The 3D Printers Are Coming: Dig More Coal?

The 3D printers are coming.  And fast. The only debate is over how fast.

Velocity matters for stock pickers following the small world of pure-play public 3D printing companies.  It is also relevant for business analysts and, perhaps surprisingly, for energy forecasters.

3D printers will — as many have observed sometimes a tad too breathlessly — disrupt a lot of businesses.  They will enable and make more profitable many others, while also creating entirely new classes of businesses.  The 3D printing ecosystem will as well accelerate the new trend of rising foreign direct investment into the United States.  And 3D printing holds the potential to disrupt China.

Most importantly, 3D printing is yet another feature in the suite of new technologies promising rising productivity, and thus in due course both wealth and job creation.

And, contrary to the claims of Al Gore among others who believe that 3D printers will cause energy use to decline as a result of the “dematerialization” trope, energy use, electricity use in particular, will actually rise as the technology goes mainstream.

As a result coal use will increase too, perhaps by as much as one billion tons a year globally.  Print a toaster and burn ten pounds of coal.  Let me explain.

But first, we summarize the technology for those not familiar with the now maturing 25-year-old class of machines called 3D printers.  They work a lot like the 2D printers that render a PC’s words and pictures as images on paper.  A computer-generated image – in this case a detailed 3D map — of a product or object can be additively built up one thin layer at a time, right before your eyes.  Thus the original and still common alternative name of “additive manufacturing.”  The machine’s print head squirts and melts plastic or metal powders not only to specific dimensions and shapes but increasingly specific material compositions.  

Such a capability not only allows rapid prototyping from trial designs, but also extreme customization for complex components, and enables clever designs that are hard or impossible to produce with conventional machining, molding or casting.   And, just like the computers that they are symbiotically married to, all these machines get plugged into an electrical outlet. But more about energy in a moment.

We can parse the state of 3D printing today into three domains; niches, toys, and the promise of wildcards.  All applications emerge from the fact that 3D printing is a remarkably flexible and dynamic technology amenable to use on desktops and factory floors.

The first commercial applications are in niche markets using expensive machines.  Some applications are rapidly expanding, such as medical devices and implants from hip joints to entire titanium jaws.  A patient-specific scan allows the manufacture of a hyper-customized device. This is true as well for many niche industrial devices where small volumes can tolerate today’s machines very slow print speeds, such as rocket nozzles and some jet engine components.

You need to work in metal for nearly any serious mainstream application.  But today’s metal-capable 3D machines run $200 thousand to $1 million.  These are not the tools we see commonly touted in the press as the low-cost desktop manufacturing revolution.  World-disrupting potential comes when you can precisely print metal things at low-cost and high-speed.  That will happen, in due course, but the physics of metals and energy are stubborn.  Meanwhile, on average, 3D printing is still much slower, in some cases untenably slower, than conventional volume manufacturing (so far).

Most of the action today is around 3D plastic printers where we find what amounts to toys, mainly including hobbyists, experiments, designers, educators and artists.  This is the market that is currently driving the forecastthat 100,000 machines will ship in 2014.  Some of these plastic-printing machines sell for as little as $500, cheaper by far than first generation 2D paper printers.

Not to denigrate the utility of printing in plastic. In many cases fascinating and even life-saving uses are springing up everywhere from doctors printing a replica of a patients’ heart to better study a surgical procedure, to on-line computer gamers making physical models of their virtual ‘people’, to pregnant Japanese women offered life-size replicas of their unborn child from a high-res ultrasound scan.

The third domain is the promise for everything, from the untapped potential of high-speed low-cost metal printers, to using the technology to print organic things from food to human tissue, to complete printed products from guns (widely discussed) to bicycles and beyond.  (Though the latter really requires a hybrid of a 3D printer and 3D assembler.)

The business model for 3D printers will be as diverse as their applications.  From home printers to high-end machines that are already starting to show up in the equivalent of a FedEx-Kinkos service model – and why wouldn’t FedEx do it?  — where you can take your design to be fabricated, or your scan, say of your daughter’s face to be printed, eerily, as an American Girl doll.  Expect 3D machines in automotive dealers or pretty much any repair shop where printing a part from a computer file will be cheaper than buying inventory.

