The overall approach to the design and manufacture of the new car was to create a strong team of preferred suppliers that, in addition to Xtrac, included Pankl, ATL and Penske Dampers, who were all persuaded to join the project. URT ultimately did all the composite work too, as we shall see later.

‘We utilised our key partners’ expertise and people to do R&D, as well as manufacture components,’ explains Travis, ‘while we just had six or seven design people.’

Byrne, who had extended his tenure as F1 designer at Ferrari in a consultative capacity, had been asked to oversee the A1GP ‘Powered by Ferrari’ project in a similar capacity. As Travis notes, that played a big part in its success: ‘Rory and I got on, and this was a strong link in our ability to do this.’

Travis, previously chief designer at Lola and at Penske Cars, had been working as technical director (special projects) at Epsilon Euskadi since 2004, and it was while testing at Estoril in 2007 that Bruce Homes, the chief operating officer of A1GP, came to see him. ‘Bruce invited me to design the new car but, to be honest, I wasn’t keen at first. However, he explained the link with Ferrari was serious, and suggested I went to meet the people at Ferrari to see just how serious they were.’

And so, in October 2007, Travis travelled to Maranello and met with Jean Jacques His, Mario Almondo, Rory Byrne, Mauro Rioli and Marco Galli, names enough to be convincing of the manufacturer’s serious intent.

‘However, they then wanted to come and see our factory.’ It was a reasonable request, except no facility existed at that point.

‘How do you sustain a building when contracts come along as and when?’ asks Travis. ‘But, by the time they visited in mid-November, we had put together a factory for them to visit.’

Travis went to URT Composites in Bognor Regis, southern England, a company he’d already had dealings with, and director, Max Cox, pointed out that the unit adjacent to URT was available, with office space and was ideal for the purpose.

‘It was just an empty shell,’ remembers Travis with what might be a guilty grin, ‘so we had to make it look like a functioning factory. Our PC suppliers loaned us some PCs, we got some office furniture and sat some CAD guys in front of them.

‘Mario Almondo, then Ferrari F1 technical director, and Rory Byrne came over, so we took them to see our UK-based ‘preferred suppliers’, which included Xtrac, and only after that showed them the ‘A1GP Technology factory’. We explained it was all new, and where everything would be.

‘The deal came together very quickly from there and we then had to set up in earnest. The decorators and builders were in over Christmas and we opened on 2 January 2008.’

The Ferrari design specification was based on key aspects and broad principles, the most important of which being, as John Travis (A1GP Technical Director) recalls, ‘the Ferrari brand be protected at all costs, so what they wanted, they got.’ From there, he went on to write a detailed specification with every single item spelled out.

‘The car was only going to run on category two circuits (Portimao, Zandvoort, Brands Hatch, etc.) and this determined lap times, power requirements and performance parameters. Importantly, the car had to have a ‘family resemblance’ to the [Byrne-designed] 2004 Ferrari F1 car. Ferrari helped us a lot with that.’

‘But F1 cars are very expensive, and we were a one-make formula with cost constraints, so there were target costs for production, including amortization of design, molds and so forth.

‘But we weren’t going to follow the notion of creating a car for the sake of it. We made a conscious decision to make it very good, with a progressive aerodynamic envelope. You have to remember that the teams were only going to see their cars when they arrived at the circuits. There was no opportunity for them to test and develop in between races.’

‘Other key areas were a rear diffuser and ground effects, wide track, but not a wide front wing because we knew how to make front wings benign when a car steered. But a really important aspect was weight distribution, due to the heavy engine.’

Because Ferrari only made bespoke race engines for its F1 program, the new A1GP racecar was to be powered by a road car V8, derived from the manufacturer’s new F430 model. However, at 160kg, it was some 40kg heavier than the previous generation’s Zytek V8.

‘Whatever we did,’ continues Travis, ‘we couldn’t get the weight distribution far enough forwards, so we were going to overload the rear tyres. We were offered the tyre sizes from the World Series by Renault car, which had about 450bhp, but we were targeting over 600bhp. That meant we were not going to be able to balance the car.’

So Travis and colleagues paid a visit to French tyre manufacturer, Michelin, in Clermont Ferrand. ‘We showed them our simulation results, which demonstrated that we would have an oversteering car, and asked them for a bigger rear tyre. With the might of Ferrari beside us, they agreed and made a new bespoke mould for us. This then tied in with buying new rims from OZ.’

Travis is in no doubt that without the Ferrari involvement this special dispensation from Michelin would not have been made on the tyres, and it was a crucial factor in helping to make the car work well.

Lean and Mean

‘Recruiting and getting infrastructure in were big issues at the start.’ explained John Travis (A1GP Technical Director).  My ex-Penske Cars colleague, Paul Baker, and I spent Christmas 2007 – working around the builders and decorators – wiring in PCs ahead of the contract designers coming in on 2 January. Thankfully, we got a good group of people in, who got stuck in and worked closely and well together.

