Hi All, I’m not an electrician, rather I’m a software developer, but I have a basic understanding of circuits. But not enough understanding it seems. I think I just cooked a Shelly Plus UNI, and I’d appreciate another set of eyes to identify my mistake. I thought I did my research and triple checked my connections, so I think its a misunderstanding of how I think this circuit should work vs how it really works. I recently completed the simpler ratgdo update for my garage door and with that confidence moved on to a slightly more complex low voltage circuit project: My doorbell.
Please forgive the circuit drawings. I imagine anyone it will hurt pros eyes to look at, but I’m hoping it conveys my point without having to dig into a proper circuit drawing tool.
This post was an inspiration-
Simple original wiring-
Chime dual wire: one wire connects to common transformer terminal and another to button wire
Button dual wire: one wire connects to hot transformer terminal and another to chime wire.
Update
For Shelly power connected VAC1 and VAC2 each to a transformer terminal (14V AC)
(Dry contact relay) to control whether a doorbell press will actually ring the chime: Split the wire coming from the button that was connected to chime to one of the OUT1 pair, and connected wire coming from chime that was connected to button to one of the OUT1 pair. (Essentially split that wire and put the Shelly between it by using the OUT1 pair).
(Voltage potential detection) for button press detection: Connected button wire that was connected to chime (and now is also connected to OUT1) to IN1 and connected GND to transformer common. This essentially mimicked what I did with my multimeter to identify which wire pair came from the button & detected the voltage during button press which was successful and measured ~4V on press and 0V normally versus hot side was a constant 14V.
Updated wiring on the left, original on the right:
Upon wiring this circuit the Shelly turned on, and I could connect. I connected it to my internet, updated firmware, and tested hitting the doorbell a few times which wasn’t chiming. I toggled OUT1 on and off a couple times trying to get the chime to chime. The last thing I did on the Shelly was update the IN1 to “detached” mode. I went back to the Shelly and when there used to be a red light (blinking on boot, but constant after a connection) the light was out, and I think I smelled a faint burnt electronics.
Because it worked initially I suspect the act of pressing the doorbell triggered the issue. So something to do with either the OUT or the IN but I think the power supplied to the Shelly should be no problem since 14V AC is within spec.
Anything glaring jumping out to anyone? Ideally I’ll get another UNI to replace this if I can figure this out quickly otherwise I’ll need to unwind this a bit to get the normal doorbell working again.
Thanks for any expertise that can be shared. HA and its community is awesome. I’m more fluent on the software side, so excuse my naivete discussing circuits such as this.
It’s very difficult to understand your wiring description.
Anyway, I don’t own Shelly Uni Plus but I don’t find any evidence you could connect AC to it’s digital input or to common GND.
Also dry contact output works like relay, just one wire connected doesn’t do anything.
Let me know if you need help to rewire it.
“2 Potential free outputs up to 36VAC and 24VDC, max 100 mA each”
I’m guessing that the doorbell will draw much more than 100mA when triggered. You’re going to need a relay in the circuit to offload that load. If there is already a relay in that diagram then I can’t see it.
Joking aside.
I expect that OP just wanted a trigger from doorbell button, not control of the chime.
That can be done multiple simple ways, for example optocoupler+esphome devboard.
Or with Shelly UNI (not plus), that gives wiring scheme for AC input.
So with Shelly Uni you could leave the doorbell circuit as it is and just feed parallel input from the button to IN.
I wanted to get the chime draw to move the discussion forward, but it appears I have damaged the transformer itself. Restoring the original circuit, I’d get .3v rather than the original 3 or 4v when I did the initial test prior to attempting the upgrade. Then I tested the voltage across the transformer terminals themselves which registered at 0v. So I’ll most likely need to get the transformer replaced (which will require calling in backup) which will probably push this past Thanksgiving. I’ll recheck this all tomorrow to make sure I’m on the right track. In times like these its more fun being an expert than a novice.
To the comments & jokes about the use case / usefulness- I realize I didn’t make my desired functionality clear! You do have a point; I wouldn’t classify this project as essential. Rather I got bit by the HA bug.
The goal is two fold:
Detect button presses
Mute/unmute the chime
The main use case is to be able to mute the doorbell and get a notification of the doorbell press instead. For a couple reasons the doorbell can be disruptive. These two things should operate independent of each other (sensing the button press when the chime is both on and off).
Assuming the amperage is under the 300mv required by the Shelly Plus UNI, @Karosm’s diagram makes sense to me. If the amperage is over 300mv then I’m using the wrong tool for the job, and there is another solution that would make more sense…
