Diffuser functions

[PlatinumZealot]

And there's a good description of diffusers in "The Aerodynamics of Race Cars" starting on page 19.

http://strangeholiday.com/oops/stuff/an ... 092016.pdf

This very good. =D>

Confirms everything.

[marekk]

@marekk:
- sparks in r31 picture catch an instant , we would have to see the video; also remeber thatlower streamlines near the ground are the least deflected
-the acceleration we are talking about is the existent centripetal acceleration needed to turn streamlines up in the car (would be down in your 2d pictures) So on the body it acts from up to down on a car (downforce from suction peak) and would act from down to up in your 2d pictures/videos.
Maybe the images being oriented opposite form a car diffuser has caused misunderstanding on the direction of the force.

I've watched this video few times, sparks almost never go higher. Of course there is flow change around beam and rear wing, so it will suck upwards, but at diffuser's exit all sparks are paralell to the tarmac.

Another ilustration could be rocket engine with expanding nozzle working in optimal (expanding to ambient pressure) mode - exhaust forms jet with perfectly parallel flow lines. But this one is supersonic and has slightly different geometry.

Lets look at the problem this way:

As the flow arrives at floor/diffuser connection point, it has no vertical momentum. There is only axial/longiutudal component of speed.

Now flow enters diffuser. Still no vertical component of speed. As any fluid, it will try to fill empty space, not by changing direction (you need external force for that), but by expanding.
Expansion doesn't change fluid momentum and direction, just decrease in pressure. No vertical acceleration of fluids molecules. If you look at flow lines, they expand to fill diffuser, but at the exit (in perfect world) they will be parallel to each other.

Suction peak is there just because the speed of the flow is max at the kink (diff inlet), and slows down quickly in semi open space. Of course air molecules, once accelerated, don't loose all the speed at the sudden, so suction peak reaches a little into diffuser itself.

Analogy to pressure decrease at bended pipe's inner wall could be usefull to explain how an airfoil works by changing flow's direction, but diffusers work different.

1.n_smikle wrote: Diffusers do not impart work. (unless you have a moving diffuser

This thread is about moving diffuser. There is not much to discuss about stationary car diffusers.

[PlatinumZealot]

I was referring movement relative to the car as in moving body work and you know it!

But OK, I am very flexible when it comes to these applications so lets do it your way, the diffuser which is on the car is moving relative to the ground.

I want to ask you this:
If you were to measure the total energy of the gas before the diffuser throat, then after the diffuser exit. Other than friction and viscous effects- are there any other changes in total energy of the air?

I really want to see how you answer the questions..

Now..
Can I show you diffusers on pump rotor?

[speedsense]

Is a diffuser really a diffuser? In some properties but not all....

Is a diffuser actually a nozzle? Again in some properties but not all....

Is the diffuser actually an expansion chamber, a vent for better lack of the right word? In some properties but not all....

Is the Diffuser actually a tunnel? Yes, but not really all the properties? What no venturi effect....

Is the Diffuser actually an extension of the nozzle effect of the floor, to vacuum the floor upstream of it and removing the interaction of high pressure pockets of air sucked into the sides of the open edges of the floor and a reduction of those pockets in front of the Diffuser to reduce the lift caused due to it. How about the ground moving underneath the car, with it's ever changing degree of effect....

Aren't we really talking a Nozzle-Tunnel-expansion chamber-that diffuses the air back to air speed, so that the wings on the car have much greater effect and PRODUCE more downforce?

Remember for any of this, the ground lingers below and changes all definitions... and to change any property of any definition or description as a device of the floor or diffuser as any of the above, change the rake, add roll, heave, warp, wheel deflection, pavement changes, ashpalt construction, intervention of turbulence Etc., Etc., and you change the official classification of the device you are naming and classifying....

Once you get to the point you THINK you fully understand a race car's aero dynamics, and you come to realize that you left something out... time for a rethink?

Only one thing has a common thread, reduction of anti-lift or a very big expensive Hoover that doesn't need a filter or a plug in the wall...

IMHO

[shelly]

@marekk: I think there is a flow momentum change into the diffuser. It does not work like a quasi 1D approximation. When expansion takes place into the diffuser, the mean flow path is deviated upwards: expansiontakes palce on both th diffuser wall and the ground. In a simplified example, if the angle between ground and diffuser is 15°, the mean flow path would deviate 7.5°.

Somewhere on the web there should be images of inviscid-potential flow solution around a wedge that show this.

Flow devition around a curve path also it is the explaination for diffuser suction peak at the leading edge -sidepod zone.


