41 replies to this topic

[jmorris]

I have found the sporting press coverage and technical analysis of aero and diffusers to be really poor. The so-called experts are long on jargon and short on conceptual explanation, and perhaps, understanding.
So I spent a few minutes thinking about the science and researching what has been written. This was done to set my own head right. I share it with you here. Sometimes real experts post on this forum, your comments are encouraged.

Diffusers:

Bargeboards, splitters, snowplows and vortex generators provide a clean, smooth, non-turbulent supply of air under the floor of car by separating the good air from the bad air. Fast moving air in this tight space (under the car) results in a low pressure air area near the middle of the car or drivers buttocks. The diffusers job is slow down this air to avoid a drag inducing turbulence behind the car, to encourage and not bottleneck the smooth flow from the mid-car low pressure area, and to induce vortices that further reduce drag at the rear and encourage flow from underneath the car.

The lower wing horizontal support bar helps the diffuser mass flow. The coke bottle shape of the side pods also works in conjunction with the diffuser. (Hint: When viewed from above, the car looks like an airfoil, air speeding around mid-car driver area, and then slowing as the body tapers at the gearbox, just like the diffuser but in a different plane. The undercut radiator pods maximize the mass of air flowing to the rear offsetting the obstruction of the rear tires.

The more freedom you have in diffuser shape/size the more you can do to maximize under-car airflow. The “cheaters” figured out that the rules refer to body parts viewed from the bottom of the car, so they added a layer or tier to the top of the primary diffuser but made this top layer just a little smaller so it could not be seen from the ground view. This is, of course, all lawyer-speak crap. A good-old boy, NASCAR inspector would, on the first day of testing, say “forget what we wrote, it’s not what we had in mind, so get rid of it before you go on the track”.

This only makes sense if you read the articles in the following order:

http://www.symscape....ts_of_diffusers

http://www.f1fanatic...le-controversy/

http://www.autosport...rt.php/id/73083

http://www.autosport...cle.php/id/1731

Regards,

[salti]

There's an excellent review on the three contentious diffusers here http://www.pitlanefa....php?f=21&t=735

[jmorris]

Salti, your right, the pitlanefanatic article is great. However, I think the last bit about the relationship about nose height and diffuser size is very debatable - the Brawn has a low nose but a big rear diffuser. I think front nose height is more a compromise between front wing effectiveness and suspension geometry. I have been disappointed in Autosport's coverage of technical issues, Anderson has (dated?) credentials but just doesn't seem to get it onto the printed page

[desmo]

I think the English language mainstream motorsports press and readerships are simply relatively tech illiterate. Watch F1 races and coverage on Italian TV or pick up an Autosprint with Piola's feature and you'll see another level of technical awareness and sophistication. Anglophones just don't seem to be up to it and are more interested in the drivers' private lives and gossip than the actual cars.

[zac510]

I wonder whether this will make teams less interested in collaborating to formulate tech rules in the future. It would be seen as quite unsporting for Ferrari, McLaren or Renault (who helped form the rules) to then build a diffuser outside the spirit of their regulations.
In short (and definitely IMO) from outside the place of formulating the rules you are better placed to exploit them properly.

[Kalmake]

Q: Because the spirit of the regulations…
Adrian Newey: There’s no such thing as the spirit of the regulations, luckily!

http://images.itv-f1...eneral&id=45080

[Paolo]

Originally posted by jmorris

Bargeboards, splitters, snowplows and vortex generators squeeze/push/cram air under the floor of car.

Actually, quite the contrary.

It is true that ground effect depends on speeding the flow under the floor, but this is not obtained by squeezing in as much air as possible under it.

In incorrect but hopefully understandable terms, there is a maximum amount of flow that can be managed by the diffuser; the greatest the diffuser exit area the more flow needs to be "squeezed in" at the front.
Having too much flow going under the floor will, beyond a certain threshold, decrease ground effect.

