10 out of 10

By reputation Audi is a paragon of efficiency in sports car racing, but its engineers admit to being naïve when they conceived the R10 TDI. Ten years on, here’s how the firm’s history-making diesel evolved to win Le Mans at its first attempt

Writer Gary Watkins

Like many off-the-wall ideas, this one was hatched in a bar. The notion that a diesel-powered sports-prototype might one day challenge for victory in the Le Mans 24 Hours would have sounded so preposterous at the turn of the century that it could only have been conceived under the influence of alcohol. And so began the story of the Audi R10 TDI. 

The participants in the boozy evening were Audi Sport engine boss Ulrich Baretzky and two of the Automobile Club de l’Ouest’s most important officials of the time, Daniel Poissenot and Daniel Perdrix. The subject of diesels appears to have come up in light of a previous conversation Audi Sport boss Wolfgang Ullrich had conducted with the powers-that-be at Le Mans about the need to incorporate new technologies into the LMP rulebook. 

The remarkable thing was that Audi was still a newbie at Le Mans when these first tentative diesel discussions were taking place. Baretzky remembers that his “night in the bar with the two Daniels” was as early as January 2001. That means the German manufacturer had only claimed one of its 13 victories in the 24 Hours at that point, and it would be another five and a half years before it would take win number six with the R10 turbodiesel. 

The alcohol-fuelled idea of racing a diesel engine in a sports-prototype took a long time to take hold in Baretzky’s mind. So outlandish was the thought that he tried to suppress any hopes he might have had of trying to convince Audi
to take a turbodiesel to Le Mans. Then, perhaps as much as two years later, came the eureka moment. 

“I realised that if someone else other than Audi, as the pioneer of direct-injection diesels for the road, should bring this technology to Le Mans and win, I would regret it for the rest of my life,” he says. “It was like a bolt of lightning.”

Baretzky might have convinced himself about a turbodiesel Le Mans programme, but he knew that winning over the Audi board would be an altogether more difficult process. 

“They must have thought I was a little bit crazy,” he says. “Whenever I suggested it to anyone, they looked at me as though to say, ‘You are an honourable man and have been successful, but isn’t that a little bit too much,
a little bit too strange?’”

One of Baretzky’s key allies was an engineer who would subsequently become an adversary at Porsche, Wolfgang Hatz. The then head of engine development at Audi, who would later take Porsche back to Le Mans with the 919 Hybrid programme as research and development director, was in favour of the idea. He was in the throes of taking the FSI direct-injection technology pioneered on the petrol-powered R8 LMP900 in 2001 into Audi’s production range. 

“He really supported the diesel idea because it was road-relevant,” says Baretzky. “He was about to prove that there was a direct link between road and race technology with the introduction of direct-injection for our street cars.”

This technological cross-over that would become the calling card of Audi’s sports-prototype programmes gave it an ace to play when it came to getting the necessary go-ahead at board level. The American
Le Mans Series had been part of its LMP programme since 2000 and there was a long-term plan to introduce its diesel range into the USA. The R10 could be a means to change perceptions about cars disparagingly still known as ‘oil-burners’ in the States. 

“We wanted to push the sportiness of the diesel,” says Ullrich, “and the best way to do that was through motor sport.”

The ACO, meanwhile, was working on its plans to allow diesels to race in the top prototype class and would announce that the technology would be admitted from 2004 in the March of the previous year. It was late in 2003 that Audi Sport started working on diesel technology.


A turning point for the programme came at Le Mans in 2004, and not because there was a privateer on the grid with a turbodiesel – of all things a Volkwagen engine taken from a Touareg SUV by British team Taurus Sport. It was deep in the night of the 24 Hours that Baretzky got the go-ahead he and Audi Sport needed to pursue their concept in earnest. 

Baretsky was in Audi’s VIP area, high in the old tribune across from the pits, when company chairman Martin Winterkorn suggested a meeting with someone he described as ‘the boss’.

“I didn’t know who he was talking about,” Baretzky says, “because I believed Martin was my boss.” 

The man in question was Ferdinand Piëch, chairman of the supervisory board of Audi parent company Volkswagen and, of course, the father of the Porsche 917. Baretzky explained Audi Sport’s ideas to the committed endurance racing fan, and the reply he got was, “You should do it”. 

