3D Printing Revolution: The Complex Reality | Hacker News Reader

3D Printing Revolution: The Complex Reality

blog.makezine.com

74 pointsjrockway58 comments

I have a bit of a quibble with this point: "CAD is genuinely difficult." When I was in the university and had access to SolidWorks, it took me less than ten hours to design something that was moderately complex. Eight years later, when I tried to design something for my side project using cheap and free 3D CAD programs, I had the hardest time doing so because the programs I tried were simply not powerful enough. The problem is not the CAD concept, it's the programs. There's no reason why CAD can't be easier than ever, and it might be if you purchase a multi-thousand dollar program.

Honestly, my concern is that something like the GIMP will take over in the open-source CAD space, and stymy efforts for making something as useful and powerful as Solidworks. We've seen this happen before--let's try to avoid it.

There is: FreeCAD, based on the OpenCASCADE kernel.

SolidWorks is just so easy to pirate, that no one has much motivation. It's a massive effort to build something like this.

Also, it requires a multi-disciplinary team: math, programming, MechE / ID, UX, etc. - it'd be hard for an open source project to do this without some type of funded entity.

Hey, FreeCAD allows python scripting, I'm game. *

*: got about 30 minutes into trying it a couple times before banging my head on a wall, goes back and plays with Blender.

You can try OpenSCAD - but it's its own 3D language, but pretty intuitive.

I agree. A great example of a "CAD tool" that lets you design complex, believable organic creatures easily is Spore Creature Creator [1]. It's easy to imagine similar kinds of tools for many other object types.

[1] https://www.youtube.com/watch?v=ZRr3lgckIAM

Totally, if you're motivated you can get SolidWorks do give you good output in days or even hours. I've heard some are worse at 3D visualization, so maybe we're outliers, but I doubt it.

I also had to use SolidWorks at school and while didn't find it to be anywhere near as hard as the author makes it out to be.

Nice article, the bit about casting is pretty spot on. If you want to make a number of things currently casting is a lot more efficient. Which is not to say you can't 3D print an exemplar, use that to make a mold, and then cast copies, very doable, But just that then you're making parts with casting resin not 3D printing.

I really like my Replicator but have yet to do anything significant with it (which makes me sad, so many projects so little time).

I can say this though as more of this stuff gets going (see the article I submitted on printing in metal (http://news.ycombinator.com/item?id=5220032)) the difference between the RepRap and the Cupcake printer, the Cupcake and the Replicator, things are progressing along nicely. People joked about my my 2K of RAM in my Digital Group z80 system but 10 years after that point in time enterprise business was using PC's in their day to day operations.

Guess I'm mostly saying "Patience, its coming along nicely."

That's true to some extent, although it's useful to keep two things in mind:

1) The improvements in the output of FDM printers is in good part due to switching to lower-strength materials (e.g., ABS -> PLA). It's great news if you want to make casting molds - but not so great if you want to directly fabricate durable parts.

2) There is no gradual progression from the familiar FDM extruders to SLS, SLA, and similar technologies that produce high-accuracy parts or can work in metals. These technologies are inherently messy and have other surprising trade-offs, and are suited chiefly for very dedicated hobbyists and for quasi-industrial applications.

1) Not true. The resolution and printed parts in ABS and PLA are essentially identical. The only reason you see a switch from ABS is that PLA doesn't require a heated build bed and companies like Makerbot are trying to simplify things to make them cheaper and more plug and play for mom and pop.

2) Why does there need to be a progression from FDM to SLS and SLA? The progression is there on the software side with things like slicing, 3d modeling, etc. There isn't a progression from inkjet to laser printers, but that didn't stop anything.

I'm going to need to save this article for the next time I get into an argument about 3D printing. Like the author, I love the technology and have enjoyed the parts that I've made with it. However, the technology is over-hyped and often treated like a Star Trek replicator. I've seen discussions of 3D printing nuclear food, microprocessors, and even living things.

To a degree the technology reminds me of Visual Basic. The technology makes it very easy to make something that looks like what you want, but it doesn't bring you any closer to actually making what you wanted.

