[Brook Drumm] of Printrbot is teasing a new 3D printer. This is no ordinary 3D printer; this is an infinite build volume 3D printer, the Next Big Thing™ in desktop fabrication.
The world was introduced to the infinite build volume 3D printer last March at the Midwest RepRap Festival with a built by [Bill Steele] from Polar 3D. The design of [Bill]’s printer began as simply a middle finger to MakerBot’s Automated Build Platform patent. This was patent engineering — [Bill] noticed the MakerBot patent didn’t cover build plates that weren’t offset to the plane of the print head, and it just so happened a printer with a tilted bed could also build infinitely long plastic parts.
While [Bill Steele]’s unnamed printer introduced the idea of an infinite build volume printer to the community, a few pieces of prior art popped up in the weeks and months after MRRF. Several years ago, [Andreas Bastian] developed the Lum Printer, an unbounded conveyor belt printer. A month after MRRF, Blackbelt 3D introduced their mega-scale tilted bed printer and later started a Kickstarter that has already reached $100,000 in pledges.
Right now, details are sparse on the Printrbelt, but there are a few educated guesses we can make. The belt of the Printrbelt appears to be Kapton film attached to some sort of substrate. The hotend and extruder are standard Printrbot accouterments, and the conveyor is powered by a geared stepper motor. All in all, pretty much what you would expect.
We do know that [Brook] and [Bill Steele] are working together on this printer, apparently with [Brook] in charge of the hardware and [Bill] taking either his slicing algorithm or firmware modifications (we’re not exactly sure where the ’tilt’ in the Gcode comes from) and getting this printer running.
While the Printrbelt isn’t ready for production quite yet, this is a fantastic advance in the state of consumer, desktop 3D printing. You can check out [Brook]’s teaser videos below.
I removed the editorializing for you. It’s still neat, just doesn’t quite sound as amazing as you make it out to be when you actually evaluate it for what it is.
This is a potential small advance in the state of consumer, desktop 3D printing but it’s still consumer grade, desktop 3D printing based on point based extrusion technology. So that means long builds, more error prone due to poorer quality components and cost lowering engineering, low resolution, low strength, limited materials and build geometry. But it can potentially print “endlessly” so that is at least an upgrade of sorts.
This is assuming it actually works for lengthy unattended builds, which has also yet to really be determined in any medium I have come across.
I worry about the kapton printing surface. On static print beds it requires replacing eventually. I assume the same is true of the belt based printing surface too. Given that it takes some patience/practice to resurface a stationary bed with Kapton, I’d also assume that doing the same for a belt based surface would be a lot trickier.
That said, I love the idea. It’s cool to see such a small tweak on the traditional cartesian design enable so much more.
I also think this design could be used to enable as much room in the z-axis as you’d like by either increasing the angle (which you could adjust for in software) or increasing the length of the x axis (and platform).
So you removed the editorializing just to insert your own. Good job!
Video of it actually printing: https://www.youtube.com/watch?v=UYAYLEVVLT8
keywords from the video:
Cloud – someone elses computer
“you come in the next morning and everything is all done” – and by “everything” he meant the house fire
I share your objection to cloud services.
But what else than “coming in the next morning” would you do wehn you run a 3D print for several hours? You would watch it, like some babys mistake the window of a washing machine with a TV? For sure not. Of course I had already some misprints, walking in next morning and could have saved some filament. But you can scrap a misprint part anyway, so most of the used filament is used up anyway when a misprint happens, have you watched it or not.
OK, regarding a fire danger I should probably buy a smoke detector or better two with radio link. One in the sleeping room and one in the room where the printer stands.
I don’t understand the term infinite build volume, when this design seems much more suited to the title of serialized printing. Good for printing many normal sized things, not for single infinite sized things..
It’s probably better to say “infinite in at least one axis.”
Because it is. Given infinite filament, electricity, and time, it could print an infinitely long part in the x axis. The z and y axises are limited.
I have to agree with ctag.
