The most accurate F1 car CFD released to public, ever

[astorphobis]

Earlier this year, I extracted 3D model of Williams FW14B from their AR app, which in the end never got released (they blamed it on hackers, but they just didn't finish the Android version in time).

Ever since that, David Penner, together with three anonymous F1 aerodynamicists, has been working on running CFD simulations on the model. This is what he wrote:

At the start of this F1 season, Williams mistakenly released a model of their current car, the FW43B, to the public via a mobile app of theirs. While this model had some issues, it's the most accurate model of an F1 car that the public as ever had access to, and with a bit of work, can be used for CFD. Given how secretive F1 engineering is, this has opened the door for the public to finally start to see how F1 aerodynamics works, which is something I've been working on in the background for the last 7 months.

This is the first of 3 planned articles as I work toward "proper" results, and is entirely on the CFD methodology I'm using on this car. CFD is hard, and I'm not qualified to know if my results look good or not, so I've consulted with a number of ex-F1 aerodynamicists to critique my methods, and to identify areas with questionable accuracy.

Enjoy!

This is what Aerodynamicist 2 said about the results:

This is by far the best and most realistic CFD of an F1 car I have seen available to the public. Miles better than the old Perrin car CFD.

Without comparisons to wind tunnel deltas I can’t be certain that this model is useful but, in some areas like mesh quality, it looks even better than the CFD we had at my former team. I can’t wait to see what comes of the second and third articles.

This is incredible work by David, read the full article here:
https://www.linkedin.com/feed/update/ur ... 228979713/

[DP_CFD]

Thanks for sharing this!

[hollus]

That model is still incomplete, it misses the strakes under the front wing and under the diffuser (he promises to add them and re-run). But still, very interesting.

One thing that baffles me a bit, though, is where the Y250 vortex is going. I always imagined it going on through the waist of the car and at least to mid-sidepod lenght. In his figures, it dies early (well, it is only CFD and it is missing a few key features for now), but more surprisingly, if goes to the top of the floor, just barely on top, probably shaven by the floor. Weird and, wouldn't it actually create lift if it goes there to die?

This is the video just at the moment when the Y250 vortexes are about to die, the two brown circles when you press play.


Is this a CFD artifact?

[jjn9128]

I'd say partly ride height but I think there's a lot of myth around the Y250. It's there to drive downwash to the front of the floor, by there it's really done it's job - that's also the core there's still a big region of rotating air around it.

[Dynamicflow]

Good effort by the person doing the simulations, however, if what I see in that Youtube clip is correct then I am afraid the numbers are not that valid nor are the flow fields - I do not see the flow entering the engine bay via the cooling packs? Has he just ignored the engine bay and defined an outflow condition at the radiator front face? That is a big no no, it will change the flow field around the car and the distributions of mass-flow to the different areas which never happens on the actual wind tunnel model. Mix that with the lack of diffuser strakes and other components and you are only headed in one direction

[hollus]

It is work in progress. The very long LinkedIn post centers about explaining his methods and tuning the parameters.
A full simulation with strakes will come later.
Lots of details on his treatment of internal aero in his post in LinkedIn.

[Just_a_fan]

As hollus says, it's the first step in a process of improvement.

[DP_CFD]

That model is still incomplete, it misses the strakes under the front wing and under the diffuser (he promises to add them and re-run). But still, very interesting.

One thing that baffles me a bit, though, is where the Y250 vortex is going. I always imagined it going on through the waist of the car and at least to mid-sidepod lenght. In his figures, it dies early (well, it is only CFD and it is missing a few key features for now), but more surprisingly, if goes to the top of the floor, just barely on top, probably shaven by the floor. Weird and, wouldn't it actually create lift if it goes there to die?

This is the video just at the moment when the Y250 vortexes are about to die, the two brown circles when you press play.


Is this a CFD artifact?

Here the Y250 is being used to shed vortexes off of the bargeboard footplate area to create underfloor downforce. This footplate area is labeled "7" in the geometry delta gifs.

Good effort by the person doing the simulations, however, if what I see in that Youtube clip is correct then I am afraid the numbers are not that valid nor are the flow fields - I do not see the flow entering the engine bay via the cooling packs? Has he just ignored the engine bay and defined an outflow condition at the radiator front face? That is a big no no, it will change the flow field around the car and the distributions of mass-flow to the different areas which never happens on the actual wind tunnel model. Mix that with the lack of diffuser strakes and other components and you are only headed in one direction

All mass is conserved, and the radiator flow was adjusted to give an appropriate total pressure at the chassis outlet, although the missing strakes may affect that. This was also in the range of expected values an ex-F1 aerodynamicist gave me. To quote another ex-F1 aerodynamicists that critiqued my results "Adding some body internals may also be something worth considering, although that is not absolutely critical for what you are doing. "

The goal of this first set of results was to simulate the car mostly as-is, and fleshing out the engine bay has potential to be painful in the future. Unfortunately, even if I flesh it out, I'll need to specify flow rates since I don't have any info for F1 radiator resistance.