Barge boards essentially deviate the excess flow around the sides and, in doing so, generate low pressure vortexes that help shield the flat bottom from the outside air, preventing too much lateral flow spilling.
Splitters essentially work as barge boards, dividing the flow in the vertical plane instead than in the horizontal one
Splitters also create some downforce by an augmented pressure on their upper face.

About "snowplow" if you use this term to refer to the Brawn GP thingie, it appears to be a sophisticated form of splitter.

[GeorgeTheCar]

There is another element that greatly effects the efficiently of the diffuser and that is the relationship between the diffuser and the wing.

The diffuser acts to accelerate the air ahead of it. This air then slows down as it crosses the diffuser and slows further as it exits the diffuser.

At this point the vortex being shed by the wing pulls the underbody air away away from the diffuser enhancing the speed of air flow under the car, increasing the downforce.

Decreasing the size of the wing, raising it and precluding any elements being formed by the suspension all serve to make the diffuser less effective and significantly decrease downforce.

Turning that around, anything that you can do to move the diffuser closer to the wing vortex, in effect coupling the diffuser to the wing, will greatly enhance the package efficiency.

[phantom II]

Nice.

Originally posted by GeorgeTheCar
There is another element that greatly effects the efficiently of the diffuser and that is the relationship between the diffuser and the wing.

The diffuser acts to accelerate the air ahead of it. This air then slows down as it crosses the diffuser and slows further as it exits the diffuser.

At this point the vortex being shed by the wing pulls the underbody air away away from the diffuser enhancing the speed of air flow under the car, increasing the downforce.

Decreasing the size of the wing, raising it and precluding any elements being formed by the suspension all serve to make the diffuser less effective and significantly decrease downforce.

Turning that around, anything that you can do to move the diffuser closer to the wing vortex, in effect coupling the diffuser to the wing, will greatly enhance the package efficiency.

[GeorgeTheCar]

Once you understand it you can see it working in NASCAR too.

The wing on the COT acts not like a wing on a F1 car to generate downforce but to create a massive vortex behind the car and act like the diffuser on an F1 car.

My friend and car designer, Tom Johnston, uses the analogy of an inverted shoe box with the top and one end removed.

In NASCAR the splitter acts like the closed front end of the box and endeavours to keep as much air out from under the car as possible. Thus the push to run the splitter as close to the track as possible. The more this strategy succeeds, the less air has to be removed by the trailing vortex.

This begins to explain aero-loose. As a car approached the rear bumper it both blocks the flow of air under the car and the trailing vortex which generated it. It also explains how a car sitting off the left or right rear quarter can have an aero effect on the leading car. In both cases the bow wave of the following car spills into the vortex being shed by the leading car, changing the vortex and thus the undercar airflow which reduces the downforce.

[kilcoo316]

Originally posted by GeorgeTheCar
There is another element that greatly effects the efficiently of the diffuser and that is the relationship between the diffuser and the wing.

Indeed.


I am somewhat surprised that the lower wing (wing mount) on the rear wing is not much more complex, with significantly reduced endplates around them to induce these vortexes which entrain the underfloor air out.

[Wuzak]

Some diffuser pictures from Melbourne

Brawn

http://forum.planet-...getfile&id=807


Ferrari

http://forum.planet-...getfile&id=808



Toyota

http://forum.planet-...getfile&id=809

[OfficeLinebacker]

Originally posted by jmorris
A good-old boy, NASCAR inspector would, on the first day of testing, say “forget what we wrote, it’s not what we had in mind, so get rid of it before you go on the track”.

Not to mention, the team would be fined $100,000 and the crew chief suspended for five races "for actions detrimental to stock car racing."

Nice review.

[jatwarks]

I would like to make some, admittedly inexpert, points about diffusers.

One factor that is easily overlooked when thinking about car aerodynamics is that the air does not flow past the car; the car passes through the air. This obvious statement might seem insignificant.

When a ground effect car passes through (supposedly) still air, the air beneath the car is compressed. Drag will pull the air along in the direction of the car, spilling some from beneath the floor pan as it goes.

What you want, of course, is for the compressed air beneath the car to accelerate along the underbody towards the rear, lowering the pressure with respect to that above the car. This is where the diffuser, amongst other things, helps; by drawing the air towards it.

