Thank you again, looks like it should work in theory as I have similar distance than you between electrode and glass, and bigger pads.
I'll try to replace the TTP224 IC on my test board, maybe it's defective/damaged during soldering. I have the same markings than you on the ICs I bought.

Thank you.
What voltage are you using ? Can you try with 3V or at least 3.3 ?
And what is the distance between the electrodes and the glass ? Because on the switch I use there are a few mm, it's the thickness of the conductive foam + plastic reflector that you can see in the description in openhardware.io.
I am not using any Cs capacitor.

@Baoshi-Zhu said in 💬 Battery powered glass touch switch:

I'm doing something similar but using ESP8266. Saw your article and I had to order a TTP224 touch module to test. I found it is ok. I need to remove Cs to increase sensitivity. Now the button activates when my finger is 5mm away from the pad, even without glass.

That's interesting, can you show the PCB you are using ? I've done something wrong with mine but I'm not sure why, seeing a PCB that's running fine would surely help me. Best would be board file, but picture would be fine also if you don't want to share your board.

@mfalkvidd said in using pins A6 and A7 of arduino pro mini to read push button state:

You can perform a digitalRead on an analog pin if you want to though, see https://www.arduino.cc/en/Reference/DigitalRead
Not on A6/A7, they are exclusively analog pins.

@NeverDie if I remember well in the datasheet they say you have to put the softdevice on it first.

I've been using Domoticz for a while, because it was the most obvious system to use with my RFXTRX433 and my 433MHz modules.
I agree with what has been said: very easy to setup, nice graphs/history but limitations in UI are very frustrating. There's a nice API too, but I'm not feeling like reprogramming the full UI to have something that I enjoy.
I'm getting frustrated with the programming of events too, for very basic scenarios it's fine, but the way it works can quickly make it a mess. For example if you switch a light in a script it will re-trigger scripts based on light switching and you have no way to know if it was triggered by user action or by script, so you you have to start using global variables (declared in another part of the UI) to keep that information etc etc

So, I'll start my controller shopping season too relatively soon I think

@Jonathan-Roberts said in 💬 Simple AC/DC Switch WeMos:

@nca78 bummer. So I think 2 options...

1. What if the Wemos is raised off the PCB. That would be almost 5/8". In your experience would that help?
2. And maybe the best option, slide the antenna module straight down,. move the relay up and spin it around so the connections are on the outside edge.
   I'm trying to make this multi-purpose and allowing both headers to be fully functional to other projects. thx for the advise.

1 would help but not that much. It's already not a good thing to have the transformer just next to it on the side, so you should avoid blocking one more direction with tracks.
I don't understand your necessity to have connectors available for something else on the relay that is too high to put anything on top ?

2 sounds like a much better option, make sure you have nothing right and left of the antenna when you move things around.
The problem you will have with this layout is the USB in that situation...

It doesn't sound like a great idea to me to use Wemos on this board especially for the relay part, it's just a cheap chinese relay with 3/4 components so why not just put them directly on you board to make the layout so much easier ?
I have the feeling that you're trying to make things "easier" by using Wemos and its shields, but in the end you seem to lose more than you gain. And neither does it seem "safer" to use their relay shield as it forces you to have extra wires between main of the HLK and relay shield connector.

I just had a look at your design and I'm afraid you'll have to change all your routing between wemos and relay.
You cannot have tracks below the wifi antenna, this will make the radio reception extremely bad.
Check the pins that are necessary for the relay board and only connect those to the wemos board (from the pictures it looks like you only need power + D1, but check it I'm not sure at all !). Then route them going around the antenna, staying as far as possible from it.

@Jonathan-Roberts I just checked EasyEDA and you can chose the "Panel by EasyEDA" option, if everything is less than 100*100mm (I suppose it means only 2 copies/PCB for you) then you will pay only 2$.
EasyEDA is nice but with shipping cost it's not so cheap, over 10$ for first PCB and after that other PCBs cost 8$, and you have no option for color, thickness, etc...

This is getting really small, it's great to get power consumption low enough to be able to use a CR1632 for a long time.