How Do F1 Cars Generate So Much Downforce?

[Quada]

Alright, new guy here. I am working on designing a car that I eventually want to build. I have sorted out the suspension for my tire choice and am now trying to figure out what the body will look like with the help of CFD. After some looking, I read that F1 cars can develop 12kN of downforce at 200kph and upwards of 16kN at 240kph. I set this as a goal (a very high goal, but a goal nonetheless) because my car will have similar dimensions and weight to an F1 car.

I started with a pretty crude model of the car with a geometry that would satisfy the suspension and engine mounting points as well as being able to safely fit one person as this will be a single seat car. So after a quite few iterations I am not getting numbers anywhere close to what I want to get. It got to the point where I just put massive wings at a massive angle of attack on it trying to get moar downforce (l/d ended up being close to 1). Even after that the most downforce I could achieve was around 3kN @ 200kph. Obviously my model isn’t refined but I’m still down 9kN. I would have absolutely no idea where that much downforce could be added. I’m using Solidworks Flow Simulation which is supposed to be a pretty accurate software package utilizing navier stokes equations that DS purchased.

I fear that I may not be setting up the model correctly but all I’m really able to change is the fluid type, mesh size, inflow parameters, and a checkbox that I have checked that reads “advanced narrow channel refinement.” If I leave the undertray, undertray splitter (penis looking thing under the tub), and diffuser but suppress the front and rear wings I get a ridiculous 300N of downforce. I modeled a simple sphere in space to verify the simulation results with hand calculations and came up more or less the same.

Can someone help explain to me how the cars generate that much downforce? Also, if you believe I am missing a step in the CFD trials I would appreciate to know.

[BenjaminF1]

Have you got any images of the rear of the car, a considerable amount of downforce is affected by the rear especially the diffuser, an image of the diffuser would be great.

Also the profile of the rear wing doesn't seem to be that great , it doesn't look like it has much of an aerofoil profile.
Your front wings also look like they block most of the airflow that would go around the surface of the car and over the diffuser. This blockage would create an area of low pressure above the surface of the car, creating lift.

[Just_a_fan]

The images suggest the car is very close to the ground. If it's too close then the underside will choke and you lose all of the downforce from that surface.

All of those turning vanes and wings just look like you've thrown things at it to try to get it to work. You need to start simple and get the basic conditions right before you add elements.

[flynfrog]

Will you have enough power to use that much DF?


I would recommend starting the other direction. Figure out what lap time you are looking for. Then you can balance weight, drag, power, and down force.

[flynfrog]

this may help get a wing sized and shaped to help you start

http://www.grc.nasa.gov/WWW/k-12/airplane/foil3.html

[flap]

Hey guys,

as well my first post on this forum. I'll try to give you some advise on designing your aerodynamic package. As someone said earlier, you should start simple before try complexe shapes. Figure out the available space for aeropackage (what you already did by designing your chassis ), next get a airfoil catalog (http://airfoiltools.com/ just as an example) select some airfoils for your application and do some 2D-CFD. The 2D simulation will help you to adjust the gaps and angle of attack of your flaps and the ratio of the flaps relative to your mainplane. The rest of your development should be done in 3D-CFD. Carefully analyse each of your cases, plot wall shear stress on the surface to indicate where separations occur or if your close to separation. Section with total pressure coefficient (usually x-z plane, but I can't see both axis of your coordinate system in your second picture ) are very useful to show if your rear wing (and undertray) is getting supplied with fresh air. Plot pressure coefficient or static pressure on your surface. I hope this helps.
Maybe you can post some pictures (section) of your mesh and some information on your boundary and initial conditions.

[pwlucas]

If you want to build an aero car, you need to learn the basics. Your airfoil angles are ridiculous and are stalled. Just get a basic car model up and going, read some books, and start applying knowledge. Also pick up some copies of Racecar Engineering and Bernoulli. Lighter reads and lots of goof info. Good luck and keep it up.

[Quada]

All great responses so far, I appreciate the information.

Have you got any images of the rear of the car, a considerable amount of downforce is affected by the rear especially the diffuser, an image of the diffuser would be great.

