Mount/ enclosure design challenge

  • administrators

    I was supposed to put this post up earlier but have been busy prepping for the TCT show next week.

    We have some beta test PCBs left and as we have already seen there are so many different extruders/effectors/bowden setup etc out there that there are loads of different way that the monitor can be mounted. To that end we are going to offer some PCBs to those people who want to design a mount or enclosure for their machine.


    • The mount is what holds the filament monitor in place on the extruder, effector or elsewhere.

    • The enclosure is what holds the mechanical parts and PCB in the right place so the filament monitor works.

    • There is potential to combine the mount and enclosure into one object for certain mounting options.

    • It is possible to use a different scheme from a hobb to rotate the magnet over the sensor (for example maybe use rubber wheels)


    Design a mount, enclosure or integrated mount and enclosure for your specific printer/extruder/effector/other and post your design on thingiverse/youmagine and link from this thread. You need to show the enclosure/mount on your specific printer/extruder/effector/other. The dimensions of all the components are in the instructions or the Youmagine files:

    After the TCT show the Duet3d team will review the designs and send out the Betas. If we have more entries than Betas we will make a call on the ones we like the most, our decision on this will be final.

    Feel free to ask questions below, there is probably something I have forgotten.

  • I’m in 🙂

    With the existing feedback, I've got few questions :
    #1 does the filament easily fill into the sensor? Just to know if I start before or after the extruder
    #2 could we add pneumatic coupler? ( Sorry if I don't have the name right)
    #3 could it be associated to another mount? It's kind of answered above but just to be sure. Like building a mount for the sensor and a titan extruder for vertical 2020?
    #4 what's the clearance needed between the disc magnet and the sensor ?
    #5 does the custom bolt with the magnet keep the height of a standard M3 hexagon bolt head like 0.7*3mm ?

    Last question could someone generate a STEP, IPT or STL file of the PCB (with sensor), custom magnet bolt and hob ? not familiar with openscad … more Fusion360 ... 😞

    EDIT : found the hob dimensions link in the source of SCAD file

  • administrators

    One point to note: when the extruder drive reverses during retraction, the filament must be forced to reverse through the filament monitor. If the monitor is on the inlet side of the extruder drive, this means that the filament should be constrained in a PTFE tube or similar so that it doesn't have the chance to bend instead of reversing through the monitor.

    I found this brass piece and a convenient way of terminating the PTFE inlet tube of a Titan extruder, in a way that could be fastened to the filament monitor housing.

  • Here we go.
    this version is intended to be mounted directly above the hotend , maube a clean solution but Idon't know if it can be a trouble manage the filament.
    I am going to post a mount for put the sensor right after the extruder this is the reason for the "pneumatic" version
    let me know what do you think about 🙂

  • Here is My version in 1st Phase of design.

    The final version uploaded and completely different than my original design that was along the lines of the original. This model is more along the lines of my extruder filament driver. Utilizes a single bed leveling spring to cause tension.

  • I'm in.
    by the way I have designed a Duet Wifi / duex5 enclosure which can be found here

  • Is this still open or have all be accounted for?

  • administrators

    its still open, I will review the designs later this week!

  • I have completed my first phase of the mount. I intend to mount it inline of a bowden system while mounted to a 2020 rail. I added a way to add bowden couplings to the system.

    If you have any suggestions or questions please let me know!

  • administrators

    Cheers pogo1213, do you have a illustration of how it assembles?

  • I can get that posted tonight

  • Hi
    My attempt at a design for the filament monitor, still a work in progress but installed tested and operates very well.
    Hope you like it 🙂

  • Here is an exploded view of how it will mount to a 2020 rail

    Required Hardware
    Standard hardware for the Filament Monitor
    1-M3x16 with magnet fitted
    3-623ZZ bearing
    1-Hobbed insert
    1-M3 full net
    1-M3x16 socket cap screw
    2-3d printed spring
    Addition Hardware
    4-M3x10 socket cap screw
    4-M3X3mm Female Threaded Brass Insert (heatset them into the Mechanical Enclosure to screw the Bowden Coupling Mount with M3x10mm screws)
    2-M5x8mm screws (to mount to 2020)
    2-M5 T-nuts (to mount to 2020)

  • Got my design ready to post, uploaded it to Thingiverse, went to publish… New users have to wait 24 hours. So my design is coming, but this time tomorrow...

    Details though - this design is primarily intended for my D-bot, which when completed will be running 4 E3D Titans feeding a Kraken. However, it should work just fine on just about any bowden setup using a Titan. It fits in the groove mount in place of the aluminium adaptor and then has a pneumatic fitting underneath to retain the tube, and rather than use a printed-spring-loaded bearing it uses a notched bowden tube and a printed pressure pad to apply pressure to the hobb. Naturally, none of this has been tested though...

