Nick Wirth: Leading the CFD revolution

Meet Nick Wirth and you can't help looking up at him. Standing close on two metres (in his socks), the 43-year old (he moves up another gear exactly a fortnight after this season's 14 March opener) is the youngest-ever Fellow of the Royal Institution of Mechanical Engineers. He cuts an imposing figure.

As is common knowledge, back in the early nineties he was a partner in Simtek Research with former FIA president Max Mosley, before founding the Simtek Formula 1 team. Thereafter, spells as technical director of Benetton and designer of an email-literate, football-playing robotic dog (Robo Dog) led him to form yet another research company, this time named after himself.

In mid-2009, Wirth Research, which designed the American Le mans Series-winning Acura that was designed entirely using computational fluid dynamics, was named as technical partner to the fledgling Manor Grand Prix F1 operation. Manor mutated into Virgin Racing, and hence our meeting in Virgin Racing's hospitality area during the Friday of F1's test session in Barcelona.

The previous day I interviewed Timo Glock, who professed to a 'strange feeling' when he first jumped into Wirth's brainchild, the VR-01, and headed around Silverstone Stowe circuit.

"In the first moment... you know it hasn't been in a wind tunnel, it's just CFD. If the computer was in the wrong setting the car could take off after 200km/h, which it didn't!"

Did it feel like a 'Playstation' car or just normal?

"The first feedback was good, the feeling was okay. So it shows the way of CFD is working. It's a completely different mentality of working and developing the car than by a wind tunnel. In terms of driving the car, I didn't feel any difference," said the German, whose last grand prix result was a fine second in Singapore with Toyota before injuries suffered during qualifying in Suzuka saw him sit out the rest of the season.

That 'feeling' seemed a good place to start the interrogation of Wirth, whose latest design will make history as being the first modern F1 car to be designed with the aid of CFD - without seeing the inside of a wind tunnel.

Had it been a stressful period, particularly given that a front wing blew off during the car's first test, and how much scepticism did he need to overcome in the paddock?

"The car is on the track and we're very happy with where we've got to," he smiles. "The stressful time was prior to building the car, but I guess the stressful time for our partners and all those other people was actually about until three weeks ago (before its first forays), but since then we have got the first amount of data out, and we got a very good correlation between the CFD and what we've seen on the track."

The car is still around five seconds off the front runners, so is Wirth not concerned that there may be a feeling in the paddock that unless Virgin scores pole position and wins in Bahrain that CFD doesn't work?

With that, Wirth leans forward as he responds to this insinuation that CFD may not be all it is cracked up to be.

"I think that is the danger, why it's important to set the record straight," he asserts. "CFD is a tool to get a set of aerodynamic answers, just like a wind tunnel is, just like any of the ways you developed aerodynamics in the past.

"Ultimately it's down to the group of human beings who come up with the experiments and design and try different shapes, and the tool - whatever they're using - gives them answers telling them 'that's better, that's worse, that isn't' or whatever. You do those experiments thousands and thousands of times during the car's development."

These experiments then provide sets of answers, which enable engineers to cross-check against actual readings - and Wirth is adamant the two tally perfectly, stressing no bold claims had been made by Virgin Racing along the way.

"We have never have made any exaggerated claims as to where we think we are going to be with regard to the rest of the paddock," he says. "We have huge respect for our competitors and the talents and abilities of their very large aerodynamic departments. We never expected to be able to come up with something in our first season with our limited resources, wherever they be deployed.

"But I'm very proud of the job we've done," he adds quietly, "and we wanted to gain the respect of the people up and down the pitlane and fans, and that is exactly what we have our heads trying to do. At the end of the day results on the track are what prove things."

For Virgin to have elected to go with CFD, the team must have believed it to be cheaper or faster than deploying traditional tools - or both. But how much faster or cheaper?

"Well," Wirth replies, "it has been quite interesting for me to look at various comments that have been made about it, the question's been asked of me: 'How good is your CFD relative to other teams?'"

After a good chuckle he admits: "I don't know, because I don't know what the other teams are doing."

Yes, but what about the speed issue: after all, like the rest of the incoming quartet, Virgin had limited time to design and develop the VR01, and, unlike two of those, is present at tests. So, it must be faster, but by how much?

"In terms of speed, I really don't want to talk about it," smiles Wirth, "because it is very clear to me that we are doing something different compared to other people, based on what I have read in the press, and I will say no more than that."

He is, though, more open about the cost benefits of CFD versus wind tunnels.

"It is a hugely expensive process, to do this," Wirth explains to put to rest any ideas that CFD is bargain basement stuff. "(But) It's a matter of public record (that) we've said our budget was tailored to the FIA's (originally mooted one-size-fits-all) budget cap which was 45m euros, and we have said we are going to carry on within that budget.

"So actually, with that amount of money, by the time you look at parts and all the other bits and pieces required to run a team, the money you have available for R&D is a few million pounds. Just to build a wind tunnel model to Formula 1 standards costs between one and one-and-half million pounds before you've even started."

By way of comparison he refers to earlier Acura ALMS cars designed for Honda in North America.

