your electrical board then just buy a contactor
https://www.se.com/ww/en/product/A9C22012/ict-16a-2no-12v-50hz-contactor/
so your out1 and out1l from the uni goes to this contactor A1 and A2
1 Like
I have a doorbell that is wireless and when pressed I pick up the 433mhz signal using sonoff bridge and portisch software so the bell rings and I also send mqtt saying the doorbell has rung and can send email or anything else I want to do with it.
I also have it set up with a photocell unit at the front gate so if anyone enters and breaks the beam the doorbell rings.
I use the doorbell for other things as well eg to confirm the gate button was actually pressed etc.
2 Likes
Thanks for the encouragement! 
I will give this a try in the afternoon, now I need to go for lunch to my in-laws, otherwise my wife will not let me play with automation anymore.
One more thought: could I maybe make the solution with the bell in the conventional circuit work when connecting the switch to the ANALOG IN of the Shelly instead of the SW1/IN1? There I may be able to distinguish then between the little power that flows through the bell all the time and the higher power when it rings?
no that would not work. you need to isolate yourselft from the ding dong so that contactor will help you .
1 Like
Isnât the 200 under V~ good enough? That should measure AC up to 200V right?
1 Like