Open Battle
Baku 2018
Azerbaijan Grand Prix
Virgin escape tunnel vision and gamble on radical car
News
Virgin escape tunnel vision and gamble on radical car
Posted By: James Allen  |  03 Feb 2010   |  12:52 pm GMT  |  96 comments

The new Virgin Racing team today became the first of the new teams to unveil their 2010 car.

It is quite appropriate timing as the debate intensifies about new teams, their viability and the possibility that they are to be allowed to miss up to three races this season.

Picture 58
Virgin have distanced themselves from that debate with this new car, which in itself is a mould breaker as it was designed entirely using Computational Fluid Dynamics (CFD) with no time whatsoever in a wind tunnel. It is the first time in F1 that this approach has been taken.

At the launch of the new team, Sir Richard Branson made a big fuss of the fact that he likes innovators and pioneers and he has fully backed Nick Wirth’s revolutionary approach. It is a big gamble though and if it backfires and the car lacks downforce or has aerodynamic stability issues, then it will be embarrassing.

But Wirth is very confident that this will not happen.

“I am well aware of exactly what it takes to be successful in this sport, ” he said. “When you see what the existing teams have achieved using the conventional but proven design approach, it is unsurprising that there is a great deal of scepticism about our all-CFD approach.

“But we are competing in a sport that is undergoing significant change having come face to face with today’s harsh economic realities. Under resource restriction, convention will become too costly and necessity really will be the mother of invention. I have absolute belief in the digital design process and the opportunity to put the all-CFD approach to the test at the highest level – to demonstrate that this could be the way for the future of F1 – is very, very exciting.”

By not having a wind tunnel and a department to run it Wirth and Virgin have saved themselves a significant amount of money and around 60 to 100 jobs. Among the technical team assembled by Wirth is Christian Silk, the highly experienced ex Renault engineer.

The car will run for the first time in a shakedown test at Silverstone on Thursday and Friday.

Team principal John Booth said, “It was always intended that we would miss the first all-team test in Valencia this week and very early on we targeted the second Jerez test in two weeks’ time for our public testing debut. It is a testament to our methodical approach and the sheer hard work of the team that we are heading to Spain a week earlier than planned to take part in the first Jerez test next week.”

Assuming reasonable reliability, this will allow them to put up to 4,500 kilometres onto the new car before the first race in Bahrain.

Featured News
Editor's Picks
Share This:
Posted by:
Category:

Add comment

E-mail is already registered on the site. Please use the Login form or enter another.

You entered an incorrect username or password

Sorry that something went wrong, repeat again!

96comments

by Oldest
by Best by Newest by Oldest
1

Am I the only on that thinks Sir Richard will build extras and offer F1 the wealthy an F1 experience?

2

I must admit this cars simplicity makes it look stunning. Simple designs have always been my favourite, don’t they always seem to go well? I hope this does. Good luck Richard, Nick & Timo

3

Being from Yorkshire it is nice seeing the White Rose painted on the front wing. It would be great if the team gave Justin Wilson a go in the car, especially as the cars are longer now.

4

Can anyone tell me the difference between a “shakedown” test and a normal test. I presume its the first run of the car, but is this not included in the testing allowance restrictions?

5

Just wondering if any pictures of the rear diffuser has been seen, as far as I can tell virgin have kept the back end rather a secret so far, anyone fancy a guess as to something exciting akin to last years debate? It was after all one of the defining factors in the early pace of last years cars. just an idea…

6

can the Apple iPad Design Your Own F1 car app be far behind?

7

It’s a nice looking car that, to me, seems like it’s come straight out of Indycars!

8

It’s not the computer or the wind tunnel but what you feed it with that matters?

9

Congrats to Virgin for being the first of the newbies to get their car out there. Let’s not forget though that they are not the only ones relying solely on CFD for the design of their car, as USF1 is doing the same thing, though they have the “luxury” of having the full-scale Windshear tunnel as a back up.

If USF1 does make the grid, it will be interesting to see how both of the CFD teams stack up against each other and against the rest of the pack.

Joe A.

10

My understanding is that they have been using Windshear all along..I’ll check

11

A nice looking car. Was it Tony Southgate or Len Terry who said that a car shouldn’t be ugly to be fast, or something that effect? Kudos to Wirth.

