## Diffuser functions

Here are our CFD links and discussions about aerodynamics, suspension, driver safety and tyres. Please stick to F1 on this forum.
shelly
144
Joined: Tue May 05, 2009 11:18 am

### Re: Diffuser functions

If less density = less lift for airfoil, that holds true for floor also.

About vortices; afree that you have to pilot them where they are useful on the car. Lower bargeboard vortex goes under the floor and is exploited not for upwash, but for low pressure core. Too many vortices will give too much total pressure loss and general efficency decrease

hollus
362
Joined: Sun Mar 29, 2009 12:21 am
Location: Copenhagen, Denmark

### Re: Diffuser functions

Lift from an airfoil, downforce from the floor, any of those air pressure driven forces, depend on the pressure (difference) in 2 different surfaces, generally one facing down and one facing up.
If you lower the density on both surfaces, you lose overall effect, there are less molecules pushing from both the high and the low pressure side so likely you will struggle to create the same pressure difference you would get at normal density.
In this case you are lowering the density in the floor (sucking) side, but leaving the normal pressure in the sky (pushing) facing side, which is most of the car. So you have the same number of molecules (and pressure) pushing down from above the car. Low pressure under the floor only affects one of the two surfaces involved, and not the one pushing down, so it is compatible with full downforce. Low pressure all around the car, that would be a problem.
Santa Claus is drowning this summer.

marekk
2
Joined: Fri Feb 11, 2011 11:29 pm

### Re: Diffuser functions

hollus wrote:If you lower the density on both surfaces, you lose overall effect, there are less molecules pushing from both the high and the low pressure side so likely you will struggle to create the same pressure difference you would get at normal density.

Less molecules, but at higher (thermal) mean speed - momentum still the same, and at molecular level pressure is really just how much of air molecules momentum normal to the surface you have.

At nice sunny winter day you can have 100kPa atmpospheric pressure and more air molecules around you then on nice 100kPa summer day. In both cases if you create pressure difference of 5% between upper and lower floor surface, you will get the same downforce (500kg/m^2), but obviously less drag, because dynamic pressure does not depend much on thermal properties of air, and depends linearly on density.

Downforce/lift creation mechanism on diffuser and wing are very different.
Wing is all about air circulation in free space, diffuser downforce is from Bernoulli effect in constrained space.

hollus
362
Joined: Sun Mar 29, 2009 12:21 am
Location: Copenhagen, Denmark

### Re: Diffuser functions

marekk wrote:
At nice sunny winter day you can have 100kPa atmpospheric pressure and more air molecules around you then on nice 100kPa summer day. In both cases if you create pressure difference of 5% between upper and lower floor surface, you will get the same downforce (500kg/m^2), but obviously less drag, because dynamic pressure does not depend much on thermal properties of air, and depends linearly on density.
Yet in winter testing F1 cars are typically faster than in summer, and this is largely due to increased downforce from the increased air density.
Although the downforce comes from a difference in momentum of the molecules perpendicular to the surface, the wings (or floors) are working on the molecules themselves, and if you have less molecules to work on, you'll have to push them harder to achieve the same 5% difference in momentum. What can be interpreted in terms of pressures, can also be interpreted in terms of good old conservation of momentum, and in the end, both sets of math must hold in the real world.
I cannot quantitate this, at least not from the top of my head, but I agree with Shelly than an airfoil (the wings) will produce less downforce operating in less dense air. Fact is the cars are (much) faster in colder days.
I am not so sure that downforce from the floor should be affected much by air density fed under the floor, since as you put it, it is a restricted space. Probably there is a smaller negative effect, and possible positive effects from cooling and boundary layers. But the effects are somewhat different from those affecting an airfoil, that's also why there is less drag penalty for downforce from the floor; and again the phenomena are different from an inertia point of view, in one you pull-push air, in the other the tarmac. My point was that the top of the car, the second half of the "floor airfoil", is still operating in air at the normal density, unaffected by the escapes, and hence the airfoil analogy is not quite valid.
Santa Claus is drowning this summer.

malcolm
0
Joined: Thu Aug 28, 2008 3:45 pm

### Re: Diffuser functions

Ok. You've left me no option but to bust out MS-Paint.

There are two ways you can accelerate flow under a car. The first is to create a low pressure area behind the car, which encourages more air to go under the car and ultimately fill that low pressure area. The other is a big nozzle that redirects more air under the car, thus accelerating it.

Marekk and Shelley are explaining the former case, whereas N Smikle seems to be focused on the latter.

