Dec 7, 2018 | By Cameron
Professors Jamison Go and John Hart of the Massachusetts Institute of Technology (MIT) Mechanosynthesis Group have developed new hardware that enables what they call FastFFF (fast fused filament fabrication). And it’s fast, see for yourself.
Desktop 3D printers are fantastic at creating high-quality and complex parts on demand, but their greatest weakness has always been speed. They can only print one object at a time, one thin layer at a time. And there are several speed-limiting factors to FDM/FFF 3D printers, with the main four being: the amount of force that can be applied to the filament as it’s pushed through the nozzle, how quickly heat can be transferred to the filament to melt it, how fast the printhead can move around the build area, and the rate that the material solidifies after it’s extruded because it needs to support the next layer.
The solidifying problem they solved like most other developers, by blasting air at it. The remaining hurdles required more creativity. When filament is pushed, it’s typically done by running it between a drive gear and an idler; tension is put on the drive gear, which has little teeth that bite into the filament and push it down as the drive gear turns. If there’s too much tension on the filament, the drive gear eats into the filament and builds up with plastic before eventually losing grip. Too little tension results in slippage and gaps in extrusion.
Go and Hart decided to thread the filament and run it through a threaded nut; when the nut is turned by a motor (via belt), the filament goes down. Anti-twist rollers prevent the filament from twisting as the nut turns. This method of extrusion is not only faster but also much more precise than the typical drive gear setup.
The next holdup of heating the filament fast enough to melt it completely was addressed with lasers. A quartz chamber is lined with gold reflectors, and as the filament goes through the chamber a laser is bounced around inside and pre-heats the filament before it goes through a traditional heating block. All technologies are improved with lasers.
Finally, Go and Hart designed a servo-driven parallel gantry system that rapidly and accurately moves the printhead around with little backlash, the shake or ripple movement that most desktop 3D printers exhibit when printing too fast. Here, the speed is mostly enabled by using a heavy-duty frame and powerful motors rather than a novel solution.
The new printer smoked the competition in speed tests, including a $100,000 commercial 3D printer. The 3D printer built by the research team cost $15,000 so this isn't likely to hit the market any time soon. This extrusion method is seven to 10 times faster, producing up to 127 cubic centimeters per hour. The quality of the prints could be better, likely improved by tuning the retraction and pathing settings, but the quality is still very good considering the speed at which they were 3D printed.
Hart also worked with Sebastian Pattinson, a lecturer at the University of Cambridge, to demonstrate a technique of 3D printing with cellulose. Cellulose is cheap, renewable, and has desirable mechanical properties, but 3D printing with cellulose has proven difficult due to its tendency to decompose when heated. By treating cellulose with acetate, they could dissolve it in acetone and successfully extrude it with a 3D printer. The acetone evaporates (and is captured), leaving just the cellulose acetate. A final bath in sodium hydroxide removes the acetate and a cellulose part is the result. The duo 3D printed a set of medical tweezers and even infused 3D prints with an antimicrobial dye that proved to be 95% more resistant to bacteria.
Posted in 3D Printer
Maybe you also like:
- S-Squared 3D Printers unveils patent-pending Autonomous Robotic Construction System
- Purdue researchers use ultrasonic vibrations to improve 3D printing with viscous materials
- CELLINK and Prellis Biologics announce $1.2M high-precision Holograph-X Bioprinter
- Study on functional mechanical properties of 3D printed lattice structures
- Get a grip with ORIGIBOT2, a 3D printed telepresence robot
- Using plasma jets to promote bone integration with 3D printed implants
- MIT Technology Review's Erin Winick 3D prints her dream wedding
- UC San Diego's easy-to-use 3D bioprinting technique creates lifelike tissues from natural materials
- Genius: 100 Visions of the Future, world's first 3D book printed on space station
- Update: world's first 3D printed plastic bridge completed after 35 days
- Clemson researchers develop ceramic laser 3D printing technique for energy storage
Sounds like a dot printer ... load and very noisy
cain wrote at 12/13/2018 2:45:26 AM:
this is amazing
Craig wrote at 12/12/2018 2:23:37 AM:
The YouTube video is dated Nov 27, 2017
Ian wrote at 12/10/2018 10:30:24 AM:
Would it be worthwhile having contra-travel between bed and head. (Head moves at half the speed while maintaining feed rate - avoid resonance issues?)
Skip wrote at 12/9/2018 6:48:45 AM:
Cool stuff. The issues I have with FDM are poor part strength and low accuracy compared to my usual fabrication methods and materials (both machining and injection molding produce way better parts). I hope 3D metal printing gets to the consumer level someday.
Scott wrote at 12/9/2018 1:11:17 AM:
Love working at MIT
Anon wrote at 12/8/2018 6:57:41 PM:
Isn't this from a year or two ago?
David Merna davidpmerna@gmail.com wrote at 12/8/2018 4:38:05 PM:
Very clever. New design and looks robust. Hey, who put the line "All technologies are improved with lasers."? Haha. Thanks for the laugh. Someone keep an eye on that person .
PWNING wrote at 12/8/2018 3:46:06 PM:
Aaron Swartz, we miss you. :(
John Cena wrote at 12/8/2018 2:24:27 PM:
This seems like the future!
I.Am.Magic wrote at 12/8/2018 12:57:25 PM:
why not use a second 40W cartridge?
Sebastian wrote at 12/8/2018 5:11:52 AM:
"The quality of the prints could be better, likely improved by tuning the retraction and pathing setting" The high amount of liquid material makes it look hard to get the retraction working well. Because of this I would expect quite some stringing and prints with several smaller structures to look not that great on the surface.
Old reprap user wrote at 12/8/2018 4:56:21 AM:
clearly they have not heard of bondtech gears.. The pre heating system is interesting and I have played with preheating a number of times.. not with lasers though! Threaded filament is also interesting...
Grace wrote at 12/8/2018 1:45:34 AM:
That picture is neither Jamison nor Professor Hart; and Jamison Go is a PhD student, not a professor
Steven Sanderlin wrote at 12/8/2018 1:37:09 AM:
Awesome job you guys! Looks very impressive! Keep up the great work!
3Ders.org provides the latest news about 3D printing technology and 3D printers. We are now seven years old and have around 1.5 million unique visitors per month.
