PDA

View Full Version : check out this aero package



coolhand
05-20-05, 02:57 PM
http://www.motorsport.com/photos/f1/2005/mon/f1-2005-mon-xp-0253.jpg

http://www.motorsport.com/photos/f1/2005/mon/f1-2005-mon-xp-0255.jpg

complicated stuff

tllips
05-20-05, 03:42 PM
I remember when the F117 first came out and they said that the reason it had so many straight lines was due to the fact that their modeling software couldn't model curved surfaces very well. So they could not guarantee that the plane would be stealth if it had curved surfaces. Ergo, the production unit had all the straight lines.

I don't understand why the jagged edge on that car...

pfc_m_drake
05-20-05, 03:44 PM
I remember when the F117 first came out and they said that the reason it had so many straight lines was due to the fact that their modeling software couldn't model curved surfaces very well. So they could not guarantee that the plane would be stealth if it had curved surfaces. Ergo, the production unit had all the straight lines.

I don't understand why the jagged edge on that car...
To defeat the pitlane speed radar?

j/k :D

cameraman
05-20-05, 04:02 PM
I remember when the F117 first came out and they said that the reason it had so many straight lines was due to the fact that their modeling software couldn't model curved surfaces very well. So they could not guarantee that the plane would be stealth if it had curved surfaces. Ergo, the production unit had all the straight lines.

I don't understand why the jagged edge on that car...

Probably the exact same reason. I would guess that they have maxed out their modeling software or the computers running it.

Joe in LA
05-20-05, 04:08 PM
Looks like that horizontal piece on the barge board is designed for cutting opponents tires. :)

RichK
05-20-05, 04:08 PM
Sometimes a hard edge or corner creates a vortex, and sometimes they want that. Remember the Reynard speedway barge boards with the "teeth" on it?

RacinM3
05-20-05, 04:49 PM
I won't even pretend to make someone think I'm intelligent enough to have ANY CLUE as to how they ended up there. All I can think of when I see that is some guy who spends far, far more time in the wind tunnel for Ferrari than he does, say, oh, at home.

Complicated stuff is an understatement.

Don Quixote
05-20-05, 04:51 PM
Sometimes a hard edge or corner creates a vortex, and sometimes they want that. Remember the Reynard speedway barge boards with the "teeth" on it?
Yes, I remember from my aero engineering class, many many years ago, that sharp corners shed boundary layers in a consistent and predictable manner, whereas curved corners do not. But that zig zag wing defies the imagination. Sometimes I think Ferrari does these things to mess with the other teams.

pfc_m_drake
05-20-05, 05:05 PM
Probably the exact same reason. I would guess that they have maxed out their modeling software or the computers running it.
Not a chance.

Dirty Sanchez
05-20-05, 06:39 PM
Sometimes a hard edge or corner creates a vortex, and sometimes they want that.ding :thumbup:

24/7 wind-tunnel. all of it devoted to making your car more aero-efficient and/or ruining the air for the guy behind you.

maxed out their computers :gomer: :rofl:

coolhand
05-20-05, 07:43 PM
maxed out their computers :gomer: :rofl:

lmao :rofl: :rofl:

RichK
05-20-05, 08:00 PM
Those are good pictures, by the way. I guess Monaco, with its garages far from the pitlane, gives the photogs new access to the cars. :thumbup:

cameraman
05-20-05, 08:54 PM
Not a chance.

Ferrari uses virtual wind tunnels on top of their physical wind tunnel. You don't think that they model the parts before they fab them and load them into the physical wind tunnel? You think they came up with that shape by trial and error? They will check a thousand variations before they build one and that takes processing power than even Ferrari finds expensive. Those wind tunnel packages are the most processor dependent software packages out there. Ever heard of the Numerical Wind Tunnel at the Japanese National Aerospace Laboratory? The fastest and most expensive computer on the planet and it isn't fast enough to do all that they want. That barge board is behind the wheel and they have to model the turbulance coming off the wheel and suspension components before they can even attempt to look at the effect of the shape of the board.

