Cfd

[Idesign]

For fun we took our contest entry and ran a CFD on it to see what we had. I thought it would be interesting to share the results.

130 mph, 0 degrees yaw produced a Cd of .338 and Cl of .226 Our balance of down force is biased to the rear in part because of the race wing and large tunnels and diffuser. Now that we have the grid set up we can iterate design changes quickly to dial in a good balance front and rear. We'll probably run a few more cases with various trim and I'll post those results too.


3689

[riptide motorsport]

Interested........whats a CFD?

[Mike N]

CFD = Computational fluid dynamics
CD = Coefficient of drag
CL = Coefficient of lift (negative would be downforce)

[kach22i]

That is pretty cool, I have to get my design into a computer 3D model.

I too have done some follow up on aerodynamics, pretty crude stuff compared to the above, but I do what I can.

http://ecomodder.com/forum/showthread.php/how-line-up-template-sports-car-design-18632.html
http://i184.photobucket.com/albums/x295/kach22i/Industrial%20Design/GKA-Design-FFR-green1.jpg
http://i184.photobucket.com/albums/x295/kach22i/Industrial%20Design/A-SIDE-CLAY-CAR-greenprofile.jpg

That light green overlay represents the "idea" aerodynamic shape for a road vehicle........from this........
http://ecomodder.com/forum/showthread.php/aerodynamic-streamlining-template-part-c-9287-9.html
3690

[kach22i]

I just had to see how close Idesign was to the old school template.

http://s184.photobucket.com/albums/x295/kach22i/Automobile/
http://i184.photobucket.com/albums/x295/kach22i/Automobile/C3X-green-overlay.jpg

Did pretty good.

[kach22i]

produced a Cd of .338 and Cl of .226

For reference
http://en.wikipedia.org/wiki/Automobile_drag_coefficient

0.32 Porsche 997 GT2 2008–present

For a template comparable, the new 991
http://s184.photobucket.com/albums/x295/kach22i/Automobile/
http://i184.photobucket.com/albums/x295/kach22i/Automobile/991-aero-overlay-kach22i.jpg

[thebeerbaron]

For a template comparable, the new 991

What I see is that the 991 matches the template less than the Idesign car, yet it has a better Cd.

[kach22i]

What I see is that the 991 matches the template less than the Idesign car, yet it has a better Cd.
I see that too (but the numbers are for the 997 not 991), and I also see that Idesign has inlets at the front corners which wash air on to the front tires. These front corners as I understand it should be large radius corners, not inlets.

Cd numbers do vary depending on the source.

http://ecomodder.com/forum/showthread.php/cars-compared-wind-tunnel-1924.html


The German car magazine "Auto Bild" (2008-04-11 issue) tested the aerodynamics of cars in the Daimler wind tunnel.
Drag factor = Cd*A, where Cd= drag coefficient, A= frontal area

Sportscar:;;;;;
Porsche 911 Carrera;0.27;2.00;0.54;;
Lamborghini Gallardo (15 cm flatter than Porsche);0.33;1.91;0.63;;0.6

For Idesign to include the Coefficient of lift is very interesting, and not always easy to find on our favorite cars.

EDIT.............................................. ......

Nifty tool:
http://ecomodder.com/forum/tool-aero-rolling-resistance.php

Explained here:
http://ecomodder.com/forum/showthread.php/online-tool-shows-aerodynamic-drag-rolling-resistance-power-6341.html

[D2W]

What I see is that the 991 matches the template less than the Idesign car, yet it has a better Cd.

Huge openings in the front and a rear wing create a lot of drag.

[kach22i]

I also notice that there is no air flow under Idesign's car, therefore no air under the car is coming up to meet the air going over the top (see below for examples).
http://ecomodder.com/forum/showthread.php/some-thoughts-aerodynamic-shapes-11183-4.html
37303731

I'm sure any FFR 818 track or street car will be easy to convert to a smooth belly. This can only help the Cd numbers.

As pointed out months ago the 1st place design was rather high in the back, and may conform to the aero template well. In coupe form even more so. I expect that design to have a very low Cd, and might make a good high mileage candidate.

X's design may not do well on a template basis and a Cfd analysis might come in handy.

