The PCB will control a garage door controller. It will basically acts as an "intelligent" switch, augmenting the one used in the typical Craftsman garage door to manually open the door.
So I opened the craftsman motor cover and found two limit switches with a middle contact endless screw (ground) that indicates the main controller when it is Closed, in between and Opened. Closed or opened is when the gnd contact touches one of the limit switches (0 volts). In between you get 5Vdc, I don't know how many amps.
So I thought it would be simple to connect a optocoupler isolating both worlds (Craftsman controller and Esp32) by having on the Optocoupler (4 ports btw) the CLOSED switch connected to the Anode and Craftsman GND to the cathode(same for the Open switch another port of the Opto) and the esp32 section with emitter to gnd and collector with pullup to GPIOxx.
The setup was tested in the "lab" with Hals sensors acting as the Close and Open switches (did try real limit switches same results). 3.3V, GND from the ESP32 and the return signal to a GPIO. It worked flawlessly. It's going thru the Optocoupler thought it does not need it, but I thought it should reflect as much as possible the real scenario.
So when I connected the Craftsman three signals (CLOSE to Anode1, OPEN to Anode2 and it's Gnd to opto GND1 and GND2) the following happened:
First it would not start the Motor. Disconnect the Open or Close wire and it starts...
Tested when the 2 wires are connected and I get 0V and 1.6V.
If I disconnect one end the voltage goes to 5V I guess the Crafstman controller needs 0V in one end and 5V at the other end to start.
I tried this many many many times. No shorts or anything.
Changed optos anode resistor from 1K to 10K to 100K to 0 same result. Checked the opto Led section, all ok. Not blown.
Then bypassed the Opto and did a Voltage divider (to get 3v) and connected to the ESP32 GPIOs (open and closed), exactly the same problem, no start, voltage drop.
Any guess as to why the voltage in the Crafstman controller drops so much voltage, hence it does not activate the Opto or directly the GPIO (will be detected as 0 digital).
Did read about CTR in opto but it makes no sense.
I could send the schematic of this section but its really very simple (TLP281-4 opto and also 4N25s same result).
Can it be done and how?
Help me Obi Wan Kenobi