Methanolandbrats
05-20-05, 09:57 PM
Ferrari uses virtual wind tunnels on top of their physical wind tunnel. You don't think that they model the parts before they fab them and load them into the physical wind tunnel? You think they came up with that shape by trial and error? They will check a thousand variations before they build one and that takes processing power than even Ferrari finds expensive. Those wind tunnel packages are the most processor dependent software packages out there. Ever heard of the Numerical Wind Tunnel at the Japanese National Aerospace Laboratory? The fastest and most expensive computer on the planet and it isn't fast enough to do all that they want. That barge board is behind the wheel and they have to model the turbulance coming off the wheel and suspension components before they can even attempt to look at the effect of the shape of the board. I worked on three dimensional turbulence at the top of thunderstorms when I was in school...............what actually gets maxed out is the researcher's brains. 3D turbulence is one of the most difficult problems there is.

coolhand
05-20-05, 10:49 PM
Ferrari uses virtual wind tunnels on top of their physical wind tunnel. You don't think that they model the parts before they fab them and load them into the physical wind tunnel? You think they came up with that shape by trial and error? They will check a thousand variations before they build one and that takes processing power than even Ferrari finds expensive. Those wind tunnel packages are the most processor dependent software packages out there. Ever heard of the Numerical Wind Tunnel at the Japanese National Aerospace Laboratory? The fastest and most expensive computer on the planet and it isn't fast enough to do all that they want. That barge board is behind the wheel and they have to model the turbulance coming off the wheel and suspension components before they can even attempt to look at the effect of the shape of the board.

there is a difference in maxing out your software in the 1970s on a large aircraft that is trying to balance performance and a "stealth radar signiture" and using software to desgine a 10 inch barge board that has implications on about 5 feet of car behind it.

coolhand
05-20-05, 10:50 PM
anyone see the coincidence of the ferrari getting uglier and slower? it seems that way.

the Renault and Mclaren look "clean" not to junky and complicated, thus they are fast.

racer2c
05-20-05, 10:53 PM
anyone see the coincidence of the ferrari getting uglier and slower? it seems that way.

the Renault and Mclaren look "clean" not to junky and complicated, thus they are fast.

You should email them. :)

coolhand
05-20-05, 11:16 PM
You should email them. :)

whats e-mail :saywhat:

cameraman
05-21-05, 12:00 AM
there is a difference in maxing out your software in the 1970s on a large aircraft that is trying to balance performance and a "stealth radar signiture" and using software to desgine a 10 inch barge board that has implications on about 5 feet of car behind it.

Sure is. Modeling the aerodynamics of a barge board like that one was impossible in the 70's. Modeling turbulence is one of the most difficult mathematical problems out there. What is absolutely amazing is that they can do it at all.

coolhand
05-21-05, 12:07 AM
Sure is. Modeling the aerodynamics of a barge board like that one was impossible in the 70's. Modeling turbulence is one of the most difficult mathematical problems out there. What is absolutely amazing is that they can do it at all.

i am sure they can "max out" their software, but that would be down to a negledgable point in the cars performance,

Lizzerd
05-21-05, 01:40 AM
Maybe I'm just stupid, but I really can't imagine how a relatively small part with those squared cut out bits can make for a significant improvement in performance. But then again, I'm not an aerodymam... aerodynamic... aerodynamcy... ummm... wind tunnel guy.

cameraman
05-21-05, 01:53 AM
Well, if you consider how expensive wind tunnel time is and how expensive it must be to design & make such a carbon fiber widget, you would have to think that they are doing for a reason...

Ankf00
05-21-05, 04:39 AM
Ferrari uses virtual wind tunnels on top of their physical wind tunnel. You don't think that they model the parts before they fab them and load them into the physical wind tunnel?

welcome to the world of continuum mechanics :gomer: congratulations :gomer:

Ankf00
05-21-05, 04:40 AM
anyone see the coincidence of the ferrari getting uglier and slower? it seems that way.

the Renault and Mclaren look "clean" not to junky and complicated, thus they are fast.

because you know how perception is directly proportional to quickness and speed

Ankf00
05-21-05, 04:45 AM
i am sure they can "max out" their software, but that would be down to a negledgable point in the cars performance,

and I'm sure their abilty to reach the limits of CATIA has nothing to do with your perception of their abilities nor the point at which they reach such a level.

i doubt they've climaxed their computing power, but that's a real possibilty no matter how advanced and complicated the system your running.... s'called engineering. s'how it goes. i doubt their programming guys are dumb enough to waste their CPU power on worthless computations though... but every .001 matters, and if making a funky ass template earns you .005, you go for it, ESPECIALLY if you've got those bastards' budget.

