Everyone involved in this should be ashamed of themselves.
The machine pictured is a Stratasys Fortus 900MC, which costs $379,900. An annual lease on a machine of this type is usually about a fifth of its purchase cost, so let's be generous and call it $70,000 per year. Each body therefore costs about $20,000 in machine time alone, before you factor in the cost of materials and the not-inconsiderable amount of finishing that an FDM part requires once it comes off the machine.
There's nothing magical about 3d printing, so we could just as easily injection-mold these parts in exactly the same ABS plastic at a tiny fraction of the cost. There'd be no point, because ABS is a terrible material for car bodies.
Car bodies are largely stiffness-critical structures. In most practical materials, they're much stronger than they need to be once you've made them sufficiently stiff. That's why carbon composite is such an excellent material for performance cars - it's immensely stiff. You don't need to be an engineer to know that ABS plastic isn't very stiff.
In the replies to my comment, there are a few of "what ifs" about the usefulness of 3d printed parts. All such questions can be answered with the following:
If 3d printing produced useful parts, it would be absolutely ubiquitous in motorsport. In reality, it's of no real interest.
Formula 1 teams add new parts to their car at almost every race, iterating as fast as their wind tunnel and computer models will produce data. It's commonplace for an engineer to arrive at a racetrack hours before a race starts, carrying a brand new part. They almost exactly match the ideal use-case for 3d printing - they produce lots of one-off parts, where speed of iteration is vital and money is no object. In reality, most F1 teams use entirely conventional manufacturing processes, even for non-structural components like wind tunnel models.
3d printing is extremely useful in a small number of applications (mainly mold-making) but it's not new and it's not revolutionary. Most manufacturing processes are dipshit-simple, even if you're producing something very complex. Just as in programming, we combine lots of simple processes into a sophisticated whole.
WIRED continues to be rubbish. Every time they report on anything I have any domain knowledge about, I've found that it's full of lies, exaggerations and half-truths. It's best just to assume the stuff they publish that I have no domain knowledge about, such as this story, is the same. It's like they're a tech tabloid.
I agree with most of what you said, but 3d printing DOES allow shapes that you couldn't easily injection mold, such as undercuts and ridges, and you also wouldn't have the requirement for some degree of draft angle for mold/part seperation.
With the exception of SLS type machines (lasers which melt powder) or Z-corp machine (powder with applied binder) many under cuts will require additional support structures to help bear the weight of the part during construction which require manual removal after the fact. Cleverly orienting the structure can help in some cases, but not all. For example, you could create a hollow sphere because you always have a counterweight, but an upright arch would require some scaffold.
The scaffolds are usually easily removed (made of a different material then the main structure), but still require manual removal.
I disagree, but not because I think you're wrong about the feasibility. I've no way to judge, but let's assume you're right: this car doesn't make sense either mechanically or economically.
And the cost to produce it? A few hundred thousand dollars of printer, a few hundred thousand dollars of designers and engineers, a bit of plastic...call if 5 million to be generous. On the scale of a wealthy nation like the United States, 5M is not much money. But the (perhaps tiny) chance that this project leads to discovery or innovation in manufacturing, fuel efficiency, 3d printing, etc. is worth far more.
This man knows what he's talking about. This article is BS and so is the car idea. There's sooo much koolaid in 3D printing right now - even though there's a ton of great stuff in manufacturing in general. It's not the panecea the media are reporting.
(Source: I used to own a 3D printing company doing every type of 3DP process)
I'm not an Engineer, but aren't there limitations on what kinds of shapes you can injection mould? Different limitations than those on 3d printing? For example you could 3d print a closed cell ABS foam-like structure, or honeycomb, which would be lighter than a solid plastic part of equivalent strength.
Every time I see these futuristic cars with continuous forms and seamless integration of body and glass I wonder: how much a collision repair is going to cost?
Sometimes, smart shapes are too smart for their own good. They are at the border of bad engineering.
You could, and can, but the problem becomes the feedstock. ABS, PLA, PC et al just aren't as strong as any of the FRP composites, and are weakened still further by the (lack of) bond strength between print layers.
FDM is great, I love it, but it's not a process for producing structural parts, and won't be for quite a while.
Its not a real car so don't call it one. Until it passed actual crash tests I don't are how much weight savings they claim to have.