Visual Basic earned large numbers of people large amounts of money and allowed them to make exactly what they wanted. Specifically, what they wanted was a solution, quickly that they could cobble together themselves without having to hire some person that would tell them all the time how what they wanted was not possible.

Ah, I hadn't meant to denigrate Visual Basic, though I can see how my comment could be taken that way.

My reference to Visual Basic rather comes from a large number of projects I saw that equated the GUI with the functionality. For example, I played with am "80% complete" program to translate between English and Latin texts. The New, Open, Save, and Exit buttons were all working. As soon as someone wrote the method for the Translate button, the program would be complete.

Visual Basic and 3D printing have both given thousands of people the opportunity to do things quickly and independently that they'd have never had the opportunity to try before. However, they've both also suffered the hype of those who assumed that their understanding of the surface was an understanding of the whole.

Um, it can. I work in a lab that is 3D printing live mammalian cells to build living tissues.

Yeah, but that's different. Just because it's based on additive manufacturing techniques doesn't mean that it's even close to being the same kind of machine as a Reprap or Makerbot. While they rarely outright lie, what mainstream tech news love to imply is that desktop hobby 3D printers will soon do that stuff.

It's about as ridiculous as implying that because cars are built by robots, soon your Roomba will be building a car.

I hated using VB as much as the next guy (VB6), but I built a number of business systems that ran for years. Your analogy is false.

Ouch. Defending VB in this case really speaks poorly of 3D printers.

No it doesn't. The analogy was flawed from two sides. Firstly, VB was (is?) a perfectly good solution for producing functional systems which solve problems. It's not very nice, but to say that it doesn't bring you closer to what you wanted is downright false. It does - that's why so many functional applications are written in it.

Secondly, I'd argue that the same statement applied to 3D printers is also false. Just because the output of hobby level plastic extruding printers is limited to certain shapes and materials doesn't mean that the process as a whole is flawed. I've played around with titanium springs, brackets and other interestingly shaped pieces which simply couldn't be produced by any other manufacturing process. 3D printing (in particular selective laser sintering) is going to produce a step change in the aerospace industry.

Finally, as an aside hobby and low cost plastic 3D printers are very useful for some applications. Producing brackets and mounting plates is considerably streamlined with a 3D printer. It's a useful technology for particular applications. Its not the fault of plastic additive manufacturing that journalists are getting more excited than they should be about it.

If you haven't already seen Michal Zalewski's "Guerrilla guide to CNC machining, mold making, and resin casting" [1] check it out!! He talks about how to design and make the molds from scratch. There is also the "The making of Omnibot mkII (work in progress)" [2] listed in the article.

[1] http://lcamtuf.coredump.cx/gcnc/

[2] http://lcamtuf.coredump.cx/omni2/

The "manufacturing things entirely using software" dream died a little for me after I looked at this guide (which is excellent and clearly a labor of love, BTW). The amount of variables in meatspace that you have to account for (that I simply had no clue about) -- like total indicated runout, reach length, cut direction -- was mind-boggling.

To me, those variables sound like things that software can do a much better job of dealing with than people can.

When you've got thousands of dollars worth of tools at risk, you don't let software guess what to do - you can program most machine tools to damage themselves without a complaint. Software can help, but suggesting that a computer can produce a flawless CAM program for a CNC tool is like suggesting that computers should write web browsers or filesystems.

CAM programming is an art, and it requires intimate knowledge of the machine you're coding for if you hope to get a final product without too many trial and error iterations.

Your reasoning seems to be, "Since poor G-code can waste a lot of billets or even produce a hundred-thousand-dollar CRUNCH, we should write it by hand instead of generating it using computer software." That sounds exactly backwards to me.

(Computers do write web browsers and filesystems; the programs that do this are called "compilers".)

I never said they weren't! However, you still need to know about them and understand them to write the software -- and to figure out why your CNC job went wrong if the software screws up.

