Everyone seems to either want, have or use a 3D printer now a days. I bought my first printer kit in 2012 and I quickly noticed I wanted a one with more capabilities but since all of the high performance models were so expensive I designed my own. Eventually I designed a $4,000 printer that I could sell for $1,400 on KICKSTARTER.
There were a few things that I wanted the printer to have: two nozzles, Bowden extrusion, enclosed build area, and a moving xy gantry.
The two nozzles seems like an obvious choice for anyone who has printed before simply because it allows you to create way more interesting prints than you could otherwise. For example dissolvable support, multiple colors and multiple materials (which is pretty cool). I did notice on other dual nozzle printers, however, that having two big stepper motors created a massive moving mass so if I wanted any sort of respectable speed I was going to have to go Bowden style.
Another pet peeve I have is a moving y-axis. There are big debates on the RepRap IRC and even in the forums about what is better for the print, a moving Y or Z. I will stand firm on the fact that moving your print rapidly back and forth is a ridiculous idea. It definitely looks cool and makes life easier in some regards, but having a variable moving mass that is semi-molten does not seem like a good idea…
Now that I knew what I wanted to do, I starting creating a CAD model of the overall design of the printer. This wasn't too difficult just time consuming. I used V-Slot for my frame and thankfully they have Sketchup models that anyone can download so it gave me a good starting point.
Once I started building the frame it became quickly apparent that extrusions require some effort to get them in tight tolerances for a square frame. This was the process that was recommended by Paul F.
How To Cut 80/20 Extrusions (within 0.002”):
2. Make square first cut on each side using the mill
3. Bolt an endstop for the extrusion at the required length to the mill
4. Cut the first piece. Now that you have an endstop at the length of your first piece all of the subsequent pieces should be the exact (or close) length
5. Take your second piece make a first pass then flip it butt it up against the endstop and then cut the excess with the mill.
|Super square frame on a flat sheet of glass|
With my frame done the next step was to create the x and y axis. This was a little bit difficult because I did not just want to copy existing printers. I don’t particularly like the Ultimaker’s X/Y set up even though it does produce great results. I wanted to be a little more creative so I decided I was going to use my frame as the rails for my Y axis. Since I had V-slot already I decided to use it for my X-axis. The details were a little weird to work out, but I think it turned out nicely.
Building the Z-axis was a little bit more difficult because the print bed was so large and I was concerned about the print bed bending as a cantilever beam. I eventually settled on a design using a tri-point mounting system with V-Rails. For the production model, however, I will change the design to something with less flex. For a production model I would focus on making the whole z-axis assembly of only 2 or 3 pieces to minimize play between the interfaces. After using the printer for several months I think a 4 point mounting system is a better choice because with cyclic heating and cooling the print bed appears to sag in the corners where it isn't supported.
|Laser engraved logo on the side wall. Fun Fact: all the radii in|
this design are different ratios to pi
With the basic mechanics completed it was time to add the print bed, nozzles and enclose the printer. The print bed you will notice is actually white plastic and not glass. This plastic is a plastic from GE called Ultem 2300 and I thank the delta printer google groups because they were definitely helpful in brainstorming ideas for a new print surface. Unfortunately nobody really sells this plastic in small sizes so in small quantities it is really expensive. However, it is an amazing print surface so far I can say that it is great for printing PLA, ABS, HIPS, Carbon Fiber, NinjaFlex and LayWood filament. Nylon really should be printed on Garolite simply because it is impossible to remove from Ultem (I may have destroyed a couple prints proving this).
Enclosing the printer was actually pretty interesting because I got to work with a laser cutter. I laser cut most of the parts based off of my CAD but I ran into trouble when making the top cover. The Bowden tube requires a large amount of room so that it doesn’t break the filament or create too much friction so I need a tall top cover. Initially I made a square box which was pretty ugly, so I figured it needed to round the edges. My first attempt was to learn how to bend acrylic and that didn’t turn out so well….
I spent a day trying to get the acrylic to bend at a 90 degree with a 4” radius. Unfortunately I failed pretty hard but that is what Makerspaces are for, failing and learning!
I did some digging and figured that the best way would be to make the bend out of acrylic a laser cut living hinge and I think it turned out pretty well.
The last part and probably the most interesting was the extruder. I needed an extruder that fit the E3D DMfit connectors and I really wanted it to only have one bolt for ease of use. Actually, the first one I made was very similar to the makerbot extruder, without a tensioning screw, but because I am printing in so many different materials it didn’t work since each material needs a different amount of idler tension. It took around 4 different prototypes until I finally created a mechanism that works.
Overall I think the printer turnout splendidly and it makes some really great prints too. Check out the time lapse of a massive D20 below. There are also many example prints on www.lathon.net and the kickstarter . I really enjoyed this project and I hope I can make LATHON’s for other Makers and engineers because, for us, a 3D printer is just as important as a hammer.