Think about a megaphone exhaust on a single cylinder motorcycle for a moment.

If you have 1 litre of gas passing through a pipe of 10 cm2 cross sectional area the gas will occupy 100 cm length of the pipe. If the pipe diameter doubles then the length of the gas will shorten by 50cm. In other words, the back end of the gas will catch up the front by 50cm. The gas following will also be drawn forward as the initial quantity expands into the larger diameter. The effect will be reduced slightly by the slowing of the flow as it expands.

The net effect is that the gas at the exit of the exhaust pipe draws along the flow.

Back to diffusers. You have compressed air beneath the car wanting to escape, and tending to be dragged along with the car. With the correct relationship between the underbody and the diffuser the air can be caused to accelerate towards the rear, into the diffuser, where it can expand and slow, prompting further flow along the underbody.

Jim.

[cheapracer]

Originally posted by Paolo




In incorrect but hopefully understandable terms, there is a maximum amount of flow that can be managed by the diffuser; .

I'll go one step further, theres only a max amount of air that can be managed by the whole car. Wind turbines work best with 3 blades because thats all they can manage, more blades and the air simply goes around the whole thing and decreases in efficiency - makes me wonder if Brawns low nose is for this reason, if it can't go under then make what goes over more efficient.

I'm hoping like hell that the others copy the nose and we have more traditional looking cars, the red Bulls etc. are effin ugly - does function follow form after all?

[ray b]

I have seen the pictures of the back ends of the diffuser
but how does the extra doubled bit start under the car

[Wuzak]

Originally posted by ray b
I have seen the pictures of the back ends of the diffuser
but how does the extra doubled bit start under the car

Good question. I always thought that the underbody had to be continuous, without holes.

[GeorgeTheCar]

One factor that is easily overlooked when thinking about car aerodynamics is that the air does not flow past the car; the car passes through the air. This obvious statement might seem insignificant.

That little insight is going to put a lot of wind tunnel companies out of business.

[GeorgeTheCar]

You have compressed air beneath the car wanting to escape, and tending to be dragged along with the car. With the correct relationship between the underbody and the diffuser the air can be caused to accelerate towards the rear, into the diffuser, where it can expand and slow, prompting further flow along the underbody.

I have to differ.

You don't have compressed air under the car. If you did the result would be lift not downforce. The object of the exercise is to through a combination of splitters, diverters, and vortices and the diffusers accelerate the air under the car so that it is at lower pressure than the air above the car.

It is incontrovertible that the teams have generated downforce and I tend to belive that their views on airflow.

[jatwarks]

Originally posted by GeorgeTheCar


That little insight is going to put a lot of wind tunnel companies out of business.

If that is the case then explain to me their investment in rolling road wind tunnels ?

Originally posted by GeorgeTheCar


I have to differ.

You don't have compressed air under the car. If you did the result would be lift not downforce. The object of the exercise is to through a combination of splitters, diverters, and vortices and the diffusers accelerate the air under the car so that it is at lower pressure than the air above the car.

It is incontrovertible that the teams have generated downforce and I tend to belive that their views on airflow.

I didn't explain my point very well. The air is accelerated because there is too large a quantity in too small a space beneath the car: If it could not escape it would be compressed.

Something that I have never fully understood, and would welcome guidance on, is the different properties of static and dynamic pressure.

In dynamic terms the faster flowing air beneath the car is at a lower pressure. But, the accelerated flow is caused by cramming more air into the available space.

This in turn draws more air into the confined space due to the higher velocity.

If the car wasn’t moving the flow would be in the opposite direction; from beneath the car, outwards.

Jim.

[GeorgeTheCar]

Rolling floor wind tunnels are necessary to properly treat the air that is "sticking" to the ground, aka the ground effect.

The air doesn't care whether it is moving or the car is moving, it cares about speed differential so the relationship between the ground (wind tunnel floor) is important.

I believe that the flow is accelerated by having a low pressure area behind the car and limiting the amount of air that reaches the underside of the car.