For instance, maybe the software is smart enough to throw a red flag if your reach length is no good for the CNC job you're going to run -- but you still have to know about it and understand the concept to set it [correctly].

We won't all be designing things. There will be a few people whose designs are widely used because they work the best.

We'll all be trading and downloading and filesharing the CAD files. Think of future banner ads: "Need a new plate? Go browse the 10 million CAD files on 3d-CAD-plates.com to download the floral plates of your dreams!"

Torrent networks and TOR will be used for the underground and black-market filesharing of CAD files for restricted objects, such as gun parts and other weapons.

In the past 10 years the only thing I bought that any current 3D printer could have made was plates and silverware. IMO, that says a lot more about the the real state of 3D printing than anything I have read online.

I wouldn't use plates or silverware that I printed on my 3D printer unless I coated it.

Do you have one? I couldn't think of that many things to print when I made one, but now that I have one (kind of like the hammer and the nail), I can see lots of things to print now.

Relevant observation: "hammer and nail" effect seems to be the same thing as pg's Blub paradox[0] or the phenomenon that when you start carrying Duct Tape with you, you'll suddenly find some use for it every other day.

Because of that I think we should learn not to dismiss things just because we can't imagine what we can do with them. The imagination will most likely come with use of the tool.

[0] - http://en.wikipedia.org/wiki/Paul_Graham_(computer_programme...

Humans learn by playing

That's actually touched upon in the article (I'm the author). In short, it's possible - but not given:

1) The culture of knowledge- and-design sharing has been prevalent in the DIY community for decades, but hasn't produced anything comparably grand. This is despite the fact that we can already easily source or make custom parts of all sorts. Will this suddenly change? Who knows.

2) There is extremely little emphasis on materials science and mechanical engineering in the 3D printing community at this point - which makes it difficult to reach a "critical mass" of good, practical CAD models to reuse (there's plenty of Yoda figurines to choose from, though). The success of the open source community owes a lot to the broad availability of high-quality reference materials and the self-documenting nature of most of the existing code.

3) The dominant FDM approach produces inherently crappy parts, which makes it poorly suited for direct manufacturing. There is no clear path to fixing it. The alternative technologies work better, but aren't as simple to operate. Of course, better approaches will eventually arrive at some point - but would require major breakthroughs, not incremental tweaks.

4) There is very little "portability" of designs not only across printing technologies, but even between individual printers - which again, is an interesting handicap.

Only on your point 4,

Assuming there would be several mainstream 3D printing technologies (like HP mainstream), someone would find way to design a middleware for them. Portability is not an issue when there are standarts and communities.

People should realize that the flip side of mass customization is actually just better search tech. Do you really need to design a different collared shirt when there are 10K already out there to buy?

I every conversation I've had lately when I tell people what I do, they mention 3D printing. I have to gently tell people that I don't do 3D printing yet as products generally fall into one of two categories: 1. Inexpensive and low quality. 2. Expensive and high quality.

My current plan is to go after market #2 in a year or two when I can raise the capital to get a really good machine.

I started with, and plan to continue, waterjet and laser cutting as the limitations of these technologies are easy for most designers to understand. What you enter as a CAD file is pretty much exactly what you get out as a final piece.

As an aside, it's amazing that using a robot to blast away raw material into a finished part using an ultra high pressure stream of water and garnet or a frikken laser beam is considered "traditional" machining.

Traditional machining in the sense that you need industrial space where you can make a mess and a lot of noise as opposed to something you can conceivably use in normal home/office space. Waterjet cutters need special FLOORS because of the weight of the tank. Laser-based SLS and SLM emit both nasty fumes and spread dust everywhere. This is why the tech is not sexy, not because it's less advanced.

What are you thinking with 3DP? Don't try to be Shapeways, et al. who compete on price and make no money...

You are very wise.

I want to do DMLS or another technology that will make metal parts.

I haven't heard from you guys in a while. Shoot me an e-mail.

Just hit me up at: nickpinkston at gmail.

I always get excited about 3d printing. Then I remember how much 2d printing sucks.