I am impressed that someone took 3d printing this far, it shows the many uses traditional automakers will be able to derive from it, mostly on the component sides. There is no reason dash assemblies could not be done this for current makers other that costs and and that printing time. You can bet they have or are looking into such technologies.
There are many tricks to forming car bodies in steel that allow for variances of thickness and such. Hot form steel manufacturing being introduced now allows for thinner and stronger stamped steel parts; essentially a water cooled press.
My understanding is that 3-D printing is still only optimal for prototyping. Once you get into production of any size, traditional methods are much cheaper per unit.
I drive an old Montero. I frequently must wait a week or more for parts to be shipped - if I can even find them online. I dream of a day when I can walk into a shop and pay a premium for parts-on-demand.
You're dreaming of today? Because that's what it is. You can walk into just about any machine shop and get your part as long as you have drawings and material spec. You probably won't be happy with the cost, though.
The point is that the spare parts you need can be made with traditional manufacturing processes because that's how the original was done. Just because you only need one, doesn't necessarily make it easier to produce by a 3D printing machine.
The Rapid Prototyping Center at MSOE has several patents around creating mold casts with 3D printers. IIRC, one of their techniques allows building a tetra lattice coolant path around the part cavity, allowing more optimal cooling than traditional methods (increasing water turbulence, getting closer to the cavity face). They weren't printing the tools directly, but the tools could be made from the parts they generated.
That was about a decade ago, so I'm sure they've moved onto more exciting things.
production: noun. 1 the action of making or manufacturing from components or raw materials, or the process of being so manufactured.
I'd say if you make a factory that takes raw materials and creates a series of one-off items, that you are engaging in production.
It seems like what you're really saying is, "this is not how production has been done in the past". With 3-D printing and related automation technologies, this will start to change very quickly. (It's already started, in fact).
The advantage of a three wheel car is that it is not a car for regulation purposes, it's a motorcycle. Same reason there are tons of small motorcycle manufacturers.
The title is extremely misleading, as the structural parts of the car are meant to be made of a metal tube frame, quote from article:
"The design puts a tubular metal cage around the driver, “like a NASCAR roll cage,” Kor claims. And he also mentioned the possibility of printed shock-absorbing parts between the printed exterior and the chassis. "
Interesting demonstration of the technology, but if you want a light, non-structural body, carbon fibre (or even fibreglass if worried about cost) would almost certainly be cheaper and faster to produce.
It's always the same damn thing with guns. Headlines talk about 3D printed firearms, but then when you actually read the article, the entire extent of the worry is "eventually hobbyists hope to print almost an entire component with a 3D printer." Shudder.
> The title is extremely misleading, as the structural parts of the car are meant to be made of a metal tube frame
It sounds sort of like a classic fiberglass kit-car... IIRC, those things were often essentially cool-looking fiberglass shells on top of dune-buggy-like tubular frames.
Until this company has reached the size required to hire enough lawyers to easily squash wrongful death suits with intimidation and endless court appeals on a regular basis I doubt I will feel safe enough to ride in one.
Half the weight is not always a good thing. Good for fuel efficiency, but anyone that has ever passed a semi truck in the opposite direction while driving a small car has probably felt the force of it push against your car. Heavy winds can also make it somewhat difficult to drive as well.
There's other safety concerns, but those have already been mentioned.
Smaller cars actually tend to do better in heavy winds than tall, heavy vehicles, as there is less surface area for the wind to buffet.
Also, being buffeted by the wind is massively affected by the suspension design. If the car is on soft, wallowy suspension, high up in the air (think Toyota Echo), you'll be fearing for your life. However, bring the car a little lower, improve the shock absorber design, work on the steering and suspension geometry, calibrate the tires to the car (not too wide, but not too narrow), and you can significantly reduce the effect of wind.
The question is how you'd compensate. I don't know of any widely-deployed variable-strength suspension. There is tech around to do it, but I don't think it's generally available.
I'm guessing you wouldn't want to do it with the steering or brakes...
I know nothing about anything on this subject, but could adding downforce (thereby adding friction/grip between the tires and road) help compensate for crosswind? If yes, then surely there must be some reasonable way for a front-end and rear wing mechanism to automatically adjust to create extra downforce during short bursts of crosswinds.