If you look at Porsche's + 200 mph Carrera Gt, there is not even a pretense to conforming to the template I've linked to. My conclusion is that the template is fine, but if you have a Porsche 10 cylinder car, you can do anything you freaking feel like, you have the HP to get away with murder.;)

[olpro]

As D2W wrote, that Corvid front end configuration and the big rear wing are very draggy. A good example of a good aero front is a Mosler http://www.moslerauto.com/ which has very small air openings and lots of smooth plan view curvature. A front like the Mosler doesn’t need all the dumb little fixes like fins, splitters, etc.
That rear wing definitely produces lots of downforce. You can see the vortex spiraling the opposite direction most car rear quarters do. The high pressure, which is on the top of the (inverted) wing (red flow), spills over at the edges creating that vortex, turning the opposite of an airplane wing’s.
The bottom line: Lift = Drag (whether the lift is + or -). The more lift the more drag and performance cars benefit from downforce so a low cd is not the norm. On the other hand, a high mileage coupe is another matter and low cd is important.
The usual modern approach to under-car flow is to put airdams on the front to force the air to go around or over. This is much more cost efficient than a smooth belly surface and just as good except maybe at Bonneville. This is especially true since racing cars can get away with much reduced front ramp clearance.
Another point, rear “diffusers” are generally just styling tricks and added weight, despite the claims often made for them.
The Corvid design roof has a sharp break in profile at the top, which is undoubtably intended to reduce the high velocity/low pressure area (blue flow). If it followed George’s template, that blue area would be much bigger and lift might be bad. I think the template is not much more than a oversimplified general guide and if they have access to a tunnel or that software they are using then they are ahead. Still that Corvid front is just horrible for drag and everything else aero.

[Flashburn]

For fun we took our contest entry and ran a CFD on it to see what we had. I thought it would be interesting to share the results.

130 mph, 0 degrees yaw produced a Cd of .338 and Cl of .226 Our balance of down force is biased to the rear in part because of the race wing and large tunnels and diffuser. Now that we have the grid set up we can iterate design changes quickly to dial in a good balance front and rear. We'll probably run a few more cases with various trim and I'll post those results too.


3689

Don't you prefer more downforce on the rear? after all those are the drive wheels and you want to be able to put power down in a corner.

[crobin4]

Don't you prefer more downforce on the rear? after all those are the drive wheels and you want to be able to put power down in a corner.

Most often one needs to turn into and out of a corner, as well. :confused:

[bromikl]

I'm sure any FFR 818 track or street car will be easy to convert to a smooth belly. This can only help the Cd numbers.



To my knowledge, they are anticipating a smooth belly. Can't remember where I read it, but it was very early in the concept/goals announcement. Nothing else has since been suggested.

[Idesign]

3796

In response to George, there is in fact airflow calculated over the entire car, we plant "seeds" so we can pinpoint the effect of specific areas of the car, the flow lines you see are a result of that, it is user defined what you see. If we want to know where the flow goes after it hits the mirror for example, we plant a seed on the mirror and it traces back. Conversely, if we want to know where the air is coming from that is entering the intercooler we plant a seed back there and trace it forward. The body is also broken into discrete panels so that force measurements can be taken and a contribution of down force and drag for each component, doors, hood, Mickey Mouse antenna ball, etc. can be assessed.

As far as the under tray and diffuser goes. Most road cars don’t require them since it is costly and inhibits service of routine maintenance parts. You also don’t need a bunch of down force on a road car since the average motorist isn’t threshold breaking and late apexing corners to the grocery store. Air dams are an economical way to manage airflow under the body and deform when you drive up too close to a curb. Most cars on the road today are engineered so that they just don't produce lift. I would imagine that for the 818 target buyer and enthusiast that is building their dream sports car in their garage and is going to subject it to part time track duty they would put up with the inconvenience of having low ground clearance to increase its looks and performance.

I’ve included some more views showing that there is indeed underbody flow and what it is doing. The cloud around the car in the one figure is showing how the air is shedding from the vehicle. The nose does present some drag but not as much as one might guess. Low speed aerodynamics (sub-sonic) is not something you can simply look at and say “the air is going to have a hard time going around that shape” The flow recirculates and some flow on the car makes a turn and heads back up stream. Sometimes you want the flow to stagnate or shed vortices to draw pressure from other areas such as under the car. That said, my next step is to open the radiator duct and vent it through the hood slots and up over the car. This should relieve some pressure\drag on the front fascia and increase down force on the hood. It will likely feed the rear wing differently too which will shift the balance and characteristics at speed. Opening the intercooler duct will further reduce drag on the rear.

Current down force totals 220lbs.

Stay tuned.

[kach22i]

next step is to open the radiator duct and vent it through the hood slots and up over the car.
Excellent, I bet this helps move everything in the right direction.

Thank you for the detailed description of how this program works.

I'm not sure how you did it, but the attachment can be blown up for more detail, very nice detail. The rear of the car is very sexy, black as a color is flattering on this car.

[Cooluser23]

I hope FFR will also post the CFD for the 818. I'd be very curious what cd, the various 818 models have, and where the drag (and/or downforce) is.