Of course, you can simulate ground effect by mirroring the body. I think this will help some people visualize why a diffuser drives the floor. The diffuser affects the air that is ahead of the leading edge of the floor. The low pressure area reaches quite far forward. Since the air needs to fill that big diffuser/expansion area, more air needs to flow under the floor. More air under the floor? It must have to accelerate... and we all know accelerated air = lower pressure.

Case 1:
1) This part of the streamline has just "felt" the low pressure region under the floor, so it accelerates down toward the floor.

2) It reaches the floor, the curve to where it goes straight again creates a localized low-pressure area (first suction peak).

3) The accelerated air reaches the diffuser, curves up a bit (giving another suction peak), and slows down, filling that big void.

4) Air has slowed back down to ambient.

Case 2:
5) big nozzle accelerates air by redirecting it under the floor

6) accelerated air shoots out the back, all pissed off because it wasn't diffused.

Case 3 (flat floor with no diffuser; I didn't want to draw a straight line by itself!):
It's not going to create any downforce because nothing will accelerate the air (other than boundary layer expansion, but lets leave that alone for now). There is no diffuser that sucks more air under the car due to the big volume at the back that needs to be filled, nor is there a nozzle that accelerates more air under the car.

So basically, with the diffuser case, you end up with the diffuser driving the entire floor. If it wasn't there, there would be no reason for the air to accelerate as there is no nozzle, and therefore it wouldn't need to be returned to ambient pressure/velocity.

Therefore, the diffuser doesn't create downforce, but merely makes the rest of the floor create downforce.

Note: similar idea to adding elements onto a wing; the latter flaps create little-to-no downforce, but they drive the main-plane of the wing to create much more downforce than if it was just there by itself.

That make sense?

marekk
2
Joined: Fri Feb 11, 2011 11:29 pm

### Re: Diffuser functions

hollus wrote:
marekk wrote:
At nice sunny winter day you can have 100kPa atmpospheric pressure and more air molecules around you then on nice 100kPa summer day. In both cases if you create pressure difference of 5% between upper and lower floor surface, you will get the same downforce (500kg/m^2), but obviously less drag, because dynamic pressure does not depend much on thermal properties of air, and depends linearly on density.
Yet in winter testing F1 cars are typically faster than in summer, and this is largely due to increased downforce from the increased air density.
...
I cannot quantitate this, at least not from the top of my head, but I agree with Shelly than an airfoil (the wings) will produce less downforce operating in less dense air. Fact is the cars are (much) faster in colder days.
marekk wrote: Floor/diffuser downforce is just pressure difference from Bernoulli effect, no significant flow circulation.

It's true that less density = less lift for an airfoil, but this is because both working surfaces of the wing "see" less dense air, and dynamic pressure (responsible for air acceleration around top surface and high pressure buildup on the lower surface) changes linearly with density.
F1 cars have of course more power in more dense/colder air due to better engine efficiency, and better lift/downforce coefficient of wings and demand less cooling.

The whole thinking behind exhaust heat use is to use it locally, where you can have gains.

Floor/diffuser/body system (one continous surafce) is hardly an airfoil and we should not try to explain aerodynamics of this part of the car using the same thinking as we apply to front/rear/wing wings. Physics behind both cases is very different.
Although the downforce comes from a difference in momentum of the molecules perpendicular to the surface, the wings (or floors) are working on the molecules themselves, and if you have less molecules to work on, you'll have to push them harder to achieve the same 5% difference in momentum. What can be interpreted in terms of pressures, can also be interpreted in terms of good old conservation of momentum, and in the end, both sets of math must hold in the real world.
Don't think so.
For stationary car (no flow) vector sum of horizontal momentum for all air molecules under the floor is 0. To lower the static pressure under the car you have to accelerate all the molecules in horizontal direction - and this is actually easier at lower air density (less mass, less inertia).
Doesn't really matter how big vertical component of molecules speed/momentum is.

marekk
2
Joined: Fri Feb 11, 2011 11:29 pm

### Re: Diffuser functions

malcolm wrote: That make sense?
To me almost perfectly. There is a little bit more to this due to floor's rake and high pressure upfront, but basically agree.

malcolm
0
Joined: Thu Aug 28, 2008 3:45 pm

### Re: Diffuser functions

marekk wrote:
malcolm wrote: That make sense?
To me almost perfectly. There is a little bit more to this due to floor's rake and high pressure upfront, but basically agree.
Yeah, I'm not trying to go all Newton and explain everything back to first principles... just a basic outline so people get that a diffuser is the main reason why the floor makes downforce.

