Reducing the drag of a two element wing through stall

[horse]

If McLaren's system is designed to default to an unstalled state, then I guess I'd be comfortable with it if I were the FIA. I mean, things could still go wrong, but then there's a whole lot of things on an F1 car that can end in disaster if they go wrong.

Very true, but you don't need to add to them, particularly as this appears to be a bugbear of the FIA.

Default un-stalled is a "blow to stall system" (always off), I am guessing then, rather than a "blow to attach" (always on).

[impaero]

I'm willing to be proven wrong by some real figures, but until then I just don't get how stalling the wing will reduce drag - the explanations given above don't add up....if anything, McLaren are trying to prevent stall by energising the boundary layer - this will result in higher downforce for a given AoA and hence allow a lower AoA, which gives lower drag.






Read about some boundary layer research here: http://history.nasa.gov/SP-4103/app-f.htm

[volarchico]

Thanks to SLC and Pup for your explanations and patience. It's a confusing subject for those trained in classical aircraft aero, so I really appreciate you explaining how this works for an F1 car!

[ringo]

If McLaren's system is designed to default to an unstalled state, then I guess I'd be comfortable with it if I were the FIA. I mean, things could still go wrong, but then there's a whole lot of things on an F1 car that can end in disaster if they go wrong.

That was the point i was trying to get across on page 6.

We have to ask if the team will deliberately risk using an inherently stalled wing.

I would call this the "normally open" theory, where the blowing is what "turns on" the wing.
In this case the wing is always stalled and requires that it is blown to operate normally.

The base bleeding concept is the "normally closed" theory, where the wing is "turned off" by blowing the jet. In this case the wing is normally not stalled and only stalled at high speeds.

The normally open theory is what seems to be the accepted theory here , which suggests that if anything were to happen to that jet , the wing will stall. This seems to be very risky.

In the normally closed theory, the one that I and SLC have not thrown away as yet, suggests if the jet stops blowing, the wing will be in no danger of stalling because it is normally in an optimal position.

[horse]

impaero, volarchico, I'm still interested to hear your ideas on this concept, if you would like to contribute. I am also still searching for a detailed description of the flow and how it appears that the same phenomenon as if the flow speed was being reduced is occurring. Please don't get upset at this chaps (SLC, Pup), I just want to understand it more, even though I am very willing to accept the outcome. That's why I started the thread after all.

simoncm, thanks for your contribution. No harm in a bit of maths, in my opinion, although I have some doubts about some of the assumptions you have made, but it's too late to discuss it sensibly now. I'll have another look tomorrow.

That was the point i was trying to get across on page 6.

Sorry, Ringo, it's very hard not to go over old ground!

[Pup]

If McLaren's system is designed to default to an unstalled state

That was the point i was trying to get across on page 6.

What I mean is that it might be possible for the valve that controls the airflow to default to the 'blown' state without driver interaction. That is, if the driver would have to maintain pressure on something, as an example, for the blowing to be cut off. I think that's what's being suggested anyway by SLC. The alternative would be if the system acted as a switch, and the driver had to turn it off and then turn it back on, which would of course be more dangerous.

I don't discount your theory. It would seem within the realm of possibility, but without any examples or other evidence to back it up, personally I still have to consider it unlikely. But who knows, maybe we'll find out someday that you've been right all along.

The only theory I've heard that I would completely rule out is the idea that their wing sucks air instead of blows. Every example I've seen of that has multiple slots so that the boundary layer is constantly reduced along the wing.

In my mind, the basic questions of whether and why the wing stalls are pretty much settled as well. I'd like to see some more hard evidence, but even without, I'm pretty convinced not only that the wing stalls, but also as to why that's beneficial.

[Pup]

Simoncm, I'm with you up to this point...

Everything cancels nicely, therefore the ratio of the drags is
Cl/Cd

A reasonable Cl is 1.5, an angled plate in an air flow has a Cd of 0.5 to 0.8

Therefore the "drag" due to the wing generating downforce is greater (ratio > 1) than the "drag" due to the wing being stalled. In this case by a factor of 2.

