Recommendation for motorized roller blinds solution (actual motor, like somfy, rollertrol, ebay...)

[NeverDie]

Well, your way would work. I'm just not sure how well it would work.

A magnetic angle sensor is arguably a cleaner solution. There's no need for recurring calibrations. But it is an extra part. So, that's the trade-off.

[gohan]

It's the way the commercial products use since it's the most universal that doesn't require external sensors. I'm not saying it is the best solution ever but just the most widely used

[Nca78]

@neverdie said in Recommendation for motorized roller blinds solution (actual motor, like somfy, rollertrol, ebay...):

Well, your way would work. I'm just not sure how well it would work.

My neighbour has zwave roller blinds controllers and they work like that, using end stop and calculating the position. The first time you need to "calibrate" and it will measure time between the 0/100% positions, then I suppose it's just recalibrating/realigning each time you ask it to go fully opened or fully closed, as he's been using them for a few years and never had to recalibrate. In the end it seems to work very well.

[NeverDie]

Aha, I see it's my fault for the disconnect. Even though this thread clearly states it's about roller blinds, for some reason I was thinking about venetian blinds instead.

[Sander Teunissen]

Disclaimer: I work for Luxaflex/HunterDouglas.

The Hunter Douglas has motors that more or less do what you describe. The easiest way to get this running is to buy into the ecosystem (including the hub). From there you control your shades by telling them to go to 50% or something. This is not time based, but based on an encoder motor.

https://motorisation.hde.nl/products/PowerView/Roller-blinds/M25T.PV-data-sheet/

The PowerView radio system is based on 2.4gHz. And I know for a fact it can be controlled using a device like this:

https://www.nordicsemi.com/eng/Products/nRF51-Dongle

You can buy such a motor at your local luxaflex store (I think as normally we only sell complete systems.)

[dakipro]

Thank everyone, I forgot to check the topic (I was thinking email notification would arrive, but got nothing.).
It looks like most popular/easiest way is to go with using end stop and calculating the position as @Nca78 mentioned above, and this is what I have found that is closest to the commercial solution I am considering.
https://www.m.nu/rullgardin/motorisera-rullgardin-1-motor-z-wave-24v

It is a website in Swedish, but basically they have a 24v DC motor and a z-wave Flush shutter from Qubino (as well as adapter for ikea blinds f.eks.). It costs a bit over 180e per window (without the blinds).
To go with mysensors I guess it would be around 35e for motor from ebay and electronic parts (and printing or making adapters). I guess 60e + time spent on it?
I would not think much if it is just one window, but I need it for five, so if I can save ~100e for window, there might be some math in the mysensors/diy approach .

But then there is the WAF where a nice remote would be a must, f.eks. https://www.m.nu/fjarrkontroller/smart-color-button-philio
which rises the cost in both approaches, but then again these blinds would just have to work, every time.

In theory, the most difficult part is to make a motor controller that would control the blinds just like mentioned Qubino controller does (which I think I can make, with help from community if needed).
But then a nice remote would be needed as well, most likely mysensorified one so that it can talk directly to the blinds, thus still controlling the blinds even if gateway or controller er offline. Remote doesn't have to be that beautiful, but since it would be the only visible component, it is very much desirable. (it can be semi-hidden as well, or a nice casing could be used and converted to mysensors, not that huge deal, two buttons in the worst case)

Because of all this I am highly considering commercial approach, but it would be really cool to have this project realized in diy, to demonstrate how some serious money could be saved (and to save me some money as well :) where some portion would be of course donated to the mysensors again ;) )

[jacikaas]

Hello. I also thinking of my roller blinds automation and i was found this on thingiverse https://www.thingiverse.com/thing:2392856

Its very cheap project for 5-10€/window. With 28BYJ-48 stepper motor you at least need 9v psu, it's even better with 12v.

I also saw, that someone was already discusing in mysensors forum about how to start this motor and there is a code. Of course if you want to use percentage for open blinds, you should modify code and add some typen of encoder on mechanics. https://forum.mysensors.org/topic/3394/curtain-control-node/31

[Yveaux]

@jacikaas Nice find!
However, I think this thread is about outdoor roller blinds, like:

Which require a lot more power to be operated.

[jacikaas]

@Yveaux
Yes, that outdoor blinds need a lot more power :)

[Nca78]

@jacikaas said in Recommendation for motorized roller blinds solution (actual motor, like somfy, rollertrol, ebay...):

@Yveaux
Yes, that outdoor blinds need a lot more power :)

Some indoor blinds too, I will try the solution in your link but I don't think it will work for me as I have a lot of height (> 2m) and up to 1,8m width, it gets pretty heavy.

[NeverDie]

@nca78 Maybe just a regular low RPM, high torque motor would work best for you?

