I know, I know - overhyped. But you know what's a funny thing? It's incredibly basic tech. Stepper motors and thermoplastic. We could have had something like it at pretty much any time over the past century, and in its modern form from about 1960 or so. It has unquestionably revolutionized prototyping and short-run manufacturing, and we just didn't think of it. The idea was just too zany and expensive - the people with enough money to invent a 3d printer could afford to just pay someone to make their prototype by hand. It's a rare modern example of the "ancient greek railway problem"; Hellenistic culture posessed all the technology to begin making crude steam trains, and almost certainly had the technical drive to approach making practical ones, if only they had thought of it or considered it a worthwhile thing to do.
(Although there is the interesting question of data management. The .gcode for a small print can still run into the megabytes. CAD software was thin on the ground in 1970 too. So maybe therein lies the difficulty - what good is a 3D printer if you must painstakingly transcribe your blueprints into movement instructions by hand?)
I respectfully disagree. 3D printing has not lived up to its hype or potential. The process is slow and sensitive to conditions. The commercial applications are relatively limited. It has radically transformed rapid prototyping, which is very important. But it's hard to compare it to the ubiquitous daily utility of paperclips or even velcro.
I have much more hope for the next generation of 3D printing, whatever that is
It has also transformed hobby/craft markets and arts. The only thing stopping it is UX at this point (you still need to have both craft and computer skills to use them).
I have very little use for paperclips, actually, but if I needed one, I could easily 3D print it. And that ability - to imagine things and have them come out of the machine - gives the magical empowering feeling that most of the people still have yet to encounter.
I disagree that it's a simple invention, though. The hardware is simple, yes, but it's nothing without software. And the software we need to make 3D printing less of an exercise in patience is simply not yet there, even now.
CAD is not accessible. Slicing is not accessible (sure, Cura will spit G-code without you doing anything - but what do you do when your print fails or falls apart because one of the myriad settings was not set right for this particular print?). Mesh leveling was not a thing on consumer 3D printers five years ago (!). Ditto for variable-width layers. And no slicers have the printhead follow curves in 3D, it's layer by layer in everything I've used, even if the model allows for something more.
What I'm saying is that that we don't have the software to utilize the existing simple hardware to its fullest potential. And without software, a 3D printer is a glorified glue gun.
(Sure, we could've had "3D-printing" pens a-la 3Doodler decades ago. They are fun, but hardly revolutionary).
> I have very little use for paperclips, actually, but if I needed one, I could easily 3D print it.
Out of curiosity, how good would a 3D printed paperclip be compared to the metal ones I am used to? I suspect the performance would be significantly worse due to material properties.
You could make it stronger and more durable, but it will definitely be larger and heavier if you make it out of PLA, and more expensive than a bit of bent wire if you FDM it.
>and more expensive than a bit of bent wire if you FDM it
Yeah, but the whole point of 3D printing is that you can make small-scale custom runs that would be exorbitantly expensive with traditional manufacturing.
You can have paperclips that double as markers/tags/have your name on them/send a message, e.g.[1].
Same goes for other things. 3D printing is a way to make custom things at small scale on demand, but automating away most of the steps after the design stage.
CNC wire bending is a lot better than 3-D printing for small-scale custom runs of custom things at small scale on demand, automating away most of the steps after the design stage. But the sets of designs you can make with the two processes are almost disjoint.
I sort of agree, however there are ways to print metal parts via a sintering process, and if your printer doubles as CNC to cleaning up critical surfaces even better. They just cost a lot of money, but that is changing, just like 3d printers becoming cheaper as the required parts become common commodities.
The assertion that "we just didn't think of it" is not even remotely correct. The late transition to the consumer market and the broad expansion in industry in the last 20 years has far more to do with the expiration of patents in the early 2000's than anything else. 3D printing for rapid prototyping has been around for decades.
Further still, the control software has only recently become efficient enough at the same time that computation is cheap enough to make slicing 3D models a minutes task instead of a months task.
Still, I think some of my point can still be salvaged - this implies that it was patented in the early 80s, which by my reckoning is still well after it became possible.
Possible, yes. Economically viable for the vast majority of applications, no.
In undergrad in 2001 I had a professor that had a 3D printed airplane model with a wingspan of ~5" that he thought was so damn cool (it was some now-defunct military airplane he helped design or something). He was also very fond of pointing out that the model cost his former employer $500 to print, so we better handle it with care. And rightfully so, because if we treated it like a toy it would have fallen apart in under a semester. Compared to what I print today using my MK3s for, I dunno, 50 cents, his $500 printed airplane was a piece of crap.
Point is there have been a lot of technological advances, patents aside, that have made 3D printers viable for anything outside of extremely high-value rapid prototyping / mold forming / other narrow niche applications in the last 20 years.
IMO the invention of microstepping controllers made 3d printing possible; and when they started getting cheap enough for hackers to play with is when all the previous ideas became practical to implment, and then got patented.
There's something to be said for the availability of cheap, high precision stepper motors: these things were a much bigger deal to acquire and control not that long ago.
I followed the RepRap blogs closely during the late noughties, up to about 2010, and there was a lot more to it than just thinking of it, even 20 years after the FDM patent. The use of PLA simplified a lot of curling problems, but even so, rafts and brims were also necessary inventions without heated beds. The pinch gear extruder design took a lot of debugging. It took them years to get a reliable hotend design because they didn't know about Kapton, which also helped a lot with bed heating. Bed leveling, stringing, and filament diameter consistency were hugely underestimated problems. And a lot of trial and error was needed to get good slicing and PID parameters.
The end result of all this, plus some design errors from learning mechanical engineering as they went, was that for many years the mean time between failures on a RepRap was a few hours.
Other than that, yeah, you could have done it with a $30k PDP-8.
Plastic 3d printers are not that amazing. Sure you can prototype a bit faster but we have wonderful methods of mass producing plastic parts very cheaply and quickly. 3d printing of plastic is not that revolutionary for the industry.
Metal 3d printing is a different story, if that gets cheap enough it will be a real game changer. It's very expensive to make metal parts from solid and you can't use casting except for certain metals. If you could 3d print accurate steel parts cheaply and quickly it could actually replace expensive deductive production processes.
I know, I know - overhyped. But you know what's a funny thing? It's incredibly basic tech. Stepper motors and thermoplastic. We could have had something like it at pretty much any time over the past century, and in its modern form from about 1960 or so. It has unquestionably revolutionized prototyping and short-run manufacturing, and we just didn't think of it. The idea was just too zany and expensive - the people with enough money to invent a 3d printer could afford to just pay someone to make their prototype by hand. It's a rare modern example of the "ancient greek railway problem"; Hellenistic culture posessed all the technology to begin making crude steam trains, and almost certainly had the technical drive to approach making practical ones, if only they had thought of it or considered it a worthwhile thing to do.
(Although there is the interesting question of data management. The .gcode for a small print can still run into the megabytes. CAD software was thin on the ground in 1970 too. So maybe therein lies the difficulty - what good is a 3D printer if you must painstakingly transcribe your blueprints into movement instructions by hand?)