As far as CFD-only goes, does it really matter if it works as well as the conventional or not? Every possible approach to car design, from the crudest six guys and chalk-marks-on-the-floor to the most sophisticated and expensive round the clock windtunnel (with a small army of techs) and CFD route can, and has, produced bad cars. Didn’t McLaren have to build an entirely different car a few years ago, because the original one was so bad? And everybody remembers how bad the 2008 Honda was.

Despite the modern tools and the obscene amount of resources used, these were still dog cars. And that’s no different from the mistakes made in the days of eye, intuition and chalk marks on the floor and pencils on the drafting table.

Whether the all CFD approach works or not, it won’t be any worse than what we’ve seen before.

12

Front wing’s looking a bit… basic, isn’t it? Somewhere there’s a lego set with a few bits missing…

13

Greetings one and all!

First, that is one sleek looking Virgin 😉 As the current generation of cars goes, the Virgin is one of the more aesthetically pleasing. Just imagine the jokes we would have had if Virgin had a car in the ‘flaps everywhere’ era!

Second, even if this year’s Virgin turns out to be a dog, the big issue around the CFD debate is going to be the relative gains or losses in pace as the car is developed througout the course of the season. The other teams will have to reduce their budgets eventually, and if the Virgin girls and boys can show that they can develop the car through the season, CFD will become a much more attractive proposition. Even if budgets aren’t reduced, imagine what a difference that money could make if used for driver development programmes, or innovative new technology. I’ll be backing the Virgin boys this year!

14

With cfd you can do many many iterations of say a front wing design or concept and know the downforce it produces. If wirth gets it right he could be onto a winner in a few years once devleoped correclty.

Wind tunnels also have there faults remember, so each technlogy is only an estimate of what happens at speed on track.

After many iterations, even of not accurate, wirth knows one front wing design is better than another. In the windtunnel how many fron wings do you make to do the same thing?

Of course most tops teams will likely use cfd initially then confirm in the wind tunnel but there cfd might not be as advanced.

Wirth is so confident about the cfd they arent confirming in the wind tunnel, only on track so it should be fun, especially for all those engineers watching!

15

Well of course, with a parametric cad model of a front wing that is coupled to your cfd solver, you can iterate your parameters to find a somewhat optimal solution and all the major teams do that right now.

I doubt that their CFD-System would be superior to those used by any of the top teams, because I can’t imagine that there is a lot of development going into these codes from the side of the teams. That sort of work is typically done by CFD-Suppliers (e.g. cd-adapco, EXA etc.), research-centers (e.g. the German Aerospace Center DLR, NASA, JAXA etc.) or Universities.

So it boils down to the ingenuity and abilities of your engineers and the computing power you have at hand – in the end both mainly a question of how much money you want to throw at it.

And it would surprise me, if Wirth had more Computing Power than lets say BMW had with Albert2 – and we’ve all witnessed how much it helped them.

And yes, windtunnel-testing is an approximation of the reality as well, but it is definitly much more close, since you don’t have to worry about getting your physics right – it’s much more about finding the right conditions matching those on the track. With CFD you have to deal with both issues. For all aerodynamicists: Granted, with the 60% models used, you will never reach real track Reynolds-numbers (or has anyone heard of cyro-windtunnels used in F1?), but you’re still a lot closer than with most cfd-applications.

But another point I imagine using a CFD-only approach is problematic is the following. If you have developed your car and go testing and your aerodynamics don’t work out as you planned them, you almost definitly need to visit a wind-tunnel to investigate what exactly went wrong, because with a cfd-simulation it could have any number of reasons. So if you don’t have the ability to do so, you’re learning curve should be pretty shallow.

Jost

16

I have to admit I’m a little bit surprise about the amount of jobs they supposedly have saved. Sure, to operate a windtunnel 24/7 you need your windtunnel-technicians, model-makers and a bunch of aerodynamicists that interpret the data, but if you were to do all that testing in silico only, you’d need additional well-trained cfd-engineers, lots of computers and people who service them too. Ok setting up tests takes more time, but 60 – 100 jobs still seems a little bit exaggerated.

It especially comes as a surprise to me, because lately you hear a lot of the major players in the aerodynamics world admitting that cfd-only approaches won’t do the trick – at least not for another 25 years. That is because with cfd you need to have a pretty good idea what is going on in the region you’re moddeling, what effects you have to account for, what meshes are applicable, what solver you need to choose, what turbulence-model is the right one and all the other thousands of parameters that there are.

