Making Auto Parts in 2035
Light vehicles will be so different by 2035, experts aren’t even sure we’ll still call them “cars.” Perhaps “personal mobility devices,” suggests Carla Bailo, president and CEO of the Center for Automotive Research (CAR), Ann Arbor, Mich. More important will be the radical changes to the manufacturing of automotive parts.
Let’s start with a prediction that seemingly every industry insider agrees on, even though it requires a massive shift in the kinds of parts needed to build a vehicle: By 2035, at least half the cars made in the U.S. will be fully electric. And Bailo said that’s a realistic estimate some would consider pessimistic. The percentage in China and Europe will be much higher than 50 percent, she added.
Why? Governments around the world are mandating the shift. And automakers are investing so much in the technology that experts like Bailo said it’s very likely batteries will achieve the required energy density to satisfy even range-anxious Americans well before 2035.
Tom Kelly, executive director and CEO of Automation Alley in Troy, Mich., thinks most consumers will conclude that internal combustion engine (ICE) vehicles are a poor choice by 2035. “They’ll think ‘I feel bad about myself. My neighbors are going to shame me. It’s more expensive. And it has less functionality.’ So, after a period of slow growth, EVs will take off, because you’ve reached a tipping point where you’re actually embarrassed to drive an internal combustion engine.” Automation Alley is a nonprofit Industry 4.0 knowledge center and a World Economic Forum Advanced Manufacturing Hub (AMHUB).
As noted above, most experts think smaller EVs will be powered by batteries rather than hydrogen fuel cells. But the latter technology has more promise for larger vehicles. Bailo explained that rolling out a wide-scale hydrogen fuel infrastructure would be more difficult and expensive than electric charging stations. Conversely, she pointed out, heavy-duty vehicles are fundamentally different from light vehicles in that you don’t want them to stop for a long period to charge. “I just don’t know how the economics are ever going to work out for a battery-electric semi-truck. But a fuel cell could really be beneficial.” Brent Marsh, Sandvik Coromant’s automotive business development manager in Mebane, N. C., suggested earthmoving equipment as another example. “These machines require big-time power density. Maybe they move to hydrogen.”
Clearly, we’ll be building far fewer ICEs and far more—not to mention much simpler—electric motors and battery cases. Beyond that, it starts to get a bit murky.
For example, Marsh said gearing is “up in the air. There are so many different drive mechanisms being considered. You can have a motor in the front of the vehicle, or a motor in the rear driving the front and rear separately. You can have one electric motor driving all the wheels, like we do today, or a motor on each wheel. That could be a motor generator on each wheel. There can be planetary gears. …There are many different ways to develop the power transmission and electric motor pack, and it’s going to take time in the market to figure out the best way of doing it.”
Marsh added that Sandvik Coromant sees new opportunities in this environment, owing to very short product lifecycles. “Somebody is going to tool something up, make it for a couple of years, and then go a different way. We envision a lot of tooling and retooling and tooling and retooling, over and over and over.”
Automotive lightweighting has been an obsession for years and will continue, within limits. Bailo said research shows continuing progress in metallurgy, with the steel industry mounting a strong challenge to aluminum thanks to ultra-high-strength steel. “Both industries have started to provide an excellent product, allowing for significant weight reduction.” But she doesn’t envision carbon fiber composites being produced in big volumes by 2035, owing to a manufacturing cost that’s seven times higher.
Marsh said anything related to power transmission that must be made from steel, to include “gears, shafts and even bearings, is shifting to ultra-clean steels with an extremely low sulfur content. Some call them ‘IQ,’ or isotropic quality steel. The reduction in sulfur greatly increases the fatigue strength of the steel. So you can produce a smaller shaft, a smaller bearing and a smaller gear that handles the same power density. This reduces the weight and size of the components, but it’s more difficult to machine.”
Sandvik Coromant is working with steel producers to develop suitable tool materials, geometries and coatings, Marsh added. And chip control is a bigger problem than usual. “They have to be relatively sharp tools, like what you’d use to cut stainless steel. But a sharp edge is usually a weaker edge, so that’s a challenge.”
In general, carbide tooling is the preferred choice for cutting these steels, explained Marsh, “unless the part is induction or laser hardened for a bearing surface or something like that. In that case, we’d use advanced tool materials like CBN or ceramics.” On the other hand, Marsh also called attention to the high demand for cobalt in the production of batteries, which will raise the price of carbide. “We know there’s a somewhat limited supply of cobalt. So we and others are trying to figure out if the carbide of the future will be binderless.”
Bailo said CAR’s studies have shown that over the last decade, material improvements that enable weight reduction have, to some extent, been offset by the addition of new features for comfort or safety. Likewise, batteries with a higher power density will lessen the need to push for more weight reduction. Marsh also indicated that weight reduction reaches a point of diminishing returns, given the nature of automotive transport. “You’ve got to have weight for gravity to keep the vehicle on the ground. We’re not building an airplane. You can make cars only so light.”