The high pressure up front adds to the nozzle effect, and the rake adds to the diffuser effect, of course, but I didn't mention that to keep it simple to explain the effect.

Thanks!

hollus
362
Joined: Sun Mar 29, 2009 12:21 am
Location: Copenhagen, Denmark

### Re: Diffuser functions

@Marekk,

I think we are agreeing 99% and I just got you confused with semantics. In the end we are trying to describe the same phenomenon, you with Bernoulli, me through the kinetic model of gases with molecules pushing each other around. After all you and me are the main defenders of the theory of the exhaust gases actually going under the floor in the R31, and we both appreciate very similar reasons why that should be so.
marekk wrote: For stationary car (no flow) vector sum of horizontal momentum for all air molecules under the floor is 0. To lower the static pressure under the car you have to accelerate all the molecules in horizontal direction - and this is actually easier at lower air density (less mass, less inertia).
Doesn't really matter how big vertical component of molecules speed/momentum is.
And this is the 1% were we differed, and I think you have now converted me. The higher temperature air coming from the exhaust is different from colder air, it has different properties, but applied properly, and accounting for the differences to "normal" air in the design of the car, it is not worse than colder air, just different, with both advantages and drawbacks. This "easier to accelerate" is new to me, and it makes a lot of sense, after all the ultimate motive force for that air is the lower pressure area behind the car, and that is a hole made in "normal" colder air.

Actually, for the first time, that gain of horizontal momentum description has allowed me to make a first principles connection between my kinetic model of gases and Bernoullis' description that so loved is over here at F1T. That horizontal net momentum of the molecules is not independent of the individual vertical momenta, hence the effect of acceleration in pressure. And of course, both descriptions arrive at the same result .
Santa Claus is drowning this summer.

marekk
2
Joined: Fri Feb 11, 2011 11:29 pm

### Re: Diffuser functions

hollus wrote:@Marekk,

I think we are agreeing 99% and I just got you confused with semantics. In the end we are trying to describe the same phenomenon, you with Bernoulli, me through the kinetic model of gases with molecules pushing each other around. After all you and me are the main defenders of the theory of the exhaust gases actually going under the floor in the R31, and we both appreciate very similar reasons why that should be so.
marekk wrote: For stationary car (no flow) vector sum of horizontal momentum for all air molecules under the floor is 0. To lower the static pressure under the car you have to accelerate all the molecules in horizontal direction - and this is actually easier at lower air density (less mass, less inertia).
Doesn't really matter how big vertical component of molecules speed/momentum is.
And this is the 1% were we differed, and I think you have now converted me. The higher temperature air coming from the exhaust is different from colder air, it has different properties, but applied properly, and accounting for the differences to "normal" air in the design of the car, it is not worse than colder air, just different, with both advantages and drawbacks. This "easier to accelerate" is new to me, and it makes a lot of sense, after all the ultimate motive force for that air is the lower pressure area behind the car, and that is a hole made in "normal" colder air.

Actually, for the first time, that gain of horizontal momentum description has allowed me to make a first principles connection between my kinetic model of gases and Bernoullis' description that so loved is over here at F1T. That horizontal net momentum of the molecules is not independent of the individual vertical momenta, hence the effect of acceleration in pressure. And of course, both descriptions arrive at the same result .
Yeah, that's the real "magic" behind Bernoulli principle - molecules moving along slower hit this particular 1cm^2 of surface more often then molecules moving quicker.

malcolm
0
Joined: Thu Aug 28, 2008 3:45 pm

### Re: Diffuser functions

marekk wrote: Yeah, that's the real "magic" behind Bernoulli principle - molecules moving along slower hit this particular 1cm^2 of surface more often then molecules moving quicker.

I had never really thought about it on that level. Mind-bottling. =D>

hollus
362
Joined: Sun Mar 29, 2009 12:21 am
Location: Copenhagen, Denmark

### Re: Diffuser functions

And they also hit it a bit harder
Santa Claus is drowning this summer.

speedsense
13
Joined: Sun May 31, 2009 6:11 pm
Location: California, USA

### Re: Diffuser functions

hollus wrote:Lift from an airfoil, downforce from the floor, any of those air pressure driven forces, depend on the pressure (difference) in 2 different surfaces, generally one facing down and one facing up.
If you lower the density on both surfaces, you lose overall effect, there are less molecules pushing from both the high and the low pressure side so likely you will struggle to create the same pressure difference you would get at normal density.
In this case you are lowering the density in the floor (sucking) side, but leaving the normal pressure in the sky (pushing) facing side, which is most of the car. So you have the same number of molecules (and pressure) pushing down from above the car. Low pressure under the floor only affects one of the two surfaces involved, and not the one pushing down, so it is compatible with full downforce. Low pressure all around the car, that would be a problem.

