Iron Man 2's Secret Sauce: 3-D Printing | Hacker News Reader

Iron Man 2's Secret Sauce: 3-D Printing

fastcompany.com

83 pointsbinarymax45 comments

I'm a little skeptical about the magnitude of the increase in time from concept to realization.

Four hours is the printing stage, but laying out something like that in cad is fairly time consuming, on top of that every piece surely went through numerous iterations. Additionally, the medium seems hard to fiddle with, so if the part didn't turn out, you have to scrap it completely.

Plus given that it is some sort of plastic, your only finishing options are probably powder coat, some sort of vapor deposition (which that photo looks like), or possibly some sort of plating if people have figured out how to do it.

While I'm sure that the tech will make some people's lives easier, this is one of those things that seems like it could easily be over hyped.

> laying out something like that in cad is fairly time consuming, on top of that every piece surely went through numerous iterations.

"Regular" design would still go through iterations.

> Plus given that it is some sort of plastic, your only finishing options are probably powder coat, some sort of vapor deposition (which that photo looks like), or possibly some sort of plating if people have figured out how to do it.

In this case, they're using plastic, but you can also print in metal, wax, rubber, nylon, paper, and resin. (And yes, you can plate the plastic. This is how Aston Martin makes headlamps for their vehicles.)

> While I'm sure that the tech will make some people's lives easier, this is one of those things that seems like it could easily be over hyped.

It's true. People get really, really excited about all of this, and start talking about "When everyone owns a 3D printer..." It's such a sexy technology that people get irrationally excited. It certainly has downsides as well as upsides.

Yes. Normal printers haven't made everyone an artist, either.

It's sort of an imperfect analogy, though. The people who say things like that are thinking of it in the "You wouldn't download a car?" sense. You can already download 3D models from a place like Thingiverse, so if everyone had a printer, we could just print new household items rather than purchase them. It's a functional thing, not an aesthetic one.

The issue with this is that just because they make things doesn't mean they make them _from nothing_. Are you going to keep not just a printer, but large vats of materials hanging around your house? I don't really think so.

OK. And the things you make with your printer still have to compete with stuff from custom processes. So e.g. your printed tea cup will probably not as good as a normal tea cup, but it may be more convenient to just print some cups instead of going out and buying some.

If this were true, the greeting card market wouldn't be nearly as big as it is.

Also, printing will probably more expensive than buying things made via more conventional processes.

It depends on how you define "more expensive."

Traditional manufacturing processes have really expensive tooling, but then each part is really cheap. Printed parts are always the same price. There's an inflection point where if you're going to make more than a few tens of thousands of things, it's cheaper to make the tooling. But if you're not going to make that volume, printing is cheaper.

Like many things in life, it's all about the best tool for the job.

I tend to imagine the large vat of materials becoming part of the corner convenience store/gas station. "Hi, I'd like $50 of gas, $38 of aluminum composite solid #3D2, and this bag of chips."

Alternatively, it could coincide with the one-hour photo lab: "plug in your USB drive to begin uploading model." prints out receipt

You still need a volume of stuff that's larger than the thing you want to build. A convenience store sized package would only provide enough material to fab something smaller than the package itself.

Mass is more appropriate than volume here, but point taken.

Sort of. You still need support materials if you were making something hollow.

But anyway, this is pedantic details. We both get where each other is coming from.

How about getting the raw materials like tap water?

It'd be kind of possible. Not all of them use liquid materials, but even then, the ones that do look like this:

http://www.youtube.com/watch?v=nwQ5HA8sE-k

Right now, you load the material into the machine with a small crane-like thing. So theoretically, I guess you could hook up piping instead. But that's still a honking big machine, and you'd need a machine that'd be larger than the things you're building with it, really.

I think eventually, it could happen. It's just not going to happen any time soon.

My favorite theoretical material supply scheme is grabbing carbon out of the air and re-configuring it into whatever you're printing with ;)

How about grabbing stuff out of the ground?

I figure 3D printing is still in it's infancy. Sure, it can't do all you can imagine it doing, but that's just because nobody's figured out how to make it do those things. Yet.

Definitely easy to over-hype it, but it deserves attention.

Plus... one of the things that has always sucked about modeling in my opinion is the difficulty of entering a model. Maybe with a good way to get a structure out, we will soon work out a good way to get a structure in?

The best current way to "get a structure in" is through laser scanning the part, and converting the resulting point cloud to an STL file.

This isn't yet to the point of being automatic, though, as laser scanners aren't yet perfect. There's still a bit of artistry involved.

This made me curious about the economics for a 3D printing on demand service. Looks like there are already some players in the space making it work:

http://www.shapeways.com/

http://www.buildparts.com/

http://www.zoomrp.com/

They probably have a willing audience in cosplayer community. If only they also automate the painting process.

This is actually what my startup does.

http://cloudfab.com/

We actually offer more processes and materials than anyone else on the web.

It's interesting to see that Cubic is still supporting Helisys LOM machines (and that your startup offers that as a service). They always produced the most interesting parts, imho (though a pain to post process).

I worked at an RP consortium nearly a decade ago and always figured that the LOM they had was a relic. It appears they still have it running (http://www.rpc.msoe.edu/machines_lom.php).

I just talked to someone from the MSOE the other day ago, I think...

Yeah, those old LOM machines are crazy. The new LOM hotness is coming out of a company called MCor: http://www.youtube.com/watch?v=vow5XvHoFZc

I'm going to be excited to see these machines later this year. Printing with A4 paper means the parts should be so cheap! And they say the consistency is similar to wood...