This brings us to another profound change that will affect everything from the mix of materials used to build car parts, to their design, where they’re built and who builds them: additive manufacturing (AM).
By 2035, “an impressive number of auto parts will be produced by AM,” said Terry Wohlers, principal consultant and president of Wohlers Associates, an AM advisory firm based in Fort Collins, Colo. “Costs will be competitive with conventional manufacturing for some parts. This, combined with other benefits, will make the use of AM compelling to OEMs and their suppliers.” One of those other benefits is the ability to further lighten some parts, he explained. “Topology optimization and lattice structures can reduce material and weight, sometimes significantly.” Wohlers also pointed to AM’s ability to replace an assembly with a single complex part. “Consolidating multiple parts into one reduces part numbers, manufacturing processes, inventory and labor.”
Wohlers may be understating it when he says “an impressive number of auto parts.” Automation Alley’s Kelly argued that by 2035, “the only time you won’t use additive will be for a reason other than price, such as a metal stamping that’s too big. Additive is the most important technology in manufacturing to come along in 100 years, since Henry Ford created the assembly line. And that’s basically what we’ve been operating on.” In Kelly’s view, AM has many advantages over subtractive manufacturing and only one disadvantage: cost per part. And that disadvantage is rapidly disappearing, he says.
For example, consider LaserProFusion technology from EOS for printing plastic parts. Business Development Manager Jon Walker of EOS North America, Novi, Mich., said this upcoming approach is about five times faster than the company’s fastest commercially available machine, which is itself twice as fast as the previous generation.
“Current technology in plastic AM uses one or two CO2 lasers inside, depending on the size of the machine. As a general statement, you increase speed by a factor corresponding to the number of lasers you add to the system. So, four lasers would be almost four times faster than one laser. But instead of jamming two 70-W CO2 lasers into the machine, by switching to little 5-W laser diodes, we’re able to line up 980,000 lasers in the same space. Instead of using two high-powered lasers, we’re using a million little lasers that can make 100 parts across the bed, for example, with each laser working independently. Or, if you’re building one big part, all 980,000 lasers could act together on that one large part.” Commercializing this technology will be a “huge turning point for the industry,” said Walker. Yet he’s just as sure the machine will be at the end of its productive life by 2035, with even faster systems out by then.
Furthermore, as Kelly put it, “fast is relative. Even if a machine is slow, if I have 10,000 of them and I can make 10,000 parts a day, that’s a different equation. Automation Alley just stood up a network of 300 printers at different manufacturers, called Project DIAMOnD. Each manufacturer owns the same printer, and they use it to make money on their own. But when we need to use all 300, we can make 300 parts at a time. And we expect this network to grow into the thousands. At that point, it’s not a part problem anymore, it’s a logistics problem—how to aggregate the output from all these suppliers.” Not only is that a solvable problem, Kelly argues, this sort of distributed manufacturing has advantages—and it’s the future.
“I think manufacturing is going to go from centralized, expensive and capital intensive to democratic, agile and independent. …The reason we’ve gone with these big assembly plants, or big manufacturers, is because they have to be set up to make one part really well. The advantage of additive is it can make a widget from nine to 10 o’clock, then make cartilage for a knee from 10 to 11. Then it can make a tray for an airplane backseat from 11 to 12. Once you have the capability of 3D printing, depending on the materials needed, you can make anything in the world, in any industry, at any time.”
EOS’ Walker likewise thinks factories might orient themselves around a material, rather than an industry like automotive. “Bridgestone now has a division that makes golf balls, tires and industrial roofing—three industries that have nothing to do with each other. But Bridgestone’s core competency is the chemistry around these elastomeric materials. Even a small company can get unbelievably efficient at 3D printing a particular material. And if they can find common uses for that material across different industry verticals, that’s where manufacturing on demand comes into play.”
What’s more, Kelly postulated, Wall Street is not going to fund businesses that make one thing really well, with a production line that’s profitable only if it keeps making that thing for four years. “Those companies will be forced out of business. … Additive will get the capital, even if it’s inefficient for years and years. Wall Street will fund additive because they are projecting where the world is going. It’s like funding Tesla versus not funding GM.”
Lest you think you can avoid this tsunami, or that it’s only the fever dream of some misguided hedge fund manager, Kelly said he recently spoke with an auto OEM executive who said his company is deeply into AM and very disappointed that the Tier 1 suppliers don’t understand what’s happening. “They’re not coming to us to talk about their additive farm and how it can be used to make our products, … how they’re innovating new ways to do it,” the exec told Kelly. “They’re fearful rather than opportunistic.”