[kach22i]

I hope FFR will also post the CFD for the 818. I'd be very curious what cd, the various 818 models have, and where the drag (and/or downforce) is.

That would be nice.

It would also be nice if someone with these tools can tell me what these trendy inlets at the front corners do to the Cd numbers.

Spotted the other day.............don't ask me why, I don't know the owner of this Miata. At least they don't wash on to the front wheels.
http://s184.photobucket.com/albums/x295/kach22i/Automobile/
http://i184.photobucket.com/albums/x295/kach22i/Automobile/Miata-mouth.jpg

EDIT: This is just plain cool....................Wind Tunnel Superbird
http://www.youtube.com/watch?v=s-t5WKkgMDk&feature=player_embedded#!
http://youtu.be/s-t5WKkgMDk

[olpro]

Someone had a "theory" on airflow with that Miata. It probably didn't do anything to improve things.

The youtube videos are cool but they don't seem to prove anything. Yea, that old Plymouth wasn't very aero... The hood scoop takes in some air when it is up... the rear wing was completely for looks...
The test set up doesn't seem to be very scientific. The fans are too close to the vehicle for the airflow to be normalized, the car is too big for the test section and will plug up the flow... a smoke test is okay for preliminary looksee but not much else.
A lot of so called aero work is BS.

Am I getting too cranky?

[David]

Very cool! What software are you using?

David

[kach22i]

Very cool! What software are you using?

David
If you right-click his image it lists Corvid.

http://www.corvidtec.com/

http://www.corvidtec.com/fluid_dynamics.html
3991

[Idesign]

We use proprietary software developed in-house for CFD I’ve generated some more images based on the CFD results. These equate to mostly pretty pictures since I just took the flow lines from the CFD solution and imported them into a renderer.

399239933994399539963997.

[kach22i]

The images above would make some nice promotional posters, maybe nice tee shirts too.

Looking like a real car, that's for sure.

[vozproto]

How much does a seat of the Raven software run (assuming that is the Corvid CFD program you are runnin)

[Idesign]

Raven is cheap compared to all the other things you need to run a high fidelity CFD case. This includes a grid generator to mesh the model, a massively parallel super computer and an A/C unit the size of a city bus to keep it all cool. This assumes you have a detailed CAD model to input.

[crobin4]

3796

In response to George, there is in fact airflow calculated over the entire car, we plant "seeds" so we can pinpoint the effect of specific areas of the car, the flow lines you see are a result of that, it is user defined what you see. If we want to know where the flow goes after it hits the mirror for example, we plant a seed on the mirror and it traces back. Conversely, if we want to know where the air is coming from that is entering the intercooler we plant a seed back there and trace it forward. The body is also broken into discrete panels so that force measurements can be taken and a contribution of down force and drag for each component, doors, hood, Mickey Mouse antenna ball, etc. can be assessed.

As far as the under tray and diffuser goes. Most road cars don’t require them since it is costly and inhibits service of routine maintenance parts. You also don’t need a bunch of down force on a road car since the average motorist isn’t threshold breaking and late apexing corners to the grocery store. Air dams are an economical way to manage airflow under the body and deform when you drive up too close to a curb. Most cars on the road today are engineered so that they just don't produce lift. I would imagine that for the 818 target buyer and enthusiast that is building their dream sports car in their garage and is going to subject it to part time track duty they would put up with the inconvenience of having low ground clearance to increase its looks and performance.

I’ve included some more views showing that there is indeed underbody flow and what it is doing. The cloud around the car in the one figure is showing how the air is shedding from the vehicle. The nose does present some drag but not as much as one might guess. Low speed aerodynamics (sub-sonic) is not something you can simply look at and say “the air is going to have a hard time going around that shape” The flow recirculates and some flow on the car makes a turn and heads back up stream. Sometimes you want the flow to stagnate or shed vortices to draw pressure from other areas such as under the car. That said, my next step is to open the radiator duct and vent it through the hood slots and up over the car. This should relieve some pressure\drag on the front fascia and increase down force on the hood. It will likely feed the rear wing differently too which will shift the balance and characteristics at speed. Opening the intercooler duct will further reduce drag on the rear.

Current down force totals 220lbs.

Stay tuned.

I don't know why I didn't this post before. This quite interesting. At what airspeed are those downforce numbers produced?
Or did I miss that reference as well?
Incidentally, These numbers are getting to (in my mind) the threshold of setting the spring rates for high speed corners (Downforce) and setting them for lower speed corners (below say less than 200 lbs. total).
It is my perception, that at around 20% the total weight of the car, the balance of frt/rr downforce has a large effect on the cornering balance which is exacerbated by pitch and yaw (whether from braking,acceleration, uneven tarmac).
Do you guys concur?