What assumptions are you making on the angle of the wing? i.e., where are you getting your numbers for Cl and Cd?

[volarchico]

And with a double element wing with that much camber, I'd imagine lift coefficients in the 2-3 range when they aren't stalled. Based on other's descriptions though, this might not be the right way to think about it since so much of the drag coefficient is actually being created from the downforce (induced drag). Anyone have any numbers on the kinds of downforce and drag coefficients you would see in a typical F1 wing not stalled? How about stalled?

[SLC]

This whole issue is not related to induced drag or skin friction type drag (which is my way of saying it’s not related to anything particularly fancy aero wise). 90% of a Top Rear Wing’s drag is pressure drag – or just the horizontal component of the wing’s load vector.

SLC, I think your terminology might confuse some people who are using the wikipedia articles on drag to follow along. What you're calling 'pressure drag' is in fact referred to on wikipedia as 'induced drag': induced drag; and I think what you call 'induced drag' is the same as what they call 'form drag': form drag

I think you might want to re-read that Wiki article. My usage of the drag terms is correct.

Parasitic drag = skin friction drag + pressure drag + interference drag (interference is more aircraft specific, though).

Induced drag = lift dependent drag (due to the trailing vortex system changing the effective incidence onto the wing element - this is NOT really relevant to our current stall/drag discussion).

Total drag = Parasitic drag + Induced drag

[hollus]

I'm willing to be proven wrong by some real figures, but until then I just don't get how stalling the wing will reduce drag - the explanations given above don't add up....if anything, McLaren are trying to prevent stall by energising the boundary layer - this will result in higher downforce for a given AoA and hence allow a lower AoA, which gives lower drag.






Read about some boundary layer research here: http://history.nasa.gov/SP-4103/app-f.htm

The problem is that those pictures, from plane wings, are simply the wrong example. The angle of attack of that wing is tipically what, 5 degrees? The rear wing of an F1 car is more like 50 degrees. You would probably find that if you force a plane to fly with the wings always turned by 50 degrees to the direction of movement, it will simply not fly! But then, if you put the rear wing of an F1 at 5 degrees, it won't create much downforce either. They are two inherently different systems, with different design goals.
Would you compare the design principles of a parachute and of a race motorbike? Yet they both work within the same laws of physics...

[SLC]

I'm willing to be proven wrong by some real figures, but until then I just don't get how stalling the wing will reduce drag - the explanations given above don't add up....if anything, McLaren are trying to prevent stall by energising the boundary layer - this will result in higher downforce for a given AoA and hence allow a lower AoA, which gives lower drag.






Read about some boundary layer research here: http://history.nasa.gov/SP-4103/app-f.htm

The problem is that those pictures, from plane wings, are simply the wrong example. The angle of attack of that wing is tipically what, 5 degrees? The rear wing of an F1 car is more like 50 degrees. You would probably find that if you force a plane to fly with the wings always turned by 50 degrees to the direction of movement, it will simply not fly! But then, if you put the rear wing of an F1 at 5 degrees, it won't create much downforce either. They are two inherently different systems, with different design goals.
Would you compare the design principles of a parachute and of a race motorbike? Yet they both work within the same laws of physics...

No, the angle of attack is not 50 degrees - you might want to brush up on your definition of that phrase. What is different with an F1 top wing is that its a highly cambered element (dual element, in fact) and as a result its resultant force vector is rather angled.

By the way, impaero, you are still stuck thinking about the classic lecture hall example of a single element, low camber aerofoil. This section will see an increase in drag when it separates, so you are correct in that respect. As I've said several times now, this does not apply to an F1 style top rear wing.

And PS: does your username suggest you go to IC? If so, go ask Prof Morrison, or if you're lucky, Prof Bearman. They will explain this to you (though Bearman might snigger slightly!).

[tok-tokkie]

This has been a very informative discussion for me.