Dr. Corkski
05-21-05, 04:45 AM
anyone see the coincidence of the Bridgestone getting uglier and slower? it seems that way.

the Michelins look "clean" not to junky and complicated, thus they are fast.:gomer:

Ankf00
05-21-05, 04:46 AM
there is a difference in maxing out your software in the 1970s on a large aircraft that is trying to balance performance and a "stealth radar signiture" and using software to desgine a 10 inch barge board that has implications on about 5 feet of car behind it.

only 5 feet? uhhh.... wrong. it's probably not best to tell others what the limits of modern CFD software is when you've never even used it yourself

Ankf00
05-21-05, 04:48 AM
Sure is. Modeling the aerodynamics of a barge board like that one was impossible in the 70's. Modeling turbulence is one of the most difficult mathematical problems out there. What is absolutely amazing is that they can do it at all.

parametric solid models were impossible in the 70's. :gomer:

Lizzerd
05-21-05, 05:11 AM
Well, if you consider how expensive wind tunnel time is and how expensive it must be to design & make such a carbon fiber widget, you would have to think that they are doing for a reason...

If you were replying to me, I'll take my sarcastic tongue out of my cheek now...

coolhand
05-21-05, 12:49 PM
only 5 feet? uhhh.... wrong. it's probably not best to tell others what the limits of modern CFD software is when you've never even used it yourself

where was i wrong on it? you generally agreed with what i siad

ilferrari
05-21-05, 01:11 PM
Alot of it is due to aerodynamic restrictions put in place over the last few years. The areas where the designer 'can' and 'cant' put aero parts don't look streamlined.

pfc_m_drake
05-21-05, 01:41 PM
There's a lot of discussion on the topic, so let me interject a few things that people might find interesting.

Ferrari use Fluent (http://www.fluent.com/about/news/pr/pr77.htm) as their CFD package of choice. For those who are not familiar with Fluent, it's a commercially available package - in fact Fluent is the most used CFD package worldwide.

As an aside, this (http://www.fluent.com/about/news/pr/img/pr77_2_lg.jpg) is a nifty picture.

In any event, one of the beautiful things about today's CFD packages is that they take advantage of parallel processing very well. As Intel and AMD move towards dual-core CPUs for the home PC, regular users will become more familiar with SMP and parallel processing - but the idea is basically this: Let's say you have a huge problem you're trying to solve. Instead of solving the huge problem with 1 CPU, you break up the problem into a bunch of smaller chunks and use a network of CPUs to speed up the solution time.

There's basically 2 limitations in CFD problems:
1) The amount of memory required
2) The amount of time it takes to solve the problem

However, because today's CFD codes parallelize so well, these two limitations 'go away' - and the limiting factor becomes pre-processing (setting up the problem) and post processing (obtaining useful data from your solution).

Ferrari has a partnership with AMD (http://www.ae-plus.com/Key%20topics/kt-software-news5.htm) - part of which AMD provides clusters of Opteron based PCs to Ferrari to solve their CFD problems on.

To make a long story short, (as Ank points out)it's very easy to make parametric type changes to a model and run them through a CFD simulation. If the CFD solution shows promise, you go to the next step (usually building a model and testing it in the wind tunnel).

A few final thoughts:

Ferrari are not interested in doing turbulence research, and that really has no no point in the discussion at hand. Other numerical methods such as DNS and LES (http://www.aero.hut.fi/Englanniksi/CFD_text/turbulence_simulation.html) are used to study turbulence.

It is certainly possible to simulate the effects of aerodynamic devices several feet behind the car. Have a good look at the stream ribbons in the picture I linked above and you'll probably convince yourself of that.

