Coanda effect

[B767Driver]

It sure seems that they are both saying the same thing to me. As I read the passages from Anderson...I don't see any appreciable differences from what Carpenter is saying. He actually, reinforces what I learned from Carpenter.

The trailing edge is definitely a player in circulation...and it appears to me the starting vortex is simply the result of the boundary layer separation...and not the cause of anything. Expand on this if you think differently.

From Anderson's illustrations...I can see how it looks like the air separates from underneath. But...it shows the starting vortex curling up rapidly after the intial downward momentum that causes that illusion.

Back to Coanda....do you feel this is any part of that effect?

[tgrayson]

Originally Posted by B767Driver

It sure seems that they are both saying the same thing to me. As I read the passages from Anderson...I don't see any appreciable differences from what Carpenter is saying. He actually, reinforces what I learned from Carpenter.

The Anderson material says that the starting vortex starts the circulation:

Anderson: ...The sense of circulation of the starting vortex is counterclockwise in Figure 15.10., which means that it must leave behind a clockwise circulation around the airfoil.

<<The trailing edge is definitely a player in circulation...and it appears to me the starting vortex is simply the result of the boundary layer separation...and not the cause of anything. Expand on this if you think differently.>>

There would be no circulation if there were no separation. It's what starts the circulation happening. This is what causes the Kutta condition to be true. A vortex forms to change to circulation in whatever way is necessary to ensure that the airflow leaves the trailing edge smoothly. This is a form of stability. Any change in AOA generates a new starting vortex.

<<From Anderson's illustrations...I can see how it looks like the air separates from underneath. But...it shows the starting vortex curling up rapidly after the intial downward momentum that causes that illusion.
>>

The flow cannot separate from the top, because it is the bottom flow that is trying to move around the trailing edge, and it will continue to do so until the separation causes the quantity of circulation to be such that the rear stagnation point reaches the trailing edge. The top flow never reaches the trailing edge until the Kutta Condition is satisfied.

And I'm not asking you to rely on the picture, but to rely on the text that describes what the picture shows. If the flow were coming from the top, the starting vortex would rotate in the opposite direction.

Carpenter also says that the flow separates from the bottom going to the top, so they both agree on this. Separation happens before circulation.

Originally Posted by B767Driver

Back to Coanda....do you feel this is any part of that effect?

Nope. Carpenter himself says:

This centripetal force mV^2/r which is required to get the particle of air round the corner can only be provided by a pressure force from the air in streampaths further out--it could not be provided by viscous forces since these would be tangential to the flow.

[B767Driver]

Originally Posted by tgrayson

Carpenter also says that the flow separates from the bottom going to the top, so they both agree on this. Separation happens before circulation.

.

Associated with illustration 7.10, Carpenter states that separation occurs cleanly at the trailing edge. 7.9, I believe is only a starting condition...where the stagnation point is on the upper surface of the wing. Once the airflow begins...the stagnation point changes to the trailing edge where separation occurs. Do I have this wrong?

Also...your point about the vortex causing circulation...not the trailing edge. Wouldn't this be like a wake? A wake doesn't "drag" the airplane down. The adverse pressure forces on the wing drag the airplane down. The wake is an effect of the cause.

How is the vortex the cause of circulation and not an effect of it? Once it leaves the surface...I don't see how it's causing the circulatory flow?

[B767Driver]

Originally Posted by tgrayson

The flow cannot separate from the top, because it is the bottom flow that is trying to move around the trailing edge, and it will continue to do so until the separation causes the quantity of circulation to be such that the rear stagnation point reaches the trailing edge. The top flow never reaches the trailing edge until the Kutta Condition is satisfied.

I'm confused on this, then. The circulatory flow is clockwise. So, to me, the clockwise flow continues in streampaths around the radius of a circle....the center of the circle a point known as the trailing edge. As the radius of the circle gets smaller...and eventually reaching the aft stagnation point, the trailing edge...wouldn't the air on top keep trying to accelerate and make the turn towards the lower surface?

Is it the higher pressure air from the lower surface that dominates the confrontation...causing the lower surface flow to separate?

[tgrayson]

Originally Posted by B767Driver

Associated with illustration 7.10, Carpenter states that separation occurs cleanly at the trailing edge.

Not quite. He says "Very rapidly the stagnation point moves rearwards from its starting point [on the upper surface] until the flow separates totally from the sharp trailing edge."

The reason the stagnation point is moving is due to circulation. The airflow is already separating as the bottom layer moves around the sharp edge.

Originally Posted by B767Driver

Once the airflow begins...the stagnation point changes to the trailing edge where separation occurs. Do I have this wrong?

No, but I'm thinking that Carpenter's use of the word "separation" is a bit sloppy here. Most books talk about the airflow leaving the trailing edge "smoothly" or "cleanly".

Originally Posted by B767Driver

Also...your point about the vortex causing circulation...not the trailing edge. Wouldn't this be like a wake? A wake doesn't "drag" the airplane down. The adverse pressure forces on the wing drag the airplane down. The wake is an effect of the cause.

Agreed, the starting vortex trailing the aircraft is not relevant. What is relevant is where the starting vortex was when it begin. Newton's third law states that for every action there is an equal an opposit one; this is why the starting vortex creates another vortex in the other direction. But once the vortex is created, it will continue until something stops it. It doesn't need the starting vortex any more. The only time a new vortex is needed is when the circulation must change.

[tgrayson]

Originally Posted by B767Driver

I'm confused on this, then. The circulatory flow is clockwise.

Initially, there is no circulation. When the airfoil starts moving, the air flows around the airfoil almost as if their is no viscosity. This is when the rear stagnation point forms on top of the airfoil, as Carpenter describes on p. 172, last paragraph. Only once a certain speed is reached does the airflow start to separate around the trailing edge. THAT establishes the circulation that moves the stagnation point to the trailing edge.

