Is circulation real?

Discussion in 'Hydrodynamics and Aerodynamics' started by Mikko Brummer, Jan 25, 2013.

  1. Sailor Al
    Joined: Feb 2021
    Posts: 650
    Likes: 27, Points: 28
    Location: Sydney

    Sailor Al Senior Member

  2. Doug Halsey
    Joined: Feb 2007
    Posts: 640
    Likes: 212, Points: 53, Legacy Rep: 160
    Location: California, USA

    Doug Halsey Senior Member

    There are a number of misconceptions here:
    1. There is no boundary layer. The velocity distribution that I posted (p 28. post #412) is for an inviscid-flow case (and is labelled as such). Since we don't know the exact shape of the airfoil or the flow conditions, there's not much point in getting a more "accurate" solution.
    2. The grid shown on the plot is analogous to what's on a piece of graph paper. The point spacing used in the calculations is entirely different (clustered to both leading and trailing edges.)
    3. The calculation method has distributed vortices & sources on the airfoil surfaces only - no grid points going in perpendicular direction. Although it's my own implementation, it's virtually identical to what's in the well known (& respected) xfoil code of Mark Drela's, which handles boundary layers just fine, even when used with relatively sparse point spacing. That's because it uses viscous-inviscid interaction techniques, rather than solving the more general Navier-Stokes equations.

    And one more thing: You keep asking for my sources, but the last time I recommended things for you to read, you said you had too many already.
     
  3. Sailor Al
    Joined: Feb 2021
    Posts: 650
    Likes: 27, Points: 28
    Location: Sydney

    Sailor Al Senior Member

    You are correct. But that was before I realised I had to get into the books. And there are so many, which is why I am asking you for your recommendation now.
     
  4. Doug Halsey
    Joined: Feb 2007
    Posts: 640
    Likes: 212, Points: 53, Legacy Rep: 160
    Location: California, USA

    Doug Halsey Senior Member

    Theory Of Wing Sections by Ira H. Abbott & Albert E. Von Doenhoff

    Flight Vehicle Aerodynamics by Mark Drela
     
  5. Sailor Al
    Joined: Feb 2021
    Posts: 650
    Likes: 27, Points: 28
    Location: Sydney

    Sailor Al Senior Member

    @Doug Halsey
    And before I realised that I could download the texts for free from vdoc.pub I assumed I would have to fork out hard-earned cash or traipse to the State Library to access them. I do not understand how copyright laws are not being transgressed, but am grateful for the service.
     
    Last edited: Oct 27, 2022
  6. Sailor Al
    Joined: Feb 2021
    Posts: 650
    Likes: 27, Points: 28
    Location: Sydney

    Sailor Al Senior Member

    The preface to Drela's book includes:
    "Most of this book is based on the lecture notes, handouts, and reference materials which have been developed for the course Flight Vehicle Aerodynamics (course number 16.110) taught by the author at MIT’s Department of Aeronautics and Astronautics. This course is intended for first-year graduate students, but has also attracted a significant number of advanced undergraduates."
    and:
    "This book assumes that the reader is well versed in basic physics and vector calculus, and already has had exposure to basic fluid mechanics and aerodynamics."
    and:
    "little or no space is devoted to introduction or discussion of basic concepts such as fluid velocity, density, pressure, viscosity, stress, etc."
    (my emphasis)
    As an aid to a racing yachtsman, even one with a strong grounding in undergraduate maths and physics attempting a qualitative understanding of how a sail works, offering this text as a "foundation of aerodynamics" is laughable at best and deliberately misleading at worst.

    A similar criticism can be levelled at your offering Abbot & Von Doenhoff's text.

    What is your source of wisdom about the source of the pressure differences around the sail ? And please, don't join Prof. Babinsky in telling me I have to undertake a 3-year undergraduate course in aerodynamics to gain a qualitative understanding.
     
    Last edited: Oct 27, 2022
  7. Alan Cattelliot
    Joined: Jul 2021
    Posts: 505
    Likes: 209, Points: 43
    Location: La Rochelle (Fr)

    Alan Cattelliot Senior Member

    Generally speaking about the boundary layer, you are right Gonzo. Catching its developpment along a surface requires several points. For instance, the "old" Fluent V6 code creates by default boundary layers with 10 points, as many student are been taught during their studies.

    The points 1/2/ are excellent reminders of the difference between the circulation theory and the Prantl's theory. The fluid viscosity is not part of the equations of the circulation. In that case, the velocity distribution around a profile is obtained "virtually" at the surface of the profil. No boundary layer. This is why i feel so unconfortable in making any comparison between the Babinsky video and Doug's velocity distribution. Because there is indeed a boundary layer around the profile in the wind tunnel. And, in theory, the velocity outside a boundary layer should only be the upstream velocity. What velocity should then be compared ?

    In my opinion, the circulation theory should not be pushed too much regarding the determination of the local velocity around a profile, because it fails to be realistic. On the other hand, with all the respect that is due to Babinsky, his demonstration lacks of precision regarding the experiment he shows. A timer would have been handy to check the time of flight of the smoke trails. In addition, there is some strong boundary effects that have been pointed out by Doug and Latestarter. Moreover, we don't have any characteristics on the flow in the wind tunnel : flow speed, angle of incidence, fluid viscosity, turbulence rate, at least.

