Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Our 3d printer is better than yours (tuwien.ac.at)
212 points by anigbrowl on March 13, 2012 | hide | past | favorite | 24 comments


While this is firmly in the "holy crap" territory, I dread to imagine what the consumables costs are.

I can't wait until microfluidics people get hold of this. It'll be (ahem) huge.


Judging from the video, the resolution / dot size is about 1 µm (one micrometre). For comparison, the thickness of the tape in a 120-minute compact cassette is 9 µm, the diameter of a typical bacterium is 1–10 µm, and the diameter of the nucleus of a typical eukaryotic cell is 7 µm (http://en.wikipedia.org/wiki/1_micrometre).


Right. There's a bit of use of the "nano" prefix in the article, but it really should be "micro".


Am I the only one who thinks about the implications of these 3D printers? I read somewhere that an entire cell phone built with a 3D printer (circuits & all) will be possible soon. No "put it together yourself" stuff, either - entirely built and assembled at once.

Doesn't that create a disruptive influence for any company creating smaller products and devices? Not suggesting that it's a catastrophe...but where does this road eventually lead?


I'm patiently waiting for the meta moment when a 3D printer can fully print a functional copy of itself in one piece. At this rate, it's going to happen.

But more seriously, seeing all this happen gives a very clear glimpse of the future. Stores full of merchandise will be slowly replaced by online catalogues where you can preview your object, optionally modify it as you like, and send it off for printing. Items may be eventually printed at home since every family will have a 3D printer just as every family now has a computer, but for short-term future there may be whole companies that will set up printing factories with high-end printers, and they will mail more complicated objects to you.


When you read articles like that it helps to ask how much it will cost compared to doing it the "normal" way. The answer, for most things that are produced in volume, is generally "far too expensive for general consumption." Make a great demo for the printer company, though.

Think of it this way: a 3D printer is a general purpose tool; it really can't compete with individual tooling optimized to pump out millions of units of a single product. For prototypes and engineering testing, it's awesome (sometimes[1]); for mass production not so much.

[1] I have seen instances where a design engineer had to make a "hybrid" prototype part that was partially done on a 3D printer and partly on a traditional milling machine to make up for deficiencies in the 3D printer output.


I doubt the CPU and many other components will be produced by 3D printers, or at least cheap 3D printers, any time soon.


While I do see it being possible sometime, I agree that it's probably true that we won't see that for a while. The Ikea model would work well at this stage where you'd expect a small booklet that guides you along the way: "When the printer's green light is flashing, open box A5 and place chip on the area shown in Figure F2. Then press the green button on the printer's console."


There is plenty of fiction thats explored some side of the idea. Neal stephenson "the diamond age" is probably a good example.

To my mind though, this is only 1/2 of the problem. most fiction has focused on nano fabs + nano dissemblers. perfect recycling + perfect fabrication.


You pay for designs, not for the final product. It's not different than buying programs for your computer.

Of course at the beginning the products will be sold assembled from shops with 3D machines for rent.


See also:

(March 31st, 2011) https://www.youtube.com/watch?v=snOErpOP5Xk

(November 11th, 2011) https://www.youtube.com/watch?v=PV0BpoRTv38&feature=rela...

Dramatically lower resolution, of course, but probably $100,000 cheaper. No granite optical tables required, for one.


"The mirrors are continuously in motion during the printing process. The acceleration and deceleration-periods have to be tuned very precisely to achieve high-resolution results at a record-breaking speed."

Isn't this a solved problem in laserprinters by using polygonal mirrors?


With polygonal mirrors you have to wait for the perfect moment to pulse the laser.

I think these are just galvanometers. Seems like they have to 'predict' the motion of the mirrors for accuracy and speed.


Interestingly, similar work is currently happening in MakerBot community: http://www.thingiverse.com/thing:17855


So, if it takes ~ 90s to fabricate a 300x130um element that is 100um thick then how long to do a 1cmx1cmx100um thick element?

Scale factor is 10000/300*10000/130 = 2564, so it takes ~ 64 hr.

i.e. it takes a long time ...

I will stick to CNC machining and photolithography for now. ;-)

If they speed it up then perhaps 3D elements could be added to microfluidic networks with it.


Hmmm, well for some things 64 hours isn't as bad as I first thought... This could be sped up and hopefully is highly paralyzable so with some more research you could have something while slower the resolution quality would be so much better is would be worth it for some situations. Imagine printing watch parts or pieces for musical instruments, etc.


I wonder if this technology will be used to add "watermarks" to objects, to guarantee authenticity, like print a logo on a clockface of your rolex


Surely not, the "watermark" is more easily added using methods suitable to mass production. Possibly it could be used to produce counterfeit watermarks though...!


hey jan, if you read this. great work! all the best, philipp


somebody tell steven spielberg "innerspace" is within reach in next couple of years.


Their title is better than yours.


Sure beats the hell out of the RepRap!


Well the Ultimaker already did (https://shop.ultimaker.com/).


And we're one step closer to the Singularity.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: