Mark Ganter and his team at the University of Washington Mechanical Engineering Department’s Solheim Rapid Prototyping Laboratory just cut the time it takes to make a RepRap Prusa Mendel 3D printer. Instead of the usual week it takes to output all of the printable parts that comprise a Prusa Mendel, Ganter and students, Scott Tandoi and Travis Nicholes, created a set of silicone RTV molds to produce the parts in a mere fraction of the time. He states that all of the plastic parts for a complete Prusa can be produced in under 30 minutes. You’ll still have to finish some parts with a drill press, but to have all the parts ready in under a hour, rather than a week, is pretty impressive.
They’re calling their version of the Prusa Mendel a Clonedel and plan to “release working STL files of the mold plates to the community at large (hopefully, within a week).” [Thanks, Matt!]
30 thoughts on “30 Minute RepRap Clone”
Comments are closed.
ADVERTISEMENT
Join Make: Community Today
Snap together lengths of plastic rails would help make more of the machine replicable if they can be made sturdy enough. Also would allow for different sizes of machines by snapping together more sections to create a rail.
I think the conditional is the snag in your statement: They can’t be made
sturdy enough. Or, I should say, lots and lots of people have tried and non
one has yet succeeded. Eliminating the need for precision-machined metal
shafts and/or threaded rods is one of the biggest hurdles toward full
self-reproduction in these printers, but so far nobody has found the way
’round. Doesn’t mean you couldn’t be the one, though, so don’t let me
discourage ya.
Snap together lengths of plastic rails would help make more of the machine replicable if they can be made sturdy enough. Also would allow for different sizes of machines by snapping together more sections to create a rail.
I am pretty confident these molds are made of silicone, not silicon That would make them pretty fragile.
Aye, but an easy typo to make! Fixed it! TY!
So this is more of a RepStrap than a Rep Rap, if you’re going by “official” taxonomy. Still, this is a good way to get a larger number of “seed” RepStraps out there to gain overall momentum.
The Economist features 3D printing on its title page and in several articles in this week’s magazine. http://www.economist.com/node/18114327?Story_ID=18114327
It’s pretty widely known in the reprap community that molded parts are more prone to failure than standard printed ones. Printed parts, by nature, are very robust due to the many individual lines of plastic interlacing the part. A molded part on the other hand can fail with a single defect in the structure.
Also, Prusa is an 8 hour print, not a week.
I would like to challenge this “belief”. Where is the tensile test proof? Show the data. There is a whole engineering community attempting to determine just “how strong AM really are”. ABS has around 5000 psi tensile strength, but it is easy to use pourable resins past 10000 psi tensile.
As for the layer-to-layer interlacing, I’ve seen many AM parts fail due to a single layer-to-layer adhesion failure. Can we use some science to remove the myths?
I was merely relating the fact that the majority of parts available on ebay that are molded are of inferior quality. I am sure there are very robust materials out there that are suitable for molding RepRap parts. The problem is that people currently making the molded parts are producing an inferior product.
If you want stress tests and other scientific data the burden falls to you. I, personally, do not have the equipment to test tensile strength of a product. Additionally I don’t feel inclined to spend $150 on inferior parts someone is producing in the current market just to prove something that people have observed when using these parts.
If you can find a pourable resin that has the properties you describe that can produce an entire set of Prusa parts for under $50 usd then you may change the way people make parts. Otherwise people will stick with the tried and true method of printing.
Produce a part using a mold and do the tests yourself if you are want people currently in the reprap project to accept it. The whole spirit of the project is people doing these sort of tests and releasing their results to the community.
tried and true? Uuuuh, people have been molding plastics for several decades. Many molded plastic parts are integral in items you use daily, like a car, power tool, and even your smug machine.
Listen, 99.99% of Mendel-variant users are doing it for the fun, experience, and an effort to try new things.
Noone is making parts for NASA, much less producing items designed for moderate tolerance and stress.
I have to say this claim is counterintuitive to me, as well. Which doesn’t
mean it’s false. But are there data to back it up? If not, surely the
claim of “wide knowledge” has to be taken with some skepticism. Also,
“molded part” encompasses a potentially vast array of materials, some of
them potentially composite. Does the claim against molded parts hold for
all the convenient room-temperature molding resins? What about if they’re
reinforced with glass fiber fill? Etc., etc….
As for the threaded rod, Lowes sells 6 foot sticks of 5/16″ for $3. Is it really going to get cheaper?
Also, these cast parts remove the SAE/Metric debate as you drill the holes you want. You are allowed to make the choice based on what is available to you in the commodity market.
People have been selling molded plastic bits for the traditional “Sells” Mendel for some time and most people who have tried them say that they’re of much lower quality than a printed set. This isn’t really anything new.
Moreover, this doesn’t solve anything that’s really a problem. A well tuned Prusa can print three copies of it’s plastics in a 24 hour span, and you can buy sets of them all day any day for $100-130. If you know someone with a bot, even a Makerbot, you can print them for a few bucks in plastic.
The barrier to entry for most people is not the plastics, it’s $200 for the electronics. RAMPS + Arduino Mega and all the stepper controllers, even assembled in kit form are still the single most expensive piece of the puzzle.
Not only that but the Prusa spec is still a work in progress. Looking at the github the OpenSCAD definitions of the X Carriage and the Frame Vertices have had changes commited within the last sixteen hours.
These molds, even if functional, are “out of date” in a way that printed parts wouldn’t be. Print ten sets and you can update the models as you go, mold ten sets and they all look like the mold.
This is, really, solving the wrong problem. I applaud the effort, but if you want to help the reprap project, figure out a way to cut the cost of the electronics.
I have a powder printer running Open3DP recipes (hydroperm/vodka/water). This brings the cost of printing all the molds and even pouring the silicone down to under $100 and a day. The problem of going out of date simply isn’t here. If a part changes, print and pour a new mold for basically nothing. The cost to iterate these is simply gone. I think you’re confusing small run pours with production runs. I can pour 10 sets today, make a change, and pour 10 different sets tomorrow.