[theVortexCreatorY250]

That model is still incomplete, it misses the strakes under the front wing and under the diffuser (he promises to add them and re-run). But still, very interesting.

One thing that baffles me a bit, though, is where the Y250 vortex is going. I always imagined it going on through the waist of the car and at least to mid-sidepod lenght. In his figures, it dies early (well, it is only CFD and it is missing a few key features for now), but more surprisingly, if goes to the top of the floor, just barely on top, probably shaven by the floor. Weird and, wouldn't it actually create lift if it goes there to die?

This is the video just at the moment when the Y250 vortexes are about to die, the two brown circles when you press play.


Is this a CFD artifact?

As mentioned by AeroGandulf the Y250 creates a massive flow field rotation around it, you can see this in the velocity animation. It therefore drives downwash into the floor which sheds multiple little vorticies. These increase mixing under the car and therefore use more of the high CpT without letting it go to waste.
A more powerful Y250 it breaks down earlier but can be useful. This is beacuse it also needs to drive the cape which co rotates to the floor vorticies, one of the most critical parts of the car. Therefore getting it powerful early can be a benefit.

[PlatinumZealot]

That model is still incomplete, it misses the strakes under the front wing and under the diffuser (he promises to add them and re-run). But still, very interesting.

One thing that baffles me a bit, though, is where the Y250 vortex is going. I always imagined it going on through the waist of the car and at least to mid-sidepod lenght. In his figures, it dies early (well, it is only CFD and it is missing a few key features for now), but more surprisingly, if goes to the top of the floor, just barely on top, probably shaven by the floor. Weird and, wouldn't it actually create lift if it goes there to die?

This is the video just at the moment when the Y250 vortexes are about to die, the two brown circles when you press play.


Is this a CFD artifact?

As mentioned by AeroGandulf the Y250 creates a massive flow field rotation around it, you can see this in the velocity animation. It therefore drives downwash into the floor which sheds multiple little vorticies. These increase mixing under the car and therefore use more of the high CpT without letting it go to waste.
A more powerful Y250 it breaks down earlier but can be useful. This is beacuse it also needs to drive the cape which co rotates to the floor vorticies, one of the most critical parts of the car. Therefore getting it powerful early can be a benefit.

Hello.
Question.

Why are we normally lead to believe that mixing is bad? Or is does it depend on the type of surface and whether or not it is chaotic mixing or a controlled mixing?

It seems modern philosophy is to not only use vortex for creating barriers and injecting air but also to inject other vortices into the space under the floor? Could you tell us how these swirling flows interact with the boundary layers to reduce the pressure?

Regards

[theVortexCreatorY250]

That model is still incomplete, it misses the strakes under the front wing and under the diffuser (he promises to add them and re-run). But still, very interesting.

One thing that baffles me a bit, though, is where the Y250 vortex is going. I always imagined it going on through the waist of the car and at least to mid-sidepod lenght. In his figures, it dies early (well, it is only CFD and it is missing a few key features for now), but more surprisingly, if goes to the top of the floor, just barely on top, probably shaven by the floor. Weird and, wouldn't it actually create lift if it goes there to die?

This is the video just at the moment when the Y250 vortexes are about to die, the two brown circles when you press play.


Is this a CFD artifact?

As mentioned by AeroGandulf the Y250 creates a massive flow field rotation around it, you can see this in the velocity animation. It therefore drives downwash into the floor which sheds multiple little vorticies. These increase mixing under the car and therefore use more of the high CpT without letting it go to waste.
A more powerful Y250 it breaks down earlier but can be useful. This is beacuse it also needs to drive the cape which co rotates to the floor vorticies, one of the most critical parts of the car. Therefore getting it powerful early can be a benefit.

Hello.
Question.

Why are we normally lead to believe that mixing is bad? Or is does it depend on the type of surface and whether or not it is chaotic mixing or a controlled mixing?

It seems modern philosophy is to not only use vortex for creating barriers and injecting air but also to inject other vortices into the space under the floor? Could you tell us how these swirling flows interact with the boundary layers to reduce the pressure?

Regards

Mixing basically uses high energy air which would normally pass straight under your car. It also thickens up and boundary layer so it can overcome more severe adverse pressure gradients. Willem Toet has a diffuser article where he talks more about this.
I think the misinformation is that people talk about ‘turbulence’. Turbulent Kinetic Energy is a factor but overall total pressure losses are a larger factor. So trying to reduce the losses and increase mixing under the floor is ideal.

That’s why when you see actual aerodynamics engineers talk about cars they talk about losses and not turbulence.