"In the dying days of wind tunnel development on 50% sports cars we spent somewhere between $15,000 and $50,000 per day in wind tunnel parts to support the programme, working to similar number of hours as we do now. Plus you have all sorts of costs, huge amounts of costs to run the wind tunnel, energy costs, staffing costs, model makers; a lot of those costs are substantially lower with CFD."

Okay, but would Wirth Research have been able to design the VR01 within the budget cap's constraints without going down the CFD path, and, if so, would the car have delivered the same performance? If not, what would the shortfall have been?

"No chance," he responds emphatically to the first question before quantifying: "A huge difference, many seconds a lap slower, or we would have had to go well over the 45m (euro) mark," adding that just 16 heads were allocated to the Virgin's CFD studies.

By way of comparison, Wirth estimates mid- to high-grid teams employ 30-40 model makers and give or take 25 model designers. True, rapid prototyping has sped up processes, but with aerodynamicists and wind tunnel staff, the total number deployed by teams' aero departments can't be far short of 160.

Although little is known about F1's Resource Restriction Agreement - a sort of individually tailored budget cap whereby team headcounts and team budgets are matrixed - Wirth agrees CFD will aid Virgin in meeting the demands of the document, which forms part of the latest Concorde Agreement.

"I believe it absolutely does, because the RRA is another way of cost-capping and at the end of the day it's all about costs, and anything you can do to reduce your operating costs and improve efficiency is going to help you under the RRA, no matter which way it is written."

He does, though, stress, that salary costs for CFD specialists are not as stratospheric as some believe, simply as it offers newly-qualified graduates a perfect opportunity of getting into motorsport.

"I don't want to go into the vagaries of exactly what we pay our staff, but it takes a certain ability in CFD, certain levels of expertise, and frankly it's a very new science, so it's a very good area for us to recruit graduates who have expertise in these areas. It's very difficult to get into motor racing, so it's a good opportunity for us to offer talented young engineers that opportunity and a place to learn their trade and help the CFD process go forward."

This 'new science' is developing rapidly, and Wirth believes F1 can help move the process forward.

"F1 has been very late coming to the CFD party. CFD was a science invented in the US and has already been used for a long time in the space, military, aero industries, and is already widely used in the general automotive industry.

"We have taken this technology forward in a war-type environment; you know any technology which gets exposed to war gets developed very rapidly, and we have been in a war against very strong opposition for many years now (ALMS), and we are now in the ultimate arena, which is why we are so excited to show what we can do."

The Wirth research group consists of three limited companies - Wirth Research (the main design/research arm), WR Digital (CFD, aero and thermal dynamics solutions provider), and WR Technology (established specifically for the F1 programme, and thus the interface between Virgin and Wirth Research) - all based in Bicester, Oxfordshire and owned by Nick and his wife Louise. A non-shareholding CEO oversees day-to-day operations.

Unlike the first F1 team to really embrace CFD, namely Sauber - which uses one of Intel's family of Albert supercomputers for calculations - Wirth Research relies on computer clusters of "literally thousands of high performance processors linked by optical cables, and cooled by giant water radiators."

"Some software applications are unsuited to that, but others are perfect for that, and fortunately for us CFD is perfectly suited to thousands and thousands clustered together, and that is precisely what we have," he explains. "In layman's terms, (it's) a bit like a quad core processor, but instead of quad, lots and lots and lots of cores. Optically connecting them together means we can do things very, very fast.

"They're all connected to a very fast, high-performance storage system. The amount of data we generate day is truly staggering; we generate over three-quarters of a terabyte per day, which is over 750GB of data every single day, and are investing in live storage systems which have capacities in the petabyte range, which is 1000 terabytes or a million GB. That's all I want to say, but it is straining our suppliers to the limit, it excites them greatly to work on projects like this, and it is very interesting."

In keeping with the energy-efficient approach of the team - no wind tunnels drawing megawatts of electricity 24/7, remember - Wirth Research is looking at consolidating all three companies under one roof and using the heat generated by the clusters for heating purposes.

"One of the reasons we are planning to do that this year is to maximize efficiency, and by choosing liquid cooling it enables us to use the CFD clusters to heat our building in the winter, and do other bits and pieces. We're interested in minimizing our energy consumption even though we know we are lower than our competitors already."

So, will Wirth be at every grand prix to oversee F1's only CFD car?

"I will not be at all the races, one of our Wirth Research guys, Mark Herd (son of March co-founder Robin, who founded the company with Mosley), is embedded in Virgin Racing as Chief Race Engineer - vast experience and a fantastic guy - and he is overall in charge of the race engineers. We have a number of employees who are embedded in the team just to form a very, very good connection. I'll be attending about half the races."

With that, testing calls, and Wirth heads for the door, bends to pass through, then returns.

"The only thing I would add is that everybody majors on the CFD, but there are many, many aspects of the digital design process which are also quite unique. The whole process - structurally, simulations - has a lot of interesting links." Lofty stuff, indeed...

We want to hear from you!

Let us know what you would like to see from us in the future.

Take our survey

- The Autosport.com Team

Read and post comments