And if you go into aeroelastics, i.e. the deformation of your wings, rotating meshes (your tires) or even thermal models (cooling of brakes, tires and exhaust), it gets really nasty. If I as an aerodynamicist had the opportunity to design a f1 car, I’d feel a lot more comfortable having confirmed my assumptions instead of flying blind all the way, just to realize on first day of testing that I got it all wrong.

And when you’re finally testing and it becomes apparent that your aerodynamic loads aren’t sufficient, with cfd you have to ask yourself what effects you haven’t accounted for, what parameters were set wrong and there are a lot of possibilities. Well with windtunnel-testing it is basicly the same, except the list is a lot (!) shorter, because you already know your basic concept works. You only have to find the difference between race-track and laboratory.

And while I’m at it, a few words about cfd software. True, it is important to have a good cfd tool and a lot of bright minds around the globe are working on incorporating new features into these codes. But it is even more essential that the user knows what he is doing (remember all the parameter I’ve written about above). A lot of commercial tool can basicly do the same things, whether you choose TAU (our inhouse code :), OpenFOAM (which in the future Volkswagen employees will be forced to use), cd-adapco’s Star-CCM+ (if I recall correctly, Toyota F1 used to work with that one) or one of the many other advanced codes doesn’t make much of a difference – as long as you don’t choose black box codes like EXA’s PowerFlow, because there is no way to know what sort of magic this tool does on the inside.

I assume this decision to use cfd-only derives from a lack of funds and the PR deivision of the Virgin Team is just trying to sell it as innovative thinking.

Anyway I wish this team all the best and truely hope they succeed with their approach.

Jost

P.S.: I hope my comment doesn’t make for a too terrible read, since I don’t use my english that often.

17

Thanks Jost.

18

Jost… thanks for the explanation…

great insight into some of the key concepts…

hope to read more comments from u in future

19

Cheers for that – you put what I was thinking into a much better post

20

Excelent points you make, J, but we also have to credit Worth with hands-on experience with motor sport of probably more than three decades, so that combined with his CFD stuf could turn out quite good. I’m not sure how successful CFD would be if used by a non-experienced racing car designer, i.e. relying soley on computors without human input and experience.

Anyway I wish Virgin luck and some success, although I am getting tired of seeing Branson apearing in all the pics!

PK.

21

Yes it could turn out quite good, but that is always the case, whether the cfd-tool actually did what it was supposed to, an engineer had a brilliant idea or it was just plain luck. My point is that it is likely they have a smaller chance of that than the top teams.

If you have a cfd-only design process and take your car to the track and find that the aerodynamic loads are in the same area as predicted by your simulation – which is all you can hope for (with wind-tunnel testing it is essentially the same) – you still don’t know if your cfd-simulation works properly. See, aerodynamic loads are the Integral of the pressure over all of your cars surface (basicly the same as a sum of the pressure but with infitely small points you sum up). So you could have many small errors in your simulation that cancel out each other once you ask your tool for the total force and point of origin – which isn’t as odd and rare as you might think it is. Well that will probably work just ok, if next time you’re building a similar car with similar devices utilising similar aerodynamic effects. But F1 is all about innovation and if you are trying out something new, you might find that you’ve always had your cfd-simulation a little bit wrong – and without wind-tunnel testing you probably have no idea where to look for your mistakes.

To avoid that from happening you need to be able to confirm your assumptions about what is going on flow-wise around your car.

I have read somewhere they are using oil bases paints to make the streamlines visible on the track (I don’t know the english term for that, in german it is called ‘Anstrichbilder’, but neither english nor german wikipedia seems to have an article about it). That should give you a general idea of what is going on, while you shouldn’t be able to see much of the details, because when your not only driving in a straight line you get all the effects of accelerating, decelerating, turning and so on into your flow pattern. Anyway, it would be great, if someone had a link to some highres pictures of those dried out paints because you should see a great deal about vortices and flow seperation.

I wonder if they have any other means of on-track-testing – maybe old school pressure holes, though I don’t know if they are even permitted by the rules?

James, if you are at Jerez, could you look out for such measurement techniques?

Jost

P.S.: I have to admit I’m maybe a little bit biased towards wind-tunnel-testing 🙂

22

Thx for the insight, Jost …

23

It makes for a great reading. Thanks! 🙂

24

I suspect lewis is 0.7 a lap faster than button. He has to be surely. If not I will eat my hat

25

Too early to say. It’s likely Lewis is faster, but not by that amount. Remember that Jenson is new to the team, and also that Lewis has had a hand in designing the car to his liking.

