F1 2017 Aerodynamics and Car Development

[godlameroso]

Sorry for the post bombs but I really find this area very interesting, and you can see Haas has a very good design, from this angle you can tell a lot about how the air is being managed. Particularly that saddle shape on the floor, both on the leading edge and the start of the longitudinal section

[Vanja #66]

Red Bull and Ferrari have been the biggest winners of 2018 because they focused on the barge boards and streamlined the sidepods, you give up some downforce at the leading edge of the floor, but it's overall more efficient.

I don't think they give up any downforce anywhere with more streamlined and narrower side pods. When you accelerate flow above diffuser by feeding more air in that zone, you also accelerate the air in diffuser. This effect goes upstream all the way to the front wing (it's weakest at that point, but it exists), including floor leading edge. Faster air means less pressure, means more downforce on the leading edge.

Sorry for the post bombs but I really find this area very interesting, and you can see Haas has a very good design, from this angle you can tell a lot about how the air is being managed. Particularly that saddle shape on the floor, both on the leading edge and the start of the longitudinal section

Top surface of floor leading edge is exactly like top surface of (rear, for example) wing, and the flick-up on the floor that all teams use is trailing edge of course, nowadays having a flap as well.

Their 4-wing deflectors are very interesting, producing a good amount of downforce I imagine. However, those 4 vortices that shed off of trailing edges may cause a good amount of drag as well.

[godlameroso]

Their 4-wing deflectors are very interesting, producing a good amount of downforce I imagine. However, those 4 vortices that shed off of trailing edges may cause a good amount of drag as well.

Depends, I know we've beaten this topic to death but the large number of vortecies shed by feathers don't seem to cause an excessive amount of drag. As you no doubt know a feather is incredibly complex but this complexity is good, because it does something people spend billions of dollars developing in the real world. It marries small reynold number phenomena and big reynold number phenomena especially in big birds.

Well the sizes are relevant for F1, birds with 2 meter wing spans with chord lengths to match, and as many flow conditioners as there are barbs in the feathers. Perhaps having a lot of tiny vortecies reduces not only drag and energy of said vortecies, but also their greater number means higher probability of vortecies interacting with anything downstream, and by extention overall consistency of lift.

Afterall this is the principle behind adding more elements to an airfoil right? Although each element is weaker on it's own, you are allowed a greater angle of attack due to mitigating separation.

Now don't get me wrong maybe it's not a good strategy to have consistent but weaker downforce everywhere, maybe in some places you can get away with, or want stronger more straightforward lift(for a variety of reasons of course). But if I were trying to reduce drag in places I already have strong vortecies, I wouldn't make the elements more simple, I'd add more. But I wouldn't go the Williams route of just putting them on and hope for the best, I'd get my flow structures defined first, and tune from there.

To add even more strangeness, sometimes weaking some vortecies upstream can make other vortecies that interact downstream become stronger, and the inverse is also true. The so called calm before the storm is a perfect example. And how shear in the upper atmosphere can paradoxically weaken or strengthen some systems depening on the circumstances.

For example, strong updrafts and strong shear winds tend to cause tornadoes(just a strong well defined vortex), but for Hurricanes(yes I know different Re#) which have slower updrafts shear winds tend to weaken or contain the storm.

I always wondered if this was the reason to aim strong vortecies at the wheels since they themselves have a strong wind shear effect.

[Vanja #66]

Depends, I know we've beaten this topic to death but the large number of vortecies shed by feathers don't seem to cause an excessive amount of drag. As you no doubt know a feather is incredibly complex but this complexity is good, because it does something people spend billions of dollars developing in the real world. It marries small reynold number phenomena and big reynold number phenomena especially in big birds.

Well the sizes are relevant for F1, birds with 2 meter wing spans with chord lengths to match, and as many flow conditioners as there are barbs in the feathers. Perhaps having a lot of tiny vortecies reduces not only drag and energy of said vortecies, but also their greater number means higher probability of vortecies interacting with anything downstream, and by extention overall consistency of lift.

Bird feathers are interesting, but there's more to their evolution than just aero - "renewable" structure and lightness come to mind. And in my view, barbs are more structural than anything else, like many tiny branches of hair that can be re-grown quickly, lightweight and lots of them form a macroscopically solid surface. The best thing for an aero surface is a perfect microscopic finish. Nature doesn't really achieve perfect, but it achieves best possible and improves in time.