Also the profile of the rear wing doesn't seem to be that great , it doesn't look like it has much of an aerofoil profile.
Your front wings also look like they block most of the airflow that would go around the surface of the car and over the diffuser. This blockage would create an area of low pressure above the surface of the car, creating lift.

Here are a couple pictures of the underside and diffuser:








The diffuser geometry was largely driven by the space constraints of the rear differential. I will try some other geometries to see if it is worth relocating the differential; perhaps it would be better suited in the front of the car!

The images suggest the car is very close to the ground. If it's too close then the underside will choke and you lose all of the downforce from that surface.

All of those turning vanes and wings just look like you've thrown things at it to try to get it to work. You need to start simple and get the basic conditions right before you add elements.

Yes the undertray was ~25mm from the ground. I reanalyzed at 50 and 75mm but the car generated about 100N less downforce in each situation. However, you are right, I am pretty much throwing things at it at this point. I did start with just the safety cell and analyzed the flow field around that and added the wheel covers and front wings after refining the necessary parts.

Will you have enough power to use that much DF?


I would recommend starting the other direction. Figure out what lap time you are looking for. Then you can balance weight, drag, power, and down force.

Nope. The goal is to have maybe around 400hp at the crank. The original goal was 750 but I recently deemed that unnecessary. However, I am still intrigued how I was not able to generate a substantial amount of downforce and I am determined to learn something from this even if I don't go through with physically building the car.

this may help get a wing sized and shaped to help you start

http://www.grc.nasa.gov/WWW/k-12/airplane/foil3.html

Yes. This is what I needed a few weeks ago! Thanks!

Hey guys,

as well my first post on this forum. I'll try to give you some advise on designing your aerodynamic package. As someone said earlier, you should start simple before try complexe shapes. Figure out the available space for aeropackage (what you already did by designing your chassis ), next get a airfoil catalog (http://airfoiltools.com/ just as an example) select some airfoils for your application and do some 2D-CFD. The 2D simulation will help you to adjust the gaps and angle of attack of your flaps and the ratio of the flaps relative to your mainplane. The rest of your development should be done in 3D-CFD. Carefully analyse each of your cases, plot wall shear stress on the surface to indicate where separations occur or if your close to separation. Section with total pressure coefficient (usually x-z plane, but I can't see both axis of your coordinate system in your second picture ) are very useful to show if your rear wing (and undertray) is getting supplied with fresh air. Plot pressure coefficient or static pressure on your surface. I hope this helps.
Maybe you can post some pictures (section) of your mesh and some information on your boundary and initial conditions.

As it was mentioned before, my wings are not airfoils, they are simply flat plates -- mostly for the sake of simplicity right now. I realize there will be a fair amount of separation and increased drag (also outlined by the visualized air fields), but it was just something to get a rough estimate with.

If you want to build an aero car, you need to learn the basics. Your airfoil angles are ridiculous and are stalled. Just get a basic car model up and going, read some books, and start applying knowledge. Also pick up some copies of Racecar Engineering and Bernoulli. Lighter reads and lots of goof info. Good luck and keep it up.

I'm not British so I have no idea what "stalled" means in that context. If you are referring to the large angle of attack, it is meant not only to generate downforce at speed but also a ridiculous amount of drag as well. This will (hopefully) be an active aero car so having the drag help braking at higher speeds will be beneficial on a lighter car. I think I had mentioned earlier that the l/d ratio was approaching 1ish. If I sustain a more "normal" angle of attack, l/d is somewhere around the 4 mark. The whole point of me doing this is to learn. I have only taken a basic fluids class so far so I am familiar with the basics but I will be the first to admit that I have a lot to learn yet.

[BrunoQuiocca]

If you want to build an aero car, you need to learn the basics. Your airfoil angles are ridiculous and are stalled. Just get a basic car model up and going, read some books, and start applying knowledge. Also pick up some copies of Racecar Engineering and Bernoulli. Lighter reads and lots of goof info. Good luck and keep it up.

Also, I don't think he actually used airfoil profiles at all. Therefore, I assume he only has drag on that car.

If you're using flow simulation, set a global goal of the force in the same direction as your fluid flow, it will measure your drag.