It is a very delicate situation to create and maintain the correct conditions but the results are clear.

[PNSD]

My understanding of the rolling floor windtunnels were to prevent the boundary layer buildup that would occur on a static floor.

Also, the role of the diffuser is to return the pressure to static pressure value.

The diffuser takes the high velocity flow and slows it down to the velocity at which it entered at and the pressure.

One main rule to remember in the way in which a diffuser works is mass air flow must remain constant..

so, p*A(1)*V(1) = p*A(2)*V(2)

With p of course being equal, so it is simply A(1)*V(1) = A(2)*V(2)

[GeorgeTheCar]

ith p of course being equal, so it is simply A(1)*V(1) = A(2)*V(2)

With the goal of the exercise to create a pressure difference between the top of the car and the bottom, I remain unsure that the assumption that "p" remains constant is a safe one.

[tahadar]

Originally posted by GeorgeTheCar


I have to differ.

You don't have compressed air under the car. If you did the result would be lift not downforce. The object of the exercise is to through a combination of splitters, diverters, and vortices and the diffusers accelerate the air under the car so that it is at lower pressure than the air above the car.

It is incontrovertible that the teams have generated downforce and I tend to belive that their views on airflow.

guys, at the speeds we are talking, air is assumed to be incompressible. only when you get near Mach 0.3 (around 230 mph) does compressibility have an effect.

George, the "p" refers to 'rho', which is density. at F1 speeds density can for our sake be assumed constant.

The air sticking to the ground is NOT ground effect. That is just the concept of skin friction. Ground effect is when you have a body moving a certain velocity in the vicinity of a surface (like the ground), which causes the creation of imaginary vortices that act to increase downforce (or lift if you're flying a plane real low).

[Paolo]

Originally posted by Wuzak


Good question. I always thought that the underbody had to be continuous, without holes.

From what I gather the underbody must have no holes when seen from below.

The air intakes for the upper part of the double decker diffuser are placed on the sides of the underbody step; so they are vertical, and cannot be seen from below.

[rojread]

Are the three current diffuser designs all individual 'in-house' versions by Williams, Brawn and Toyota designed with the new rules in mind, or are they possibly three versions of an aerodynamic tweak 'bought in' from one of the various composite support companies in 'F1 Valley'? Seems odd that three so similar could just pop up.

[GeorgeTheCar]

Shouldn't be any surprize.

It jiust requires a single, even off the cuff, conversation to send people down a very similar development process. Add to that the knowledge base in F1 is a pretty tight one.

[Ross Stonefeld]

It seems to be more a case that the teams are split into two groups depending on how they interpreted the letter and intent of the aerodynamic limitations. Everyone who doesn't have that diffuser is saying/thinking "err, we would have done that too, we thought it wasn't allowed"

[zac510]

It wouldn't really be so bad if they hadn't all banned themselves from testing during the season, running the wind tunnels extensively, etc etc. Really shot themselves in both feet numerous times with that one.

Having effectively ruled out developing a new part quickly, the only real option is to knock down your opponent via the 'courts'.

[Wuzak]

The Brawn diffuser explained.

http://vbox7.com/play:9c408f41

Shows holes in the floor - surely that is illegal?

[jmorris]

Hey, the guys from Autosport must be reading this stuff and using it! In the March 26 issue their (Gary Anderson) discussion of the Brawn diffuser was much better than their earlier meager efforts. Do you think McLaren could use our help? Keep writing and sharing.

[jmorris]

Wuzak, thanks for a great video link. The hole in the diffuser would seem to encourage flow around the side pods. As the air accelerates around the side pods it creates a low pressure area just above the edge of the undertray helping to maintain the low pressure under the car by shielding it from high pressure air spilling in. The Brawn diffuser perhaps works to encourage flow both beneath and along the side of the car. McLaren may have been trying to get the same effect by cutting notches in the undertray in front of the rear wheels - but it obviously didn't help them.

[kilcoo316]

Originally posted by jmorris
Wuzak, thanks for a great video link.