    In the mean time, here's a pic...
    I'll be interested to know what you think!

    Edit - now up on Thingiverse…

  • administrators

    Thanks all for the work.

    ssrp - nice and minimalist, might struggle with slightly flexible filaments though. Will be interesting to see. I like the flange bearing for the idler - makes it harder for the filament to slip out. How do you adjust the idler tension?

    pogo1234 - I see now, I take it there are pneumatic style bowden couplers on both sides?

    edwardo - two great ideas in this, a groove mount adaptor for a titan and the PTFE almost completely enclosing the filament. Something like this will probably be really important for flexible filaments. I have not done enough experimenting with flexibles and the filament monitor yet but really flexible stuff is difficult so this is probably part of the solution (another part being a very smooth running filament reel mount so the filament does not stretch at all.)

    I have time set aside tomorrow to assemble and test the last of the beta boards. I am then going to contact as many people as I have spare boards to work out how we are going to get you one.

  • @T3P3Tony:

    pogo1234 - I see now, I take it there are pneumatic style bowden couplers on both sides?

    Yes, I use the grove mount style from here (
    It seems they are pretty easy to get and I have had great success with them. I will add them to the hardware list. Thanks for the feedback.

  • I had some issues today with some movement only using two of the mounting holes on the Smart Effector, I only used 2 due to the clip on the 6 way connector being on the inside. I will at some point reverse the connector to allow better use of that mount hole, it needs to use all 3.
    Optimizing for flexible filaments shouldn't be too difficult, the the ptfe tube was originally further through the bottom of the mount and cut angled to get closer to the hobbed bolt and bearing.
    Tensioning is done by a small torsion spring visible just below flanged bearing. The hobbed bolt and bearing have so little friction i didn't deem that much pressure on the filament to be required, although appreciate it must not slip!

  • Sorry, I am a bit late to the game, been way too busy designing adapters.
    Here is my version:

    It is a simple stacked design. Originally the stacks were identical, but the bowden tube need made it a little more tricky and now there are 2 separate stacks or bearing housings.
    The top most stack has smaller bolt holes so that thread can be tapped in it, to remove the need for nuts on that side.

    The two ends for the bowden tube are the same, so it is reversible making sure the user can always see the LEDs.
    The box that holds the PCB is easy to modify, so that any mounting system can be used. For now I have a simple clip that clips it to the drive cable of the Nimble.
    Doug has printed the version without the bowden tubes and it looks pretty.

  • I have a different design, but that needs to be SLS printed, it cannot be printed using FDM.
    It only uses 2 bearings and the magnet screw with hob, nothing else.
    But I will need to add bowden points to that.
    The big advantage of this design is that the filament can be slipped in, no need to thread it through the housing. In other words, it can be slipped into place after the filament has been loaded.

    If you are still interested, despite the fact that it is SLS printed, let me know.

  • Hi Lykle,

    That's a great idea, I knew there was a neater way to do it, mine has been working but it makes getting to the breach a PITA. I will print one now and try it.

  • Just one thing, the Duet3D version retains the bearing of magnet shaft, in your version it does not. I found this with my version also, and solved it by applying a drop of adhesive to the outer of the bearing to stop it sliding out slightly which brings the magnet away from the PCB and causes it to stop working, maybe worth placing a small flange over the front of bearing block 1 like this:

    Of course, it might be that the bearing will be an interference fit and this won't be necessary.

  • I had a devil of a time getting the bearings in. I doubt if they would walk out by themselves
    But, always good to add some safety measures

  • Cool, I printed it and assembled it yesterday, you're right please ignore my addition above, it doesn't give quite enough room for the magnet shaft to sit above the sensor chip.

    It's much easier to get to the breach and change filament etc.. than with my version.

    I ran calibration but was getting rather vague results as the unit visibly moves up and down when retracting/un-retracting as its only held by the clip onto the drive cable. I might add a second clip to prevent this unit from rotating like this.

    Also how best to attach the Bowden tube at the Nimble end? Is there a mod for a Bowden tube to fit to the nimble? So far I've just been pinching the lower end of the tube behind the breach,

  • Yeah, there is no real solution for the bowden tube to the Nimble, yet.
    This was more the general design for use in any situation.
    For the Nimble I have a different design, that needs to be SLS printed, but clicks in properly.
    How the PTFE tube sits between the two units, I dunno yet. Still working on that, as it was a new requirement for me. I should have realised of course, but hey.

Log in to reply