About airfoil profiles. All F1 teams develop their own profiles to the max, and this is quite a challenge, it demands people with extensive knownlegde in fluid mechanics, wind tunnels and so on. This is heavy, heavy engineering.

To skip this part, you can start out by choosing an already made airfoil profile. I recommend you to start with an Eppler423, or a Selig1223. These are high lift airfoils, but simple to use as they are not too aggresive. Then you can use very simple 2D softwares like Xfoil to determine their stall point (stall angle) and maximum Lift Coefficient (being rough, turn them upside down later for downforce). And just then, with your 2D goals set, you can start the 3D tour. Make a 3D model of the wing, test chords, geometries, endplates and so on.
I'll put myself as an example: I am an engineering student working on my course conclusion project. Wich is the development of a high lift airplane wing for a engineering competition we have at Embraer's "yard" here in Brazil. I'm working since january and it will take me until about july to finish one fine-tuned wing, and still I have the rest of the team to share information with about other parts of the airplane.
This leads us to another thing. You can't simply stuff the car with wings. Front and rear wings must be balanced, otherwise you will have grip issues, throwing your (already made) suspension setup on the trash.

And then, you have floor and diffuser to think about.

Well, you have a lot of reading to do, good luck and keep posting the development, mate! =D>

[Greg Locock]

According to the ferrari 200o book at medium downforce setting, the biggest part of the downforce in an f1 car is generated under the body, although the wings do have some effect on that. As such the airfoil properties of the wings themselves are rather small potatoes, and i doubt airplane sections are of any real relevance.

Read Katz.

[gixxer_drew]

F1 cars really dont make all that much downforce compared to what they could. The rules are constantly changed to reset development and reduce the effectiveness. You are just starting out, be patient and work hard at it.

Also getting CFD to correlate to reality is no simple task... until you have done a lot of work there do not trust the absolute values. Even getting an expensive wind tunnel to correlate is difficult.

[Greg Locock]

Just reading the f2000 book, Cd is about 1 and CL is about 3.

So if you want emulate a brick on wheels, then copy f1.

Approximate L/D ratios front wing 8 rear wing 3 body 10 (I got front and rear the wrong way round in a recent post)

then there's a whole bunch of bits that generate lift as opposed to downforce and drag as well. Hence the overall l/D of 3

So rather than chasing silly numbers for CL, and hence CD, what do you really want this car to do, how much power do you have, and why aren't you considering a fan car?

[flap]

If you want to build an aero car, you need to learn the basics. Your airfoil angles are ridiculous and are stalled. Just get a basic car model up and going, read some books, and start applying knowledge. Also pick up some copies of Racecar Engineering and Bernoulli. Lighter reads and lots of goof info. Good luck and keep it up.

I'm not British so I have no idea what "stalled" means in that context. If you are referring to the large angle of attack, it is meant not only to generate downforce at speed but also a ridiculous amount of drag as well. This will (hopefully) be an active aero car so having the drag help braking at higher speeds will be beneficial on a lighter car. I think I had mentioned earlier that the l/d ratio was approaching 1ish. If I sustain a more "normal" angle of attack, l/d is somewhere around the 4 mark. The whole point of me doing this is to learn. I have only taken a basic fluids class so far so I am familiar with the basics but I will be the first to admit that I have a lot to learn yet.

In this context "stalled" means, that the flow ist (almost) completly separated from the surface. Maybe adjust the maximum value in the shear stress scene to 2 Pa, to get a better idea where to flow starts separating (0 Pa wall shear stress means separated).

[Greg Locock]

FFS read Katz, best 30 bucks ever.

[gixxer_drew]

Just reading the f2000 book, Cd is about 1 and CL is about 3.

So if you want emulate a brick on wheels, then copy f1.

Approximate L/D ratios front wing 8 rear wing 3 body 10 (I got front and rear the wrong way round in a recent post)

then there's a whole bunch of bits that generate lift and drag as well.

So rather than chasing silly numbers for CL, and hence CD, what do you really want this car to do, how much power do you have, and why aren't you considering a fan car?

F1 uses a frontal area of 1.5m2 I believe?