Hmm, printing paper is not exactly super-cheap per weight unit.

Yes, but weight isn't what matters, it's volume.

I don't see prices on the web site. Would you mind giving me a rough idea of what it costs to print/create various items? (Not a serious prospect, just curious.)

The reason that there aren't prices is that just like the objects, they're all custom!

The cost to make something depends entirely on the exact geometry of the object, the material it's made out of, and various other bits of overhead of the shop that makes it. Sometimes, the orientation of the part in the printer can matter!

Generally, on average, something that an industrial designer would be making would end up costing between $200 and $500.

But again, this is incredibly price sensitive. There's also a trend lately of people making knock-off materials that cost a tenth of the 'real' ones, and some people use those...

I've literally seen orders of magnitude between sellers on the same part. Some people specialize in getting stuff done really quickly, and charge a premium!

Looks interesting. As a potential member of your target audience, I consider a valuable/essential addition to be tolerances and minimum wall thickness for the various processes. Phrases like "moderate surface detail" and "rough surface finish" are pretty useless.

Examples of models produced in the various technologies, what they cost and their characteristics and surface finish (photos) would be very helpful.

I think you need to reduce the barrier between having an stl file and knowing the cost of a part with given tolerances, strength and surface finish. Perhaps this means offering at least one higher priced, longer lead time service with a fixed cost per volume of material. Shapeways does this very well.

Did you see FabFacts? ( http://cloudfab.com/fab_facts/ ) Your first paragraph sounds like a criticism of it, but your second implies you haven't seen it.

In any case, yeah. With FabFacts, we were trying to strike an appropriate balance between too much and too little information, which is why we did the summary + full data sheets. If this is a bit too lopsided from one side to another, I'd love some more detailed suggestions, but it's a _really hard_ problem.

> I think you need to reduce the barrier between having an stl file and knowing the cost of a part with given tolerances, strength and surface finish. Perhaps this means offering at least one higher priced, longer lead time service with a fixed cost per volume of material. Shapeways does this very well.

We're currently (mostly) serving the industrial design/engineering market, which is already reasonably knowledgeable about the field, or at least parts of it. It's easier to start with an informed userbase, you know? But decreasing the expertise necessary is certainly in our plans, and where we want to go... but with only two people, you have to prioritize your time. Most of our current customers want _more_ options, not less.

You're right. I did see FabFacts, but did not see the "More details & datasheets" link within it. Thanks.

Another http://www.redeyeondemand.com/

Zoom RP is amazing. We upload and order parts by noon, they are delivered by Fedex at 10am the next day in Portland. And their Polyjet printers make the best quality small parts.

Can't wait though until these printers get sub $5000 and it makes sense for every designer to have one next to their desk though. HP is getting into the mix, so prices will surely be driven down. http://develop3d.com/blog/2010/04/the-new-hp-designjet-3d-pr...

How much does it cost to print something like a cup?

It's entirely dependent on the exact process and material. Cups are also a bad example, due to their geometry.

You'd probably be looking at roughly $50-75ish+, if I had to guess.

You can see the prices here:

http://www.shapeways.com/gallery

Open source 3d printing machine: http://reprap.org/wiki/Main_Page It can make many of its own components so is at least partially capable of self replicating.

I own a uPrint and run it all the time. In the same day we can print a new version, test it out on the bike, make some changes with a Dremel, measure the changes, redesign in Solidworks and repeat. Once you add time into the cost savings equation it's really a no brainer. We are looking at a much larger one that has almost a cubic meter envelope. BTW the reliability of the Stratasys printers is unbelievable.

I wonder if the same thing will happen to 3d printing as what has happened to normal printing: using high ink prices to subsidize the actual units. I sure hope not: if industries actually build goods using this process the demand will bring the "ink" cost down.

That's exactly how the industry works. If you look at the manufacturers' 10Ks, it's very much the 'razor and blade' model.

And you're also right, it's a real problem when other people make knock-off materials. Lots of people are doing it, and the manufacturers are trying to stop it by making the cartridges super proprietary and making unauthorized materials void your warranty, but that's not stopping everyone...

New (awesome) line of merchandising: put the 3D models up for sale on http://www.Shapeways.com/ and let people order them in various materials & colors.

Home fabrication is very inspirational to read about. The fabrication process maybe extremely time consuming at the moment, but it's a glimpse of what to come. Video on the internet was ridiculously slow once, now it's becoming our main source of video. You saw this video on YouTube, not on your telly. The advantages of 3d printers are so many, that it will only be a question of time before the'll get cheaper, faster and better enough to go mainstream. Amen.

> Maybe the best example are the gloves that Downey wore--which were no thicker than a dime, and could be worn for hours without getting so hot that the dude needed some Colombian Marching Powder to take the edge off

Really?

Indeed. That was quite a non-sequitur in the middle of what was a pretty cool tech article.

Its a reference to his old dug habits that he was pretty famous for. I agree it didn't need to be mentioned.

obligatory mention: MakerBot makes $750 3D printer kits that cost pennies to operate. While the results are not comparable to what is shown in the article, the age of the 3D printer in every home is almost here.

It's already priced low enough that every designer of 3D objects can own one.

Not affiliated with MakerBot. (And they are good guys -- all the plans and source code are available for you to download and build yourself).

We use this things at my job all the time. Amazingly useful when you need a fast engineering prototype, or something to get your head around.

As the costs come down, you'll see a continued move towards 'rapid manufacturing': actually using these technologies to build the final goods.

Reply on news.ycombinator.com