The problem for a Tier 1, Kelly explained, is that AM is very well understood. “It’s time and material, and that’s public knowledge. You can’t hide behind the cost of your production line. The OEMs know exactly how much time it’s going to take to print it and how much powder it’s going to take. And they know the spot prices for the powder. Therefore, you’re just arguing over what margin you need to make, and that’s a very tenuous position for a Tier 1, because most of the time they’re organizing the Tier 2’s and 3’s. But now a Tier 2 or Tier 3 sees a golden age coming. They can actually have a relationship with a GM or a Ford, because the computers will handle all the complexity.”
AM is also “tied at the hip” with the move toward EVs said, Walker. “There are probably five companies within a 10-mile drive of our office in Novi that have a lot of experience in designing something like a crankshaft. And they probably have had that competency for 100 years. But with EVs, there are tons of new parts we’ve never had to make before.” This opens the field to new entrants of all kinds. Walker also referenced the skateboard architecture being used with EVs, in which the electric motors, batteries, suspension and steering are embedded in a few standard configurations, while the body and everything humans regularly contact can be customized. “Additive is perfect for specific niches, when we have low volumes and higher cost per part.”
Both Bailo and Kelly think that because digital manufacturing enables mass customization, the customer will demand it. Or perhaps more accurately, only those companies that take advantage of the constant improvement and customization enabled by AM will survive.
It’s already happening, said Bailo. The Hongguang Mini is quickly filling the streets of China, easily surpassing Tesla sales in recent months, in part because the company is willing to do whatever the customer wants in terms of styling. (See photo of the Mini on the first page of this article.) And it’s not just color. Want your car to be covered in a wallpaper pattern? No problem. Cartoon characters? Ditto. Bailo said she’d read about an owner who spent over $2,000 to cover the car’s interior with brown velveteen, plus dozens of sparkling lights in the roof liner. The Mini costs only $4,200, so this buyer was willing to pay an extra 35 percent just for customization.
“People are not going to wait for a five-year life cycle, or even a two-year life cycle for a minor change,” said Bailo. “Look at what Tesla’s doing: Smaller volumes, changing products rapidly, short development cycles, which then negates the need for hard tools. Soft tools that are made from additive can be used. And people are going to want these products customized just like they can customize their phone today. You’re going to need short run parts at different colors. For ride-sharing services, you’re going to need replacement parts that are going to have to be made fast and onsite. A lot of delivery companies are going to do their own maintenance. So there will be a role for additive.”
Unlike Kelly, Bailo doesn’t necessarily see AM taking over the high-volume parts—much of the skateboard, for example. But for the human interface, it will be essential. She doesn’t think most buyers are all that concerned with who made what under the hood now. And “in the future, the propulsion system will become even more commoditized. It’s something everyone thinks of as their secret sauce, because it’s so competitive in terms of mileage and range. But eventually it won’t be, like the internal combustion engine has become today.”
She expects to see platform optimization and platform sharing, with customization occurring in the “top hat.” Said Bailo, “The way that vehicle interacts with you, the creature comforts, that’s what’s going to drive you to that brand,” Bailo explained. “And more and more, it’s the human-machine interface. Twenty-five percent of car buyers today do not test drive their vehicle, but they do want to make sure their phone will pair.”
As Bailo sees it, “the companies that are going to succeed in the future are those that understand how to analyze risk and then put supply chains in place to manage that risk. … It doesn’t mean that everything is going to local manufacturing. But [companies will] do that very strategically, based on the elements that they consider put them at risk if they don’t have it localized.” Kelly’s notion of a distributed network of AM sites would be a huge help.
Wohlers agreed that “additive manufacturing will help to simplify supply chains for some types of parts,” but cautioned that “it will take years to certify suppliers. The pandemic has motivated OEMs to move in this direction, so the process is underway.” One would think automotive certification for many additively produced parts will be mature by 2035. After all, as Walker pointed out, we already have additive parts in our bodies and in commercial aircraft (including critical jet engine parts). If the medical community and the FAA can certify AM processes and parts, so can automotive.
There’s another, nearly hidden, aspect of AM that helps secure the supply chain: its simplicity and stability relative to subtractive machining. As Walker put it, “our systems are very repeatable because it’s all laser technology. It’s not like a CNC machine where ball screws move and wear over time. … And each ball screw, from serial number to serial number, is going to move a little bit differently. And maybe the motor driving the ball screw wears out, and so on. ... There aren’t really any moving parts in our machines. You have a laser and galvos, and once you’re happy with your setup, you can transfer it to other systems and it’s going to repeat incredibly well. AM is going to enable a lot of companies that aren’t first tier automotive manufacturers today to become automotive suppliers of scale in the future.”
Connect With UsEd Sinkora