26
las amazonas folladoras

i doubt the car will be a ferrari beater, but it is something new to check out. If it’s slow, we can still compare the drivers.

27

If the slower Cosworth car this season kiss good-bye to exclusively CFD design.

My hat off for the team, very brave move to try a new approach to save costs.

Best of luck

28

I said this in another thread, the Computer Software is only as good as the people who wrote it, every piece of software is usually riddled with many bugs which are hard to iron out. In theory if you could make a piece of software that was really really good then you could in theory reduce your workload in the wind tunnel

29

Times move on I know, but I can’t help remembering that the last F1 car to be designed entirely without a wind tunnel was the Lola….

30

Looks simple and tidy but does’t look as tightly packaged as the others around the rear? Lets hope it goes well. Color scheme reminds me of IRL though, not sure why, would’ve preferred all black, but looks good nonetheless and good to see they have plenty of sponsors and new ones too.

31

A great-looking car. I suspect that the one that goes to Jerez will have a more developed front-wing though – no sense allowing your rivals to steal a march on you by showing all your cards at once, plus the simpler front wing shows off the launch livery & sponsors rather more elegantly than the (likely) race one will.

Hoping for a form-book upset from these guys!

32

I like the livery also like the fact that the front end bucks the trend for the ultra high nose with the nose mopunting points cut back under the nose (the most extreme example being the sauber). And if looks are anything to go by this will be a flying machine ( how sculpted and thin is that nose cone!!!! ). But for some reason (and I hope I am wrong) I think it will be a slow car. Remember Nick Wirth designed (and owned/ran) the Simtek team and lets not forget it was a really pretty car but was as slow as a dog with two broken back legs.

33

I genuinely hope this works out. We need this kind of innovative approach, and it is exactly what the new teams should be about.

If the car isn’t perfect that doesn’t mean that CFD is worse than a wind tunnel, or the Wirth made the wrong decision.

How many times have we heard engineers say that the on track data doesn’t match the wind tunnel figures? I have never heard anyone say that the CFD is wrong!

34

If there no-one is willing to experiment, nothing will advance. Isn’t this is what F1 is all about…. Full mark to Wirth.

35

the front wing looks incredibly simple… surely it wont look like that when they start testing

36

Yep. I totally agree. I think its a place holder. They either didn’t want to show the front wing or it isn’t ready yet. Also what is interesting is that all but one of the pictures released are a 3D rendering of a model. One of the 3D renderings shows the diffuser, and that also is not a finished diffuser. There is virtually nothing to it.

All the teams would have CFD software which they are using to design their cars. The advantage they would have also is that they have had years to calibrate it so that it gives the same results as the wind tunnel and the car in the real world. I assume that many teams cocked it up last year because the rules changed so radically that it was harder to get the CFD to produce the real world numbers. …That’s just a guess though.

…Anyway it will be interesting to see how Virgin go. I hope they do well. It’s a nice colour scheme.

37

Wirth must have an idea of what he is doing tho seeing that he has design a winning ALM car just using CFD.

I just hope it pays off for Virgin

38

Sometimes simple is good in engineering – can the same be said for aerodynamics?

39

Well in some cases yes, in some cases no.

But the flow around the front of a F1 car is anything but simple, especially the interaction of wing and tyres. So you definitly have to come up with some tricks to avoid loosing much needed downforce or creating drag.

But honestly I have yet to look closely at the wing and even if I have, my expertise of front wings is far from sufficient to pass judgement on one just by looking at it.

Jost

40

i dont know tbh

i’ve just watched an interview on the beeb and they have said this is only a basic version of there car.

its very much like R30, that had the old wing on it from last year until it rolled onto the track

41

It’s a good sign that at least one new team’s showing some sense of readyness…

The car looks great in terms of hte livery… but i’m really surprsied that the front wing looked very basic… and the same can be said to the rear diffuser as the pictures seem to imply a single deck as opposed to a double deck…

hopefully when they do their shakedown, they’d bring an update package similar to the way renault launched their R30…

Top Tags
SEARCH News
JA ON F1 In association with...
Multi award winning Formula One photographer
Multi award winning Formula One photographer

Sign up to receive the latest F1 News & Updates direct to your inbox

You have Successfully Subscribed!