The Big Project
Express delivery: inside the Bloodhound supply chain
Engineers and suppliers are uniting in the effort to make Bloodhound a reality.
Around the country, something big is starting to stir. Painstakingly crafted 3D designs are beginning to take real form; carefully researched components are arriving at test benches; and fuel and lubricants are starting to flow. Engineers are turning the much-simulated shape of Bloodhound SSC, the vehicle that will aim to surpass a thousand miles an hour in South Africa’s Hakskeen Pan desert in 2014, into a reality, and the project’s headquarters - the riverside industrial unit in Bristol known as the Doghouse - is getting ready to receive its long-awaited resident.
Like every engineering project, Bloodhound is a collaboration; between engineers of different disciplines and with different backgrounds, but perhaps most of all between the core team, which is led by chief engineer Mark Chapman, and more than 200 suppliers. However, in many ways the project is unique: there is only a single product, very few spare parts, and the relationship between client and supplier is driven by mutual publicity rather than the more common commercial considerations.
Putting together a supply chain for this project was a tough challenge, said Conor La Grue, Bloodhound’s engineering lead and supply chain manager. On the one hand, the draw of being involved in a high-profile project with such a tantalising, attention-grabbing goal and where publicity is guaranteed was a powerful one. On the other, the engineering demands of making a car that can go almost 30 per cent faster than anything that’s ever travelled on land are considerable. The suppliers would need to be a select bunch.
Early on, the team was focused on the design of the car. ‘You can’t design without tools, so very early on we had a partnership with Siemens and Unigraphics software, and also with Intel; they provided us with a multi-million dollar solution for CFD stations and infrastructure, which allowed our designers to work,’ La Grue said.
The focus then shifted to finding organisations that would be able to build Bloodhound’s primary structure - the chassis, body, supporting structure for the jet engine and hybrid rocket that will propel the vehicle, and the wheels. ‘In the first year of the project , we had something like 200 meetings and covered an awful lot of miles trying to find those key partners,’ La Grue said. ‘They had to be the right kind of people to meet our core engineering needs, which are absolutely paramount. And doing that during the deepest recession in living memory was no small challenge either.’
Although Richard Noble has been involved in two previous land speed record projects - Thrust 2 in 1983, which he drove himself, and Thrust SSC in 1997, the current record holder at 763mph - the engineering demands of Bloodhound meant that the team had to start from scratch. Only Noble himself, the driver of Thrust SSC, Andy Green, and chief aerodynamicist Ron Ayers came over from the Thrust SSC team. ‘It’s really a new generation of car, with very little in common, technically, with Thrust SSC,’ La Grue said. ‘The level of technology brought to bear on the 1,000mph target is a leap further on again, so we had to start from scratch. When I came on board in August 2008, we had a list of maybe 65 suppliers, and they were all for some of the test rigs, nothing for the car itself.’
The project set out to be cost neutral, and the partners for the primary structure needed to be capable of meeting the stringent engineering requirements. That meant, in basic terms, looking for companies with experience of things that go fast while keeping the people inside them capable of working at a very high level of efficiency, as well as safe. That immediately suggested two sectors: aerospace and motorsport. The Bloodhound team worked with the various industry associations in both sectors, as well as calling on its own experience to identify likely possibilities. Then the road trips began.
Setting out just as the financial markets began their catastrophic crash, Noble and the team, which by that point included Formula One (F1) veteran John Piper as chief designer, found they had a new set of problems. ‘People are looking very much at their bottom line,’ La Grue said. ‘Getting involved with something that basically doesn’t want to pay market rates and is looking to generate benefit for you when you’re worried about your own profitability isn’t easy.’
It took some time for the partnerships for the long lead primary structures to come together, he added. ‘The choice always has to start with the engineering requirement. There can be absolutely no compromise on that; everything on the car has to perform exactly the way that we need it to.’
Talking to the Bloodhound team, it quickly becomes apparent that although the shape of the car and its specs are very familiar, as a manufacturing project it is still in its very early stages. Chapman recently confirmed this in his blog for The Engineer’s website. ‘The difficulty we’ve faced is that from day one we’ve always had fantastic images of the car, which gave the illusion of a finished design,’ he said. ‘And to be fair, the external shape hasn’t changed significantly in well over a year.’ But although the shape looks the same, in fact a great deal of development has been taking place, and this has affected how the supplier relationships have been built up.