“Winterkorn pulled me away and I said, ‘What does this mean?’ His reply was that now we could do it, now we could accelerate,” adds Baretzky. “Without that, no one in the group would have dared.”

There had been scepticism within Audi about a programme that would take it back to
Le Mans for the first time as a factory since 2002, not least from one of its diesel pioneers, Richard Bauder. 

“He said we had to build the engine in cast iron,” says Baretsky. “I told him that I would never voluntarily build a racing engine in cast iron and that it had to be aluminium. He said it would never last, but I replied that we’d make it last – and we did.”

Ullrich describes developing the world’s first purpose-designed diesel racing engine – as distinct from previous production-based racing powerplants – as a “real adventure”.

“No one had done it before and there was no experience to fall back on,” he says. “Most of our suppliers from the R8 thought we were completely crazy when we told them what we wanted to do. There was quite a lot of discussion before we could convince them, and only then could we go full throttle.”

Building an all-aluminium racing powerplant was a step in the dark for Audi Sport’s engine division at Neckarsulm. 

“We didn’t know anything about turbodiesel engines except that you put diesel in them and the pistons go up and down,” adds Baretzky. “Bosch didn’t have a powerful enough engine control unit, so they had to develop it from scratch. We had to understand that the pistons needed an enormous amount of cooling. We had to learn about the injector pressures. We had to learn everything.

“The worst thing is that in the beginning there was nothing and we knew that it would take 13 or 14 months before we could get everything together to have an engine running on the test bench.”

That resulted in Audi Sport taking a production V8 with a similar bore and stroke to its forthcoming race engine to begin testing on the bench. It wouldn’t be until May 2005 that the race engine would fire up for the first time. 

“We learnt a lot about turbocharging and the injectors, and we found a lot of problems very early,” says Baretzky.

Not least of Audi’s problems was building an aluminium engine strong enough within the weight constraints. The rules allowed for a maximum of 12 cylinders and a capacity of up to 5.5 litres, and that was the configuration for which Audi opted.

“We chose that because we didn’t know anything about diesels and the loads they could withstand,” explains Baretzky. “We had no idea if that was too big or too small, no clue at all. But we thought the bigger the capacity the better and that it would be better to distribute the loads to a maximum number of cylinders.”


The record books show that the R10 TDI won on its race debut at the Sebring 12 Hours in March 2006 and scored a victory first time out at Le Mans three months later. Audi ultimately dominated those races against limited opposition, but so new was the technology that the results might easily have gone the other way. 

The R10s encountered a run of engine problems in the week leading up to both events. What’s more, one of the two cars entered each time encountered problems early on, leaving Audi’s victory hopes resting on one car. 

Sebring, back then a round of the ALMS, already had an enshrined place in Audi’s Le Mans preparations. In the first few months of 2006, the car-breaker of a circuit became a home from home for its test team. 

Testing wasn’t straightforward – Baretzky knew the airline menu card by heart, so frequent were his trips to Florida early in 2006 – and things only got worse when the team returned to the Sebring International Raceway for the ALMS opener. Audi Sport Team Joest turned up with a truckload of engines, and come race day had pressed each and every one of them into service. 

“Everything was fine on the first day, except that the oil pressure slowly went down on both cars,” says Baretzky. “We’d never seen this before, but thought, ‘OK, we’ll change the engines.’ Then it was the same story the following day. 

“We had new engines for qualifying on Friday, and it was only in the evening that we found the problem. The oil tank had a sensor that was meant to be glued in place. For whatever reason, and we never found out why, it had been taped in on this batch. The oil, the heat and the turbulence meant the tape came off, and it was this tape blocking the filter that caused the oil pressure to drop.”

The problem might have been solved and another pair of engines – the final two – installed for the race, but Audi’s problems weren’t over. Oil temperatures on pole winner Allan McNish’s car soared during the short raceday warm-up ahead of the scheduled start. 

The pressure was fine, so Baretzky concluded that there must be an issue with the oil cooler within the engine vee. And Audi happened to have a supply of those attached to the line of abandoned engines. 