If you were to try to use this “infinite” X axis for a part that significantly exceeds the X axis build area, you would quickly realize that the resulting part would not only have the traditional weakness in the Z axis, but would also be weak along the x axis, because the layers would not be continuous, and/or would have insufficient adhesion between the “segments” created by advancing the platform.
The best use for this technology is to support automatic or assisted removal of parts from the active print area.
Marketing this as an “infinite” build volume is deceptive.
I agree with your statement about this machine’s best use. I also agree that the word “infinite” isn’t very accurate here.
I don’t agree that the X axis isn’t infinite. It really is. You could print a rectangular bar forever given enough filament, time, and energy.
Your assertion that it’d suffer from “the traditional weakness in the Z axis” is a pretty subjective assertion. I’ve seen my own prints suffer from weak layer bonding. I’ve also seen my own prints that have excellent layer bonding and don’t suffer at all from what you imply is a universal truth. Print Nylon with a perfectly setup machine and you’ll see pretty quickly that it’s a single cohesive object. In my tests, my testing rigs have broken before layers have separated. These are rigs made of steel expressly designed to try and pull layers apart.
The “segments” you describe are layers. Print a rectangular bar and it’ll have layers at an angle perpendicular to the angle of the print head plane. Support the overhanging bar as it prints and your problem goes away.
To be totally fair, printing an infinitely long bar is silly. I never said it was a good idea. I just said it was possible given infinite time, energy, and filament.
One degres of infinite.
It’s “infinite build volume” in the sense that along the axis of the belt, it can continue to print as long as there’s material to print and it can stay on the belt.
Do it in freefall. Lets see, one Jupiter size planet could be turned into about 6 x 10^28 tugboats. If they are 5cm each, that is about 30×10^28 cm or 300 billion light years?!? Really? Scaling is so counter-intuitive. Anyway, close enough to infinite for me, even if I have a bunch of extra zeroes in that calculation.
Better make that a massively parallel project or speed of light will limit the construction time to 300 billion years or more. Lets see….. 4 printers per square meter, each producing 1 meter of tugboats per day is 400 meters of tugboats per day. If you fill an area the size of the Earth’s orbit with printers, they print 2.8×10^16 km of tugboats a day, or 2,900 light years of tugboats a day. Cool. Only 103 million years to use up Jupiter! Great. We can finish before the Sun burns out.
Gak. 4 printers per sq meter at 1 meter per day is 400,000 meters per day. Typo.
Only one more :-) 400,000 per square km per day.
I see a big mistake: You assume that Jupiter is made out of plastic :-)
Details are here! The belt is kapton-covered steel, and the whole thing is already available (with tech specs published) at https://printrbot.com/shop/printrbelt/
Can someone explain me why you would something like this? The only reason I can see is the size you can print in one axis, but even this isn’t very useful.
If you just want to print some pieces, a normal printer can do the job.
If you want to print more pieces, for a small production for example, wouldn’t plastic molding better suit to the situation?
OK the idea is neat and it was already featured on hackaday, but why is every one wanting to create a machine that isn’t (for the moment) fast enough to do compete with traditional methods?
Because progress is often a result of many incremental steps, not one giant leap.
Or because one can.
The belt is mechanically simpler than lifting the entire build platform or the extruder along the vertical axis, which leads to a more rigid machine and higher precision at a lower cost.
I’ve never tought about this, and it is awesome. It allows printers to be more compact also. It should also be useful to print some parts without supports.
The video linked to by jasoncoon above gives a good reason: it can do multiple print jobs, one after another, without intervention. As each job completes, the conveyor belt platform moves the completed print off the end and the printer can start the next job automatically. Have a multiple-piece job that won’t fit all at once on a normal printer’s platform? No problem–this printer can print the pieces one at a time overnight and they’re all ready the next morning.
I have an idea that I may turn into a side business, if so I would use this to make the main components, so I could set it up with some ways to kill the power but then, set it to run continuously to make all the parts I need.
I know of an individual who started printing cup holders for a specific vehicle, he started with a single printer, and now has 20 printers with 4 people working for him. With this, I could do what he’s doing with a lot less printers, and less people.