Feathers and wing tips of birds (big and small) are also often different. Take an albatross for example, his wing tip feathers form a continual surface to form a single low intensity vortex (we know it's low intensity as feathers don't curve a lot, almost at all). Eagle's feathers form many smaller ones. Each solution is probably good enough for their needs, I think an albatross spends more of its time gliding than an eagle.

That's it from me on this subject in this thread, I don't want us to hijack this thread from variante.

Afterall this is the principle behind adding more elements to an airfoil right? Although each element is weaker on it's own, you are allowed a greater angle of attack due to mitigating separation.

Yes.

Now don't get me wrong maybe it's not a good strategy to have consistent but weaker downforce everywhere, maybe in some places you can get away with, or want stronger more straightforward lift(for a variety of reasons of course). But if I were trying to reduce drag in places I already have strong vortecies, I wouldn't make the elements more simple, I'd add more. But I wouldn't go the Williams route of just putting them on and hope for the best, I'd get my flow structures defined first, and tune from there.

Everything is possible, teams don't know for sure until they try it. I imagine a good aero guy makes the least number of wrong turns in aero development compared to his colleagues.

As for this solution of Haas, I'd say they use vortices to have a firm grip on airflow downstream, probably around rear wheel. Feels to me like it's very important to get as much air inboard, especially air that would turn outboard from the wheel on its own. This is why teams often go towards aggressive side pod taper and coke bottle shape - to have low pressure zone on the side and draw air inward (and also draw more air from upstream as well) - which is why I suggested to variante to make this shape less aggressive on his model at first and gradually taper it as much as possible.

To add even more strangeness, sometimes weaking some vortecies upstream can make other vortecies that interact downstream become stronger, and the inverse is also true. The so called calm before the storm is a perfect example. And how shear in the upper atmosphere can paradoxically weaken or strengthen some systems depening on the circumstances.

For example, strong updrafts and strong shear winds tend to cause tornadoes(just a strong well defined vortex), but for Hurricanes(yes I know different Re#) which have slower updrafts shear winds tend to weaken or contain the storm.

I always wondered if this was the reason to aim strong vortecies at the wheels since they themselves have a strong wind shear effect.

Thanks for this, a good piece of knowledge. I think you are right with vortices and wheels, you have a big vortex smashing against front wheel, braking apart and preventing strong turbulence from forming behind the wheel. This is especially important for under-floor aero, teams go to great lengths to prevent any low energy air from entering there - just look how they energize the frontal swept barge board vortex with many small elements at first, and one or two big ones later.

[godlameroso]

Your conversation has started the gears turning in my head, so perhaps a strong vortex if thrown head on with the tire will break up into smaller weaker vortecies, or, if you hit it just right, it creates another strong vortex(kind of like how a bottle makes resonant sound if you blow over it just right). I suppose the trick is in getting the vortex to hit the shear from the spinning tire just right, which is even harder under yaw and a turning wheel. The importance of the Y250 vortex and it's integrity and all that jazz.

So we have the side of the floor, we have the air following the coke bottle shape, the rear tire interacting with the floor, the inner wheel winglets, the rear wing end plates, as the devices tasked with creating and directing the vortecies that both interact with the diffuser and hit the rear tire in such a way as to minimize it's negative effects. So then the bargeboards both direct air under the floor by creating downforce, but also manage front and rear tire wake. My only question is if the side pod end plates are doing double(or triple) duty as well or if it's just to help the vortex coming off the side of the floor.

The bargeboard appendages must be there to, as you say, re-straighten the vortecies coming off the inner tire.

Doing the validation work for the change in tire geometry under load must be a nightmare, I wonder how the grooves from blistering affect the shear wake off the tire.

[Vanja #66]

Vortices don't break into smaller vortices, unless geometry provides it. Eddies break into smaller eddies all the time, on the other hand.

Side pod deflectors (I hate the idiotic X-wings name motorsport.com is pushing) reshape the flow i front of side pods that goes outboard and send it straight to the rear, so they catch air that would never end up going above diffuser and probably help send it that way. Haas solution induces 4 vortices on each side as well, the question is which way do they go in front of rear wheel - inside, outside or do they smash into the wheel (unlikely, in my opinion).

Tyres are the biggest aerodynamic unknown, and everything else is so fine-tuned that getting bad results of tyre wake in CFD (and wind tunnel) is very bad and can be very costly in terms of lap time. How they deflect under various loads, how fast do they loose thread depth, how bad can they get in terms of graining, that's very hard to simulate as far as I know and the contact patch is especially important - and hard to simulate. However, I think teams are getting on top of that very quickly right now.

