Venturi vs flat floor

[Silent Storm]

Ringo, you’re mixing up real world compressibility (like compressors, turbos, or engines) with what actually matters in aerodynamic modeling. They’re not the same thing.

SharkY is correct. Below 0.3 Mach, compressibility effects are so minor that air is treated as incompressible, that’s standard practice even at the highest levels of professional aero work.

[SharkY]

What is your background and experience with CFD?
Let's start there.
Air is very compressible. Familiar with the humble air compressors, turbo, or the humble piston engine?
Or air pressure at sea level vs on top of a mountain?
No Mach speeds there.

Edit: What website/ap can be used to post images here?

My background is in aerospace engineering, however my CFD experience is quite humble (especially compared to some on this forum).
I've never said, that the air is not compressible. Compressors or piston engines usually take advatange of the fact, that the air there is enclosed in a confined space and can't escape. And most axial and centrifugal compressors actually speed air up to Ma 0.8-0.9.
As for atmospheric conditions, it's due to the gravity: air is compressed by its own weight and also can't escape, as the gravity is holding it down.

That is a bit different to a free flow through a tube/around a car/over a wing. It's because in that case the air can escape and there is no mechanical device to pump more energy into a system (an isentropic process).
The air always wants to take a path of the least resistance. If there is a free outflow, it won't linger and cram into greater density, when it can simply increase its speed (that is a simplification, as without increase in temperature the molecules don't actually speed up, but just start to move more in the same direction and the average speed increases). However, the closer the flow is to the speed of sound, the more difficult it is for air to speed up, so there is some cramming in place and the density rises. That relation to the speed of sound is polynomial and so the rise in density for Ma 0.3 is less than 5%, as can be calculated with the ideal gas equations:
See: https://www.grc.nasa.gov/www/k-12/airpl ... ntrop.html (eq. (8)).
That rise in density will be regardless of the shape of the channel.

Now, that is an idealised model. I do know that the real life is a bit more quirky, but generally for the flow through a channel up to Ma 0.3, an incompressible regime can be used, as it will be well within typical engineering tolerances.

[Silent Storm]

What is your background and experience with CFD?
Let's start there.
Air is very compressible. Familiar with the humble air compressors, turbo, or the humble piston engine?
Or air pressure at sea level vs on top of a mountain?
No Mach speeds there.

Edit: What website/ap can be used to post images here?

My background is in aerospace engineering, however my CFD experience is quite humble (especially compared to some on this forum).
I've never said, that the air is not compressible. Compressors or piston engines usually take advatange of the fact, that the air there is enclosed in a confined space and can't escape. And most axial and centrifugal compressors actually speed air up to Ma 0.8-0.9.
As for atmospheric conditions, it's due to the gravity: air is compressed by its own weight and also can't escape, as the gravity is holding it down.

That is a bit different to a free flow through a tube/around a car/over a wing. It's because in that case the air can escape and there is no mechanical device to pump more energy into a system (an isentropic process).
The air always wants to take a path of the least resistance. If there is a free outflow, it won't linger and cram into greater density, when it can simply increase its speed (that is a simplification, as without increase in temperature the molecules don't actually speed up, but just start to move more in the same direction and the average speed increases). However, the closer the flow is to the speed of sound, the more difficult it is for air to speed up, so there is some cramming in place and the density rises. That relation to the speed of sound is polynomial and so the rise in density for Ma 0.3 is less than 5%, as can be calculated with the ideal gas equations:
See: https://www.grc.nasa.gov/www/k-12/airpl ... ntrop.html (eq. (8)).
That rise in density will be regardless of the shape of the channel.

Now, that is an idealised model. I do know that the real life is a bit more quirky, but generally for the flow through a channel up to Ma 0.3, an incompressible regime can be used, as it will be well within typical engineering tolerances.

That was spot on, SharkY.

To reinforce... Nobody said air isn’t compressible, just that compressibility is negligible in open, subsonic flow up to 0.3 Mach, which is why the incompressible assumption is valid in almost all automotive aero contexts.

Even in pro motorsport CFD (like F1), full compressible solvers are only used when absolutely necessary, and even then, the difference below 0.3 Ma is typically under 5% which is well within tolerance for most decisions.

