Scuderia Ferrari 2012

[shelly]

iirc ferrari has used dallara wind tunnel in 2009 and 2010, then went to toyota in late 2010 and used that tunnel for the end of 2010, all 2011 and 2012 till now. It is a long time since they do not rely on their maranello tunnel alone. (windshear does not count as windtunnel)

[munudeges]

Well, Montezuma insisted on building that ludicrous wind tunnel on a 45 degree angle.

[Forza Ferrari]

Speaking about windtunnel issues, does anyone around here have any info on loic bigois joining the team? he should have, since mid september according to reports earlier this year

[Hail22]

Speaking about windtunnel issues, does anyone around here have any info on loic bigois joining the team? he should have, since mid september according to reports earlier this year

I haven't heard of that, will take a keen look into it.

[Forza Ferrari]

It would make even more sense if bigois has started working for us, 'cause we are starting indeep windtunnel investigations, maybe he is the guy in charge of calibrating the tunnel for next season?

[Hail22]

Just found this:

http://au.eurosport.com/formula-1/korea ... tory.shtml

Will Massa Be retained...two weeks til its cemented I guess...

[banibhusan]

Yeah. It seems quite evident that Massa will be retained. He has proved that he can still perform well if the car is well balanced and drivable. And without doubt he is the best team player that Ferrari and Alonso can find on the grid. I read in autosport that the deal might be announced as soon as the Indian GP.

[raymondu999]

On the subject of that errant windtunnel, I have a very big question. Obviously the windtunnel helps tests parts, and returns you numbers. Generally speaking, with a working windtunnel you probably then see if the numbers are an improvement, and then decide from there whether or not the new update is worth it.

Or you can use it qualitatively, using that paraffin smoke stuff to highlight where the airflow is going in each part, to see if a part manages airflow the way it's supposed to.

In my laymen mind, a windtunnel is basicaly a big fan in a tunnel - sometimes with a rolling road of course. How exactly does it go "wrong?" In what way is it not functioning properly? I mean, isn't the job basically just to blow air over the scale model? Or am I missing something here?

Another thing is - again in my laymen understanding - a windtunnel obviously doesn't design or produce anything for you. It is a tool you use for checking and validation - in the sense that you stick it in the windtunnel, then go "Ah yes, that vane affected the airflow the way I thought it would" - or "Ah no, my thinking of what that flap would do was wrong." Whether or not an upgrade - say new sidepods, or a new wing - works, should be independent of whether or not the windtunnel works, no?

Ferrari though seem to be bringing the new parts on the track, and not working. These parts are probably designed by humans - so isn't there fundamentally something wrong in the Ferrari aero engineers' understanding of either aerodynamics, or their car's overall airflow?

This is an honest question as I'm struggling to really understand why the windtunnel loss is hampering them as much as it is. I'm not trying to flame the aero guys at Ferrari, nor its windtunnel. Any and all answers appreciated.

[Sgt_Saboo]

On the subject of that errant windtunnel,

That's whats been bugging me too. A windtunnel makes wind go fast or slow, nothing more or nothing less so what exactly could be the problem? Do they use load cells to collect data from the wings and could they be in someway responsible?

It might be far fetched and completely ridiculous but could there be something wrong with the AIR itself?

[bhall]

The problem isn't so much the wind tunnel itself as it is the correlation between the wind tunnel, CFD applications and the real world. If testing results from any one of those mediums produce data that doesn't sync with the other two, the design team then is essentially rendered blind. How can a design progress if CFD analysis says something will work but reality says otherwise?

For years, Ferrari relied on Fiorano to provide the bulk of the team's testing data, and it obviously worked very well when testing wasn't all but banned. As a result, however, I think the team's "virtual" design infrastructure was neglected. Now they seem to be experiencing the pitfalls of such neglect.

It's not an easy problem to fix, either. By all accounts, it took Newey two years to overhaul Red Bull's design infrastructure to get it to where it is today.

[Richard]

A windtunnel makes wind go fast or slow, nothing more or nothing less so what exactly could be the problem? ... It might be far fetched and completely ridiculous but could there be something wrong with the AIR itself?

You're not far from the truth, the problem is due to the air. Wind tunnels are fiendishly difficult to calibrate at the level of detail needed by F1. We're talking about scale models so size effects come into play, plus there is the difficulty of getting nice smooth air flow to mimic a car driving in the open, add in the effects from the edge and roof of the tunnel, the pulse of the fan, changing temperature & humidity. Then add in try to mimic the undulations of the car driving on a real track, ie pitch, yaw and roll.

