Finally ready to post my still unnamed IDEX design. I started this design about 4 months ago. The build volume is 300 x 300 x 300 mm. It runs off the Duet Wifi with the Duex5 expansion board and a 5" PanelDue touchscreen. All blue parts are 3D printed. See text and pictures below for more information. I’m not ready to release the CAD files into the wild yet but message me on hangouts and I might give a few select people I trust access. Feel free to ask me any questions.
I’d like to thank in no particular order @raykholo , @Rene_Jurack , @Ryan_Carlyle , and @Thomas_Sanladerer . Without all your help I’m not sure I would have been able to finish this project. I would also like to give credit to the BCN3D Sigma, Herculien, and Folgertech FT-5 because I did take inspiration from some components on those machines.
Some parts are not exact in the CAD and pics. The extruder motors are actually 26 mm long, not 38 as pictured. The LCD shown is for the 4.3" version, I couldn’t find any dimensions for the 5" version so I’ll have to modify the case I designed when I get the component.
Key Features
Reprap firmware combined with the IDEX (Independant Dual Extrusion) capabilities allow for the user to easily print nearly any material with water soluble supports. This is thanks to the independent leveling of each print head and the ability to create idle temperatures to help prevent PVA material from crosslinking.
10 micron resolution for each axis put it on par for resolution with the Ultimaker 3 and Prusa i3 MK2/Mk3
690 x 536 x 586 mm external dimension give it a footprint smaller than the Lulzbot Taz 6 with a larger build volume.
380 x 350 mm Aluminum bed with 350 mm square heater powered with 120V AC with a solid state relay helps the bed heat to 120 C in under 3 minutes. The heater and bed are over-sized so the user won’t have to worry about cold corners.
Build plate is embedded with high temperature neodymium magnets and a magnetic steel sheet is placed on top. To remove prints all you have to do is lift off the magnetic sheet, flex, and your parts pop off without any chiseling.
Deut Wifi uses 32 bit motion control and TMC2660 drivers for fast and quiet motion control
Duex5 board is added for extra stepper. This board has 4 extra connectors for potential future upgrade to add Y splitters to each head
Control board cover clips to acrylic base for easy access to the electronics.
X and Y movement is assisted with high quality genuine Hiwin linear guideways
All belting is genuine Gates 2mm pitch GT2 or GT3
Aluminum extrusion frame with 3D printed mounts for easy modifications
Dual part cooling fans for each hotend
Genuine Thomson Linear rails and bushings for Z axis
All printed parts that attach to aluminum extrusions are slotted for easier assembly
Killer design. Only nit-pick is I’m not sure on the motivation to have the X linear rail on top of the stage? I’d have it on a side to help pre-load it against one edge of the rail (assuming clearance bearings) or on the bottom to reduce the moment between the belt and the block.
I, too, would love to float around in it. Guess I’ll wait till it’s public.
@Adam_Steinmark Looks slick, nice job. I’m in agreement with Mike Kelly that you’ve got too much distance between the X carriage center of gravity and the linear block. The moment stiffness of those blocks isn’t all that high. It’s going to wobble like a pendulum after corner jerks.
@Ryan_Carlyle Huh I thought they were stiffer, I guess that’s why most designs use 2 blocks for a carriage. The belts pull closer to the center of mass so the stiffness of the carriage isn’t as much of a concern but yeah I could see how it might be an issue for sharp corners. I guess I could re-position the rail to the front of the extrusion and move where the belts attach to the carriage, should I add a rail to the back as well or will it not make too much difference? I had the block on top because all the weight of the carriage is supported by the block, rather than being supported by the screws in the other orientations.
@Adam_Steinmark Man, screws are so ridiculously strong, you don’t need to worry about them carrying load.
They’re very stiff for XYZ loads but somewhat less so for twisting/moment loads. There’s just only so much that two sets of ball bearings 9-12mm apart can do to resist wrenching by a 100mm moment arm. Even with good preload, the balls will flatten enough to produce a small angle error that turns into a big nozzle position error. (And/or the extrusion will twist.)
The easiest improvement I think would be to drop the linear block to the bottom of the extrusion and keep pretty much everything else the same. Looks feasible from the renders, but you’d know better than me whether it’ll fit and work that way.
@Ryan_Carlyle yeah I considered adding a rail there or just moving it and it fits but then there’s very little space to slip the belts in. I’m working on a new revision of the carriage now with a rail on the front and back. The belt now attatches to the front of the carriage and the opposite side moves in between the front of the carriage and the extruder idler. I’ll have to move the X motors though and find a toothed idler or at least a smaller idler for the X axis. The redesign will move the head forward about 8 mm so I may have to move a few other components around to keep the 300mm travel and then to keep the lead screws lifting from the center of the building plate. I’ll post just the revised X axis when I’m done with that part.
