Mishap Duet 0.8.5

  • Finally finished building my CoreXY!!!
    The bed is leveled, end stop are up and running, I am almost shaking while getting the First print ready (cube), next is the enclosure for the board… it's temporarily on a block of wood... on the aluminum base...

    I didn't pay attention and it fell on the aluminum base (printer base) caused a short and blew...

    Expensive mistake, I could have avoided it with a sheet of paper, cardboard,...

    I now need to order a new board a 0.8.5 or another compatible board with with the Duex4. I know that I could go with the WIFI but since I already have the Duex4... (next project using the diamond RGB...).

    I was told some one might have a spare used one... it was meant to be a small budget Cartesian printer... I kind of went over board... just a little bit...

    Thank you

  • Nice,

    Did you mill the aluminum top plate?

    I am thinking about building something very similar using the mecadunio extrusion kits where the belts are inside the T slot but with linear rails like you have done.

    Just one difference is that my build plate will be stationary with the printhead all on the xyz axes. The 4 corners of the Z will be linear rails as well.

    I might not even do belts if i can afford the igus lead screws and linear rails. (have not started pricing it yet as i am busy with two other builds at the moment)

  • My top plate was CNC. to fit rails you don't really have a choice.

    For the part below, do not feel offended, I do not know your plan and I am curious by nature.


    • For a CoreXY car you require 2 straps at 2 different highs, I used a twisted strap idea that was suggested by Filipe Campos in the RepRap forum (not my idea). Can you fit 2 strap in the extrusion slot?

    Not sure:

    • For the rail at 4 corners, I am unsure if it's a good idea… rails are very good but also very stiff so IF everything is not well aligned, it might get stuck or cause resistance. I was "lucky" in that I was using some socket cap screws that the cap were long and they fit in the extrusion slot almost perfectly, made the alignment easier (on the heatbed). The only easy way with this would be to your whole support system milled or laser cut (can be expensive).

    • If you plan to have you whole XYZ axis on your 4 rails you will require a ballscrew, the weight should be too much for a basic threaded rod (especially if you plan to have your top plate mounted on your rails.

    • If you put the top plate within the 4 rails you will loose a lot of building room (at least the size of you car, hotend & strap system (if don't fit within slot).

    Hoping this helps

  • Thank you for your feedback, this model is what I was looking to build after,this company has kits for all types corexy, cartesan, etc:


    They did not use any rails or screws, all belts.

    I am not sure if you are familiar with the igus screws (same company that makes igus lubrication free bearings and bushings)


    but here is a concept that I was thinking about integrating into the above design:


    They claim 400mm/s

    Here is a similar concept:


  • P.S. I do have a CNC machine… in the process of building out this fun guy, it is just not commissioned yet

  • If you are interested the machine itself is $3500 before electronics. All steel construction with profile linear guides it is a 4' x 4' build area they do have a cheaper 2'x 4' model the machine is a Saturn from FinelineAutomation (no affiliation just a happy customer)

    They do sell plug and play electronics packages from CNCrouter parts but I went the custom route using Teknic Clearpath SDSK servos and UCCNC UC300ETH controller.

    I have a build log on CNCzone if you are interested in more details as to not derail the subject of your post

  • That looks like a nice project.

    I do not know much about Igus or polymer lead screw, but generally speaking, screws might tend to bend, then again the X-Y Z axis are probably lighter then the heatbed…

    I am curious on how they calibrate the Z-Axis... 1-2... can't be 4 stepper motors and the strap system... can 2 stepper motors control the Z-Axis on 4 slide while remaining calibrated? I could see this with 2 ballscrew, or with 4 motors, but then that should be complicated to calibrate...

    Thank you for the intel, I like what I just saw.

  • Mechadunio, not the same as the kit company… it is a board that turns an ordinary stepper into a servo. there is a thread on the google group where some Duet owners are working on implementing it. So far they report positive improvements over accuracy with a few issues like auto calibration to be worked out.

  • another option i seen on a leapfrog at our local makerspace was a single motor driving a belt that ran 3 screws for the Z. it isnt a very accurate printer though and i have not yet dived into what needs to be improved on it.

  • as far as screws bending, If done like the examples it should not matter. the screws are only a means to apply motion to the Z while the linear guides on 4 corners provide the stability. If you have a scew next to each guide it should keep everything in check.

    You can see this concept to the extreme if you research Project Escher and take a look at the commercial offering of Escher from Titan 3d…. that printer runs 4 independant xyz carriages over the top of a solid fixed build table.

    They do this with profile linear rails that the xyz carriages sit on and glide on. the motion is then done by having a servo attached to each side of the carriage with a 3 to 1 gear reduction pully which pushes the carriage along the screw.

    the screw itself in that machine is stationary, by not moving the same screw is able to be used to drive each carriage independantly.

  • I am not as knowlegable as most of these guys but I am a fast study and love taking on projects that push the envelope and look for ways to inprove overall design. So I plan on producing one of these mecaduino units and eventually a Project Escher type printer as well. But as always it will be as funds become available

Log in to reply