A bonus when running continuously, is that a single bad print (for whatever reason) would likely only ruin one print instead of all of them.
My parts are small meaning I could print 2 rows of “A” parts and 2 rows of “B” parts simultaneously, and if any one part of the 4 being printed concurrently failed, it would only ruin 4 prints, and the next 4 prints and those after would likely be good.
For a hobbyist, it may not be a good fit for many, but for production (printing patterns to be cast in metal) this is a big step, especially because it breaks the chains Stratasys had on continuous printing.
Becase desktop FDM has a bunch of main problems right now:
* “idle time” because of printer not empty for the next job.
* “Idle time” because end of filament.
* Failed prints because of adhesion problems.
* Accuracy of the final prints (designers do not want to adjust the model just for printing)
This works towards solving the first one, But I think at a cost of more failed prints, as kapton is not the best solution for adhesion. And a cost at extra design constrains.
As for when is FDM useful compared to other methods? When you need many one or two offs, with quick turn-around times, it beats molding. FDM is proving to be more useful then SLA and SLS in the area of material options.
Need an example, see this excellent example of VW saving millions in their factories:
https://ultimaker.com/en/stories/43969-volkswagen-autoeuropa-maximizing-production-efficiency-with-3d-printed-tools-jigs-and-fixtures
It’s not just that their jigs&fixtures are cheaper. They are also better, because they could iterate faster on them. And (IMHO) best of all, this doesn’t only reduce costs, but it also reduced work related injuries, as the tools could be made easier to handle thanks to the quick iterations. Resulting in happier workers.
(Disclaimer. I work for Ultimaker)
Severely over-hyped, I don’t expect these to become very popular.
Controlling layer orientation is important even for hobbyists, and absolutely critical for functional prints. Layer orientation affects the strength of the parts, how they look, and the dimensional adjustments that need to be made. For example even today holes perpendicular to the layers will come out undersize, and holes into the side of a part will be ovals. If you are tapping or pressing inserts into the holes you need to know how it will be sliced and compensate for that.
No amount of hype will make a printer with diagonal layers useful.
Add dual extrusion and PVA. And you can print the layers in whatever direction of your model you want.
Wow this looks very cool and interesting.
Knowing Brook he will not sell it unless it works great. I have one of his printers (the Play) and it’s a real workhorse. I also like the way he’s honest about it and doesn’t promise things he can’t meet like all the “Flawless $99 3D Printer” kickstarters that turn out to be total crap. And the steel construction really helps to make it a stable platform with very low deviations.
The play is a real workhorse and rarely fails for me. The bed leveling is a godsend especially because so many people are interested in 3D Printing so I move it a lot to do demos. Last time there was also someone with an Ultimaker and she had to spend 15 minutes levelling it and on mine it was just click and go.
Very neat. Now I can print all my long shit in one piece! So cool.
Also, infinite volume. Volume. No one cares if you understand it; it is still correct.
The concept of 3d printing is great, and I really want to get into them, but I never need to build anything that would call for one (I mean genuinely call for one). Maybe I need to start making things that have gears and things like that. I always end up making things from flat pieces of plastic that can be cut and have holed drilled in in a matter of seconds. Most angles tend to be right-angles, so simple brackets do the job very cheaply.
I need to print 100 widgets at work. Currently I can use two traditional printers that can make 6 each in 5 hours. So 24/day (one overnight print and one print in the morning). This would allow me more time to actually print continuously all day.
The cos players will love it. print a sword / gun / armor ect in one go.
I hadn’t thought of that. I saw it more for longer struts and supports. Now I need a sword…
I love reading these comments. Bad and good. I’m satisfied to just keep searching for the niche that this printer fits. There is no one size fits all printer. They are pretty personalized machines. The Printrbelt is strong on making a lot of parts, long parts and situations that might need a bunch of people who want parts and don’t want to get intimate with the machine- maybe one person would maintain it. Teacher? Librarian? Print service, maker space. If it resonates, it’s for you. If not, no problem.
Brook