Bloodhound is a fast jet, Formula 1 car and space rocket rolled into one. At full speed, the supersonic vehicle will cover a mile in 3.6 seconds. Andrew Hobbs spoke to engineering director Mark Chapman about how the latest smart manufacturing and data analytics technology is enabling the car to attempt the World Land Speed Record at a specially built track in South Africa.
Bolting the jet engine from a Eurofighter Typhoon and a rocket to an aerodynamic car, while protecting the safety of driver Andy Green, is no small task. It requires some of the foremost minds and companies in aeronautical engineering to collaborate, and cutting-edge digital twin and additive manufacturing technology.
We were eager to ask the engineer charged with spearheading these efforts how a diverse range of disciplines and technologies can come together to form an exemplar of Industry 4.0.
Internet of Business: What is your background in engineering and how did it lead to your role at Bloodhound SSC?
Mark Chapman: “I started off as an apprentice at British Aerospace, working on Hawk and Harrier aircraft. I then went on to do a degree in communications and electronics engineering at Bath, before transferring to aeronautical engineering.
“Since then, I’ve mainly been working on research and development projects for various design consultancies, as well as oversees for the likes of Boeing. This has involved everything from systems hydraulics to software application development.
“I was involved in the F-35B vertical take off aircraft with Rolls-Royce, Lockheed Martin and BA Systems, before joining Bloodhound nearly 11 years ago.
“When Bloodhound started there were just two engineers. Anything that looked like a car was overseen by Brian Coombes, who has a background at Red Bull Racing, and anything that looked like an airplane was my stuff – things like integrating a jet engine and a rocket, stability control, and so on.”
Is what you’re doing at Bloodhound all about breaking the World Land Speed Record or is there a wider significance to your work?
“Everyone looks at Bloodhound and sees a blood-orange car that goes really fast, but that’s just a small part of the project. The whole ethos from the beginning has been about STEM outreach for primary-aged children.
“Beyond that, as you find out more about the car and the techniques and methodologies behind it, it’s actually for inspiring engineers too. It’s become an exemplar. It’s a product that’s made by all these fantastic companies that have come together in a creative collision.
We take a lot of techniques out of their normal context, showcase the art of the possible, and push things to the extreme.
“Yes the end result is a supersonic car that will push the land speed record beyond reach for at least a generation, but in doing so, that challenge, that goal, allows you to really advance technologies – whether that’s in digital or manufacturing, or just how people think about design.
“And, actually, that’s a far more relevant, powerful goal than just going faster than anyone else has gone on land.”
Have you had interest from manufacturers who want to apply some of the technologies you’ve developed and deployed?
“We quite often get asked questions about how to become a high-performing team. We are a very agile organisation – we demonstrate how you can be agile, while still mitigating risk. We’re building a product that goes faster than a bullet and carries a person safely.
“To be able to do that in a very agile environment is quite unusual for a technology company.
“I think there are also important lessons in how we try to use off-the-shelf components and then modify them. It’s not about reinventing the wheel. There’s some great stuff out there. It might be in a different engineering niche – you just need to find and apply it.
“For example, we worked very closely with a Norwegian rocket manufacturer on how we handle the oxidiser. There’s a very strong partnership between us and they’re feeding off us as much as we’re feeding off them.”
Do you think these lessons can apply to companies with mass-production lines too?
“Absolutely. The way we approach decision-making and use computational tools is completely scalable. It’s just a case of organisations being willing to give teams a lot more freedom.
The way digital technology is moving is transforming how people work.
“We’ve had a Chinese partner company for a few years now, and going to China and seeing how they work is eye-opening. We don’t need to worry about the Chinese copying stuff, we have to worry about keeping up. They take new ideas and embrace them.”
What are the main incentives behind employing advanced technologies such as digital twin and additive manufacturing at Bloodhound? Is it all about those small gains that mean the difference between breaking a record and not, or is it more than that?
“We rely heavily on computational methods, from design optimisation tools to predictive analytics. We have nearly 900 sensors on the car to gather vital data, which gets transmitted to the cloud via our IoT technology. It’s here that the analytics and performance predictions take place.
“We test the car so that each run is a bit faster than the last. First, we predict runs and analyse that data. Then, we try to match the reality with the prediction, using that to influence what we do next.
