Javafoil

Discussion in 'Software' started by dalebirrell, Feb 8, 2010.

  1. dalebirrell
    Joined: Nov 2009
    Posts: 25
    Likes: 1, Points: 0, Legacy Rep: 32
    Location: Strathclyde Uni

    dalebirrell Junior Member

    Wandering if anyone can help...

    I've been trying to use Javafoil to calculate lift and drag coefficients of a NACA 63-412 section and been having a bit of a mare.

    I've taken section data from Abbot and Von Doenhoff's 'Theory of Wing Sections' and Im trying to get Javafoils results to agree with that of the books tunnel test results. For a 0012 section, the results agree relatively well, but if you try the 63-412 section, the numbers aren't close and the results don't even make sense. (Apparently the foil will create positive lift coefficient at negative angles of attack.)

    Does anyone have any idea why Javafoil doesn't like this section?

    If you are a Javafoil user, you can try it in a matter of seconds. The 63-412 data can also be found here -
    http://www.teknologika.com/mothblog/hydrofoil-sections/
    you can just copy and paste.

    I've been trying with a Reynolds number of about 0.52x10^6


    Cheers
     
  2. wet feet
    Joined: Nov 2004
    Posts: 563
    Likes: 53, Points: 28, Legacy Rep: 124
    Location: East Anglia,England

    wet feet Senior Member

    I can't offer any suggestions about Javafoil,but if the endeavour works I should have attached a table generated in Profili.I expect that Xfoil which is the application doing the work in the background,would give the same results.I hope it helps.
     

    Attached Files:

  3. daiquiri
    Joined: May 2004
    Posts: 5,373
    Likes: 247, Points: 73, Legacy Rep: 3380
    Location: Italy (Garda Lake) and Croatia (Istria)

    daiquiri Engineering and Design

    Just a few observations, without re-running your analysis:
    1) It is very important to comprehend well the first 2 lines on this page: http://www.mh-aerotools.de/airfoils/jf_validation.htm
    2) In Abbott&Doenhoff, the Reynolds numbers start from 3*10^6, so be careful when comparing data for different Re's.
    3) Numerical methods are very sensible to No. of discretization points along your airfoil, try increasing the number of points and see how the results vary
    4) Different turbulent-transition and stall-prediction methods will result in different lift/drag curves. No magic wand there, only experience, knowledge and good literature can help choosing the appropriate one. This is particularily true for airfoils which have a "drag bucket" at low-AoA's, like NACA 6-series.
    5) Don't expect miracles near the stall region of the lift curve. No 2D method can predict the lift/drag curve near the stall in a satisfying manner, because stall is a 3D phenomenon. If a software is well done, it's numerical output should agree well with the experimental data (and hence can be assumed as reliable) up to 7-8° AoA (if it is not a thin airfoil).
     
  4. Leo Lazauskas
    Joined: Jan 2002
    Posts: 2,696
    Likes: 149, Points: 63, Legacy Rep: 2229
    Location: Adelaide, South Australia

    Leo Lazauskas Senior Member

    It makes perfect sense to me: cambered airfoils can have +ve lift at -ve AoA.

    Follow Daiquiri's advice, and keep trying!
    Leo.
     
  5. dalebirrell
    Joined: Nov 2009
    Posts: 25
    Likes: 1, Points: 0, Legacy Rep: 32
    Location: Strathclyde Uni

    dalebirrell Junior Member

    Thank you all for your help and advice.

    Perhaps I should give this 'Profili' software a try those results are ideal for what I'm doing, I've had a look at Xfoil but it was so user un-freindly I was scared off. Maybe I just need to stop being lazy and read the manual (frankly that scares me as well.)

    As far as Javafoil is concerned I'll try playing with the No. of discretization points, although it seems to be something more fundamental that's wrong.
    The Reynolds number i've been trying is lower than 'theory of wing sections' gives data for, but the results for the lift coefficients should not be drastically different, it's the drag coefficients when plotted against lift that will change more significantly. This is the reason i've been comparing Javafoil to the book - so I can produce lift to drag results that are reliable(ish).

    And finally although cambered foils will produce lift at neg angles, the lift still should vary almost linearly before stalling. Javafoil was producing a 'V' shaped graph where the lift at -8 deg was more than that produced at say +3 deg.

    Anyway, thanks again for all your input

    cheers

    Dale.
     
