An ex-Formula One engineer has a vision for making manufacturing nimble and ubiquitous.
Michael Fuller spent more than a decade as an engineer at the pinnacle of the automotive racing industry. His experience in Formula One led him to a potentially lucrative idea: using 3D printing to create a new heat exchanger half the weight of existing designs. Heat exchangers — which move heat either into or out of some piece of equipment — are important not only in cars but in countless other industries, including aerospace, chemical manufacturing and refrigeration. When you’re building for speed or sending something into space, halving the weight of a key component is a big deal, so Fuller’s designs could be transformative. The market size for such an invention may surprise you: the heat exchanger industry is expected to be worth around $20 billion by 2020.
For Fuller this heat exchanger is just the first step. He sees his company, Conflux Technology, as part of a much larger revolution in the way we make things. With the new techniques around 3D printing that his company is helping to pioneer, large engineering projects will no longer have to outsource complex assemblies of components to be manufactured on the other side of the world. Instead, critical components and the expertise required to deliver them will be available nearby. He foresees a manufacturing base that is smaller, faster and more reactive, with capabilities a hundred times better than what we have today.
Of course, people have been talking about the potential of 3D printing for a long time. Fuller says that in its early days, Formula One used the technology for prototyping and, later, to produce small parts. Advanced manufacturing was impossible, though, as the technology could not produce the surface tolerances and tensile strengths required. But in the last twelve months, he argues, 3D printing has finally become mature enough. Incumbent companies, watch out.
I spoke to him recently about how he came up with his vision for the future of manufacturing.
[Angus Hervey] When you were young did you know what you wanted to do when you grew up?
[Michael Fuller] As a kid my father used to take me and my younger brother to the go-kart track. It didn’t take me long to realize I wasn’t going to be the next Ayrton Senna. But I still loved it, so I used to tell people I was going to be a racing car maker. After about two years of this my father sat me down and told me it was time for me to put up, or shut up. He helped me draft and send a letter to every Formula One boss saying, “hi my name is Michael Fuller, I live in Australia and I am 12 years old. What do I have to do if I want to work in Formula One?” And to my amazement, I got some replies.
At the age of 13 I started volunteering at a local motor racing team. I did some cleaning, sweeping, looked after the tires, and very quickly decided I didn’t want to be a mechanic. That left me with the simple option of becoming a senior engineer for a Formula One team. Which also meant I knew exactly what university degree I needed. And of course that made choices in high school easy for me. In hindsight it was perfect. Because while everyone else was flapping around, I knew exactly what I was doing and why. That clarity gave me an incredible sense of purpose. It made the pain of studying differential calculus bearable. The concepts may have been obscure… but the goal was always to make racing cars.
What was the Formula One Industry like?
It’s the bleeding edge for motorsport, and an innovation hotbed. That means things move quick. Take the way Formula One used to make brake ducts. An aerodynamicist would come up with a concept and shape, which were then given to designers who would sculpt in CAD. Then a model maker would craft a model to put into a wind tunnel. The engineers would review the results, and it would go back to the model designers who might create five iterations on either side for testing. That meant the model maker now had ten versions to create, and these were all inspected to make sure they were accurate before testing in the wind tunnel again. At a certain point, perhaps four weeks before the race you had to freeze the development and say, “OK let’s go with that design.” That’s because a carbon fiber composite brake duct can have more than 60 parts in a tooling assembly; there’s huge complexity involved in the manufacture of full scale car parts. Now imagine that entire process applied throughout a Formula One car.
3D printing of course, changed everything. Because now you could take a design straight from the computer to a prototype part and constantly make small improvements and changes. When it came to first order aero performance, it meant we could continue the development for longer as the manufacturing lead time was so much less. We no longer had to make a call four weeks before a race since it now took 48 hours to get the part printed. Even though the advantages of that were obvious to us, especially the younger generation of engineers, it still took a while for things to change. Probably four to five months for everyone to come on board. Incredibly fast for any other engineering discipline but glacial by Formula One standards.
When did you come up with the idea for your own company, Conflux Technology?
In my career I’ve done quite a number of engine installations, where you’re responsible for connecting all the systems. In tech speak I guess you could say it’s the physical version of systems integration. Some of the pain that I felt was in the performance of heat exchangers. That’s because there’s so many ways you can lose efficiency — in their size, their weight, thermal efficiency and through power losses due to restrictions to flows. I’ve always been really interested in exploring the potential of metal additive manufacturing, or 3D printing, where you have metal powder laid down and fused layer by layer. It was something I’d experimented with in Formula One many years ago but back then the sizes and densities they could achieve weren’t quite ready. The technology wasn’t mature enough.