If someone is getting large aero deviations at 200 km/h and blaming compressibility, they’re likely looking in the wrong place. Boundary layer behavior, turbulence modeling, surface curvature, or vortex interaction are far more likely culprits.

[ringo]

I am not going into the semantics and tit for tat.
The original poster queried about the turning vanes under the car and why through a small gap the air passes.
There are density changes, and that is a fact, and is actually observed in any finite element flow study.
If you look on any thermodynamic table for air properties and look on the specific volume column.. the numbers do change. So not sure why there is an opposition to what I am saying.

Example: https://media.cheggcdn.com/media/4ac/4a ... QV6PqO.png

So that needs to be clarified. Other people do read these forums, and it would be misleading to ignore something fundamental. Regardless of if you think it doesn't make a difference to a car moving at 200km/h. Because it may well do, and it is in fact of a core concept as to why the air behaves as it does under the car. There would be no need for wind tunnels if these "little things" are negligible.

[Hoffman900]

dbl post

[Hoffman900]

No one has answered the question.

The strakes are vortex generators. First seen on Reynard Indy Cars in the 1990s. Bobby Rahal going inverted at Motegi exposed them to everyone and Paul Tracy going inverted at Road America in a Penske Lola also showed them.

https://www.researchgate.net/profile/Jo ... dGlvbiJ9fQ



As explained by Dr Katz:

this section we discuss simple modifications that can be added to an existing car to increase downforce. One of the simplest add-ons is the vortex generator (VG). VGs were used for many years on aircraft, mainly to control boundary-layer flows. The size of VGs in such applications was on the order of the local boundary-layer thickness, and apart from influencing boundary-layer transition, they served to delay the flow separation on a wing’s suction side. The use of such devices in automotive racing is quite different. Here the focus is on creating a stable and long-tip vortex, which in turn can reduce the pressure along its trail. A simple option is to add VGs at the front of the underbody and the long vortex trails of the VGs can induce low pressure under the vehicle. This principle is widely used for open-wheel race cars (e.g., Indy), and a typical integration of such VGs into the vehicle underbody is shown in Figure 13. In such an application the VG is much taller than the local boundary-layer thickness and the objective is to create a strong and stable vortex which, as noted, can generate suction loads along its trail. The principle was extensively used with delta winged aircraft at high angle of attack (Polhamus 1971), but when the wing surface was not at high angle of attack, the interest was mostly diminished (see, for example, Buchholz & Tso 2000). “

No one has used a sealed floor edge in 30+ years. Cart / Indy Car because they never ditched the venturi floor, learned that you use these VG’s to energize the flow, and use floor edge leakage from in front of the rear tire to continue to feed this powerful vortex. This is a known design going back to at least 1992 and is documented.

F1 and later Indy Cars have used these VG’s ahead of the floor on their barge boards. Look under the flat floor F1 cars and you’ll see these VG’s.

Willem talks about them starting at 14:00 and you can see how all of this looks in CFD:
https://www.youtube.com/live/kixMMfEQ-F ... tvDGGXIoQF

The biggest mistake most people make when looking at these floors is thinking in laminar flow. That’s not how it works at all. That’s how they use to try to get them to work, and with sealed edges, but that’s 1970s-mid 1980s thinking. Furthermore, there is a gap between the strakes and the track surface. They’re not sealing to the road. F1 has controlled minimum ride height since Senna’s passing via a wooden plank. CART / Indy Cars have done the same, but people have used different plant materials that are less “grabby” so rules were a little less tight.

Funny enough, in addition to the VG’s, Frank Dernie has mentioned they stole the outwash front wing from CART / Indy Car too. It’s been a near spec series for a long time now, but the 1980s up until the mid 1990s, it in many ways allowed for more creativity. It also shouldn’t be shocking that Newey is a product of this world spending his formative younger years there.

[Hoffman900]

Ringo, you’re mixing up real world compressibility (like compressors, turbos, or engines) with what actually matters in aerodynamic modeling. They’re not the same thing.

SharkY is correct. Below 0.3 Mach, compressibility effects are so minor that air is treated as incompressible, that’s standard practice even at the highest levels of professional aero work.

A PhD aerospace aerodynamicist friend likes to remind me that F1 cars at their fastest are at landing approach speeds for aircraft (which in their world, is obviously not fast at all, since he’s dealing with supersonic craft, which is where things get weird).