Can anyone acknowledge that it's friggin' amazing they've gotten this far with stereolithography? Yeah, it's an expensive machine which takes a long time to crank out one instance of an object, and requires some non-"printable" components (hey, they're not printing the engine yet!), but really: they've gone from "impossible" to "push button and wait, some assembly required", from making small chincy toys to major fractions of a car.
Printer costs too much? prices come down. Long production time? machines get faster. Materials not strong enough? they're getting stronger.
Sorry if the future will be here next Tuesday instead of today.
Hypothesis: these folks made their prototype the same way everybody makes their prototype, and sold it to Wired as novel by describing a standard industrial machine as one of those 3D printer all the cool kids want.
This is simply false. If a plastic 3D print were "as strong as steel", then we could print an entire gun that would fire more than one round. But we can't, because what comes out of plastic 3D printers is not remotely "as strong as steel".
If the title says it, it doesn't matter what the article says. My objection is to the title, and to the sleazy journalistic practice of promising something in the title that can't be supported by the article.
"Cancer Finally Cured!" -> "researchers report modest progress in alleviating melanoma in cockroach fingernails." (just an example, not real)
"Marijuana use leads to psychosis" -> "marijuana use correlated with a higher rate of psychosis" (this actually happened recently, multiple examples, but correlation is not causation)
Quote: "Cannabis use can cause drug-induced psychosis, trigger the first episode of a psychotic illness, or make a pre-existing psychotic illness worse."
It's entirely false, of course. There is no scientific evidence to support these claims. It's a classic case that confuses correlation with causation.
Would be so cool if you could go to your local shop and they could print you a new fender.
Though I guess those kinds of printers are really expensive and take days to complete a fender. So I guess for a long while you'll have to have a central location do all the printing. Which defeats the whole point really.
Interesting how he wants to install a diesel engine, but run it on pure ethanol, and it has an electric motor too for good measure. (But no mention of batteries apparently.)
Seems like he's trying to mention every buzzword possible, even if they contradict.
To that end, they should hire the guys behind the WikiSpeed car (http://www.wikispeed.com) as consultants. Their carbon fiber✝ costs them $800 to make.
So if this was to scale to, say, 100 000 cars per year one would need 30 000 printing machines. That would be quite a huge factory and it begs the question: who's going to produce the printers? (and how?)
To be fair, this is using a very early technology in a new way. I'm sure if their plans work out, then the speed of production can be reduced, whether it's through customized 3D printers specific to this use case or by other means. I wouldn't knock it just because the numbers are high this early in the development lifecycle.
What would impress me is an entire car, made stem to stern from 3d printers, even if it took a year using Titanium 3d printers and plastic 3d printers.
Also he would probably make more waves/growth if he were to print up things like all-plastic parts for the Honda accord and sell it to local garages for 1/100th of the price.
I don't doubt there are all kinds of new materials out there stronger than steel, but the article clearly states that he uses "ABS plastic via Fused Deposition Modeling", which is a different material and a different process.
I believe that there are UV-resistant (sunlight) treatments for ABS. Many polymers (plastics) are vulnerable to UV light, but are still used in outdoor environments without major ill effects.
And yes, my earlier implication was that it is less strong than steel. A strength-strength comparison is probably not fair here as ABS would be considerably lighter. Even still, steel would be approximately 10X the strength, with about 7-8X the density. A further issue is that 3D printed ABS would not be as strong as bulk-material ABS, although I am told this is improving.
The machine pictured is a Stratasys Fortus 900MC, which costs $379,900. An annual lease on a machine of this type is usually about a fifth of its purchase cost, so let's be generous and call it $70,000 per year. Each body therefore costs about $20,000 in machine time alone, before you factor in the cost of materials and the not-inconsiderable amount of finishing that an FDM part requires once it comes off the machine.
There's nothing magical about 3d printing, so we could just as easily injection-mold these parts in exactly the same ABS plastic at a tiny fraction of the cost. There'd be no point, because ABS is a terrible material for car bodies.
Car bodies are largely stiffness-critical structures. In most practical materials, they're much stronger than they need to be once you've made them sufficiently stiff. That's why carbon composite is such an excellent material for performance cars - it's immensely stiff. You don't need to be an engineer to know that ABS plastic isn't very stiff.
The whole thing is hokum, absolute hokum.