I do not fully understand you explaination of suction coming from residual velocity of the flow at the beginning of the diffuser.
I do not agree with you when you say that a diffuser does not work like a wing: the floor is like a wing extradox in ground effect, whose shape is strange and not optimal because of regulations.

About rockets: the flow is made hrizontal by the shape of the final part of the nozzle; if you take an upstreams section and cut it with an horizontal plane, the upper would have an up-directed component of momentum, and the lower would have a symmetric down pointing momentum.

As far as Katz description of diffuser, I find it a very good intorduction, but not a complete description.

I have thought about the renault pictures and I think that sparks are coming from the planck rubbing the tarmac because of the failed upright; so maybe we are seeing a diffuser in a far-off project situation , not working properly.

[White Knight]

Forgive me for intruding. Not to go far off subject, however are we alluding here to a new division in aerodynamics, begging to define where the diffuser stands? as stated, it combines so many principles, and has been adapted so each team utilizes different principles within the parameters allowed to achieve certain results.

Chastise or redirect if necessary. Otherwise, i'm interested.

[PlatinumZealot]

I wouldn't say it combines many principles. A diffuser is simply a gradual increase in the CSA of a duct or pipe etc. That's all it is.

I think you mean you can use the principle of the diffuser to do many things.

A pump rotor
Compressor rotor
Turbine expanding section
A Saxophone
An ac duct
A musket gun
A drain into a pond
etc..

Anywhere you want a gradual expansion.

[speedsense]

I wouldn't say it combines many principles. A diffuser is simply a gradual increase in the CSA of a duct or pipe etc. That's all it is.

I think you mean you can use the principle of the diffuser to do many things.

A pump rotor
Compressor rotor
Turbine expanding section
A Saxophone
An ac duct
A musket gun
A drain into a pond
etc..

Anywhere you want a gradual expansion.

If we look at a purely flat floor (non F1 stepped style) and we put this flat floor on an F1 car and can keep it perfectly parallel to the ground (impossible in real life) we will reach a point (with speed) that the boundry layers of the ground and the bottom of the floor will rise in pressure (due to drag) and create lift under the car. At some point of added speed, the strength of the increasing pressure will start lifting the car, changing the ride height thus allowing an increase in volume under the floor and higher amounts of lift until eventually with even more added speed will lift the car off the ground by overcoming the downforces (above the car).
Adding a diffuser to this equation allows the volume of air increase (before the diffuser) with the speed increase and any lift generated from it to be on a less gradient scale as the volume has a chamber to expand into lessening the lift generation. So in this respect, the diffuser is performing an "anti-lift" property, not down force production, except in allowing more leverage of the wings and assorted spoilers, splitter etc. to maintain a "higher" advantage...
This leads us to the "importance" of extraction of the volume of air in the diffuser allowing faster refill, further reducing the volume under the floor.
Further reduction of lift by designing the floor to lessen even further (back to current F1 design) and smooth transition into the diffuser (radius of the throat) thus keeping the pressures from raising too rapidly on entry into the diffuser as it fills the volume chamber.
Scooped bottoms of the side pods prevents spilling of air flowing around them from entering the sides of the floor. Illegal Ground contact skirts would prevent all of it, so the next best solution (legal one) is the Renault solution with a front blown exhaust, that creates a heated vortex (more compressed than an ambient vortex) along the bottom of the side pods. This acts like a "virtual" side skirt thus reducing intervention of turbulent higher pressure, (than under the floor) from entering the sides.

Lastly, the exhaust blown diffuser. As the best performing one so far is the RBR's and now the Mclaren's solution. Again we have the "problem" of side intervention, through the side gap of the diffuser. The rear wheel is an air pump, pumping air inboard directly at the side gap of the diffuser, adding turbulent, higher pressure air to the volume in the diffuser. Again the best solution is the illegal one, side skirts, stopping all intervention. The RBR/Mclaren solution, blow a heated vortex down the sides of the gaps, blocking the air from the wheel from entering (virtual skirt).

So the diffuser, is an air volume storage area, that is under constant refill at speed, and it's efficiency is based in extraction of the volume of air AND blocking out invention of side "deposits".

So the diffuser might just be defined as an "air bank" (kinda like the ones we put our savings into ) with the volume needed to be extracted as quickly as it was deposited.