[Silent Storm]

I am not going into the semantics and tit for tat.
The original poster queried about the turning vanes under the car and why through a small gap the air passes.
There are density changes, and that is a fact, and is actually observed in any finite element flow study.
If you look on any thermodynamic table for air properties and look on the specific volume column.. the numbers do change. So not sure why there is an opposition to what I am saying.

Example: https://media.cheggcdn.com/media/4ac/4a ... QV6PqO.png

So that needs to be clarified. Other people do read these forums, and it would be misleading to ignore something fundamental. Regardless of if you think it doesn't make a difference to a car moving at 200km/h. Because it may well do, and it is in fact of a core concept as to why the air behaves as it does under the car. There would be no need for wind tunnels if these "little things" are negligible.

Ringo, no one’s disputing that air is compressible in general, but in subsonic aero, that’s not the point. At F1 speeds (200–300 km/h), compressibility effects are tiny, less than 5% density change and well within tolerances. That’s why engineers treat air as incompressible in this regime. It’s not a guess, it’s a standard simplification used by professionals, including in aerospace and F1.

You keep referring to air compressors and thermodynamic tables, that’s fine in closed systems or supersonic flows, but this is open channel external flow where pressure gradients, vortex behavior, and flow attachment dominate the physics not density shifts. Now you’re trying to reframe the conversation to appear right on a technicality but that doesn’t change the fact that compressibility is negligible in the context originally discussed.

Wind tunnels aren’t needed because of minor compressibility shifts they’re used to handle turbulence, vortex behavior, transient effects, and complex interactions that CFD can’t perfectly model.

Yes, air is compressible. But invoking that here distracts from the real physics at play. It’s misdirection, not contribution.

[Silent Storm]

Ringo, you’re mixing up real world compressibility (like compressors, turbos, or engines) with what actually matters in aerodynamic modeling. They’re not the same thing.

SharkY is correct. Below 0.3 Mach, compressibility effects are so minor that air is treated as incompressible, that’s standard practice even at the highest levels of professional aero work.

A PhD aerospace aerodynamicist friend likes to remind me that F1 cars at their fastest are at landing approach speeds for aircraft (which in their world, is obviously not fast at all, since he’s dealing with supersonic craft, which is where things get weird).

Exactly... And that’s why context matters. In aerospace, F1 car speeds are trivial, and below Mach 0.3, compressibility effects are so minor they’re not worth obsessing over. That’s not opinion, that’s standard engineering practice backed by physics.

Some people confuse the existence of compressibility with its relevance. That’s how you end up with arguments about air compressors in a conversation about underbody aerodynamics. It’s a classic case of knowing just enough to sound confident, but not enough to be correct.

Just for context... Some of us have already gone down this road with OP in earlier threads. The conversation tends to spiral into circular logic and cherry picked doubts, which is why I’ve learned it’s best not to overinvest there. Your post adds value though, and I hope it lands better than some of our previous attempts did.

[Hoffman900]

Ringo, you’re mixing up real world compressibility (like compressors, turbos, or engines) with what actually matters in aerodynamic modeling. They’re not the same thing.

SharkY is correct. Below 0.3 Mach, compressibility effects are so minor that air is treated as incompressible, that’s standard practice even at the highest levels of professional aero work.

A PhD aerospace aerodynamicist friend likes to remind me that F1 cars at their fastest are at landing approach speeds for aircraft (which in their world, is obviously not fast at all, since he’s dealing with supersonic craft, which is where things get weird).

Exactly... And that’s why context matters. In aerospace, F1 car speeds are trivial, and below Mach 0.3, compressibility effects are so minor they’re not worth obsessing over. That’s not opinion, that’s standard engineering practice backed by physics.

Some people confuse the existence of compressibility with its relevance. That’s how you end up with arguments about air compressors in a conversation about underbody aerodynamics. It’s a classic case of knowing just enough to sound confident, but not enough to be correct.

Just for context... Some of us have already gone down this road with OP in earlier threads. The conversation tends to spiral into circular logic and cherry picked doubts, which is why I’ve learned it’s best not to overinvest there. Your post adds value though, and I hope it lands better than some of our previous attempts did.

Another poster here who use to do the same admited he got his ideas from “conversations” with ChatGPT. A lot of these actually wrong but psuedo smart posts / ideas, likely originate from that.