[Nca78]

@neverdie said in Recommendation for motorized roller blinds solution (actual motor, like somfy, rollertrol, ebay...):

@nca78 Maybe just a regular low RPM, high torque motor would work best for you?

Yes there are 220V motors you put inside the tube (same concept than the link given in the first post).

[krejcarek]

Hello, let me shortly give you my 5 cents about this topic.

I have since 1.5 years a roller blind solution up and running. The logic is based upon the solution of @scalz.

It uses standard Becker (similar quality as Somfy) motors 4 wired, that normally has an indoor switch unit. I chose this since significantly cheaper than any radio controlled solution, and absolutely robust. Definitely a plus for something firmly built into the walls.

For control I have hubs for 4 windows each. They include mysensors, Arduino, and an 8 pcs relay board. Enabling of setting individual positioning by counting time. The total run time is adjustable through mysensors commands. This proved to be for me a simple but very reliable method. I can set any percentage, or even stop, and then it reports actual state. The complete ecosystem is within an openHAB controller, thus I can set automated rules, e.g. go down at the local dusk time, or gently wake up just by go up a bit.
Code is here available: https://github.com/krejcarek/mysensors

Overal very happy with the setup. Two issues I experienced:

• some struggle with range, and unreliable transfer in mesh
• the units are in the attic. In winter I experience some unreliable operation, some units "freeze". Can be literally, or due to some solder quality issues, not sure yet what is causing it.

[jacikaas]

@nca78
Yes, thats quite heavy. Need to check it practically. I started to test the motor (now only on the desk), but my 28BYJ-48 motor is shaking with AccelStepper library. With some simple code without included any library - working fine, so I will look why thats happening.

[parachutesj]

@dakipro I do have somfy motors wired to the control panels everywhere. The motors itself are very good. There are also newer ones I think which can be controlled wireless in case you do not have the wires yet or would need to wire it. then it only needs power at the window.

For controlling, I used to have mysensors, but never had very reliable results. Potentially my bad electronic skills and switched to comercial solution lately (Homematic).

for controlling, you might need to look into this:
https://github.com/marvinroger/arduino-shutters

[dakipro]

At the end I purchased and installed z-wave roller blinds from Swedish company m.nu .

So far so good, and if all is great, I will order three more motors (and two knobs/switches) for living room and terrace doors (which should also automatically keep the shades open if doors er open, to prevent damage).

Made a basic node-red flow that uses xiaomi lux sensor to get them down, but ultimate goal is to have them understand when the light is very bright and when it is "cozy", perhaps even knowing to get the blinds just low enough. Found one good explanation on the internet, will check it out (perhaps use multiple lux sensors strategically placed around the room to determine sun position and strength). Any tips and tricks?

(gif of the action, cannot upload gif to the forum https://media.giphy.com/media/5aY6vSwVn1hsdYC6nd/giphy.gif gif is twice the speed)

[wallyllama]

This post is deleted!

[gohan]

@dakipro what did you use as remote controller?

[dakipro]

It is a Smart Color Button - Philio, got it as a package with the motor itself, from m.nu

https://www.m.nu/rullgardin/motorisera-rullgardin-1-motor-z-wave-24v?nosto=productcategory-nosto-3

It is wireless, gyroscope based "button" that works directly with motor itself. They say battery lasts for 6months, but it is usb rechargeable, so not a biggie. It reports battery level to the controller so that is a plus.
As far as WAF goes, it is wireless and looks nice so that is a plus, but to use it it is a bit, well.. different to operate as it is just held by magnet on the holder/wall and has no rotation limitats, so it is not 100% intuitive for first time use as you just turn it and then nothing happens, then you have to wait almost 2s for it to settle and send new position to the shades, and one cannot change direction (at least I didn't find how) so "left" might be counter intuitive as if shades are open or closed, so it is a bit fiddly to operate. Also it has a tolerance rotation, meaning that fine adjustments for the shades are a bit hit and miss, so...
But I am hoping that blinds will mostly work automatically so that would not matter that much. And I didn't find some much better remote dimmer-like controllers for the blinds, so.. it will work for now. Easy to replace I think

[parachutesj]

@dakipro why bother with individual sensors at the windows? You can calculate the sun position (azimuth and elevation) any time during the day and check if it would be possible for the sun to shine into the room. And in second step you only need to know if sun is shining or it is cloudy. Simplest way is using an online weather service. However I noted that this is not very reliable. My solution is based on MySensors:
It is a weather station with some sensors (rain, temp, pressure). I've started with a light sensor but found it very unreliable also protection from environment is difficult.

So at the end my sun senors is quite simple. One dallas sensor exposed to the sun and a second one in the shade. If the difference is high enough, sun is shining. Depending on your geographical location and preferences values might vary but for me (Zurich, CH) 10 degree celsius seems to be a good value.