Hydrofoil exercise to validate CFD analysis

Discussion in 'Hydrodynamics and Aerodynamics' started by quequen, Oct 1, 2014.

  1. quequen
    Joined: Jul 2009
    Posts: 370
    Likes: 15, Points: 28, Legacy Rep: 199
    Location: argentina

    quequen Senior Member

    Fortunately! great contribution!
    I've tryied all the standard Global Goals to capture drag and still can't get it. Are you using Ecuation Goals?

    Here are my last results with a 1000000 grid (almost my limit). Drag is completely bad, as usual. Better topside pressures are at the hinge, the place where Oracle paid special attention.

    Velocity: 8m/s
    AoA: 4º
    Leeway: 0º
    Total Lift: 1434 N
    Total Lateral Force: 1080 N
    Vorticity: 2%
     

    Attached Files:

  2. quequen
    Joined: Jul 2009
    Posts: 370
    Likes: 15, Points: 28, Legacy Rep: 199
    Location: argentina

    quequen Senior Member

    I can make the hulls, but need linesplans or some hints for the foils.
     
  3. Mikko Brummer
    Joined: May 2006
    Posts: 574
    Likes: 83, Points: 28, Legacy Rep: 258
    Location: Finland

    Mikko Brummer Senior Member

    Here you go Quequen, lines & foils plans and full scale measurements as well, for validation work.

    I will get back to your Flowsin analysis later with more time - I think you are not doing the meshing the best/economical way - I always use a local initial mesh, with plenty of refinement on partial cells in particular. Also, the goals are really for convergence, although you can read forces from them too. I use surface parameters for reading the results, there you can look at forces on individual faces or the whole thing, tick from the parameters what you want to see.
     

    Attached Files:

  4. PatHanley
    Joined: Oct 2014
    Posts: 7
    Likes: 1, Points: 0, Legacy Rep: 23
    Location: United States

    PatHanley Junior Member

    Attached Files:

  5. daiquiri
    Joined: May 2004
    Posts: 5,371
    Likes: 258, Points: 93, Legacy Rep: 3380
    Location: Italy (Garda Lake) and Croatia (Istria)

    daiquiri Engineering and Design

    So we are all in the ballpark.
    My XFLR5 results (preliminary - don't have time to refine the model):
    Lift (vertical) = 1691 N
    Lift (lateral) = 1640 N
    Drag (total) = 132 N​

    Mikko's CFD:
    Lift (vertical) = 1643 N
    Lift (lateral) = 1623 N
    Drag (total) = 153 N​

    The difference in the lift value data is within 7%, and the drag differs by 13%, but only because of my imperfect model.

    Guess that we can conclude that VLM tools are very suitable and competitive to N-S solvers for this type of problems. The required time for performing the calcs is measured in seconds, compared to N-S solvers which require hours.

    Thanks to Quequen for allowing us to perform this comparison. :)
     
  6. Joakim
    Joined: Apr 2004
    Posts: 892
    Likes: 53, Points: 28, Legacy Rep: 422
    Location: Finland

    Joakim Senior Member

    How did you do it with XFLR5? I did a quick try, but had no success defining such a "wing". I have used XFLR5 previously only for 2D foils, thus I have no experience making 3D shapes with it.

    Why is quequen still getting completely wrong drag? It seems there is no pressure drag. Is the domain way too narrow? Actually looking closer to the table he shows, the Z (drag) force is very close to zero, but there is -41 N frictional force. Is the pressure force in the other direction (about +42 N) or is the Z force without friction (=only pressure)? Positive Z is to the bow.
     
  7. markdrela
    Joined: Jun 2004
    Posts: 307
    Likes: 30, Points: 28, Legacy Rep: 324
    Location: MIT Aero & Astro

    markdrela Senior Member


    The Z axis is not aligned with the freestream here, so the Z force is not the drag, and the Y force is not quite the lift. These are computed by

    L = Fy cos a + Fz sin a
    D = Fy sin a - Fz cos a

    With Fy = 1433.96, Fz = 0.65, and a = 4 deg, I get

    L = 1433.3
    D = 99.37
    L/D = 14.42

    Edited ... I misunderstood the printout of the forces. Now assuming that Fz_friction is already included in Fz.
     
    Last edited: Oct 10, 2014
  8. Mikko Brummer
    Joined: May 2006
    Posts: 574
    Likes: 83, Points: 28, Legacy Rep: 258
    Location: Finland

    Mikko Brummer Senior Member

    Looking now at your latest attachments, I see that your foil is vertical and the flow is not in Z-axis (like Mark Drela notes). That's not a good idea, combined with the minimal domain, your foil is like working inside a tube.

    Turn the foil at 4 degrees, set the flow in Z-direction (-4 m/s), and enlarge the domain maybe 1,5 times, and don't use the automatic evenly spaced mesh distribution, but rather tweak it so that the cells are much larger at the domain walls and decrease towards the foil. At the foil, use a refined local initial mesh, and you will get good results with less than 1 million cells.
     
  9. gigi
    Joined: Feb 2013
    Posts: 7
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Croatia

    gigi Junior Member


    Please tell us more, what "unorthodox tweaking" have you done in XFLR?
     
