Options for integrating motorized windows

[Cyknight]

11 hours ago, DLite said:

Thanks, @Cyknight. Just to be clear, this is for windows, not blinds.  The two windows in question are awning windows that are hinged at the top and opens out from the bottom.  Each window has a DC motor installed, with wiring running about 15 feet from that motor to the window itself and wiring running about 5 feet from the motor to a wall switch.  
 

If I am understanding you correctly, we would need input sensors so that manual switch activation results in C4 controlling the window. And, effectively, the actual control of the window would be exclusively through C4.

Am I right though that all these issues go away if I get rid of the manual switch and maybe replace it with a C4 keypad that is bound to the Shelly relay? Or are there additional complexities still to navigate?

 

shade, awning, window the principals are the same.

No, the issue I see more is that the shelly can't even DO the reverse polarity in and of itself to begin with. It's got two relays, but you're not energizing one line or another with a shared common/neutral, you actually need to 'make' 1 side hot and 'make' the other side neutral, and reverse that action to go the other way.

So not only is there the issue of shorting if you parallel the wires to the switch: there's still the actual control itself.

Looking at the pictures you put up, I can't quite tell how the wiring is likely to be. Which makes it problematic to say how you could approach it.

Perhaps you can answer this clearly though.

First the meanwell piece you posted is NOT the motor: it's a power adapter.

So, is the wiring like this:

IGNORE the line voltage 110/220 for now

Is there a low voltage wire from the power adapter to the switch, then a wire to the window (motor)?

Or is it power supply to switch, then back to powersupply location and on to the window (motor)?

From there we can maybe figure out what options there are.

 

[DLite]

3 hours ago, Cyknight said:

shade, awning, window the principals are the same.

No, the issue I see more is that the shelly can't even DO the reverse polarity in and of itself to begin with. It's got two relays, but you're not energizing one line or another with a shared common/neutral, you actually need to 'make' 1 side hot and 'make' the other side neutral, and reverse that action to go the other way.

So not only is there the issue of shorting if you parallel the wires to the switch: there's still the actual control itself.

Looking at the pictures you put up, I can't quite tell how the wiring is likely to be. Which makes it problematic to say how you could approach it.

Perhaps you can answer this clearly though.

First the meanwell piece you posted is NOT the motor: it's a power adapter.

So, is the wiring like this:

IGNORE the line voltage 110/220 for now

Is there a low voltage wire from the power adapter to the switch, then a wire to the window (motor)?

Or is it power supply to switch, then back to powersupply location and on to the window (motor)?

From there we can maybe figure out what options there are.

 

Sorry for the confusion. Yes, the motor is at the window itself, which is a good 25-30 feet off the floor -- hence, I wasn't able to snap a picture.

I should have a full set of pictures later this summer, at which point we are scheduled to have the windows removed and reinstalled (for reasons unrelated to the motors).  Thanks very much for all the advice here.  Hopefully I can prevail upon the good folks in this forum to take a look at the details this summer and help me think through a smarter way to design the motorization and integration.

 

 

[booch]

So in order to redeem myself from my bad initial advice (ha) and because I hate unsolved problems, I thought about this a little more and think I have an easy solution -- combining the Shelly Plus 2PM with a 2 DPDT relay module (which are availible on Amazon for $10). See wiring diagram below:

Notes:

• I got my engineering degree from YouTube, so the graphic is a little rough (ha)
• This makes some assumptions about the existing wiring and switch capabilities which should be validated. (If the existing switch can't do it though, many can.)
• I based the labeling on the specific relay linked above (because its wiring diagram was posted vs. in the box), but if the motor draws >1A, a different model will be needed. There are many availible with greater capacity that will work identically

Cynight -- what do you think/anything I didn't think of? Surprised I hadn't thought of this before, as I love Shellys and this combination makes them much more capable.

[Cyknight]

This could be simplified quite a bit (and I feel there's an error but I can't put my finger on it).

I don't have access to my Visio right this moment so I can't promise when I could create the diagram.

I would suggest TWO relays as you could regain any space lost by using half the wires. Pay a bit more and you could get small format easy wiring ones and even be able to stuff it all into a single gang box, though it may be worth changing it to a two gang and use a blank plate, or with a 4" square box with a nice single decora cover for it (flat square to single decora adapter and a GFCI adapter cover plate are usually available at local hardware stores, there's more options online) to give more space. Say something like these for the relays: https://a.co/d/hhN97Lw

 

The relays would then be simpler SPDT relays, and you connect the + to the NO on both, - to the NC on both, while COM goes to the motor. Now the shelly AND the switch could be wired to the activation inputs on the relay in parallel safely.