Real-time analytics is vital for our car health monitoring; things like looking for fluttering vibrations. We’re very reliant on such techniques.
“We also run a digital twin of the car. Should the car look like it’s deviating from the model by over 10 percent, our driver Andy gets an alert saying to abort the run because that shows our prediction isn’t good enough. In these cases, we re-run the twin and work out where the errors are and why the model is diverging from reality.”
Will that sometimes be down to a missing data source?
“It can be. We know a lot about running the car to just over 200 mph, from our testing in Newquay, UK, so we’ve got a reasonable amount of data. But when it comes to the first runs in South Africa, it’s very difficult to accurately model how the wheels will interact with the ground there.
“The car is like a rolling testbed. We use it as the experiment and that changes how we design, model and analyse it.
“This is different to the automotive industry, where, if you design a new suspension system, you can go onto a rolling road and there’s a huge amount of data that you can put into a model and see the precise effects, using the advances of computational fluid dynamics (CFD) and wind tunnels.
We’re pushing the boundaries, we are the outlier. We can’t run the car in a wind tunnel. It’s physically impossible under the conditions we’d require.
“In everything we do, we have to use the car as the test subject and then design a prediction analysis case that backs that up.
“But we also pull from different specialities. For example, we’ve been working with MATLAB. We did a lot of aerodynamic optimisation using a design of experiments method, taken from how the engine control units in cars are programmed in Latin America.
“They used a method that allows you to match ethanol engines to altitude. We just used different variable inputs to learn how to keep the car on the ground at high speeds.”
How does data standardisation fit into this?
“We have struggled with this. We use four or five analytical tools and two or three modelling programs. You find yourself wasting hours cutting and pasting to CSV files.
“For the first 16 months on the project I was spending time writing scripts in a CAD software program to extract data in a format that would work with our CFD software, because there wasn’t that compatibility.
“I doubt it’s deliberate; they’ve got their own data transfer methods, but it’s very difficult to move data between them.”
Is there simply not a solution out there that can take care of data compatibility? Do you have to build your own custom solution?
“We’ve worked with a couple of companies that provide that service and we were a test-case for them. They’ve written scripts that allow us to work between various software packages.
“Unless you’re an enterprise-size provider or you have a very good relationship with a software company, like we do, you’re left with the lowest common denominator, which is the CSV file.”
How has additive manufacturing enabled you to refine the design of the car?
“We use additive for a number of elements. We’ve got some complex ducting along the car, and the easiest way, because we’re a one-off, is to use additive manufacturing.
“The cost of tooling is huge and can quite easily dwarf the cost of the actual components. Therefore, we use as little tooling as possible. Additive manufacturing is a fantastic solution for this.
“While we have used it to optimise some components and it’s enabled us to manufacture stuff we couldn’t do any other way, it’s also great for creating short-run components very economically and in special materials.
“We’ve got parts made from carbon, titanium, and alloy steels. That kind of flexibility allows us to create custom components. For example, when it came to the steering wheel we scanned Andy’s grip and then ran an online competition to source a design.
“We printed some of the results in ABS nylon, for Andy to try, and the one that was most ergonomic for him, and provided the functionality we needed, has been printed in titanium.
“However, I would say there isn’t enough experience in using additive manufacturing amongst designers yet, so most aren’t applying it to its full potential.
Traditional machining is sometimes still the best solution, but if you have additive in your toolbox, and you know how to make the most of it, it’s a incredibly powerful technique.
Bloodhound announced last month that it was entering administration. What’s the current outlook for the future of the project?
“Administration is never a good place to be, it’s not where you’d want to be. However, there are various options being progressed at the moment. And I think there will be a good outcome. It probably won’t be a blue and orange car that comes out of it, but the ethos and the goal will be the same.
“I think it will become much more technology focussed. That Edwardian ambition of being the fastest is worthy but it isn’t as relevant today.
“What is relevant is showing companies what technologies they can use and becoming an exemplar for things like digital twin, 5G, and interoperability – future-looking technologies.
“We’ll see what happens over the next week or so, but the project is definitely continuing. It will just be in a slightly different vein.
“This time last year, I was talking to Elon Musk about technology and the question of how you make that step up – its vital to set bars for others to strive for.”
You can hear more from Mark Chapman on digital twin, predictive analytics, and data standardisation at Internet of Manufacturing 5-7 February 2019, Munich, Germany.