[marekk]

If we look at a purely flat floor (non F1 stepped style) and we put this flat floor on an F1 car and can keep it perfectly parallel to the ground (impossible in real life) we will reach a point (with speed) that the boundry layers of the ground and the bottom of the floor will rise in pressure (due to drag) and create lift under the car. At some point of added speed, the strength of the increasing pressure will start lifting the car, changing the ride height thus allowing an increase in volume under the floor and higher amounts of lift until eventually with even more added speed will lift the car off the ground by overcoming the downforces (above the car).

Speed of flow relative to ground will be 0 on calm day. So no boundary layer on air-ground interface.
Boundary layer's thickness for a real big wing (Boeing 747-400 - 12m chord at root) at really high speed (800 km/h) is about 15cm at maximum, far on the trailing edge.

In F1 car we can talk about few milimeters.

Adding a diffuser to this equation allows the volume of air increase (before the diffuser) with the speed increase and any lift generated from it to be on a less gradient scale as the volume has a chamber to expand into lessening the lift generation. So in this respect, the diffuser is performing an "anti-lift" property, not down force production, except in allowing more leverage of the wings and assorted spoilers, splitter etc. to maintain a "higher" advantage...
This leads us to the "importance" of extraction of the volume of air in the diffuser allowing faster refill, further reducing the volume under the floor.

Do you suggest that diffuser is better "chamber" to expand into then the whole rest of earth's atmosphere ? Why ?

Scooped bottoms of the side pods prevents spilling of air flowing around them from entering the sides of the floor. Illegal Ground contact skirts would prevent all of it, so the next best solution (legal one) is the Renault solution with a front blown exhaust, that creates a heated vortex (more compressed than an ambient vortex) along the bottom of the side pods. This acts like a "virtual" side skirt thus reducing intervention of turbulent higher pressure, (than under the floor) from entering the sides.

Why is heated vortex "more compressed" then ambient vortex. ?
Renault exhaust goes under the floor (at least at first). How does this flow go to the bottom of the side pods ?

So the diffuser, is an air volume storage area, that is under constant refill at speed, and it's efficiency is based in extraction of the volume of air AND blocking out invention of side "deposits".

Don't think so.

Some of estimed forum members are trying to understand diffuser function from the wrong end IMO.

F1 diffuser isn't driven by high pressure at inlet. There is no higher then ambient pressure under the floor.

It's really that simply (at least until we want to know the details):

A solid shape (diffuser) is moving through the fluid, creating low pressure area behind.
Air around fills this void, part of this air is coming from diffuser's inlet. Inlet's area is smaller then exit's, so air at intake has to go quicker - this in turn decreases static pressure of air in this region.
Pressure difference between top of the floor (at ambient pressure for flat floor, slightly above for raked one) and bottom of the floor creates downforce.

No more magic needed.

[shelly]

@speedsense: I partly agree and partly disagree with you.

The basic idea I disagree is the flat floor+diffuser not being a downforce device, but being a anti lift device instead.
I also disgree on the mechanism you explained for lift form the floor due to boundary layer growth.

I agree with you on one of the function of eahusts being sweeping away wheels wake and squish vortices; I also agree on the function of scooped-undercut sidepod inlets

Like marekk, I would like you to elaborate on the properties of heated vortices.

@marekk: locally the flow o a floor is divern by the pressure differential between the rear of the car (low pressure in the wake of bluff body + rear wing effect) and the high pressure zone being the stagnation on the vertical wall of the sidepod between the radiators inlet and the floor. When this wall is not present, because the rdiator opening goes all the way down to the floor, there is still an high pressure zone due to the blockage of the radiator being felt upstream.
Te flow is force to escape from high pressure to under the floor along the raidus of the laeding edge (which is limited to 50mm -in r31 is partly solid and partly formed by the side of the ex jet).
Acceleration and change of direction to horizontal make the leading edge suction peak.
After that stream lines tend to diverge laterally (pressure rise to a higher level, but still lower than ambient); further downstream the effect of sction at the beginning of the diffuser starts acting and streamlines are deflected inward.
Second suction peak at the diffuser kink line, with deflection upwards of the flow; then increase in pressure and decrease in velocity through the difuser

[marekk]

@marekk: locally the flow o a floor is divern by the pressure differential between the rear of the car (low pressure in the wake of bluff body + rear wing effect) and the high pressure zone being the stagnation on the vertical wall of the sidepod between the radiators inlet and the floor. When this wall is not present, because the rdiator opening goes all the way down to the floor, there is still an high pressure zone due to the blockage of the radiator being felt upstream.
Te flow is force to escape from high pressure to under the floor along the raidus of the laeding edge (which is limited to 50mm -in r31 is partly solid and partly formed by the side of the ex jet).
Acceleration and change of direction to horizontal make the leading edge suction peak.
After that stream lines tend to diverge laterally (pressure rise to a higher level, but still lower than ambient); further downstream the effect of sction at the beginning of the diffuser starts acting and streamlines are deflected inward.