I’ve long since deleted reddit, but you’re seeing the same there in what were technical subs. People posting with questions / responses that are clearly “AI” generated and full of errors.

[hsg]

i am out

[Hoffman900]

No. Are you a bot?

[Silent Storm]

A PhD aerospace aerodynamicist friend likes to remind me that F1 cars at their fastest are at landing approach speeds for aircraft (which in their world, is obviously not fast at all, since he’s dealing with supersonic craft, which is where things get weird).

Exactly... And that’s why context matters. In aerospace, F1 car speeds are trivial, and below Mach 0.3, compressibility effects are so minor they’re not worth obsessing over. That’s not opinion, that’s standard engineering practice backed by physics.

Some people confuse the existence of compressibility with its relevance. That’s how you end up with arguments about air compressors in a conversation about underbody aerodynamics. It’s a classic case of knowing just enough to sound confident, but not enough to be correct.

Just for context... Some of us have already gone down this road with OP in earlier threads. The conversation tends to spiral into circular logic and cherry picked doubts, which is why I’ve learned it’s best not to overinvest there. Your post adds value though, and I hope it lands better than some of our previous attempts did.

Another poster here who use to do the same admited he got his ideas from “conversations” with ChatGPT. A lot of these actually wrong but psuedo smart posts / ideas, likely originate from that.

I’ve long since deleted reddit, but you’re seeing the same there in what were technical subs. People posting with questions / responses that are clearly “AI” generated and full of errors.

Yeah, I’ve seen that trend too... Especially in car threads. Sometimes discussions get flooded with confident sounding takes that don’t hold up when you look closer. It’s like people discovered buzzwords without understanding what any of them mean. Makes it harder to have real technical conversations, which is a shame because there’s a lot to learn when things stay focused and grounded. That’s why I usually stay out of it now. I see you got to experience OP for yourself.

[ringo]

Run a study my friends.
I did not say density dominates what happens under the car. None at all. I am saying it is not a negligible parameter.
This is exactly why at the Mexican GP, in less dense air, the cars behave differently around the track. Am I right or wrong? And there are many engineers on this forum; maybe not aerospace but dime a dozen. Chat GPT would probably give me nice long paragraphs make me look smarter; I should try it.
As to be specific about the floor under the car. It would be helpful with an actual study of a ground effect floor model.
Or if there were some guide from an insider as to what objectives drives the shape of the front, and middle prior to the diffuser.

[Hoffman900]

Run a study my friends.
I did not say density dominates what happens under the car. None at all. I am saying it is not a negligible parameter.
This is exactly why at the Mexican GP, in less dense air, the cars behave differently around the track. Am I right or wrong? And there are many engineers on this forum; maybe not aerospace but dime a dozen. Chat GPT would probably give me nice long paragraphs make me look smarter; I should try it.
As to be specific about the floor under the car. It would be helpful with an actual study of a ground effect floor model.
Or if there were some guide from an insider as to what objectives drives the shape of the front, and middle prior to the diffuser.

Density changes through a weekend, session to session, track to track. The racers that race the 24hr races know this, and Daytona especially as the race has had cold fronts blow through during the race causing a big change in temp (and leading to some blown engines in the process).

That said, no one is changing their aero concepts around the change in air density. Mexico, they just slap on what would be a draggier / higher df set up from a slower average speed track at a lowe elevation, and maybe some cooling tweaks to deal with the less dense air. Does it matter? Sure. I was just at 14,100’ last weekend and have raced bicycles that high as well (Pikes Peak and Mt Evans bicycle hillclimb races in CO). You feel it for sure.


Also, people seem to have little appreciation for how prescriptive the F1 rules are. There is variation in the details, but the general tunnel geometey, strakes, floor inlets, wings, suspension arms, etc are all dictated in narrow boxes.

You can see what these boxes look like here:

https://www.racecar-engineering.com/art ... gulations/

Also a good explanation of the floor. For clarification fences = strakes = vortex generators.

As for learning aero, this is one of the best books for someone getting into it:
https://www.scribd.com/document/6536968 ... amics-Katz

As I showed above, that illustration of a Champ Car venturi floor with vortex generators looks just like the current F1 rules. That was published in 2002 in the SAE article, and that design was at least 7 years old by that point.