  10. quequen
    Joined: Jul 2009
    Posts: 370
    Likes: 15, Points: 28, Legacy Rep: 199
    Location: argentina

    quequen Senior Member

    Prof. Drela, Mikko, thanks for the hints!. I'll try to fix them on my next attempt.

    I finished a set of foils for the Vendor2, they look good but still don't know how to solve the flap's hinge: details from the document seems to be pretty poor hydrodynamically, mainly the low side. Are this things really done on that way? A detailed picture would be helpful.

    I'm going for the hull, if nobody has done it yet...

    -
     
  11. gigi
    Joined: Feb 2013
    Posts: 7
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Croatia

    gigi Junior Member

    Interesting so far.
    To join the party, I analyzed this case with free 3D panel code APAME.

    APAME is free, licensed under GPL
    I have started very recently to use this program so I am still in learning faze. Result are somewhat different compared to what other people get.


    V 8 m/s
    rho 999
    Alpha 4
    Side slip 0

    Drag force 1.7395110e+001
    Side force -1.9365950e+003
    Lift force 1.9615790e+003
     

    Attached Files:

  12. daiquiri
    Joined: May 2004
    Posts: 5,371
    Likes: 258, Points: 93, Legacy Rep: 3380
    Location: Italy (Garda Lake) and Croatia (Istria)

    daiquiri Engineering and Design

    Sorry for the late reply - yesterday I got back from a trip, during which I didn't have time to write.

    XFLR5 can only model symmetrical wings, that is true. But it can model asymmetrical fins and calculate the aerodynamic forces created by them. ;)

    So I have done the following steps:
    1) create an airfoil similar to H-105 (I didn't have the coordinates back then)
    2) perform a direct airfoil analysis for Reynolds numbers of interest
    3) create a new plane made of:
    - a very small wing (b=0.05 m, c=0.1 m), placed sufficiently far from the origin of axis. For example, in this case you can place it at z=-10 m.
    - a fin, which is the L-foil we want to analyse, placed at the origin of axis.​
    4) perform the analysis of the plane for a range of AoA's. Be aware of the orientation of the fin in the global coordinate system. In my case, the fin was upside-down, so that negative AoAs corresponded to positive pitch attitude. For convenience of extrapolation of numerical values of lift and drag forces, the starting AoA in the range should be the AoA of particular interest (in this case -4°, please note the negative value).
    5) plot the Fx. Fy, Fz forces vs. Alpha
    6) right-click on the desired graph and select the "Current graph"-->"Define Graph Settings"-->"Scales" option and read the maximum value of the force.
    7) be aware that in the coordinate system of the fin Fx is the drag, Fy is the sideways lift, Fz is the vertical lift.​

    In the step n.3 it was necessary to create a small fictitious wing because XFLR5 won't do the analysis if it doesn't see a wing in the model. The wing has to be very small (compared to the fin) and placed far away from the fin in order to not perturb the flow field and to not create a significant lift and drag forces. For this reason, and since XFLR5 calculates the reference area from the wing geometry (and not from the fin geometry), the aerodynamic coefficients will be wrong - though the global forces will be correct. So you'll have to manually calculate the aerodynamic coefficients, if you need them.

    That's it, more or less. :)
    Cheers
     
  13. quequen
    Joined: Jul 2009
    Posts: 370
    Likes: 15, Points: 28, Legacy Rep: 199
    Location: argentina

    quequen Senior Member

    daiquiri, thanks for the smart xflr5 trick.

    gigi, thanks for the interesting Apame code, two features caught my attention:
    1- it has the ability to import meshes
    2- there's a Pulki II in the frontpage!

    -
     
  14. gigi
    Joined: Feb 2013
    Posts: 7
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Croatia

    gigi Junior Member

    Great daiquiri, thank you for this trick!
     

  15. quequen
    Joined: Jul 2009
    Posts: 370
    Likes: 15, Points: 28, Legacy Rep: 199
    Location: argentina

    quequen Senior Member

    This one is a bit closer to the ballpark. Now the flow is parallel to Z and the Comp Domain is bigger.

    H105 - v7

    Velocity: 8 m/s
    AoA: 4º
    Leeway: 0º
    Turbulence intensity: 1%
    Grid: 1.35 millon cells (using nested local meshes)
    External analysis, no added boundary conditions (no mirroring)


    Lift: 1399 N
    Drag: 152.68 N
    L/D: 9.16
    Lateral force: 1098.18 N

    Total Drag = Friction (73.589 N) + Induced? (79.094 N) = 152.683 N

    -
     

    Attached Files:

Loading...
Similar Threads
  1. Maarten88
    Replies:
    5
    Views:
    1,003
  2. S V
    Replies:
    0
    Views:
    1,876
  3. Eytan Levi
    Replies:
    0
    Views:
    1,688
  4. container
    Replies:
    7
    Views:
    4,086
  5. vejas
    Replies:
    34
    Views:
    11,952
  6. revintage
    Replies:
    8
    Views:
    2,605
  7. revintage
    Replies:
    14
    Views:
    17,342
  8. rallyhybrid
    Replies:
    39
    Views:
    6,411
  9. optiwings
    Replies:
    1
    Views:
    2,485
  10. Tommifin
    Replies:
    5
    Views:
    3,103
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.