Help with Shelly 2 PM Calibration

[Zaskart]

Hello.

I recently set up a Shelly 2PM to control a pool cover.

The cover moves by starting an air compressor which operates one of two pulleys (controlled by 2 solenoid valves) that drag the cover in the desired directions.

Previously the cover was started by a key switch that connected both the solenoid valves (one for each position) and the motor circuits (in both positions).

I wired the 2PM By connecting both the valve and the motor to the outputs and isolating the motor circuit via another relay (just to decouple the motor circuits, otherwise I'd have a short between the 2 motor output wires, thus between both of the outputs)

Everything works, I added back the original key circuit, no fuss, everything seems right.

The real problem is that I cannot calibrate the cover position. I was able to get a decent range of motion taking the closing and opening times, but, as far as my understanding goes, i can't calibrate my cover, as the compressor doesn't change it's power consumption at all when the cover reaches its ends. No end stop is present or supported by the original cover design.

Is there any way in which I can redefine the calibration process (even via scripting, why not) or any other way I can use to access the state of my cover in term of %, maybe using the opening & closing times?

Also, Any suggestion regarding my circuit organization is well accepted, this is the best version I could come up with, but any feedback is appreciated.

Thank you in advance.

EDIT: I reposted the question to it's proper topic. I'm having troubles navigating the forum as part of it is in my native language and part of it is in German. Also google seems to think there's no need to translate anything and wouldn't help. Feel free to delete the topic, as I seem to not be able to do it. Sorry for the inconvenience.

Edited August 26 by Zaskart

[thgoebel]

Please draw a little sketch which shows the circuit diagram of your appliance!

At first glance, two limit switches have to be added which change the load of the Shelly plus 2PM at the end of the movement by decoupling the compressor relay. 

Edited August 26 by thgoebel

[Zaskart]

This is the relevant part of the circuit.

For clarity, the OP1 uses the normally open side of the relay (so it connects when the Line is presented to OP1), the OP2 The normally closed side of the relay. Again, all of that fuss just to disconnect the 2 Line wires going to the motor, which have to be started in both cases alongside the corresponding solenoid valve.

As far as adding some switches, I don't think I'd be able to add wires that close to the pool rim to have something in the closing side. Is there any way to add both switches to the motor side of the cover? As far as I can understand the motor engages one of two sets of rollers both connected to the same rope, one rolling up when opening, the other when closing.

By the way, a switch that varies the load by a little (for context it was suggested in the original reply) won't work, the motor has a power consumption of 1.2KW and fluctuates a lot when started (800 to 1300W). The motor is wired through 2 Capacitors to stabilize it.

I don't know much more than that (Motor type, stabilization circuit and all that jazz) as it's not my pool and I don't have access to anything but the original circuitry. The power consumption I'm referring to it's the one i can read via the shelly app.

Edited August 26 by Zaskart

[thgoebel]

My diagram:

Annotations:

LS1, LS2: Limit switches

M1, M2: Motor relays (each with a NO contact: m1, m2)

V1, V2: Valve solenoid

Reaching the limit switch, valve solenoid and motor relay is disengaged. Movement stops, power drops to zero. This allows calibration to be effective.

 

 

Edited August 26 by thgoebel

[Zaskart]

Thank you guys for the feedbacks!

I did a bit of research on what kind of switches I could use, and unfortunately both end stop and rotary end stops are not an option, as end stops wouldn't have a place to mount to, both near the rails and near the ropes (I imagined mounting a block to the ropes to trigger the end stops, but the block wouldn't let the rope run), and there is not enough space to mount rotary end stops to the side of the roll.

So I'm back to my original question: Is there any way to program or redefine the calibration function via software using the times of the opening/closing as a guide?

Thank you in advance.

Also, please @thgoebel tell me which tool did you use to produce the diagram above :)

[thgoebel]

(an app for ipadOS)