Is the Carbon 3D printing technology a breakthrough, or just hype?
I have been talking about 3D printing both in my job and on social media for 3-4 years now, so I had many people asking me what I thought about Carbon 3D’s new process, which made a big splash on the internet this week. Carbon says it is 100x faster than traditional 3D plastic printing, and the videos are ultra-cool: the company itself says that “We were inspired by the scene out of Terminator 2 (with) T-1000.”
To examine this new technology in more depth, I have pulled in Brian Piccioni, a technology analyst who has repeatedly ranked #1 in in North America and in Canada. I tend to be more of sceptic on 3D adoption, while Brian is more bullish, so I hope our two opposing viewpoints will let you reach your own conclusions.
DUNCAN: My first issue is that there are so many potential problems and not really enough data yet. Almost all of my research in 3D printing came from discussions with those in industry who are out there buying and using commercially available and affordable machines. So they overwhelmingly talk about what is realistic, proven, affordable, and in the market today, rather than what is possible. (I.e. doesn’t tend to be bleeding edge stuff like the Carbon 3D announcement.)
There’s not enough detail in the stories or on the website. We don’t even know when it will be for sale, or how much it will cost!
That makes me wonder about the material properties of the polymer they use. Are the printed objects strong, how do they handle heat, cold, humidity or time? Can they be painted or coloured? While most industrial prototypes are not used to support a bridge, they still do have certain material characteristics in order to be useful. The range of plastics produced by other, slower 3D printers is (these days) surprisingly strong and capable of all sorts of uses.
BRIAN: Agreed. The first Stereo lithography or SLA prototype I had made at Eicon Techologies in 1991. It was as fragile as eggshell and cost a few thousand dollars for a two piece box about 4” by 3”. And somebody dropped it. More recently I had a couple of very large prototypes 4” by 16” printed by SLA in an ABS type plastic. Those cost only a few hundred dollars and look and feel exactly like an injection molded product, and are just as tough. We don’t know if the resin used by the Carbon3D material is fragile or tough. It may be something they can improve on…or it may be an inherent limitation for this kind of technology. We will see.
DUNCAN: Next, the video clips they show look amazing. But they are likely only going to film the demos or conduct the tests that their technology does particularly well at. Which is fair enough, but I am sure there may be other shapes or limitations where their advantage over existing technologies is not as great…what are they? Can this technology scale up to larger size models? Is it likely to be weirdly expensive even at scale? Is the polymer feedstock hard to make, store, or does it cost a lot?
BRIAN: Also agreed. Although the demo is impressive (fine resolution and good speed) the technology may have other significant limitations. That being said, if it really does offer this kind of quality/resolution and speed, even a 100% price premium might not matter.
DUNCAN: In the world of rapid prototyping (which makes up about 70% of the enterprise 3D printer market today) there is a huge advantage in being able to make something in 3-4 hours instead of 3-4 days/3-4 weeks. Making it in 3-4 minutes isn’t as useful: most design cycles just don’t require that kind of ultra-quick turnaround.
BRIAN: Well, not necessarily. For a lot of these things today, you start it up at night and go home so 20 minutes or 12 hours isn’t much difference. But many industrial type 3D printers are used in service centers to produce models for others. In this case, the faster the machine the more models you can make and therefore the greater your income. In general, being able to offer 1 hour turnaround is better than 1 day turnaround. I would also make the point that TODAY’s design cycles can’t make use of 10 minute print jobs…but that’s because there was no point. If Carbon’s technology works as advertised then the design process will evolve to take advantage of (to coin a phrase) rapid rapid prototyping. This reminds me of simulating a semiconductor chip years ago: the software could take hours to run a simulation, so the engineers would wander around and find something else to do. Once the simulation speeded up, they didn’t do that anymore and were able to get chips to market much faster.
DUNCAN: The main reason most people would need that kind of speed from a 3D printer is for final part manufacturing. Making a prototype or an intermediate part (tool, die, cast, mold, jig, etc.) is done in quantities of one and time to manufacture is usually not the critical factor. But if you want to make thousands of objects, getting the speed up is critically important: no current 3D printer that takes hours for making an object can compete with traditional manufacturing technique that makes thousands of objects per hour.
And here we come back to the basic problem of non-metal 3D printers. They (typically) make small, plastic objects out of a single material more slowly than and much more expensively than other plastic manufacturing techniques. The need for those kind of objects is low (of any shape), and the need for the kind of objects that have those characteristics and have shapes that can optimally (or only) be manufactured by 3D printers is even lower still. Hearing aid shells and foot orthotics remain the ONLY mass market needs for volume plastic-only customisation at this time.
BRIAN: Yes, hearing aid shells and shoe inserts may currently be the only volume markets but those are extremely high markup businesses with long turnaround times so slow, small size printers are good enough. If you introduce a high speed printer that can make larger parts, the potential for mass markets expands considerably. One could be quite imaginative, depending on the speed of the printer and the material quality. Many common car parts are plastic and could be printed on site assuming the machine, etc., was up to the task. Currently those parts are warehoused and take a typical 24 to 48 hour turnaround to get. So if you bring your car in for service and it needs a certain plastic part, the mechanic may take 15 minutes to fix your car after waiting a day or two for the part.
If you have a fast enough machine which can produce large enough parts out of good enough material and at the right price, the market will follow.
DUNCAN: That is a very good point…but there are a lot of assumptions in that sentence of yours! So I guess we will have to wait to see who is right. I have some concluding thoughts:
- As we watch Carbon 3D move towards commercialisation, I think we need to remain skeptical. A lot of press releases have come out from 3D printing companies…and they sound great. But if you check, the 3D final part produced was only a ‘test’, or was never actually used (the wrench on the Space Station). Or a 3D part was made once, as a proof of concept that it could be done, but doesn’t seem to be how they make the part most of the time.
- I worry about the finickiness of the liquid process Carbon uses. I could be wrong, but it looks super-sensitive to vibrations, level floors, and maybe even dust in the air. They may be faster, but if they need some kind of expensive clean room, the advantage may be less than we think. I also wonder if it can only be used for one very specific kind of plastic, or is this capable of being ‘ported’ to other materials?
- One of the worst things to happen to the 3D printing market was the positioning of it as “just like the Replicator on Star Trek.” Science fiction movies and TV shows are great, and using them as an example allows the public to rapidly grasp the concept. But it also leads to unrealistic expectations. 3D printers don’t make “Tea. Earl Grey. Hot” and never will. They don’t make communicators or phaser guns. And they don’t make liquid metal killer robots capable of independent thought. They make single material objects with complex shapes at a reasonable price. That’s pretty amazing, but we need to keep our expectations realistic.
BRIAN: Absolutely! 3D printer breakthroughs are like battery breakthroughs and cancer cures: there is a new one every week, and yet, remarkably, very few ever make it to market. Of course, progress is being made although a lot of that is much slower than you would imagine. You are, of course, absolutely correct in being cautious: many products consist of plastic, metal, and, especially nowadays, electronics. It is unrealistic to expect any 3D printer will be able to crank out anything with that type of complexity, moving parts, and so on. Besides, isn’t it strange that in the 24th century the replicator doesn’t know that Picard always asks for hot, Earl Grey tea?
DUNCAN: Indeed. You’d think that a technological society that could create an Artificial Intelligent life form like Data would have figured that one out! Especially for the Captain…