New IKEA Fridans motorization solution

I have created yet another IKEA Fridans motorization (YAIFM). My goal was to have a “drop-in” replacement for the orginial “magic wand”:

  1. Remove the wand,
  2. Pop in the motor assembly,
  3. Profit.

EDIT: here is the link to the GitHub Repository: GitHub - AndBu/YAIFM: An affordable drop-in motor and control system for automating IKEA Fridans blinds.

in-action-compressed

It all started because in my home, I was not able to move the brackets that hold the blinds in place as space was limited. As a consequence, the motorization had to fit into exactly the same space as the original manual “magic wand” (also, I didn’t want to unmount anything). I ended up using a DC-motor with a rotary encoder that just fits into the shaft of the blind.

You can see some details of the solution (and the main parts required) here:


Mounting and unmounting is quite simple:

drop-in-compressed
drop-in-2-compressed

Further design goals were:

  • Integration via ESPHome
  • Powered from a single 5V USB-C source
  • Worldwide availability of components
  • All plastic parts 3D printable on a stock Ender 3
  • Standard tools for assembly (screwdriver for the plastic parts, soldering iron for the electronics, crimping tools for the connectors if you are fancy)

Please let me know of your are interested in my solution, I am happy to upload the design files and document a little more. There are also a few quirks that still need to be ironed out so I am happy for feedback on the design :).

Cheers, AndBu

12 Likes

Very interested in any parts list and design file you can share!
I want to have something like this for my living room windows. And automation for when i watch movies to close it😎

Cool to hear that my design might be useful to you :slight_smile:

I will need a couple of days to wrap things up but I will upload the design files to Thingiverse / Printables and post here again.

Also: have you already checked out the existing IKEA Fridans motorization solutions? There is one which uses a stepper motor in direct drive configuration…:

…and there is one which uses a stepper motor with a 3D printed gearbox:

Cheers, AndBu

P.s. I originally planned to integrate a miniature geared stepper motor in my solution (instead of the DC motor) but I scraped this approach. Stepper motors have a far lower power density than high-speed DC motors. I was just not able to get enough torque out of the ones I bought.

1 Like

Yeah saw some of the others aswell, thought yours was more compact and liked that you are using a limit switch it seems😊

Sorry guys, I haven’t forgotton this. Sometimes life just happens in between. I should find the time to document this and upload this soon :slight_smile:

2 Likes

Oh, please do :slight_smile: Your setup is very clean, I’d love to follow your tutorial!

Can’t wait! This will be super useful. If you’re short on free time, would you be able to share the STLs and esphome config w/ us? Worst case, we can figure out the rest :slight_smile:

1 Like

Please make this a reality, I’d love to use this!
Been waiting on it for a while.

I am almost done documenting it. I changed jobs in between and my wife and kids became sick and then you know … skiing season started :roll_eyes:.

I am planning on uploading most of it tonight. The software is just a proof-of-concept until now with some harcoded stuff, but at least you can start with the hardware as that is tried and tested. I don’t know how many design iterations I went through over the past year :sweat_smile:.

Sorry that I kept you waiting for so long.

Cheers, AndBu

… aaand here you go:

Be careful with the first assemblies, there might be some errors in there so unless a few more people confirm that it works, use your brain before switching things on :grin:.

Cheers, AndBu

2 Likes

Really cool idea, i’m back and forth between this and Automated roller blinds (IKEA Fridans/ESPHome/HA/MQTT/HomeKit/HomeBridge/REST @ 200cm wide) but i’m not sure which might be the better solution, they both seem from the videos/gifs to be of similar speed.

I’m concerned with the sound from both even for the other one i was considering using a forked esphome with TMC2209 drivers to make it silent, and maybe even figur eout a way to do sensorless endstop for closing the blind, but not sure if that was possible.

And for this one i’m worried as you mention that it’s got a continuous run issue with heat, had you considered maybe using a brushless motor?

Either way really cool project!

The heat issue is purely hypothetical at this point as I don’t want people to print one of these in PLA in the desert, place it behind a closed window with aircon off and then have it fall on their heads. I have never had any heat issues myself (24°C ambient) and if you worry about heat, print it in ABS or similar. This is for automating blinds so you are anyways not going to roll it up and down 10 times in a row, are you?

The noise is audible, though. There is no denying in that. Brushless motors require more sophisticated electronics and are potentially harder to source in that size. The noise comes from the gearbox, not from the motor so a brushless motor would not have been of help.

In then end, build what suits your needs … and show us your build when you are done :).

Cheers, AndBu

Have you thought of trying to grease them?
Oil might help too but grease will most likely absorb gear noises better then oil and sticks for a longer time.

The gears and the motor come greased. The motor itself and the gears run pretty silent on their own. As soon as you mount them into a stiff plastic housing (and apply torque), the vibrations get transferred to the housing and the housing functions much like a speaker. This is how the noise is emitted.

Ok, maybe something else to try: use the inner tube of a bicycle (or car) to cut slabs and use this as damping material to mount between the mechanical parts and the housing.

Yep, that sounds like a good solution to me, or some shrinking hose.

Yeah, shrinking hose might do the job too.
With a bicycle or car tube, I think it, absorbs the vibrations better since it’s rubber and thicker.

The only thing is: does your enclosure still fit everything with a layer of rubber…

Unfortunately, it won’t fit. Tolerances are really tight. I am not saying, it can’t be done, but the current version doesn’t foresee that.

Update: just uploaded step files for making derivate works easier (was requested via GitHub). If anyone builds this, I’d be interested to see results here. For me, it still works flawlessly!