  6. daiquiri
    Joined: May 2004
    Posts: 5,373
    Likes: 247, Points: 73, Legacy Rep: 3380
    Location: Italy (Garda Lake) and Croatia (Istria)

    daiquiri Engineering and Design

    An open-source alternative to Profili (which is a commercial software) is XFLR5:
    http://xflr5.sourceforge.net/xflr5.htm
    It uses the Xfoil engine for numerical analysis and adds a graphical user interface - hence should be more user-friendly. It does have a learning curve, though...
    All the warnings and advices given for Javafoil are valid for Xfoil too, of course.
    Good luck! :)
     
    1 person likes this.
  7. dalebirrell
    Joined: Nov 2009
    Posts: 25
    Likes: 1, Points: 0, Legacy Rep: 32
    Location: Strathclyde Uni

    dalebirrell Junior Member

    thanks daiquiri!
     
  8. Guest625101138

    Guest625101138 Previous Member

    Dale
    I would be surprised if the Javafoil data was not close to any test data.

    I have attached some images and the data for your required foil at the specified Re#. This is one of the standard sections that Javafoil can generate so I used that rather than load your data.

    As Leo pointed out it is a cambered foil so will have lift at zero AofA.

    I will be interested to determine the cause of any differences if your test data does not agree with the attached. Note that this data is for an infinitely wide foil having no induced drag. You can set the aspect in the Options page if you want data for a foil of finite width. I have my own correction for induced drag and have never checked my method against JavaFoil. I have often checked the unbounded 2D data on mostly thin foils and it gives good agreement.

    Rick W
     

    Attached Files:

  9. Joo
    Joined: Apr 2010
    Posts: 3
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Bremen, GER

    Joo New Member

    I have a question concerning the software javafoil. Does the program only calculate a 2-dimensional flow or also a 3-dimensional flow? In other words: Is the induced drag included in the coefficient Cd? When calculating a NACA-foil with javafoil, the corresponding drag is always much higher than drag in the diagram of the book “Theory of wing sections” (Abbott & Doenhoff).

    Does anybody know an answer my questions?

    Thanking you in anticipation.
    Joo
     
  10. daiquiri
    Joined: May 2004
    Posts: 5,373
    Likes: 247, Points: 73, Legacy Rep: 3380
    Location: Italy (Garda Lake) and Croatia (Istria)

    daiquiri Engineering and Design

    Hello Joo,
    Javafoil is an applet for analysis and design of 2D airfoils only. There is a correction for finite aspect-ratio wings, but it is not a 3D analysis.
    It is all written at the Javafoil site, I'm citing it here:
    "If the aspect ratio is not zero, a correction is applied to lift and drag coefficients. These corrections are suitable for aspect ratios above 1.0 and try to simulate the effect of a finite wing. This is not a good replacement for an analysis of the three dimensional flow. The feature is intended to allow for the comparison of older wind tunnel tests, which have been performed with finite wings."

    I invite you to read the Javafoil manual before using the applet - the theoretical backgrounds (and hence limitations of the applet) are explaind pretty well, imho, and I believe one should try to understand them before using the software.

    As about your question about the drag curve produced by the Javafoil, it has been brought up in this discussion: http://www.boatdesign.net/forums/boat-design/naca-sections-32067-2.html - from the page 2 on. The conclusion was that Xfoil software performed much better for the analysis of the reference NACA0012 foil. But it has a disadvantage of not being able to handle multiple foils (which Javafoil does), you decide how important is that feature for you.
    Cheers!
     
  11. Joo
    Joined: Apr 2010
    Posts: 3
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Bremen, GER

    Joo New Member

    Hey Daiquiri!

    Thanks a lot for your help! Apart from CFD-Analyse is there any possibility to calculate accurately the lift and drag of 3-D-foil?

    Cheers!
     

  12. daiquiri
    Joined: May 2004
    Posts: 5,373
    Likes: 247, Points: 73, Legacy Rep: 3380
    Location: Italy (Garda Lake) and Croatia (Istria)

    daiquiri Engineering and Design

    Joo,
    an airfoil (or foil) is by definition a cross-section of a wing and is hence a 2D shape, with the relative 2D airflow (or waterflow) around. There is no such thing called 3D foil.
    However, an airflow around an infinite-span rectangular wing simulating an airfoil can be three-dimensional. It happens, for example, at high (near-stall) angles of attack. Is that what you intended? Or did you perhaps mean to say simply "wing" instead of "3D foil"?
     
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.