For example, the rear primary structure and chassis of the car is being made by Hampson Industries, a Wigan-based supplier of tooling and composite structures for the aerospace industry. Hampson is building the spaceframe at the rear of the car, which will support both the jet engine - a Rolls-Royce Eurojet EJ200 of the type that powers the Typhoon fighter - as well as the hybrid rocket, a new design produced by the Falcon Project. This partnership has been in place for more than two years, La Grue said, but the contract to provide the car’s towering rear fin is only now being put in place. ‘The build of the rear chassis is now well under way, but it had to reach that level of maturity so that we had a clear definition of the fin.’
Although it might seem like decisions are being taken at the eleventh hour, this is not the case. ‘We’ve only been able to fully engage with partners once the project reaches the right level of maturity anyway,’ La Grue said. ‘Ideally, we mature the design in the last stages in a design-for-manufacture way. That makes sure that we’re making all the parts as easy as possible for the suppliers to manufacture, while also never compromising on the engineering requirements.’
The reality of the project is that the designers keep improving all the parts of the car for as long as possible until the design is frozen and it goes off to be built. ‘Yes, publicly we might say that we have an answer on a particular part, but where there’s any opportunity to keep improving that part before it has to be handed over to the design-for-manufacture team, then I’ll take that opportunity,’ La Grue said.
This does not mean that the suppliers are shut out of the process; indeed, as far as the Bloodhound core team is concerned, the suppliers play an essential role in the design process. ‘Whenever we’ve grown the team, we’ve looked for suppliers with real breadth, who can bring us a high level of understanding of materials, processes and practices; that’s vital for safety on this project,’ La Grue said.
“We’re facing a big unknown in trying to break the record, and the way we tackle that is to stack up an awful lot of unknowns”
Conor La Grue, Bloodhound
Peter Watt, group design manager for Umeco, which is building Bloodhound’s forward chassis and monocoque, is a part of this process. ‘It’s mainly been advising on design for manufacture, and some materials recommendations for us,’ he said. ‘I have a fair bit of background in automotive structures and motor racing, and I’ve made all the mistakes. We’ve built hundreds of monocoques before for all levels of racing. You can pass on that knowledge to help them out.’
This kind of experience with the supply chain is vital, La Grue said. ‘We’re facing a big unknown in trying to break the record, and the way we tackle that is to stack up an awful lot of knowns. For example, with the composites, we’re using three- or four-year-old resin weave and core technologies that have been crashed umpteen times and are very well understood. We’re stacking up people’s experience, the history of the material, good design and the validation that comes with that experience and history to provide a whole lot of knowns to solve the very difficult unknown of what’s going to happen to the car when it’s travelling across the desert in excess of 1,000mph.’
This approach is part of Ayers’s design philosophy. A veteran aerodynamicist who worked on the Victor delta-bomber - for many years the basis of the RAF’s in-flight refuelling aircraft - and the Bloodhound surface-to-air missile, Ayers decreed while designing Thrust SSC that the priority of the project was to design the safest vehicle possible and then make it go fast, rather than designing the fastest vehicle possible and making it safe. ‘He’s enforcing the same philosophy on Bloodhound,’ La Grue confirmed.
Because of this, Bloodhound is a mixture of custom-developed and tried-and-tested off-the-shelf products. ‘On the main structure, pretty much everything is bespoke,’ La Grue said. ‘If I had a full bill of materials for the car, which is still evolving, it would be of the order of 3,000 components, and at least two-thirds of them will have been drawn and produced specifically for Bloodhound; the rest, even though they are nuts and bolts and connectors and so on, many of them would be tailored in some way.’
One example of off-the-shelf components are the on-board processing chips that will run Bloodhound’s computers: low-power Atom chips from Intel, the project’s IT partner.
A second example is the wheel bearings, which come from US supplier Timken. The first components to arrive at the Doghouse, these were supplied by the company from stock after receiving the data on the wheel speed and the loadings they would have to endure. As the design has evolved, the weight of the car has gone up - from around five tonnes to nearer six - which means the loading on the bearings has also increased. Timken has recalculated the lifetime of the bearings based on these new figures and found that it is about 50 hours - still way more time than the team needs for a series of record-breaking runs.
In contrast, some of the systems are unique to the car, and in this, the team is working closely with suppliers to find the appropriate solution. The jet engines are a good example of this, La Grue said. ‘These are cutting-edge pieces of kit, just about the most advanced jet engines that are around at the moment, and part of the agreement was that we had to be able to use them. Bloodhound goes much faster than a Typhoon does in the air, and it does it on the hard deck, so we’re presenting pressures at the surface of the turbine blades that are way in excess of those you’d see in normal flight,’ he explained.