When stalling was first suggested as being advantageous I immediately thought that is absurd. Stalling a wing kills the lift (downforce in our case because the wing is inverted) and maximises the drag - just about the worst thing it seems.

Autogyro pointed out that the forces act very differently on a car compared to an aircraft because the car is prevented from moving under the aerodynamic loads (unlike an aircraft). The forces on the rear wing have leverage about the rear tyre. Here is a diagram posted by slimjim on page 4

The lift vector & the resultant vector both cause downforce on the rear wheel by the mechanical leverage they have. Those vectors are far from downwards so downforce is not apparent if you consider them in aircraft terms.

If you kill the lift vector through stalling the upper element you will get the induced drag of the stalled wing which is a smaller force so the downforce is reduced BUT SO IS THE DRAG. That is what Christian Horner has confirmed.

SLC made some strong hints about stalled wings which I did not believe at the time. He also mentioned fluidics. On the Maclaren thread there was conjecture about the little snorkel so I in jest suggested it was the switch for the slot on the rear wing. I thought SLC was being absurd in suggesting stalling reduced drag so I wanted to propose something just as absurd. Seems like I was absolutely wrong on both counts. Very informative this has been.

The picture of the flow viz also made me believe that stalling was absurd since it shows excellent streamlined flow on the upper element.

For safety the normal flow regime must be with the slot blowing, attached flow on the upper element and maximum downforce (and drag). Interrupt the blowing and the upper wing stalls reducing downforce & drag for increased speed down the straights. Maclaren must have been confirming that attached flow was the stable condition - remember they blocked of part of the air inlet at one time.

The blowing increases the downforce when it is on. This allows the upper wing to be at a lower angle (for the same force). When the wing is stalled it creates less drag because it is at a lower angle. So the controlled blowing allows the drag to be reduced not only by promoting stall but also by lessining the amount of form drag from the stalled upper element.

[autogyro]

Exactly, it is why the intake to the slot on all the cars using this development, is left open on the front of the wing section when the shark fin feed is not used.
The open air flow continues to feed the slot and maintains smooth flow over the rear of the upper wing section.
To get a higher top speed with lower DF they use the fin and devise a method of blocking the flow to the slot intake when needed.

There may be a better way, so I am keeping an eye on the Renault, tum ti tum.

[ringo]

What i have found out is that the velocity coming through the slot is not multiples of the free stream. I also found that the inlet has a ram effect.
The reason the velocity is low is because of some choking, friction and end and bend losses.

This is a fin i added, it's not what exactly on the Mp4, but i think it may have a nozzle in the fin like this.

pressure is very high because of the dynamic pressure and flow restriction of the slot.



the flow re accelerates as it approaches the low pressure behind the wing, but still not enough to multiply speed by a big factor.
Ram effect can increase speed and pressure at the same time. Whenever both happens you know the air is being compressed by the car. Effectively Mclaren have an air compressor without the need for a secondary power source.
There is a drag penalty associated with the this as well, since the car is doing the work, Mclaren probably did a good job of finding a viable outcome with the drag saved by the flap.


I am still clueless about the design of the ducting of the slot. I know the cross section may be close to the coanda duct, in order that the air can leave the slot in some order.
As can be seen, the damn arrows are barely leaving the duct.
The duct can only be a cavity as well, since the wing has to have 1 cross section beyond 75mm from center line.
The system Mclaren is obviously a refined system and may take a while to duplicate much less get the same effect. There are just too many things that have to be done right.

What i notice is that, blowing speeds of about the free stream barely change the flow on the element. Only bulging the velocity gradient. The flow momentum is so high that it takes about 3 times the free stream velocity to rip it of the element to stall it. The farther away from center the harder to stall as well.
I am puzzled by this whole system and think there is more to it than we know.

Next time i will just force it to stall and just take the readings.
I think that is more important than trying to duplicate the thing.

[ringo]

I'll keep modifying until i get something. I think i can change some of the geometry of the ducting and the slit it self.