Some claim 3D printers will soon be as common in homes as computers.  A better analogy would be the dishwasher or clothes washer since you only need one per house.   One analyst calculated a homeowner could save $300 to $2,000 a year, all costs considered, printing rather than buying stuff.  They calculated this for a basket of products that included an iPhone case, garlic press, razor, spoon holder and, strangely, a perogi mold.   You just have to buy cartridges of plastics and metals for your 3D printer, and download (one assumes, pay for) the digital code for a specific product.

Which business model wins?  Odds are all of them will thrive.  Eventually 3D printers will be found in every factory, classroom, hospital, restaurant, nearly every business, and maybe even most homes.  Will the number of 3D printers shipped ever match the 120 million 2D printers now sold each year globally?  Perhaps.  There was a time not so long ago that no one had a printer. But even if 3D printers only hit a fraction of that scale, it will change a lot of things. But it won’t change the laws of physics.

3D printers still use materials and energy.   In fact, they will use roughly the same amount of material, and likely more energy than the process they replace.  How so?

It would be wildly unrealistic to think that 3D printers will only replace one-for-one things that are currently manufactured the old way, from bicycle pedals, buckles, fuel nozzles and jewelry to iPhone cases and perogi molds.  Even the few examples noted earlier make it clear that printers will be used increasingly to fabricate things that one just never bothered to manufacture or build before – and thus a lot of material and energy will end up being consumed to make things that are completely new.  But such a constellation of future demand is impossible to guess.  Who thought smartphones would end up replacing cameras and generate an astronomically higher volume of photography – which in turns consumes vast amounts of hardware and energy?

So for now we can only reasonably explore the energy implications of producing with 3D printers what is already fabricated.  Let’s look at the $2 trillion manufacturing sector in the United States.

Electricity and natural gas supply nearly 90% of the primary energy used for U.S. manufacturing; 37% from gas and 51% from kilowatt-hours.  Switching to 3D printers means replacing gas with electricity.  3D printers no more burn gas than does your PC.  You plug them in to map out and then heat, melt, fuse, bond and build raw materials into a finished product using electric heaters, or lasers and even electron beams.

Consider, for the sake of a ranging estimate, what happens if everything made in America were fabricated instead using 3D printers.  Exclude the production of stuff you can’t 3D print like raw chemicals, refined fuels and paper, and you find manufacturers burn about 2 trillion cubic feet of natural gas.  You need at least the same amount of overall heat energy delivered into the raw materials inside the 3D printers (more on this in a minute) – but now it comes from electricity, about 600 billion kWh worth of it.  That amount would nearly double the quantity of electricity used for manufacturing today.  It would also, given the grid we have, lead to over 150 million tons more coal burned annually.

Extrapolate this result globally where total manufacturing output is 5-fold that of the U.S., and where coal’s share of the global grid is the same 40% as here (and will be in two decades still, according to the Energy Information Administration).  So if you were to 3D print everything that is today made conventionally, you increase the global coal burn by something approaching one billion tons a year.

Of course this is an overestimate since it is based on the zero chance we’ll see a 100% flip from conventional to 3D manufacturing.  But at the same time we are egregiously underestimating the actual electricity that will be needed when printing replaces burning, or incredibly efficient injection molding by assuming a one-for-one substation of heat energy.    3D printing a plasticobject uses 5 to 10 times more energy per pound compared to conventional industrial injection molding.  You have to make elliptical arguments about optimizing the utilization of 3D printers to offset that energy deficit.  Then there are the metals.

The energy needed to make a pound of metal into a product using a 3D printer can be as much as 100-fold greater than using conventional casting or machining. On average, printing metals requires 10 to 100 kWh per pound.  The world uses metals by the gigaton.  Do the math.

Nonetheless, 3D printing, when it is faster and cheaper, will be wildly embraced because it will be so productive for so many applications.  The extra electric cost for printing the perfect titanium hip joint is worth every dollar and kilowatt-hour.  So a lot more kilowatt-hours will get used when the technology spreads widely.

There are those, as earlier noted Mr. Gore amongst them, who claim 3D printing will offer energy savings because of their precise use of materials, eliminating waste; hence the “dematerialization” legend.   There is a two-fold problem with this argument. It ignores the remarkable and improving material efficiency of existing manufacturing.  And it ignores the unavoidable waste in 3D printers too, given practicalities of machines.  In fact, one detailedstudy of inkjet type 3D printing found some 40% of the non-recyclable ‘ink’ is wasted.  No conventional manufacturer would tolerate that.