This means that the design of the intake for the engine has to be very different from that used on the Typhoon. ‘We need an intake that is very efficient at standstill, but that is also fully efficient at Mach 1.4 and all the other points in between along our 10-mile track. That’s a unique design constraint and we’ve had support from Rolls-Royce - from people who understand the complexity of the engine and its performance.’
Other suppliers are using the unique demands of the project to develop technologies that could have further commercial applications. Cosworth, for example, is working on the systems that will harvest data from the car during its record-breaking runs and relay it back to its audience in Bloodhound educational outreach, an important facet of the project that involves hundreds of schools, colleges and educational institutions.
Cosworth’s head of marketing, Pio Szyjanowicz, said that getting the data off a fast-moving object as it passes a fixed point is quite a challenge. ‘We’re looking at a new system at the moment, based on some next-generation racing car telemetry that we’re rolling out. When the rolling chassis is ready to do some runway tests, we’ll start to plug the boxes in and make sure they work.’
“Bloodhound is not only a draw in itself. It is also the world’s fastest laboratory, and provides an opportunity to promote engineering”
Most of the suppliers are involved for promotional purposes; Bloodhound is not only a draw in itself, but it’s also the world’s fastest laboratory. A component that can survive the rigours of travelling at a thousand miles an hour while in the hostile conditions of a desert will be able to cope with most things. But it is also an opportunity to promote engineering itself, and Szyjanowicz believes that it could help people to understand the culture of the industry.
‘There’s no guarantee it will succeed,’ he said. ‘Everyone’s giving it their best shot and we’ll learn a lot in the process, but we still need to find a way to get everyone engaged with the challenges, and particularly to communicate that we learn more from mistakes than successes. If we can generate an acceptance that failures along the way aren’t a bad thing, we’ll have come a long way in lifting the lid on what engineering is about and what we do.’
That said, the prospect of failure is not one that is entertained much at Bloodhound headquarters. ‘Everything on the car has to perform exactly the way that we need it to,’ La Grue said. ‘There’s no trivial part of the car on the project, so you have to have buy-in from absolutely everyone. All the suppliers have to feel part of the team.’
The ins and outs of the car on which the Bloodhound team’s record hopes are pinnedThe Bloodhound Project aims to set a new land speed record of more than 1,000mph (1,600km/hr) at Hakskeen Pan, in South Africa, early in 2014, beating the record set by Andy Green in Thrust SSC, 763mph (1,228km/hr), in 1997.
- Length: 13.4m
- Height at fin tip: 2.8m
- Wheel diameter: 0.9m
- Mass fully fuelled: 6.4 tonne
- Design speed: 469m/sec
- 0-1,600km/hr: 42sec
- Wheel speed: 10,000rev/min
- Rolls-Royce EJ200 turbojet thrust: 90kN (0-300mph)
- Fuel: Kerosene
- Falcon Project hybrid rocket motor thrust: 122kN (300mph max speed)
- Fuel: hydroxyl-terminated polybutadiene; high test (86 per cent) peroxide oxidiser
- Cosworth V8 2.4-litre F1 CA2010 APU: 588kW (rocket oxidiser pump power)
One of the world’s leading suppliers of computer chips, California-based Intel was among the first partners to join the Bloodhound project
Intel’s involvement in Bloodhound was a product of a single person’s enthusiasm, said spokesman Alistair Kemp. ‘One of the managers in our platform technology engineering group is very interested in the land speed record,’ he explained. ‘He was looking into Thrust SSC, found out about Bloodhound, got in touch with Richard Noble and started a discussion about how we could help.’
Initially, Intel’s involvement was in the design of the car. The team had been using a computer at Swansea University to run CFD simulations of airflow around the car, but the scarcity of available computing time meant it was running behind schedule. Intel stepped in to offer aerodynamicistRon Ayers the use of a supercomputing cluster to process the CFD data, helping him get back on track and complete the basic design of the car.
‘From there, we started looking at other areas where they were using computer technology, such as the internal management of the on-board systems,’ Kemp said. ‘We’d just launched the low-power Atom chip and we thought it might be useful for them. These chips are a standard product.’ Also on board are solid-state data drives, which will act as Bloodhound’s ‘black box’.
In most cases, Intel has supplied technology that Bloodhound has found a use for, rather than the team requesting the company’s input, Kemp said.
Umeco supplies materials to a variety of industries and manufactures tooling for the aerospace sector
Working on the Bloodhound project represented a return to Umeco’s motorsport roots, according to group design manager Peter Watt. ‘We still have a lot of expertise in motorsport; we sell materials to 80 per cent of the F1 grid,’ he told The Engineer. ‘But in terms of making parts, this is like old times. Bloodhound is a one-off project and our expertise and track record are the main reasons they approached us.’