Shelly out 1 and switch UP go to SPDT1 and Shelly out 2 and DOWN go to SPDT2.

As the NC state means the motor is connected to - on both ends, it means it's stopped. Activating Shelly 1 OR UP on the switch would connect + to one line, causing it to go up (you'd have to figure out which is which by testing but easy enough), while the other option now puts + to the other side, and make the blind go down.

In this scenario, should for whatever reason BOTH the SPDT relays get energized, all that happens is that both sides of the line to the motor get + connected, which in DC doesn't matter as there's still no active path for the current, and there is no shorting happening.

 

[booch]

12 hours ago, Cyknight said:

This could be simplified quite a bit (and I feel there's an error but I can't put my finger on it).

I don't have access to my Visio right this moment so I can't promise when I could create the diagram.

I would suggest TWO relays as you could regain any space lost by using half the wires. Pay a bit more and you could get small format easy wiring ones and even be able to stuff it all into a single gang box, though it may be worth changing it to a two gang and use a blank plate, or with a 4" square box with a nice single decora cover for it (flat square to single decora adapter and a GFCI adapter cover plate are usually available at local hardware stores, there's more options online) to give more space. Say something like these for the relays: https://a.co/d/hhN97Lw

 

The relays would then be simpler SPDT relays, and you connect the + to the NO on both, - to the NC on both, while COM goes to the motor. Now the shelly AND the switch could be wired to the activation inputs on the relay in parallel safely.

Shelly out 1 and switch UP go to SPDT1 and Shelly out 2 and DOWN go to SPDT2.

As the NC state means the motor is connected to - on both ends, it means it's stopped. Activating Shelly 1 OR UP on the switch would connect + to one line, causing it to go up (you'd have to figure out which is which by testing but easy enough), while the other option now puts + to the other side, and make the blind go down.

In this scenario, should for whatever reason BOTH the SPDT relays get energized, all that happens is that both sides of the line to the motor get + connected, which in DC doesn't matter as there's still no active path for the current, and there is no shorting happening.

 

Ah -- I'm tracking -- hadn't occurred to me to use the COM ports in that way. And yeah, that is a little bit simpler, ha.

Updated diagram below to save you from Visio. Note that I listed the through-Shelly switch wiring as optional, as it'd allow for button tracking/integration in the driver (which I'll want to do when I go to reference this at some point.)

Feel like my relay game is a lot stronger now!

[DLite]

10 minutes ago, booch said:

Ah -- I'm tracking -- hadn't occurred to me to use the COM ports in that way. And yeah, that is a little bit simpler, ha.

Updated diagram below to save you from Visio. Note that I listed the through-Shelly switch wiring as optional, as it'd allow for button tracking/integration in the driver (which I'll want to do when I go to reference this at some point.)

Feel like my relay game is a lot stronger now!

Very cool.  Will any DPDT relay module -- say, on Amazon -- do the trick here, or are there specific ones I should be looking at?

[Cyknight]

1 hour ago, booch said:

Ah I see what felt off in the original now, I hadn't checked the linked relay and the way it was (and is) without knowing it has a dedicated ground (and not a DC- which you marked it) makes it appear as if you're permanently energizing the relay. Do also note that on the linked relay there, you would be using the xx1 and xx3 connections, not xx1 and xx2 (or xx3 and xx4). The linked board is actually TWO DPDT relays on a single board (and you'll be using the as two SPDT relays).

Note that a Dual Pole relay is still a single activated relay - it just isolates two circuit that get switched at the same time. While you could use that to reverse polarity, it wouldn't allow you to reverse it AND have it disconnected.

 

As a small note, you could use a single DPDT relay to reverse polarity, and a SPST (or any type of) relay to cut the power to the circuit in series (as in the SPST cuts the DC+ connection in the diagram above to the DPDT relay, but you would have one NO/NC combo hooked up +/- and one -/+.

 

Oh and yes, if the Shelly has two sensor inputs, you could wire the switch to the sensors so the system can keep track of the state, or you could wire parallel which has the advantage of 'back up control' - even if the system is down or the shelly gets fried, the switch keeps working (not that there's a lot of risk of that).