Something to think about regarding the floor gap, if the air is still (before the car gets there) the air just above the asphalt of the track will experience radiated heat from the higher temp of asphalt which could be much higher than the air 1 meter above it.
If it's sunny and 23c ambient, the race track maybe 36c and radiate the air just above it to somewhere in between. So an imbalance of density of the top of the car and the bottom would occur.
"Driving a car as fast as possible (in a race) is all about maintaining the highest possible acceleration level in the appropriate direction." Peter Wright,Techical Director, Team Lotus

marekk
2
Joined: Fri Feb 11, 2011 11:29 pm

### Re: Diffuser functions

speedsense wrote:
hollus wrote:Lift from an airfoil, downforce from the floor, any of those air pressure driven forces, depend on the pressure (difference) in 2 different surfaces, generally one facing down and one facing up.
If you lower the density on both surfaces, you lose overall effect, there are less molecules pushing from both the high and the low pressure side so likely you will struggle to create the same pressure difference you would get at normal density.
In this case you are lowering the density in the floor (sucking) side, but leaving the normal pressure in the sky (pushing) facing side, which is most of the car. So you have the same number of molecules (and pressure) pushing down from above the car. Low pressure under the floor only affects one of the two surfaces involved, and not the one pushing down, so it is compatible with full downforce. Low pressure all around the car, that would be a problem.

Something to think about regarding the floor gap, if the air is still (before the car gets there) the air just above the asphalt of the track will experience radiated heat from the higher temp of asphalt which could be much higher than the air 1 meter above it.
If it's sunny and 23c ambient, the race track maybe 36c and radiate the air just above it to somewhere in between. So an imbalance of density of the top of the car and the bottom would occur.
Yeah, but all downforce generating devices on F1 car are just few cm thick, so don't think temp differences between both surfaces could be that big from that.

speedsense
13
Joined: Sun May 31, 2009 6:11 pm
Location: California, USA

### Re: Diffuser functions

marekk wrote:
speedsense wrote:
hollus wrote:Lift from an airfoil, downforce from the floor, any of those air pressure driven forces, depend on the pressure (difference) in 2 different surfaces, generally one facing down and one facing up.
If you lower the density on both surfaces, you lose overall effect, there are less molecules pushing from both the high and the low pressure side so likely you will struggle to create the same pressure difference you would get at normal density.
In this case you are lowering the density in the floor (sucking) side, but leaving the normal pressure in the sky (pushing) facing side, which is most of the car. So you have the same number of molecules (and pressure) pushing down from above the car. Low pressure under the floor only affects one of the two surfaces involved, and not the one pushing down, so it is compatible with full downforce. Low pressure all around the car, that would be a problem.

Something to think about regarding the floor gap, if the air is still (before the car gets there) the air just above the asphalt of the track will experience radiated heat from the higher temp of asphalt which could be much higher than the air 1 meter above it.
If it's sunny and 23c ambient, the race track maybe 36c and radiate the air just above it to somewhere in between. So an imbalance of density of the top of the car and the bottom would occur.
Yeah, but all downforce generating devices on F1 car are just few cm thick, so don't think temp differences between both surfaces could be that big from that.
Can't speak for a F1 car, though it is "common" to lose downforce on very hot days, where a wing change of one or two degrees greater is needed to regain the same amount of ride change (due to downforce) for the same speed. Cooler days are just the opposite where lessening wing and retaining the same downforce numbers, as observed in data acquisition.
On a hot day of say 100 F, the track could get as higher as 125-130 (depending on how dark and make up of the asphalt) Putting your hand a couple of inches above the pavement you can feel the difference in the radiating heat.
If the air temp went from 70 to 100 degree change and I have to add wing, surely the difference of 100F air temp vs a ground temp 25-30 f hotter must be creating some type of imbalance and may one of the reason as the needed increase in wing angle...IMHO
"Driving a car as fast as possible (in a race) is all about maintaining the highest possible acceleration level in the appropriate direction." Peter Wright,Techical Director, Team Lotus

### Who is online

Users browsing this forum: Vyssion and 0 guests