Flettner Rotors - Calculating Rotating Skin Friction

Discussion in 'Hydrodynamics and Aerodynamics' started by rwatson, Jun 16, 2018.

  1. rwatson
    Joined: Aug 2007
    Posts: 5,839
    Likes: 276, Points: 83, Legacy Rep: 1749
    Location: Tasmania,Australia

    rwatson Senior Member

    Dolfiman likes this.
  2. Dolfiman
    Joined: Aug 2017
    Posts: 1,022
    Likes: 433, Points: 83
    Location: NICE (France)

    Dolfiman Senior Member

    Thanks for this new paper, I read it rapidly, some comments :
    - It is about numerical experiences with various turbulence models and paying careful attention to grid meshing and numerical stability,
    - the great thing is that they summarized their results in formulations for Cl and Cd ( formulations (8) in the paper) with as input just 3 nondimensional data : Spin ratio (SR), the aspect ratio (AR) and diameter ratio of the upper plate (de/d), and giving all the needed coefficients in Table 6. From this, one can build a helpful spreasheet application for early stage project.
    - the frustrating thing is that nothing new about experimental results : they compare their results with experimental ones said given by Pearson (2014), but actually this paper refers itself to experimental works done by Jakob Ackeret 1932 ! (Fig. 3 of Pearson paper). Cascade of references in the recent papers usually fall to same old experimental (public) data.
    - And also nothing about the energy input needed to rotate the cylinders, not even mentioned.
    >>> Again, playing with sophisticated CFD tools is great but there is also a crucial need of new experiments aiming a full coverage of the issue (inc. the resistant torque) and with the aim to publish the full results, an opportunity for a University.
     
    rwatson, CocoonCruisers and Doug Lord like this.
  3. rwatson
    Joined: Aug 2007
    Posts: 5,839
    Likes: 276, Points: 83, Legacy Rep: 1749
    Location: Tasmania,Australia

    rwatson Senior Member

    And yet they said they referenced data from recent installations like the Enercon Ship ?
     
  4. Dolfiman
    Joined: Aug 2017
    Posts: 1,022
    Likes: 433, Points: 83
    Location: NICE (France)

    Dolfiman Senior Member

    To my knowledge, inthe Enercon document they referenced or others similar, there is no info on CL, CD in relation with spin ratio, aspect ratio, de/d, .... than can be compared with the numerical tests of Di Marco and al., just global info on the installation and energy saved. I suppose all these data are part of their their know how and they don't want to make that public. It is why I advocate a University investigation on that issue.
     
    rwatson likes this.

  5. Dolfiman
    Joined: Aug 2017
    Posts: 1,022
    Likes: 433, Points: 83
    Location: NICE (France)

    Dolfiman Senior Member

    I looked again to this paper and developed an application (here attached) with using their formulations for Cl and Cd ( formulation (8) and coefficients aijk and bijk in Table 6 in the paper), with :
    - as input 2 nondimensional data : the aspect ratio (AR) and diameter ratio of the upper plate (De/D)
    - computation with the spin ratio SR = 1 to 3
    - as output : CL, CD
    Some comments :
    - I first checked the formulation (8) with their numerical results : when De/D is 2 and over, there is a lot of discrepancies, so I suggest a use limited to :
    --- when De/D = 1 (no end plate ) >> AR = 2 to 8
    --- when De/D = 1 to 1,5 >>> AR = 2 to 6
    --- numerically hasardous to use it for De/D > 1,5
    - I then checked the CL, CD results with some available test results : Ackeret with AR = 5 and De/D = 1,5 (as reported by Pearson 2014) , Naca Reid tests (approached with AR = 8 and De/D =1) >>> the good news is that the CL and CD curves have similar shape and that introduction of same correction factors (1/1,45 for CL; 1/2 for CD) can fit with the results for both cases. So I adopt these corrections in the application >>> comparison attached with using the corrections.

    To make complete the application, I also compute CP cyl. and CP plate, the adimensional coefficients for the power to rotate (aero part) the cylinder and the end plate if any, as investigated previoulsy in this thread (pdf attached as a remind), by introducing these definitions :
    CP cyl = Power /(Cf ½ Rho Swcyl U^3)
    CP plate= Power/(Cf ½ Rho Swplate U^3)
    And idem, I introduce a correction factor (x 1,20) for CP cyl to fit with the (only one) test results available (comparison attached).
     

    Attached Files:

    rwatson likes this.
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.