“I asked the chief mechanic how long would it take,” remembers Baretzky. “He told me it would take three hours. I replied, ‘I’m sorry, but the race starts in one and half hours.’ Without any further discussion eight mechanics started work, four on the race engine and four on the engine from the day before.”

The job was completed just in time for McNish to start the race on time, albeit from the pits. The car was pushed onto pitroad literally as the cars were leaving the grid to begin the warm-up lap. It was a good job, too. McNish and team-mates Tom Kristensen and Rinaldo Capello dominated the race on the way to a four-lap victory, but the sister car shared by Emanuele Pirro, Frank Biela and Marco Werner was out inside three hours. 

Telemetry contact had been lost with the car when Werner started to complain about a loss of power. Exactly at that moment the data connection was restored, revealing 140deg C water temperatures. The driver was ordered to the pits and the car retired to avoid an engine failure. 

“For whatever reason, the radiators were clogged with tyre debris,” explains Baretzky. “The funny thing was that one car had this problem and the other not. We were left with only one for the rest of the race, and you could imagine that I aged about 20 years over the remaining hours.” 

Audi didn’t race the R10 again until Le Mans in June, relying on the old R8 instead for its ALMS campaign until after the 24 Hours. It was a similar strategy to that which Audi had employed in 2000 when it had given the new R8 a winning debut at Sebring before going back to an updated version of the previous season’s R8R.

“It was clear with such a young project that we couldn’t continue in the other ALMS races with the car at the same time as doing the maximum for Le Mans,” says Ullrich. “We used Sebring as a testing ground, but until Le Mans was done we used the existing R8 in the US.”


Le Mans week didn’t begin well for Audi. It was hit by a series of injector problems from first qualifying. 

“We had carried a problem through testing and we were changing everything on the engine, even the loom,” recalls Baretzky. “Every morning out in the car park we were having a conference call to Bosch in Stuttgart. We knew what the problem was, but we didn’t know how to fix it.”

It is a complicated story that involves the ISA (injector voltage adjustment) values embedded in the ECU. Baretzky explains that the development of the engine through the first half of 2006 changed the way the injectors behaved, but no one understood that the values needed to be altered to compensate. 

So unsure of the reliability was Audi that it took the unprecedented step – at least according to its own practices at the time – of running one engine per car through the two qualifying days and into the race. It was a case of better the devil you know.

“We kept mileage on Wednesday and Thursday to an absolute minimum, knowing that those were going to be our race engines,” explains Audi Sport Team Joest boss Ralf Juttner. 

Audi was running just two cars at Le Mans that year, a decision taken with regard to the cost and complexity of the R10 programme. One suffered a serious delay as early as the fourth hour, when a set of injectors needed replacing on the pole-winning car shared by Capello, Kristensen and McNish. 

Twenty minutes and six laps were lost because, says Juttner, “At that time the electronics couldn’t identify which injector had failed, just which bank of cylinders it was.” The need for repairs to a front corner, following a clash with a GT, and then the replacement of a turbo left no way back for this car. 


That left Audi’s victory hopes once again on the shoulders of one driving crew, this time Pirro, Biela and Werner. They took over the lead from the sister car and were never headed. There was a delay to change the gearbox internals after a differential failure, but no recurrence of the injector problem that had hit the pole car. To this day, Audi doesn’t fully understand why.

“Maybe the winning car ran two or three degrees cooler, and that’s why the problem didn’t occur,” suggests Baretzky. “The moment the injector got too hot under load and it went outside of its operating window, a security system shut it off. I remember it taking us two or three months after Le Mans to find out the real reason.”

Pirro and his team-mates finished four laps up on the second-placed Pescarolo-Judd, while the Kristensen Audi made it home third, a further nine laps in arrears. History had been made and the top prototype class at Le Mans was set on a course that has made it the technological laboratory that it is today. 

Had there been no Audi R10 TDI, would we be witnessing the super-powerful and super-efficient hybrids that race under the latest LMP1 rulebook? Probably not, thinks Ullrich. 

“We opened the door to new technologies with the R10 TDI and, before that, the R8 FSI,” he says. “Maybe we changed the mindset by showing what could be done with future-orientated technology.”