But we should make the reasonable assumption that the efficiency of raw material use by 3D printers will, in due course, be no worse than today’s methods.  This is not to say there are no material savings: some will arise from eliminating production of spare parts.  Again this will be a bigger deal in economic terms and in convenience for repairs and spare-parts-on-demand.  It will be a de minimis energy benefit because, for example, there’s a lot more material in your car than in the pro rata amount of spare parts kept in inventory.

Anyway, in accounting for energy in the total fuel cycle we find most is used in the production of the raw plastic or metals in the first place, not in their fabrication into a final product.  And that remains the case no matter whether you choose to fabricate a part or product with 3D or old tech machines.  For plastic products, about 80% of total energy costs are tied up in producing the raw plastic itself.  It is thus relevant to note that one reason you can expect the 3D printing ecosystem to blossom in America is that $100 billion of new chemical production facilities are planned here now, because of the huge cost savings from the hydrocarbon shale boom.  But that’s another subject.

We should also dispose of the third in the triad of tropes about 3D printing and energy: savings in transportation.  You may have heard the claim that energy is ostensibly saved in printing the part on-site instead of shipping it to you.  Well, until physicists figure out how to convert energy into matter (we can do the opposite in nuclear reactors) the weight of the raw feedstock transported to your 3D printer is the same as the weight of a finished product carried by UPS or air lifted by an Amazon drone.

Productivity, precision, speed, convenience, and stunning flexibility… these are the metrics that matter and why 3D printing is taking off.  These are metrics that, in the underlying physics, extract an energy cost.  Always and everywhere.  None of this is news to serious students of energetics.

The environmental implications of 3D printing, because of its explosive growth, is becoming a subject of some academic interest.  See for example Robert Olson’s fine summary in the Policy Journal of the Environmental Law Institute, 3-D Printing: A Boon or a Bane? and Kath Kovac’s article in Australia’s national science agency magazine ECOS, How Green Is 3D Printing.

The how-green debate will doubtless continue.  The more interesting debate is found on the investment pages of Seeking Alpha where analysts argue over which, if any, company will emerge as the HP of 3D printing.  HP [NYSE:HPQ] has a 30% market share of the 120 million printers shipped globally that print 2D words and pictures on paper. So, portentously, HP has promised to soon announce a 3D printer of their own, bringing a giant player into the game dominated now by a phalanx of small companies.  Will HP also pursue an acquisition of one of the comparatively tiny but hyper-valued 3D companies? It should get interesting.

And for the energy analysts, consider that a home 3D printer uses 3 to 10 times more energy than does a paper 2D printer.

The fact that 3D printers will continue the global electrification macro should be unsurprising.  These machines are built from a combination of electric motors, electric heaters and lasers, and electric-powered computers, connected to an electric-fueled Internet.

If you listen to the enthusiasts, we are mere years away from the 3D printer becoming a standard appliance in every home.   Some claim the adoption curve could look the same as it did for PCs.  But the bigger economic story will come from the transformation of the manufacturing ecosystem.   It is not hyperbolic to consider that the 3D revolution of mass customization will be ultimately as impactful as the century-old revolution from mass production.

Investor note

According to industry analysts at Wohlers Associates, 3D printing is a $2 billion industry today and will grow to over $10 billion by 2021.  For those interested in investing in 3D printer stocks I commend the commentators, on both sides of the enthusiasm chasm, at Seeking Alpha.  Companies that are publically traded in this space include: 3D Systems (NYSE:DDD),ExOne (NASDAQ:XONE), Stratasys (NASDAQ:SSYS),Voxeljet (NYSE:VJET), Organovo Holdings (NYSEMKT:ONVO), andArcam  (NASDAQOTH:AMAVF).

Some 3D stocks are starting to exhibit Tesla-class [NASDAQ:TSLA] valuations and expectations.  But there’s a difference.  Tesla is competing in a field where extraordinarily talented and successful companies with deep engineering skills have played for a century.  There is no comparable competition in the 3D printing space.  The 3D printing industry is at the same stage of development as the auto industry was in 1914.