Really?

Maybe I am just old and remember using manual typewriters, line printers, dot matrix. Now when I get amazingly crisp nearly instant typeset output from just about everything in seconds I am still amazed.

m-m-manual?! :P no, I admit it, I'm a young. Printers jam a lot, they have ink problems, drivers are always a pain. They are probably a lot better than a typewriter though.

It's true that the mainstream media has been overhyping this technology, and some of the mainstream tech media as well (Wired, for instance). From the way everything becomes "3D printing" in the news, you'd think that a 500$ reprap is a few years away from being able to print human organs, cars and houses. Some journalists need a reality check: hobby and consumer 3D printers are able to make simple plastic pieces with some caveats about overhangs. They only recently started appearing with dual extruders. They're still very fussy machines that need frequent adjustment. And those that don't are either very expensive or very limited in capabilities.

On the other hand, there are very legitimate uses for these as they are right now. I guess it's hard to get excited about replacing a broken plastic handle on a beach cooler or a broken knob on a washing machine, but you can make cool stuff with them. I printed a docking station for my Nexus 7 with mine.

Okay, so you extrapolate a gloomy future from a snapshot of the tech? I remember dozens of such articles from the early days of the web.

Subtractive processes have advantages? It's pretty obvious that at some point these devices will be able to run additive and subtractive process in whatever sequence you like.

This stuff is so complex that only a few currently understand robust design? All the better to hatch a million designers in garages all over the world.

The complex reality is that price is not there yet for CNC(including tooling) and lasers.

The Roland of the article cost more than 3 thousands dollars. I could make eight Prusa reprap (I did it for a University course) for this price, or buy printrbots.

About CAD: For some people it is genuinely easy, like some kids are "naturals" for playing basketball.

When you teach kids, it blows your mind what they are capable of doing when left alone once they learned the basics.

>The Roland of the article cost more than 3 thousands dollars. I could make eight Prusa reprap (I did it for a University course) for this price

Yeah, or you could make a small 3 axis CNC for $400 as well.

There are quite a few mills priced comparably to 3D printers, e.g.:

http://www.rolanddga.com/products/milling/imodela/

https://sites.google.com/site/cncdiyorg/2520

https://probotix.com/FireBall_v90_cnc_router_kit/

Whats your take on plasma cutters?

Some rebuttals:

1) CAD is genuinely difficult

True, but this barrier is lowering rapidly. Anyone can make simple items using programs like Autodesk 123D with no training. Low-cost 3D scanners will also lower this barrier (see projects like http://www.kickstarter.com/projects/621838643/desktop-3d-sca...).

2) There is a lot more to industrial design than meets the eye

Good design is hard, and the supply of designers is very limited. But most of the population has no reason to learn 3D design. Providing fiscal incentives to designers of novel products is crucial for expanding this base. More importantly, empowering more people to have ideas translated into physical objects will greatly increase the supply of casual designers. Have you ever shown a friend an object that you designed and 3D printed? It's magical in a way that photoshop is not. Don't underestimate the in-person viral factor for inspiring designers!

3) Mechanical engineering is a real science

As in open-source software, 3D designs often build off of each other. Both in terms of remixing existing designs, as well as sharing printing profiles for different slicers. The 3D printing community is much more collaborative than traditional manufacturing institutions. Software in 3D printing is 'eating' the mechanical engineering world slowly but surely: no longer do you need massive amounts of capital to design and start selling a new product. The barrier to entry for product designers has lowered!

4) Manufacturing processes are not perfect – and won’t be any time soon

Hence the need for curation and an easy way to find quality designers and printers....

In June I wrote a bit about this:

"When automated fabrication—the scenario where you get your next bicycle by downloading bicycle blueprints over the network and sending them to a machine that then produces a bicycle for you without human intervention—happens, it will not be by means of 3-D printers, which work by depositing layers of a small number of materials. Instead, it will take the form of automated assembly by robots of parts mostly made by other means, such as laser cutting, torch cutting, CNC machining, and planar printing processes."