‘Remember, we had no car design history, no archive of parts and no schedule. Peter Morgan, who was with me at Epsilon Euskadi, did the scheduling with a huge Gantt chart on his wall. Alan ‘Hatchett’ Harris was the buyer. We had to put him in a soundproofed room as he took no nonsense from any suppliers. Martin Dixon was production manager. Stuart Allen was in charge of composites with Paul Baker, and Mario Saccone worked in Bognor and Italy on finite element analysis. We were a very mean, lean team.

‘We also had a very good inspection department, and a very good inspector, who set up a system so all stores were bar coded and everything was made in batches to keep unit costs down. The bar coding meant we knew exactly what parts we had, and each car comprised over 2500 parts. To think that one man did all that is incredible. We practically built the inspection room around the inspector during the early stages.’

Returning to the point made earlier about utilizing the skills and resources of their preferred suppliers, Pankl designed, analyzed, manufactured and tested all of the suspension, including the uprights (made using electrical discharge machining, EDM) and wishbones.

Xtrac did the same with the gearbox. Travis’ team designed its cast magnesium casing, Xtrac then doing the finite element analysis before manufacturing it.

‘It was an unusual gearbox because of the high crankshaft height,’ remembers Travis. ‘Ferrari didn’t want to do a specialist sump for the engine, which meant we had to have a big shear plate under the engine. This meant the crank was high, 125mm above the bottom of the shear plate, so the gearbox ran input drop gears to step the gear cluster down to lower the c of g of the transmission. The drop gears also stepped down the torque, which allowed slimmer, lighter change gears, too.’

It was a longitudinal, six-speed transmission (based on Xtrac’s successful F1 arrangement in use since 1995) and the ratios were changed from the front end.

‘The sump only had scavenge ports on the right-hand side, meaning we’d lose oil pressure in long right-hand corners as the oil was not returned to the oil tank and was retained in the engine. So we had to re-design the oil tank baffling to reduce this issue. As we hadn’t got a car at that point, we were testing an oil tank lying on its side in the workshop so we could simulate the g forces!’

When a chassis was complete, and the engine installed, track testing of the baffled oil tank was undertaken at Snetterton as the team found the long linking right handers of The Bomb Hole and Coram perfect test beds for similar tracks on the proposed calendar.

In keeping with F1, carbon brakes were employed for stopping duties. ‘Brembo made bespoke calipers and discs for us,’ continues Travis. ‘We wanted long life material so, through Michael Messina, they provided endurance specification material.’

Unsurprisingly, given the Ferrari connection, Magnetti Marelli was the preferred partner for the electronics, and Anton Stipinovich the man who designed the systems for the A1GP car.

‘The electronics were bloody complicated, and included the Bosch direct injection fuel system, gearbox control, anti-stall and so forth. Every car had live telemetry on it, and Ferrari had screens of data on every car, each of which had GPS. Ferrari did not want to see an engine blow up in public, so there were two technicians per car. It was an entourage.’

All of that was led by Michelotto, who tuned the engine. They put us under extreme pressure to get everything right, especially with regard to cooling. We were in Maranello every two weeks for technical reviews. I remember they gave us heat rejection figures from the dyno, and there was a total of 190kW from the oil and water systems, and I was just thinking, ‘how are we going to cool this?

‘The friction in a production engine is a lot higher than a normal race engine. So the oil cooler was a heat exchanger in the v of the engine, and from wind tunnel work and pressure coefficients from the radiators we calculated we needed 3470cm2 of cooler area. Once we had that knowledge, Docking Engineering made bespoke coolers for the car.

‘Ferrari also had a specific warm-up procedure we had to follow, involving starting, warm up, seal the system, pressurize it with a small hand pump, and off you go. Even the coolant we used was controlled [by Ferrari].’

Having discussed the chassis requirements with many well-known composite manufacturers, it was decided to bring everything as close to ‘in-house’ as possible, by using the next-door neighbor, and experienced motorsport chassis manufacturer, URT Composites.

‘All our composites parts were made by them. We had a close relationship in all respects and it just made sense to use them and be in control, and perhaps achieve some cost savings, too. But it was all very high-quality stuff, including a composite roll hoop to pass the FIA’s standard F1 tests. It was unheard of to make a composite roll hoop for a one-make series, but we developed a process using a Rohacell ‘spider’ with a carbon exterior and an aluminum billet internal mandrel, for which we had calculated the thermal expansion to consolidate the molding. That enabled us to develop a production process.

‘Even the [40 per cent scale] wind tunnel model was bloody good, despite being for a one-make series. All the aero development was done with this in the ex-Penske tunnel at Southampton University, there was no CFD.