Unfortunately, the video link is a load of bull. Its is not even geometrically accurate. There are two holes that clearly expose the upper bodywork as seen from the ground and thus are in direct contravention of the regulations


It is not based on any numerical work, only someone with a fancy graphics package.


Diffusers mainly generate downforce by increasing the massflow rate of air under the car*. By clearing out the region around the top of the diffuser, the teams are getting some more downforce from momentum exchange, and they are entraining more flow out from under the diffuser by turbulent mixing of air over the diffuser - basically energy levels.


*As others have alluded to, the front of the floor and the upstream components - bargeboards, splitter plate, sidepod lower surfaces etc need to be designed to match this optimum *peak* massflow rate.

[Paul Prost]

For F1 external aerodynamics, the air is incompressible and density is constant.

You only have to include temperature effects (and the corresponding changes in density) when examining brake cooling or exhaust gas analysis. As was pointed out, Mach = 0.3 is considered the limit of incompressible flow. Beyond this Mach number, you have consider air as a compressible gas, not an incompressible fluid.

The purpose of the diffuser is to accelerate the flow under the car, lowering the static pressure in this region. This is accomplished purely by the continuity equation, i.e. rho1*A1*v1 = rho2*A2*v2 as posted previously. By increasing A2 (diffuser area) you increase V1 (floor velocity) and thus lower the pressure. The diffuser will help increase the strength of the low-pressure region at the beginning of the floor, as well as create a low pressure region at the end of the flat floor. Static pressure will increase inside the diffuser itself, but its the upstream effects that are important.

Barge boards serve two purposes.

1) Divert the tire wakes to stop the low-momentum fluid in the tire wakes flowing under the floor.
2) The lower edges up the barge boards create strong coherent vortices which themselves create additional regions of low pressure under the floor.

You want to keep the low-momentum tire wakes away from the floor. The sidepod undercuts also interact with the vortices cast from the upper edge of the barge board which helps 'seal the floor'.. i.e. these upper barge-board vortices roll along the edge of the floor and act in the opposite sense to those vortices formed along longitudinal edges of the floor.

The undercut also interacts with the diffuser and lower wing.

Therefore, if you change your diffuser, you will change the flow conditions upstream of the floor and the pressure gradient along the floor. Therefore, your bargeboards are going to have to be redesigned. Additionally, the interaction with the lower rear-wing and diffuser is very tightly connected as well.

[jmorris]

Since the relationship between volume of flow and pressure is linear, so is the relationship of diffuser to downforce...right?
The questionable diffusers have perhaps 10-15% greater area than the non-questionable ones, anyone have better information?
So, if (that is a Big if) you have room for the larger diffuser (no gearbox or suspension pieces obstructing flow), other than fine tuning to achieve aero balance front to rear, do you really need to "build a new car" as some the non-questionable teams claim.
If your barge boards and undercuts are working to seal the floor, I can't see where they become unsealed with 10-15% greater flow.

[kilcoo316]

Originally posted by Paul Prost
Barge boards serve two purposes.

1) Divert the tire wakes to stop the low-momentum fluid in the tire wakes flowing under the floor.
2) The lower edges up the barge boards create strong coherent vortices which themselves create additional regions of low pressure under the floor.

More than that.

The bargeboard also serves as a means of biasing the transverse massflow rates under the car. By having a taller inner section of bargeboard, the 'stagnation point*' is further up, tending to direct more flow down, under the bargeboard. The outer section has a lower effective stagnation point, 'forcing' less flow under the car.


*it doesn't actually exist as such, but its an effective (if technically incorrect) term to explain what I mean.



It is pretty much this that would require a bargeboard redesign.

[derekf1]

Originally posted by kilcoo316


Unfortunately, the video link is a load of bull. Its is not even geometrically accurate. There are two holes that clearly expose the upper bodywork as seen from the ground and thus are in direct contravention of the regulations


It is not based on any numerical work, only someone with a fancy graphics package.


Diffusers mainly generate downforce by increasing the massflow rate of air under the car*. By clearing out the region around the top of the diffuser, the teams are getting some more downforce from momentum exchange, and they are entraining more flow out from under the diffuser by turbulent mixing of air over the diffuser - basically energy levels.