I posted that on http://lists.canonical.org/pipermail/kragen-tol/2012-June/00.... What do you think? Is that correct?

This is something that has hit my consciousness only recently as a result of the gun debate (3-d printers are being used to manufacture magazines). Since then I have checked out some uses for it on YouTube and have seen everything it used to manufacture model airplanes that fly to hand-grenades (yikes!). I have already been pricing out units myself and would like to purchase one of these one day in the not-so-distant future. The change that this technology will bring to so many individuals around the world is nothing short of revolutionary. I believe that it will really impact our society over the next decade in ways we would not even realize yet- kind of like the Internet did in the 1990's.

I think that when everyone is pointing to a technology as "the future", that's a sign it might not be, at least not the way they imagine it. If 3D printing is so great, where are the huge revolutionary changes? 3D printers have been around for several years. What are we waiting for, exactly?

There is a lot of work going on to develop 3D printing as a mass manufacturing process. It is improving rapidly, but you have to understand that in the physical world progress is always slower than in software. 3D printing isn't a small change in manufacturing process (like more advanced CNC machining is) but a completely different approach. The level of investment, testing and iterating required to produce 3D printed components which demonstrate proven reliability and lower costs than existing processes is high.

The aerospace industry are actively pursuing 3D printing as a manufacturing process. Selective laser sintering allows shape-optimised parts to be built out of titanium - something that we cannot do any other way - which will lead to upwards of 50% weight savings. It will make a huge difference in the weight (and so efficiency) of aircraft. The issue is that the aerospace industry operates on the principal of proven heritage. The 787 may be the newest large airliner, but there's nothing in it that we haven't been using elsewhere for 20 years. Space is even slower at adopting new technologies - reliability is key.

> What are we waiting for, exactly?

I'd say precision. We need mature manufacturing processes for home printers so they can be mechanically precise enough to create everything. A lot of it is just that the hardware is immature, and needs a few more years of finer grained manipulation development in both the printer design itself and the resulting printers capabilities.

I mean, the precision extreme is that you can have a printer that arranges matter on a molecular scale. That is what Star Trek esque food processors are conceptualized as. If we can't get molecular, we should at least try as hard as possible to make it happen (though considering we are on a mm scale now, getting to the 0.1 nm or 100 pm scale (where a hydrogen atom is around) might take some time).

I had a chance to play with a 3d printer. The quality of output was just subpar. If it was higher quality I'd find quite a bit of use for it. In the mean time it isn't worth the cost.

On March 15-17, in Elkhart Indiana is the Midwest Reprap Featival. Josef Prusa, of Prusa Mendel fame will be there.

The festival is free to attend, no tickets or anything. If you are a vendor, they ask a small donation.

Business idea: CAD object library inspired by Lego.

Imagine being able to build your model from components in software, then print out the parts and put it together.

That's what I'm doing with my open-source Bitbeam project. I design the Lego Technic-compatible parts in OpenSCAD. Then the parts are lasercut, milled, or 3d printed. I've shipped several Bitbeam-based robots via Tindie already, and I'm looking to expand my part and kit catalog. Want to help? :-) (Bitbeam.org)

The problem with 3D printers is that, it is not a profitable market at all, hence not a single mainstream vendor would apply it commercially.

With a full fledged 3D printer, you can essentially print another 3D printer. Just like how everyone uses windows, a large sum of potential customers would be freeloading. The exception would be that, they wouldn't have a huge market that would keep the cash flowing. Heck, electric cars weren't commercial until someone found a better way to monetize it else than charging it only for the electric bill.

The current technology and budgets for technology allows an affordable 3D printer to be at your home. But it is specifically left out of plans. Think of a blueprint for a pcb, chair or car that has been developed open source. Unless they stranglehold all the resources needed you wouldn't have to rely on any manufacturers or or middle men, which would start to drown the market.

We will see our first commercial 3D printers when they find a way to enforce a solid(er) DRM on it. Until then, it will be a geek hobby and left esotheric.

Reply on news.ycombinator.com