[This is a difficult discussion to have online; I need books and whiteboards. Coming to Memphis anytime soon?]

[B767Driver]

I should pick up Anderson's text and read it.

Like you said...Carpenter didn't expand much on how the trailing edge induced circulation. But, you are saying that the lower surface airflow flows up around the trailing edge and mixes with the upper surface flow...and creates a vortex. It's this vortex....somewhere on the upper surface....that actually induces the circulatory flow...and separates at the trailing edge leaving behind the starting vortex...trailing off into wherever land of irrelevance.

If that's it....I think I got it.

[B767Driver]

Originally Posted by tgrayson

[This is a difficult discussion to have online; I need books and whiteboards. Coming to Memphis anytime soon?]

You did a good job.

The text, while explaining the concept well, left me with a big void in understanding the nuts and bolts of how the TE induced the circulation. Now, I see exactly what's going on.

And I'd have to agree....no Coanda Effect here. As the circulatory flow requires some type of lower flow to induce a vortex.

[tgrayson]

Originally Posted by B767Driver

I should pick up Anderson's text and read it.

Hmmm.....lots of vector calculus. There are only a handful of sections that I found useful. This is "Fundamentals of Aerodynamics", not his more accessible "Introduction to Flight." I could probably photocopy these sections, if you like.

<<Like you said...Carpenter didn't expand much on how the trailing edge induced circulation. But, you are saying that the lower surface airflow flows up around the trailing edge and mixes with the upper surface flow...and creates a vortex. It's this vortex....somewhere on the upper surface....that actually induces the circulatory flow...and separates at the trailing edge leaving behind the starting vortex...trailing off into wherever land of irrelevance.

If that's it....I think I got it.
>>

That sounds like pretty much it. I would change "lower surface airflow flows up around the trailing edge" to "lower surface airflow separates at the trailing edge and curls up into a vortex". That vortex then induces an equal and opposite vortex around the airfoil. (the bound vortex)

[seagull]

Wow, and me wrapped up with court stuff and unable to sit long enough to even form an intelligent response, and now there really isn't anything left that needs to be said!

Brilliant thread, though!

[caliginousface]

Alright just so I have this clear, doesn't the Coanda Effect say a fluid will follow a convex surface? So not really critical to the development of lift, but just charactertistic of fluids in general.

[tgrayson]

Originally Posted by seagull

there really isn't anything left that needs to be said!

You could always ask if circulation is "real" or just a mathematical method that just happens to provide the right answers.

[tgrayson]

Originally Posted by caliginousface

Alright just so I have this clear, doesn't the Coanda Effect say a fluid will follow a convex surface?.

Yes.

Originally Posted by caliginousface

So not really critical to the development of lift, but just charactertistic of fluids in general

The most careful statement would be that there is no evidence that the Coanda Effect is necessary for the production of lift, since lift is explained accurately through analytical methods which do not rely on the Coanda Effect.

[Realms09]

I spent some more time at the library today looking through Schlichting's "Boundary Layer Theory" which appears to be a classic reference on the subject. Once again, it did not have any specific reference to Coanda effect. I then found a book that was a collection of papers on circulation and boundary layer control, which did have a few references to the effect. I also did a search of an online aerospace paper database. In all cases these papers were regarding jets of air being blown near the leading edge or trailing edge to achieve various performance goals.

From "Experimental and Computational Investigation into the Use of the Coanda Effect on the Bell A821201 Airfoil," Gerald Angle Il, et. al:

The coanda effect can be described as the balance between the inertial and normal pressure gradient forces in a near-surface jet of a fluid. A simple case used to describe this phenomenon is a two-dimensional wall jet, which entrains the surrounding fluid. As the boundary layer is entrained, the local pressure in the boundary layer is reduced, creating a pressure gradient that pulls or entrains the jet towards the surface. From the conservation of momentum, as fluid is entrained, the jet velocity is reduced. Eventually, the jet velocity is low enough that the fluid viscosity creates an adverse pressure gradient, again separating the flow. Expanding this concept to a convexly curved surface, a pressure gradient is created, forcing the jet to bend around the surface, until the adverse pressure gradient is reached.

The key here is that anywhere in the technical literature where Coanda effect is mentioned, it is directly referring to a jet of air in air. In other words, this is a small stream of air moving quickly relative to air flowing around it that encounters a surface. There are no such jets on a normal wing, so it does not make sense to talk about this effect regarding lift production.

My speculation and incomplete thought is that it is not fundamentally different than how a boundary layer and corresponding flow outside of it develop as any fluid flows over a body, but it just caught the attention of investigators because they were dealing with jets.

[tgrayson]

Originally Posted by Realms09

I spent some more time at the library today looking through Schlichting's "Boundary Layer Theory" which appears to be a classic reference on the subject.


I spent some more time at the library today looking through Schlichting's "Boundary Layer Theory" which appears to be a classic reference on the subject. Once again, it did not have any specific reference to Coanda effect. I then found a book that was a collection of papers on circulation and boundary layer control, which did have a few references to the effect. I also did a search of an online aerospace paper database. In all cases these papers were regarding jets of air being blown near the leading edge or trailing edge to achieve various performance goals.

Yep, I have that and didn't find anything, as I recall.


<<As the boundary layer is entrained, the local pressure in the boundary layer is reduced>>

Why would that be? The increased velocity does not come from the boundary layer itself (as in static pressure), so the Bernoullli principle would seem not to apply.

[Realms09]

I agree that the paragraph I quoted is not entirely satisfying, but it's the best explanation of the effect I have found so far. Exactly how that "low pressure" occurs is something I am still trying to figure out. Now, however, it is Thanksgiving break