Since the first desktop 3D printers, people have been trying to figure out a way to manage desktop 3D printers and turn them into tiny little automated factories. One of the first efforts was a conveyor belt build plate that was successfully used by MakerBot until it wasn’t anymore. Octoprint has been a boon for anyone who wants to manage a few printers, but that’s only half the solution.
For his Hackaday Prize entry, [Mike] has come up with a solution that turns a desktop 3D printer into a completely automated factory. Not only does this project take care of removing the part from the bed when the print is done, it also manages a web-based print queue. It is the simplest way to manage a printer we’ve ever seen, and it’s a great entry for the Hackaday Prize.
First up, the software stack. [Mike] has developed a web-based queue and slicing software that ingests 3D models and spits out Gcode to a printer. This, really, is nothing new. Octoprint does it, Astroprint does it, and even a few 3D printers have this capability. This is only one part of the project though, although it is geared more as a maker space management software than simply a dedicated 3D printer controller.
You can’t have an automated mini factory without an automated build plate, though, and here [Mike] has come up with something really great. His solution for dispensing prints after they’re completed is brilliant in its simplicity. All you need to do is drop the floor out from underneath the print. [Mike]’s solution is a trap door print bed. At the beginning of the print, an inkjet printer spits out a piece of paper, with a few lines of text, onto the print bed. When the print is finished, a stepper motor unwinds a cable, and a trap door opens up underneath the print. The part drops into a bin, the door closes, and the next print is loaded up in the queue. It’s brilliantly simple.
You can check out [Mike]’s demo of this system after the break. It’s awesome and so sublimely simple we’re shocked no one has thought of this before.
Awesome.
Seems like it would work well for PLA and not much else. Anything that warps much, like ABS, would just lift that paper right up, making a darned crumby print.
If there’s enough humidity in the air (or it’s cold enough), it would warp even with PLA… Took me a year to find a way for my prints to lift less the masking tape I use.
Mmm, I wonder if vacuum suction would work in his case…
Its an interesting proof of concept that can be expanded on with different print bed materials and even auto loading whatever filament you want. With some further development It could do a lot more but the basic concept looks sound to me.
A roll of disposable bed material instead of paper pulled across a heated bed. then the bed swings out of the way dropping the disposable bed material and printed object. That way more filaments could be used.
Swapping the filament would be fairly easy with all of the multi filament solutions already in use.
Looks like a good start to me.
The problem with having a roll of material is that its all tied up in patents. That’s why this is a completely new method.
Is this the mike of espbasic, a talented chap, I suppose any sheet material could be used that a 2d printer can handle.
Yep. It’s me. Been working on this for a while.
Does anything print well on either ‘plastic food wrap’, baking paper/parchment paper, or waxed paper? I mean, they are already on a roll; just throwing out ideas that I myself can’t test out.
The trapdoor idea is surprisingly simple. I would put a basket underneath though :P (or something that holds the paper in place?) but … it works. Of course this would only be useful as far as it sticks.
I expected from the write up of the article that the printer did some pattern to hold the print ( ¿ does it stick better or worse ?). But user id … well, sure, it makes sense :P
BRILLIANT!
I am looking for some help making a better mechanical way to clamp the paper.
Electromagnets to a steel bar?
Glue.
Wait, that’s not right…
Glue would work. Add a delay for drying, or use CA with a spritz of quickener, and a mechanical lever to break it loose.
There was an article here not so long ago about an old HP plotter that uses high-voltage electrostatic forces to hold the paper to the plotter bed. I thought that was pretty nifty.
That old electrostatic paper clamping method worked surprisingly well. I’m not sure why it faded away… maybe cost, maybe no one came up with a good way to automatically align the paper. On the HP plotters you would manually position the sheet on the bed, then flip a switch to hold it in place. There were no obstructions on the top of the sheet, so the plotter could cover 100% of the paper.
Vac table?
What about two rolls? Like a belt grinder, maybe the print will pop off after it gets rolled off, or else you could add a scraper.
What about the rollers from an inkjet printer? Have one on either end pull 1cm each to make it taught.
What about perforating the paper first, and then putting some one way hooks. You push the paper onto the hooks, then pull back a little, print, then push all the way over to release.
What about some little push pads under the table, when they get pushed up by the build platform, they travel away from the center, pulling the paper as they do?
Oops, I meant that as a reply to the clamping paper topic.
What about printing on glass bed and utilizing pneumatic and/or solenoid actuated print releaser. It could be a flat sharp blade aligned to the bed which will knock off the piece from the bed once the bed is cooled. It could have long pushing action for pushing the piece off from print bed accompanied by a small sharp stroke for releasing it.
Alternatively a 5 axis robot could perform the knocking off and grabbing + moving of the piece.
The bed with a bulldozer razor edge has been patterned . . .
Well I’m not a patent solicitor, but surely someone can go around this by making some structural/functional change. Perhaps by adding an impact feature, vibration feature, changing razor blade to flat bar or some other. Companies tend to go around various situations like this all around, with varying success of course. But still I guess might be worth of investigating and searching for a loophole.
For example check some basic plastic nut type cable glands, all of the manufacturers seem to have patented their own technology for cable gripping function, yet numerous manufacturers found their own and unique way to do grip the cable and archive the same end result, and most of the cable glands even look the same if you compare them side by side.
I’m a little surprised by the paper. While it is very good for the complete workflow (people are always comforted by slips of paper it seems), it is not a good buildplatform material to print on IMHO. Small prints like he one in the vid will hold nicely, but (as others have pointed out too) bigger prints like say a straight box will put too much stress on the corners so the paper + print will warp.
For PLA I am quite content with a glass bed with a gluestick coating. If the nozzle is not set too low, the parts will just pop off after the buildplate has cooled down. Some construction will be needed to actually pop your prints off though.
Monoprice sells 30 meter rolls of super wide Kapton tape. Make a set of glass build plates with that on it. I wipe mine with rubbing alcohol between prints. Still get warping on large PLA prints despite heating the plate to 70C or more.