[Idesign]

"130 mph, 0 degrees yaw produced a Cd of -.338 and Cl of .226 "

[kach22i]

Regarding the inlets at the front corners, I've just found out that they are called "Air Curtains" and that BMW is making claims about them and how they are intended to work.

You may have to sign up for this other forum to see all of the graphics, it's a good forum, so it might be worth signing up for other reasons too.

Link:
http://ecomodder.com/forum/showthread.php/air-curtain-17418.html
4168

[crobin4]

"130 mph, 0 degrees yaw produced a Cd of -.338 and Cl of .226 "

Thanks, and not shabby at all. Once the openings are provided an exit path to bring the drag down. All should be right in the world of acceleration between 80-150 mph (128-240 kph).
Since I'm a cone-carver, my world revolves around acceleration between 25-100 mph.
A-mod cars are an example of downforce in this speed range. (very high drag) and they don't tend to look as pretty as yours, either.:)

[Liilwya]

Regarding the inlets at the front corners, I've just found out that they are called "Air Curtains" and that BMW is making claims about them and how they are intended to work.


I have a Bachelor thesis available that was designed for BM air curtains and cfd trading (http://www.cfdbroker.com/). It was a lot of work and I'd be happy to send it to anybody who's interested in it.

[VTX]

Very nice work. I hope that FFR is seriously considering this design. It's one of my favorites.

[Vman7]

As D2W wrote, that Corvid front end configuration and the big rear wing are very draggy. A good example of a good aero front is a Mosler http://www.moslerauto.com/ which has very small air openings and lots of smooth plan view curvature. A front like the Mosler doesn’t need all the dumb little fixes like fins, splitters, etc.


The Mosler MT900 is one of my all time favorite designs. The front would be great, only one problem, in order for a front to happen like that on the 818, the front would be really long in order to get the right top curve, which would be fine for a race car, but for a street car...ouch! Just the way I see it. :)

[vozproto]

I have a Bachelor thesis available that was designed for BM air curtains. It was a lot of work and I'd be happy to send it to anybody who's interested in it.

If it has some high level explanation that an engineer (without PhD level aero background) WOULDN'T get lost in... I would be interested in taking a look.

[kach22i]

If it has some high level explanation that an engineer (without PhD level aero background) WOULDN'T get lost in... I would be interested in taking a look.
Me too, I'd like to take a look at it.

[SkiRideDrive]

Me too, I'd like to take a look at it.

Count me in as well. Thanks.

[BipDBo]

I have a Bachelor thesis available that was designed for BM air curtains. It was a lot of work and I'd be happy to send it to anybody who's interested in it.

I wouldn't mind taking a look. In HVAC we use air curtains at exterior doors. They are also used in refrigerated cases. I don't know how translatable that information would be to that application. I wonder if some sort of air curtain could be applied to the perimeter winshield of roadster or targa to decrease turbulence. This may be a good topic for my thread, "HVAC/turbulence."

[BipDBo]

For fun we took our contest entry and ran a CFD on it to see what we had. I thought it would be interesting to share the results.

130 mph, 0 degrees yaw produced a Cd of .338 and Cl of .226 Our balance of down force is biased to the rear in part because of the race wing and large tunnels and diffuser. Now that we have the grid set up we can iterate design changes quickly to dial in a good balance front and rear. We'll probably run a few more cases with various trim and I'll post those results too.


3689

You stated a Cd of 0.338. Pardon by bluntness, but that's not very good. This body looks very sleek except for where it's most important, at the nose. The nose is shaped like a parachute. I bet a redesign of the nose would drastically improve the aerodynamics. Your Cl of 0.226: In what dirction is that vector? Up or down?

I'm curious what Cd and lift my design has. If I emailed you a cad file, could you calculate it? Is it that simple.

[vozproto]

That could be kinda cool. But I would imagine that could entail one hell of a creative design to get the air flowing out at any substantial velocity to be functional.

[David]

I have a Bachelor thesis available that was designed for BM air curtains. It was a lot of work and I'd be happy to send it to anybody who's interested in it.

I just want to know if those air curtains actually work:). The are the thing above the doors of grocery stores etc?

My wife was asking what they were for a couple day ago.

David

[kach22i]

FYI: Anyone interested in the topic of crosswinds should check out this video.

Volkswagen Wind Tunnel Testing
http://www.youtube.com/watch?v=UAe_zwX9NBM&feature=player_embedded
<iframe width="560" height="315" src="http://www.youtube.com/embed/UAe_zwX9NBM" frameborder="0" allowfullscreen></iframe>

[gwayne]

What is the frontal area?

gw