At the time Umeco joined the project, the design was in its infancy. ‘There was nothing to speak of on the design, bar a few basic layouts,’ Watt said. ‘Ron Ayers had been working on it for a while and he knew the basic shape of the thing, and there had already been a few major changes. It’s evolved a lot as the years have gone by.’
Umeco is working on several components. The largest is the monocoque, the carbon-fibre composite safety cell that surrounds driver Andy Green. It is also making the nose section, which runs from the front end of the monocoque to the tip of the car; the gas turbine intake duct; and the aerodynamic rear-wheel farings. ‘Those have been redesigned quite a few times and they’re still working on them,’ Watt said. ‘They’re responsible for two-thirds of the total drag on the car.’ The monocoque is a vital part of the structure. As well as protecting Green, it also has to withstand all the vehicle loading. ‘We’ve gone for well-proven systems in terms of materials,’ Watt added.
The techniques used are similar to those found in conventional motorsport. ‘The structure isn’t very different, even though there’s a big jump from motorsport speeds to land speed records; there’s just more of it,’ Watt said. ‘It’s generally a bigger structure anyway, but it’s also more of a hybrid; there’s a reasonable amount of motorsport detailing, but where the rear spaceframe joins onto the monocoque, we’ve gone for an aircraft-type mounting system. It’s less risky - we only have one shot and it has to be right.’
Watt has found the atmosphere on the project refreshing; less formal than aerospace projects, although it has necessitated some process changes. ‘We know we’re getting information piecemeal, so we have to work around that,’ he said. ‘We have drop-dead deadlines for certain bits of information and Bloodhound know when they are. We’re starting to lay out the tooling for the monocoque on CAD and in a couple of weeks we’ll starting machining bits of block for the tooling. Once we start building, the team we have on the project will start to grow.’
High-power engine supplier Cosworth has recently been growing its non-motorsport business, particularly in electronics and defence
Cosworth is a major partner on Bloodhound, providing several different systems for the car. The most important of these is an F1 V8 engine, which is an integral part of the hybrid rocket system that will propel the car from around 300mph to its full speed.
The hybrid rocket uses a liquid oxidiser, hydrogen peroxide, to facilitate the burning of its solid synthetic rubber fuel. The Cosworth engine powers the pump - derived from a component on Britain’s now-defunct air-launched nuclear missile - which squirts the oxidiser into the rocket fuel.
‘We knew we needed a great deal of power per unit volume,’ said Pio Szyjanowicz, head of marketing. ‘We proposed that we needed something current and in operation, where we know all the operating characteristics, and that suggested the F1 engine.’ Currently powering the HRT and Marussia teams, the CA2010 engine has a 2.4-litre capacity and produces in excess of 300bhp per litre.
The duty the engine will perform on Bloodhound is very different from that of an F1 race. Instead of running for two hours at varying speeds, it has to run at full speed for 20 seconds under considerable acceleration.
‘The main challenge is that it’s reverse-mounted; that is, it’s the other way round from in a racing car,’ Szyjanowicz said. ‘There’s actually a benefit in that, because it’s dry-sumped and all the baffles in it are designed to keep the lubricant in the right place under significant braking, which means that it will work in our favour when it’s facing backwards but accelerating. But still, it’s significantly greater than any braking load and the period of sustained acceleration is longer than a braking period, so we’re having to do a lot of upgrading of the oil system.’
The pump is causing some problems as well. ‘We had to do a full characterisation, because nobody knew how it would behave,’ Szyjanowicz said. ‘The pump it was based on was powered by a turbine and all we knew was that it went round fast.As it turned out, it isn’t where we hoped it would be in terms of its original specifications, so we’ll have to run the engine a bit harder than we thought.’
Cosworth will also supply the control system for the hybrid rocket, which is still to be specified. ‘We have to verify the operation of the rocket itself, the fuel pump, the engine powering the fuel pump and the control system,’ Szyjanowicz said. ‘It has to operate all the cut-outs, as well as bringing the engine online when Andy Green fires it to pump in the fuel. That system will be a one-off, and we’re working closely with the team from the Falcon Project, which is developing the rocket motor.’
While the Bloodhound work is unique, Szyjanowicz believes that Cosworth’s motorsport background is invaluable. ‘There’s an ethos in motorsport of constant development and time pressures, which is a huge part of what we do.’
What the Bloodhound project brings to the company is an element of the creativity of engineering at extremes, Szyjanowicz said. ‘It’s very much analogous with Concorde or the Apollo missions,’ he said. ‘Someone says something that is, frankly, pretty bonkers. But then someone comes along who says “Yes, it is bonkers, but…”’