About 12 months ago, though, I decided it was time. So I developed an idea for the design of a heat exchanger utilising the geometric freedoms that are only achievable by way of additive manufacturing. One morning in the shower (that’s always where I have my best ideas) a concept popped into my head and I realised I could make it work. I threw some shapes together in CAD. At this time I was consulting to the university sector in Melbourne in advanced manufacturing and heard about a Monash University spin-off company called Amaero which could provide a commercial prototyping service. So for the past six months I’ve used funding from a Victorian government grant with a co-contribution from my own funds to go through iterations of printing and functionally testing prototypes.
What’s so special about your design?
Heat exchangers are profound in their simplicity. They operate in the application of the first law of thermodynamics. Sometimes you need to add heat to a system, and sometimes you need to take it away. How you deal with that heat matters. It could be a closed loop, where a fluid takes heat away from a machine doing work then transfers it to the atmosphere. For example, a car radiator is a liquid-air heat exchanger. Water gets pumped around the engine removing some of the heat and then transfers it to the air. Our skin is another example. We take food in, convert that energy from chemical potential to kinetic, which we use to do work (like breathing or moving) but we also create heat which transfers to the atmosphere via the skin. Any time you can improve the efficiency of how you manage that heat you have more energy available to go longer or faster or work harder.
But in industry there’s been no significant innovations in this area in the last 20 years. We’ve reached the limits of historical techniques that involved subtractive manufacturing, things like etching, bending and pressing plates, brazing and welding. It’s time for the next generation of heat exchange devices. I’ve taken elements from historical designs and brought them together with new geometries. That’s resulted in a compact heat exchanger with high area density, low pressure drop and high thermal exchange performance. We’ve just finished the proof of concept testing phase and we’re already exceeding the performance of the world’s best practice, with a 50 percent weight reduction. That’s pretty incredible.
What kind of applications does this technology have?
We are in the gold rush stage of additive manufacturing technology development. 3D printing machines are getting faster, larger and more versatile as we speak. Creating a product that’s going to disrupt the heat exchanger industry isn’t the main goal though. Instead, it’s the first step that I’m using to test the hypothesis of decentralised manufacturing; the idea of making parts at the point of use. People have been talking about this for years, but we’ve only just gotten to the point on the technology maturity curve where it’s possible. The question now is whether 3D printing can be used to make parts and components that will disrupt incumbent industries at commercially viable costs and delivery schedules.
Once this model gets applied to other manufacturing industries it becomes transformative. Let me give you an example of what I’m talking about. Imagine an engineering firm that’s drilling a tunnel through a mountain. They have a certain number of components that get consumed in the process. That means the parts need to be ordered months ahead of when they’re predicted to wear out, creating these incredibly complicated global supply chains. With this technology instead of ordering out complex assemblies of components manufactured on the other side of the world by specialists, critical components and the expertise required to deliver them will be available on or near site. We’ll be putting 3D printing metal additive machines close to point of use; with the engineering designs that we’ve worked with the engineering firms to develop, and then manufacture them right there. That’s higher productivity, lower lead times, less supply chain risk and less environmental and financial costs.
What’s been difficult about this process?
Nobody in industry is ready to do what I want to do now, which is the serial production of 3D printed metal parts. And while Amaero, the company I used to manufacture my prototypes, have been great at this stage they’re not established to be a serial production facility. It’s also been frustrating to see how long things take when you don’t have the resources you have in Formula One. I’m just not accustomed to something taking this long. However I have to say in my experience the Australian innovation ecosystem has been fantastic.
Where it’s going to get interesting is the next step which is the financing of the pilot production plant. We’re looking at spending around $11 million for that. It’s not the amount that’s daunting (I’m accustomed to working with those sort of budgets) but rather the prospect of raising it in Australia. And I want to do this in Australia because it’s the perfect place for it. We’ve got great engineers and an abundance of talent which can compete globally. Remember, in nominal terms a 3D printer costs the same in China as it does here. Once you take out the high labour quotient as a cost factor the only remaining barriers are the government’s regulatory framework and raw material supply. It means we can compete with China and other countries on a level playing field.
What does the future hold for the manufacturing industry?
I think in ten years we would have just started to prove out the bigger decentralised manufacturing vision, the point of use vision. This is going to create a totally different type of enterprise. It means that suppliers don’t only supply hardware from a silo any more; they supply designs and IP manufactured under licenses by local facilities. Within a decade we’re going to see this scale. And scalability is everything here, as it means higher productivity. You’re talking about a hundred fold improvement over traditional manufacturing techniques. As we see that start to take hold we’ll see these machines spread across the globe, backed by an ecosystem of service supply companies. A new, high functioning, cooperative, cluster based cottage industry will arise with fast reaction manufacturing capabilities that have a greater capacity to value-add. The global supply chains will decentralise and democratise.
Ultimately, this technology means we can do more with less. And that really matters for everyone on the planet.