    In these conditions, increasing the resolution in Doug's calculations or SailorAI calculations won't give anything. Please, correct me if I'm wrong, but, to me, Doug's intend was to provide an example of speed distribution around a profile, and he has taken a great care to explain that his calculations were to be considered as an approximation, in the purpose on explaining more about the fact that "the flow speed is greater on the extrado than on the intrado". SailorAI calculations were made in order to figure out what are the consequences of the speed distribution given by Doug, if one interprets this velocity distribution in the context of smoke puffs in a wind tunnel. It seems to me that Doug's goal is attained, and that SailorAI has gained from his own work greater insights. Excellent.
     
    Last edited: Oct 27, 2022
  8. Sailor Al
    Joined: Feb 2021
    Posts: 650
    Likes: 27, Points: 28
    Location: Sydney

    Sailor Al Senior Member

    Oh, how I enjoy superciliousness!
     
  9. Alan Cattelliot
    Joined: Jul 2021
    Posts: 505
    Likes: 209, Points: 43
    Location: La Rochelle (Fr)

    Alan Cattelliot Senior Member

    @SailorAI : Am I wrong ?
     
  10. Sailor Al
    Joined: Feb 2021
    Posts: 650
    Likes: 27, Points: 28
    Location: Sydney

    Sailor Al Senior Member

    Despite not qualifying as a 1st year graduate student of aerodynamics, I have started to read Drela's book and, probably for that reason, I am struggling early.
    Can you confirm that his reference to "boundary layers" in Sect 1.5.4 corresponds to the boundary layers described in Prandtl's 1904 paper?
    And that the reference to "no-slip" condition in that section corresponds to Prandtl's condition that in the boundary layer between the fluid and the solid body the fluid adheres to the surface?
     
  11. Alan Cattelliot
    Joined: Jul 2021
    Posts: 505
    Likes: 209, Points: 43
    Location: La Rochelle (Fr)

    Alan Cattelliot Senior Member

    Yes, "boundary layers" in Drela 1.5.4 corresponds to the Prantl's boundary layer theory. The boundary layer is the name given at the flow region where the viscous forces dominate the pressure forces, as observed by Prantl in his experiments. In order to retrieve this behaviour in calculations, it is assumed that the flow velocity at the surface shall be null - the fluid adheres to the surface -. This is the "non-slip" condition. The Navier-Stokes equations can be simplified in the boundary layer region, assuming that the boundary layer stay attached and laminar.

    upload_2022-11-22_7-48-10.png

    In the diagram, the boundary layer is represented in pink color. For a less effort in flow calculations, not to solve the Navier-Stokes equations everywhere outside the profile, some algorithms decompose the problem in two parts :
    Viscous flow inside the boundary layer, resolution with Prantl's equations
    Inviscid flow outside the boundary layer, resolution with Euler's equations
     
    DCockey likes this.
  12. Sailor Al
    Joined: Feb 2021
    Posts: 650
    Likes: 27, Points: 28
    Location: Sydney

    Sailor Al Senior Member

    To me, the "No-slip condition" is counterintuitive. I understand that internal friction, or viscosity, will slow the fluid towards the boundary, but does it really reduce to zero? I have seen videos of very viscous fluids being moved slowly showing this behaviour, but intuition tells me that lower viscosity fluids moving at higher speeds (below turbulence) will surely have a velocity at the boundary. I know Prandtl's 1904 lecture introduced this condition, and that it has become embedded in fluid dynamics, but it bothers me.
     
  13. Mikko Brummer
    Joined: May 2006
    Posts: 574
    Likes: 83, Points: 28, Legacy Rep: 258
    Location: Finland

    Mikko Brummer Senior Member

    Just look at the dust on your car hood.
     
    Alan Cattelliot likes this.
  14. Alan Cattelliot
    Joined: Jul 2021
    Posts: 505
    Likes: 209, Points: 43
    Location: La Rochelle (Fr)

    Alan Cattelliot Senior Member

    Perhaps this video will help you get the feeling of it. If you observe the flow near the wall, in the several experiments that was done, for instance @9'50'', or @13'10'' or @15'19'', you can almost see the velocity profile of the boundary layer, that you can compare with other classical diagrams. @15'52'', the flow stall in the diffuser is observed. Look at the flow going upstream in this region. This stall is only possible if the flow adheres to the wall, like an alpinist that can climb a wall only because of his crampons. This is wall friction.



    This wall friction can also be directly observed during a windy and rainy day, by looking at the water drops on the windows.
     
    mc_rash likes this.

  15. gonzo
    Joined: Aug 2002
    Posts: 16,790
    Likes: 1,714, Points: 123, Legacy Rep: 2031
    Location: Milwaukee, WI

    gonzo Senior Member

    It would be more accurate to state that the limit of the velocity function approaches zero.
     
    DCockey likes this.
Loading...
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.