As the 3D printing hype reaches a fever pitch, you're starting to hear a few more naysayers quietly chime in that all this talk of a revolution might be just a bit overblown. As a mechanical engineer and the head of a product design firm, I've had a front row seat to 3D printing for the last 20 years. There is little doubt that additive manufacturing processes have completely revolutionized product design and development. And, in certain situations, they hold tremendous promise for manufacturing. But before you sell your injection molding machines and shutter your factory you might want to pause and take a deep breath. We're not quite there yet.
To help demonstrate the potential - and some of the limitations - I thought I'd try an experiment. I had shattered a small cooling fan blade on the hydrostatic unit for a zero turn lawnmower I own. Our firm happens to own a 3D printer. So why not "build" a replacement instead of buying one?
Shredded fan blade
In many aspects this should have been a slam-dunk for 3D printing. For starters, it's a single part made from a uniform plastic material. And that's good for our experiment. Because by and large 3D printing is limited to a pretty narrow range of materials and isn't generally applicable to highly interactive assemblies comprised of different material types - which, unfortunately, describes most all products.
But for my experiment I just needed one plastic part and it wasn't even a particularly complex one. It didn't contain multiple material types, didn't have particularly tight tolerances and didn't hold any electronics. I wasn't trying to 3D print an iPhone. This should be easy.
Before 3D printing my replacement fan, there was the minor issue of needing a 3D model - the mathematical 'blue print' of my part. Without the 3D model, the 3D printer has no idea what to produce. I guess I could have called the supplier and asked if they'd kindly send me a CAD file. For free. I could only imagine how that conversation would go. Umm, nope.
I wasn't going to let this stop me. So I pulled out the calipers, fired up my CAD system and after a while had a reasonably close 3D model of the fan.
3D CAD model of the fan blade
Not a perfect model. But for my simple test case, this was good enough. My fan was, fortunately, a very low-tolerance part - I just needed to match up some clearance holes and move a bit of air, which is a good thing, because in general, 3D printing makes some pretty significant compromises with respect to surface finish. There are exceptions but when compared with the old fashioned processes it purports to be on the verge of replacing, 3D printing often comes up lacking.
But the requirements for my fan blade seemed pretty pedestrian so 3D printing ought to do just fine. Looking good so far.
Now I just had to build the part. 3D additive processes come all manner of tradeoffs with respect to process, surface finish and material choices. With one common denominator: They are all excruciatingly slow.
3D Printing Software predicts a 16 hour build time
The 3D printed fan blade before removing the support material
Our 3D printer built the fan in a breathtaking 16 hours and 20 minutes. Now, to be fair, this is an older 3D printer - we bought it about eight years ago and paid about $55,000 for it. We recently replaced it with a comparable newer model which cost about $45,000.
Our newer 3D printer is state-of-the-art and predicted it would run the fan part about 17% faster than our older model. And the price has dropped about 20%. All good evolutionary advancements over eight years. But certainly not the quantum leaps that are going to be necessary for 3D printing to replace conventional manufacturing methods.
But none of this mattered to me for my test fan. I just needed one part and I wasn't in a particular hurry. At this point I had a bit of my personal time invested in it and I'd run our machine an afternoon and overnight. The whole process had taken me about three days and I had myself a replacement fan blade. A 3D printing success!
|The replacement fan|
|Installing it in the lawnmower|
So now for the reality check...
Assuming my personal time is worth only $30/hour and including true material costs, the 3D printed fan blade I built probably cost me about $180. Ouch. At a more "professional" rate for my time the cost of the fan rapidly escalates beyond the painful to the ridiculous.
Even assuming my time is free - and if I'd gotten a CAD model from someone for free - and if I could use a 3D printer for free - I still consumed almost $56 in ABS and support material and tied up a $45,000 piece of equipment for a day. Ouch again.
But what would it look like if we were going to use 3D printing as a production manufacturing process for this part?
Let's assume that the production rate on this is approximately 320,000 units per year - fairly modest since there are two per lawnmower and this is a pretty common hydrostatic drive unit.
Assume we're willing to run our 'factory' 300 production days a year and 20 hours per day (leaving the rest for maintenance, downtime, etc.). We'd need to produce a part every 1.56 minutes. With a conventional 300-ton injection molding machine covering about 300 square feet and a $50,000 mold that's no sweat. We'd have our choice of materials and could mold the fan with a mirror smooth surface finish if we'd like. In fact, if we wanted we probably have capacity to burn and could run another part or two on this same machine without taxing it.
To meet production rates with the current 3D printing technology, we would need to purchase just a bit over FIVE HUNDRED 3-D printers!!! The equipment cost - more than $23 million. Those 500 3D printers would probably occupy around 25,000 sq. feet of manufacturing space. For one smallish, fairly boring part. Presumably, I'd get some sort of a volume discount on an order of 500 but the gap is so huge I can't conceive of it being closed in any time soon.
But wait - there's more.
A run-of-the-mill ABS plastic cartridge for our 3D printer cost around $260 for about 56 in^3 of plastic raw material. That works out to be around $123/lb.
The street price on injection mold grade ABS right now is about $2.70/lb. So the 3D printed material is running at about a 50X penalty.
Oh yeah - and then there's the 3D printing support material - the consumable stuff necessary to support the part while it is being built - another $250 for about 56 in^3. And my part needed more support material than actual production material. And this gets thrown away.
Oh yeah - and then we have to soak the 3D printed part in a caustic bath to strip the support material. I'm not even sure I understand the cost and environmental implications of all that gunk.
So what does all this add up to?