Having spent a fortune to replicate nice laminar air flow and a rolling road, they also need to consider that the air on a track is turbulent due to other cars in front.

As bghall notes, it then needs to be compared to the CFD anslysis. What if the wind tunnel is correct and the CFD is wrong? Or more likely, there will be multiple factors and what if both CFD and wind tunnel have overlapping flaws?

Actually the reality is that the wind tunnel and CFD are both approximations of reality, hence they have inevitable limitations and flaws. So to paraphase Rumsfeld, teams need to figure out if the problem lies in known knowns, or known unknowns, or perhaps unknown unknowns of two complex and sometimes conflicting approximations of reality.

[raymondu999]

The problem isn't so much the wind tunnel itself as it is the correlation between the wind tunnel, CFD applications and the real world. If testing results from any one of those mediums produce data that doesn't sync with the other two, the design team then is essentially rendered blind. How can a design progress if CFD analysis says something will work but reality says otherwise?

Ok, I hear you.

But here's the thing I don't quite get. Windtunnel and/or CFD doesn't make a car quicker. The aero engineers' designs and parts do. Obviously Ferrari have still been bringing parts to the car - their travails with the Singapore/Suzuka rear wing have been well documented.

So say your CFD and windtunnel isn't correlating. Sure, you're updating a bit blind - but obviously parts are still getting manufactured and put on the car, without the benefit of CFD/windtunnel validation. But back before CFD and/or windtunnels were used in F1 - people still updated their cars, and while some upgrades didn't work, some did. Why is it that in Ferrari's case, most aren't? Obviously this is indicative of them actually NOT DESIGNING GOOD PARTS, rather than indicative of the windtunnel. Had they had a better, working CFD and windtunnel kit, then the difference would just be that CFD and the windtunnel (and not the track) tells the engineers that the part was bad.

I hope I'm making my question clear. CFD and windtunnel are not essential in making an actual working upgrade - they help, they expedite the process, but they won't make the car quicker.

You're not far from the truth, the problem is due to the air. Wind tunnels are fiendishly difficult to calibrate at the level of detail needed by F1. We're talking about scale models so size effects come into play, plus there is the difficulty of getting nice smooth air flow to mimic a car driving in the open, add in the effects from the edge and roof of the tunnel, the pulse of the fan, changing temperature & humidity. Then add in try to mimic the undulations of the car driving on a real track, ie pitch, yaw and roll.

Having spent a fortune to replicate nice laminar air flow and a rolling road, they also need to consider that the air on a track is turbulent due to other cars in front.

It's not easy

To be fair - that's a problem all wind tunnels probably have, and not Ferrari-specific.

[Richard]

To be fair - that's a problem all wind tunnels probably have, and not Ferrari-specific.

Yes, all wind tunnels have those issues, but obviously Ferrari have more unknown unknowns than they can cope with. Other teams appear to have a better understanding of what is going on in their tunnels and CFD.

[turbof1]

So say your CFD and windtunnel isn't correlating. Sure, you're updating a bit blind - but obviously parts are still getting manufactured and put on the car, without the benefit of CFD/windtunnel validation. But back before CFD and/or windtunnels were used in F1 - people still updated their cars, and while some upgrades didn't work, some did. Why is it that in Ferrari's case, most aren't? Obviously this is indicative of them actually NOT DESIGNING GOOD PARTS, rather than indicative of the windtunnel. Had they had a better, working CFD and windtunnel kit, then the difference would just be that CFD and the windtunnel (and not the track) tells the engineers that the part was bad.

You do know that when you are speaking about "back before windtunnels", we are speaking about a period where aerodynamics were barely known. People were back then only starting to use wings. It is obviously that when you aren't familar with something, and start to study it, that you'll gain a massive amount of information in the beginning. Back then parts where very simple, so also easy to understand and it was relative easy to bring working upgrades to races.
In present F1, that is completely different. The basics of aerodynamics are completely known for every team and where allowed, applied. Today's aerodynamics are about very subtle changes, micro aerodynamics essentially. When you try to make an aero update, you are deploying that in a vastly complex structure of bodywork, airflows, vortices,... . Therefore it is alot harder to insert working updates. You need a good working CFD and wind tunnel, AND a good correlation.

[bhall]

Exactly.