*As others have alluded to, the front of the floor and the upstream components - bargeboards, splitter plate, sidepod lower surfaces etc need to be designed to match this optimum *peak* massflow rate.

Quite. The geometry portrayed in that link just won't work but I think it is the principal of the diffusers that is being demonstated. The Brawn does indeed have holes in the diffuser which thus negates the need for a vertical transition between the reference and step planes and therefore allows secondary expansion above the plank. Clearly a hole in the diffuser is a bad idea because it would equalise the upper and lower pressures, so the diffuser must be designed so that the hole is actually closed by taking advantage of manufacturing tolerances and prestressed components. This idea is nothing new and was first exploited in the wishbone diffusers first introduced in the 2007 Mclaren and copied by virtually everyone since. It was always a grey area but in this latest interpretation it really is taking the piss.

[bobqzzi]

Originally posted by Ross Stonefeld
It seems to be more a case that the teams are split into two groups depending on how they interpreted the letter and intent of the aerodynamic limitations. Everyone who doesn't have that diffuser is saying/thinking "err, we would have done that too, we thought it wasn't allowed"

There is no intent, only the letter- at least in a system like the FIA runs. In Nascar, as someone pointed out, the tech inspector would have just said, "wow, you're right the rules as written do allow that. Now take it off because I just changed the rule." Each system has its pros and cons- the ridiculous part of the diffuser situation is that it takes a month to appeal.

As the rules are written, the diffusers are legal and the teams that were too stupid to read the rules carefully will, hopefully, be SOL. (and I'm a Ferarri fan). As for having done it too if they thought it was legal..I guess, but if they were unsure, what would be the downside of asking? Brawn didn't because he didn't want to give it away- but if you had decided NOT to do it, why not ask for a clarification as there would be no downside?

I think they just missed the possibility as strange as that may be.

[RDV]

rojread-Are the three current diffuser designs all individual 'in-house' versions by Williams, Brawn and Toyota designed with the new rules in mind, or are they possibly three versions of an aerodynamic tweak 'bought in' from one of the various composite support companies in 'F1 Valley'? Seems odd that three so similar could just pop up.

It was originally mooted by the aero team at Aguri F1, in view of the new rules, and when that closed the relevant personnel went back to Honda, plus one off to Toyota...the Williams version apparently thought up in-house.

[ivanalesi]

An interesting video I found on Ferrari's website, comparing the air-pressure along the underbody between F430 and F2008: http://www.ferrari.c...&c=Video&cat=14

[DaveW]

Srs. Briatore & Domenicali seem to have missed a trick. It is becoming clear that they should have protested pull-rod suspensions.

[michaelab]

Could someone please explain a bit more what Williams' Adam Parr said regarding the double decker diffusers and gaps between the reference plane and the step plane?

http://www.autosport...rt.php/id/74516

The bit I'm interested in:

Part of the case presented against us related to what we call as the use of multiple vertical transitions. Essentially, you have to have a reference plane, which is like the plank, and 50mm above that you have the step planes. One of the key issues in the case was: when do you have to have a transition between those two?

Essentially you have to have a vertical transition between the two when the step plane is visible directly above the periphery of the reference plane. Where you don't, it is explicit that you don't have to have one. So one of the key issues in the case was that if you don't have to have one at certain points then by definition you can have many transitions.

Ferrari's case was that you could only have one or at best only one on each side. The problem that they had was that for many years cars have had multiple vertical transitions because at the front, where they have turning vanes or bargeboards, they have had a slot in that transition that creates more than one.

Two things I haven't understood where a picture or diagram would help enourmously:
1. Where are the 'holes' that allow air into the upper deck of the double decker diffusers?
2. Any pictures of the "multiple vertical transitions" (ie holes/slots) in Ferrari's (and others) bargeboards pre-2009?

It seems to me that the key part of the disagreement is not so much the existence of the upper deck of the diffuser, as the existence of the 'holes' that feed it.

Michael.