To produce this fan blade via a production process using FDM at today's technology I'd estimate a per-piece price of $69.77. That'd be fine if I wanted to spend $350,000 on my lawnmower. I'm not quite there yet.
To produce this fan blade via conventional injection molding I'd estimate a per-piece price of $.79. That seems about right for a part of this nature.
I know I know. The 3D printing costs are inflated because at huge volumes the costs would go down - but I just don't see an 80X cost disadvantage disappearing any time soon. Particularly given what appears to be a fairly slow pace of development.
So what about these huge "advancements" we keep hearing about in 3D printing? Twenty percent improvements in cost and speed over eight years are great but they're not going to render your injection-molding machine obsolete. So where is all the 3D manufacturing "hype" coming from? Presumably from people who've never seen the inside of a factory. A breathless fascination with the ability to watch parts magically appear before our very eyes is fine - but extending that to the context of manufacturing is a fool's errand.
Am I a 3D printing hater? Hardly! For those people at the leading edge of innovation, 3D printing is a powerful tool in the design and manufacturing process that has unleashed new avenues of amazing creativity. It has simply redefined the development process as we know it. But to extend that to mean the demise of all-manner of conventional manufacturing strikes me as a bit naive. Well more than a bit.
It would seem to me that in order for us to see a dramatic uptake on the adoption of 3D printing for more mainstream (whatever that means) manufacturing processes, there would need to be non-linear, exponential improvements and gains being made in costs, materials and speed. In spite of the hype, I don't feel like I'm seeing that. FDM, SLS, SLA - there seem to be many of the same basic 3D printing options that have been around for five to 10 years and some going back as far as 25 years. There are now low-cost, low performance versions available and they're a lot more visible to the public but the basic core technologies aren't really changing that much.
I think the current manifestation of 3D printing simply doesn't overlap enough with current paradigms of mass manufacturing, distribution or sales. Of course it's possible those paradigms are outmoded and in the future we'll spec our products online and watch them get manufactured and assembled locally (or something like that) but I'm betting we're several generations and technological revolutions away from that. I believe Ford makes an F150 pickup truck every 40 seconds or so. Our 3D printer is awesome as a design tool but in 40 seconds it can make a part about the size of - well - nothing. To keep up with automotive-like production rates I'd need an army of 3D printers cranking away and when compared to a conventional injection molding machine - which can spit out a fan blade in a material of my choice in 40 seconds for a variable cost of a less than a buck - I'm not sure I see who benefits from this. I fully realize there are different paradigms for "design" and "manufacture" but I certainly don't see those overlapping within the reasonable future. I can see a few manufacturers employing 3D printed parts - as a trial or even a marketing gimmick - but I just don't see the huge benefit to either consumer or manufacturer.
Consider all the ways that just "plastic" parts can be made - 3D printing, vacuum forming, urethane casting, bulk molding, SMC, pressure forming, RIM, blow molding, glass layup, unit tooled, MUD bases, P20 tool steel injection molding and multi-cavity injection molding - and that each of these represents a unique point on the capital cost, variable cost, material efficiency trade-off. And these are just plastic parts. By the way, Matweb.com lists over 75,000 polymers in their database. My 3D printing partner offers "the widest range of materials in the 3D printing world!!" Eleven.
3D printing is another interesting tool in the chest - not the finest and last one. And its inherent tradeoffs mean a pretty limited intersection with the hyper-optimized efficiency that is required to manufacture most of the products we use on a day-to-day basis.
3D printing gives you some good stuff:
+ Exceptional design flexibility
+ Zero tooling costs
+ Unlimited part-to-part flexibility with no change-over costs
+ Fairly low capital investment
... and some not-so-good stuff:
- Agonizingly slow speed and astronomically high per-part costs
- Limited material choices with compromised properties and performance
Given the fairly stagnant pace of technological progress on the two negatives, I still see 3D printing relegated to an interesting, niche but fairly minor manufacturing role. Custom medical prosthetics, "art-to-part" duplication of singular designs (dental implants, etc.) and personalized Oreo cookies.
My bet is that 3D printing will have its place in manufacturing but that for the foreseeable future that role will be quite small and niche: ultra-low volume, highly tailored parts from a very small set of materials that can tolerate exceedingly high variable costs, loose tolerances and modest material property requirements. We just don't need that many more Yoda figurines and awesome iPhone cases.
Oh... one more thing.
It turns out that the original fan blade was molded from a polypropylene blend.That material was chosen because it is cheap and oil resistant but also very tough and forgiving. My 3D printed fan blade ran for exactly 10 minutes and then shattered into a million pieces.
I bought a replacement injection molded fan blade for $6.77 off of Amazon. It arrived the next day.
3D printing is very cool. But it has a long way to go to match the hype.
Dave Franchino is president and principal of Design Concepts, Inc., a product innovation consultancy specializing in helping clients build brands and business through product strategy, research, design, engineering and prototyping. He brings nearly 25 years of product development experience as a corporate practitioner, academic and leader of a product development and strategic innovation firm.
Drawing on his diverse experiences as a manager within a Fortune 50 business, a small business owner, an entrepreneur and a consultant, Dave coaches and mentors his clients on developing and sustaining innovative cultures and creating world-class design solutions with true business impact.
Franchino holds a master's degree in mechanical engineering - manufacturing systems engineering, from Stanford University with a focus on integrated design, manufacturing and marketing. He holds a bachelor's degree in mechanical engineering from the University of Wisconsin-Madison. You can follow Dave on twitter @davefranchino.
Editor's note: The author is a PlasticsToday contributor. The opinions expressed are those of the writer.