[variante]

I like a lot this kind of discussions! But it's been hard to keep track of all of your thesis, especially without visual help.

I think I read that Haas turning vanes generate drag when shedding vortices, when they actually produce very little because of the airflow angle they work with. The real problem with those devices is that they shouldn't produce too much upwash.
Also, as a personal opinion, I don't think those vortices are shed on purpose, as it would be hard to make good use of their energy downstream. They probably burst way before reaching any meaningful part of the car.

Talking about that area of the car, I find it interesting how they manage to harmonize the need for inwash (to increase airflow to the rear) and the need for outwash (to "seal" the floor, especially around the rear wheel volumes).

And since I mentioned the "sealing effect", it would also be interesting to understand up to which extent you want to seal the floor, instead of increasing flow mass through the diffuser, as well as the beneficial inwash that generate diffuser vortices.
For instance, for MVRC I did find very useful NOT to seal the floor. Strong vortices would build up in the diffuser, allowing me to run higher AoA. This is the outcome:



Hopefully my next post will come with a few F1 CFD results

[godlameroso]

If we want to take it up a notch have any of you guys heard of Helicity? Think braided vortecies.

Back to the show, here's an oldie






How does the front bodywork and bargeboard deal with the turbulence that happens at the bottom of the tire?

As the tire moves towards the ground the air it's running over creates a high pressure pocket, likewise the air that fills in the void it leaves behind creates a low pressure pocket behind the tire. At the top of the tire where it's in free stream air the tire has little effect, possibly due to it being enveloped by a boundary layer. The centrifugal force means the boundary layer changes with speed. Under the boundary layer closer to the tire is a shear zone which will alter the direction or hurt the integrity of any vortex in its vicinity(depending on how close to perpendicular the flow is). The effect is dependent on the angle which the vortex hits this shear zone.

Could be the reason we see little winglets on the front suspension. Instead of the tire weakening the flow structures, you use the energy from the tire spinning to redirect the flow to your advantage.

[Vanja #66]

variante, I think floor sealing has more to do with keeping front tyre wake away from floor and diffuser, than with ground-effect sealing. Diffuser and ground effect floor don't work in the same way, so I never understood why teams seal them until I found out how important it is for them to keep turbulent, low-energy air away from the rear. As you car has completely closed wheels, there is no need for this - which is most likely why in WEC we don't have this floor sealing as well.

godlameroso, I think winglets on front suspension redirect non-turbulent air to mix with and weaken vortices forming on front tyres. The biggest difference with rotating and non-rotating tyre is in the boundary layer trip on top of it, the rest of flow structures are close enough. Boundary layer on tyre surface going forward clashes with on-coming air, so it causes a number of effects. This effect can be simulated in wind tunnel with stationary wheels with trip strips around 15mm high on top of wheel and a bit to the rear (10-15 degrees), when you don't have moving ground in wind tunnel. Like we did 2 years ago:

https://www.linkedin.com/pulse/wind-tun ... ovi%C4%87/

[CAEdevice]

The thread is going on and it gets more and more interesting. Thanks!

I would like to know your opinion about the cuts on the sides of the floor I have seen on the Mclaren. Something like that is present on Variante's car.

I have to admit that I am testing something similar on my 2017 MVRC car.

[Vanja #66]

godlameroso mentioned that I think, it's to enhance the sealing vortex all the way along the floor edge.

[CAEdevice]

Maybe they can deviate the flow entering the floor from the sides, and make it go over the upper surface? About the same way an old styled* front wing did "obscuring" the floor? I mean the Mclaren version, not sure about Variante's the pictures are not detailed enough in that area.

* old styled = before the vortex management era, when it was a wing and not a diffuser/vortex generator.

[Vanja #66]

Those vortices are spinning the wrong way for that to happen, if that's what you are talking about.

[CAEdevice]

Those vortices are spinning the wrong way for that to happen, if that's what you are talking about.

No, I am not considering vortices, but simple flow. The section of the Mclaren cuts remembers me a small wing.

[Vanja #66]

Yes, it's exactly that (a flap actually, to be precise) and it's made to raise the pressure above and in front of them, to energize the flow under the floor edge so that the vortex would be stronger and provide better sealing. Flow isn't going perpendicular over them on both sides, it's going sideways under them, so when upper and lower flows meet they form a vortex.