I do like the 'Beauty vs Speed' debate too...it leads to some interesting discussions, but I've typed enough for now :D :D

Ankf00
05-21-05, 01:48 PM
pre/post-processing == The Suck. :mad:

pfc_m_drake
05-21-05, 04:00 PM
pre/post-processing == The Suck. :mad:
I couldn't agree more :mad:

pchall
05-21-05, 04:05 PM
aero--day--damned--nicts

dando
05-21-05, 04:25 PM
I think I saw body work like that in The Grinch. :gomer:

-Kevin

Steve99
05-21-05, 04:26 PM
It is certainly possible to simulate the effects of aerodynamic devices several feet behind the car. Have a good look at the stream ribbons in the picture I linked above and you'll probably convince yourself of that.

That is some seriously dirty air behind the car.

pfc_m_drake
05-21-05, 07:24 PM
That is some seriously dirty air behind the car.
It gives you a good visualization for what they're talking about when they say

"I could get close to him, but because of all the dirty air, I just couldn't get close enough to pass him."

SID
05-21-05, 07:55 PM
It gives you a good visualization for what they're talking about when they say

"I could get close to him, but because of all the dirty air, I just couldn't get close enough to pass him."

It also shows you why Schuey was so hell-bent on getting to T1 first.

pchall
05-22-05, 10:48 AM
It also shows you why Schuey was so hell-bent on getting to T1 first.

That looks like the perfect solution allowing a team not to race against the opposition: design your car to mess with their aero so much the are completely screwed when trying to pass. :shakehead

ilferrari
05-22-05, 01:21 PM
At least that wasn't the worst piece of aerodynamics on show this weekend :eek:

http://news.bbc.co.uk/media/images/41173000/jpg/_41173697_paris300.jpg

Ankf00
05-22-05, 01:22 PM
That looks like the perfect solution allowing a team not to race against the opposition: design your car to mess with their aero so much the are completely screwed when trying to pass. :shakehead

that's the goal :D ;)

TorontoWorker
05-26-05, 04:13 PM
I remember when the F117 first came out and they said that the reason it had so many straight lines was due to the fact that their modeling software couldn't model curved surfaces very well. So they could not guarantee that the plane would be stealth if it had curved surfaces. Ergo, the production unit had all the straight lines.

I don't understand why the jagged edge on that car...

I don't pretend to know that much about aero stuff - but as far as the F117 Nighthawk - read Ben Rich's book about the Skunkworks and how exactly the F117 came about. While we all take for granted the current power of PC's today, when the F117 was being mocked up for tests at Edwards, it was stated that at the time, rounded surfaces was out of the question for the near term. Remember, the F117 is not a new aircraft - it was being built in the early 80's after 5+ years of study before hand with the Have Blue project. Fives years in PC time (and power) makes quite a bit of difference as everyone has seen with the B2 design.

One wonders what they are building right now useing 05 model computers? I guess we'll find out in 10 or so years from now.

coolhand
05-26-05, 06:23 PM
Ferrari uses virtual wind tunnels on top of their physical wind tunnel. You don't think that they model the parts before they fab them and load them into the physical wind tunnel? You think they came up with that shape by trial and error? They will check a thousand variations before they build one and that takes processing power than even Ferrari finds expensive. Those wind tunnel packages are the most processor dependent software packages out there. Ever heard of the Numerical Wind Tunnel at the Japanese National Aerospace Laboratory? The fastest and most expensive computer on the planet and it isn't fast enough to do all that they want. That barge board is behind the wheel and they have to model the turbulance coming off the wheel and suspension components before they can even attempt to look at the effect of the shape of the board.

ever heard of Tacit Blue?

it was the stealth concept design that Northrop developed in the 70s and 80s. It was their stealth concept.
http://www.edwards.af.mil/articles98/docs_html/splash/mar98/cover/tacit.htm
http://www.edwards.af.mil/articles98/docs_html/splash/mar98/cover/images/tacit.jpg

it had round edges.

G.
06-01-05, 01:54 PM
I don't understand why the jagged edge on that car...For cutting down trees when you get pushed off the race line. :p

http://www.speednation.com/store/product975.html

(actually, I think the sharp angles are specifically to remove vortexes in the slipstream.)

Ankf00
06-01-05, 10:45 PM
One wonders what they are building right now useing 05 model computers? I guess we'll find out in 10 or so years from now.

Low-observable UAV's...

stealth UAV's which are weapons capable give me the creeps, dont know about you...


(actually, I think the sharp angles are specifically to remove vortexes in the slipstream.) I would think the sharp corners would be creating vortices rather than eliminate them