Agree. That's about what i wanted to say - front of the floor driven by high pressure area under the nosecone and in front of bargeboards and sidepods (blowed), rear driven by lower pressure behind the car (suction), both areas connected forming closed lines for the most of the flow, low pressure wake behind front wheels diverging flow lines outside.

Second suction peak at the diffuser kink line, with deflection upwards of the flow; then increase in pressure and decrease in velocity through the difuser

Maybe it's just semantics, but i don't think you can describe diffuser like an airfoil in ground effect.

Airfoil shape with non zero angle of attack has 2 working surfaces facing outwards to each other, lower (flow facing) and upper (behind flow/airfoil interface). Flow speed decreases near lower and increases near upper surface. As both flows meet after trailing edge, resulting flow (downwash) will curve down (circulation) due to shear forces (viscosity). This change of flow direction/momentum is reacted by lift force. Energy used to accelerate air molecules downwards is (in rough aproximation) induced drag.

Diffuser has 2 working surfaces (diff top wall and tarmac for a car) facing inwards . In car's reference frame, upper surface (diff) see flow at speed = speed of flow under the floor + speed gained from acceleration of air molecules, lower surface see only this "accelerated" part, and has no angle of attack, so is not really working at all.

For any vertical cross section in diff, pressure distribution is something like: high in boundary layer of tarmac (but as tarmac see only "accelerated" value, not much of this), lowest in the axis of flow from the inlet, highest at diff wall.

Vertical component of expansion will change flow lines direction upwards, but positive pressure gradient will prevent acceleration in this direction, and as both vectors compensate, flow at the exit of the diff will be more or less uniform across the area and more or less paralell to the tarmac (perpendicular to diff exit).

I failed again to find illustrations for diff working as sucking device, but scramjet engines use this sort of diff geometry:





And i am quite sure, they don't want thrust vector to deflect upwards.

[horse]

Double post.

[horse]

Maybe it's just semantics, but i don't think you can describe diffuser like an airfoil in ground effect.

Airfoil shape with non zero angle of attack has 2 working surfaces facing outwards to each other, lower (flow facing) and upper (behind flow/airfoil interface).

And the floor (i.e ground effect) making that 3 working surfaces by your labelling 2 facing inwards and a third outwards. A car's diffuser also has a third outward facing surface, that being the top of the car.

As both flows meet after trailing edge, resulting flow (downwash) will curve down (circulation) due to shear forces (viscosity). This change of flow direction/momentum is reacted by lift force. Energy used to accelerate air molecules downwards is (in rough aproximation) induced drag.

I also think your angle of attack is wrong in this description. A car diffuser is like a wing with negative angle of attack (or neutral) in ground effect.

[marekk]

Maybe it's just semantics, but i don't think you can describe diffuser like an airfoil in ground effect.

Airfoil shape with non zero angle of attack has 2 working surfaces facing outwards to each other, lower (flow facing) and upper (behind flow/airfoil interface).

And the floor (i.e ground effect) making that 3 working surfaces by your labelling 2 facing inwards and a third outwards. A car's diffuser also has a third outward facing surface, that being the top of the car.

Right, but due to low pressure area under the beam wing (pressure lower then under diff), top of the car doesn't contribute to flow circulation/bending.

As both flows meet after trailing edge, resulting flow (downwash) will curve down (circulation) due to shear forces (viscosity). This change of flow direction/momentum is reacted by lift force. Energy used to accelerate air molecules downwards is (in rough aproximation) induced drag.

I also think your angle of attack is wrong in this description. A car diffuser is like a wing with negative angle of attack (or neutral) in ground effect.

I meant classic airfoil (airplane wing) in free stream, to explain where flow bending around airfoil comes from. For F1 wings flows bend upwards of course.

[bravefrog]

[...] A car diffuser is like a wing with negative angle of attack (or neutral) in ground effect.

This was exactly the principle used on the first "ground effect" cars (e.g. Lotus 78) which had channels in the sidepods shaped like inverted airfoils.

When the FIA mandated flat floors in 1983, diffusers were allowed, to create low pressure under the floor, as Scarbs explained recently here: http://scarbsf1.wordpress.com/2011/03/2 ... explained/