Activity for André

  • André André posted a comment on discussion Open Discussion

    Thanks, Still, this is sort of a rare bug. It usually occurs when the save process is interrupted - for instance the program is closed during the save process, or the disk is full during the save operation. The file ends up corrupted. To prevent this I'll force the save to a temporary file, and only if successful will overwrite the existing file. However, if the save process wasn't interrupted in your case, then it might be something more problematic. If you get the issue again, I'll be very interested...

  • André André modified ticket #103

    issue with local cl in VLM with winglets compared to AVL

  • André André posted a comment on ticket #103

    This seems to be related to the issue descripbed in ticket #102. There is something wrong with the calculation of tilted geometries. However, the main and recommended method is to run without the tilt activated. This is what AVL does. The tilt option was essentially an experimental option which I implemented to see whether there would be a benefit, but it turned out not to be the case. I'll restore it correctly nonetheless.

  • André André modified ticket #104

    Total loss of project which had been opened/saved multiple times

  • André André posted a comment on discussion Open Discussion

    Sorry to hear about that. Agreed, this is not acceptable and I'll try to do something about it in the next version. André

  • André André posted a comment on discussion Open Discussion

    What happens if you run inviscid analysis only ?

  • André André posted a comment on discussion Open Discussion

    Sure, please share them and I'll be glad to upload them to the website.

  • André André committed [r946]

    - Improved the display management of boundary layer graphs in XDirect

  • André André posted a comment on discussion Open Discussion

    One other reason for the results to be different is that the mesh is not the same in both cases. Have you ruled this out by refining both meshes to the same panel size?

  • André André posted a comment on discussion Open Discussion

    Not sure wwhat the question is, but the confusion might come from the development history. Operating points came first, using traditional wind axis. I implemented stability analysis later on, and introduced the stability axis. Agreed, fig. 43 might be confusing, but I'm not very good at 3d drawings in Word. To apply slideslip, the program rotates the whole geometry by beta (or -beta, need to check) around the body z-axis, then rotates the wind vector by -alpha.

  • André André modified a comment on discussion Open Discussion

    Did you install the Qt libs as part of the QtCreator package ? If you are using the system's libraries, check using your package manager, that the QtScript development lib is installed. It's called libqt5-qtscript-devel in OpenSuse, but the name will be different in Ubuntu I guess. Note that you need the -devel library to compile, not only the runtime library. Please share the error message if you need further help.

  • André André posted a comment on discussion Open Discussion

    Did you install the Qt libs as part of the QtCreator package ? If you are using the system's libraries, check using your package manager, that the QtScript developement lib is installed. It's called libqt5-qtscript-devel in OpenSuse, but the name will be different in Ubuntu I guess. Note that you need the -devel library to compile, not only the runtime library. Please share the error message if you need further help.

  • André André posted a comment on discussion Open Discussion

    On the matter of forced and free transitions, I tried to explain things here. https://www.youtube.com/watch?v=C9RDw9o-Oig There is also this highly instructive experiment by Matthieu Scherrer made back when he was a student. No better example of forced transition. http://scherrer.pagesperso-orange.fr/matthieu/aero/nimbus4e.html

  • André André posted a comment on discussion Open Discussion

    Lift and drag are in wind axis. In the guidelines, there is also Fig. 43 which describes the stability and body axis used in stability analysis. André

  • André André posted a comment on discussion Open Discussion

    Can you share the .dat file?

  • André André posted a comment on discussion Open Discussion

    Type 1 to 3 analysis are explained in the xfoil documentation. Re, Mach dependence ------------------- A few comments are in order on the TYPE command, which allows the user to set the dependence of the Mach and Reynolds numbers on CL. Any CL-CD polar can be of the following three types: Type parameters held constant varying fixed ---- ------------------------ ------- ----------- 1 M , Re .. lift chord, vel. 2 M sqrt(CL) , Re sqrt(CL) .. vel. chord, lift 3 M , Re CL .. chord lift , vel. * Type 1...

  • André André posted a comment on discussion Open Discussion

    I'll look into it. In the meantime, you can download the dll pack from the files section.

  • André André posted a comment on discussion Open Discussion

    First you need to compile the code, then look for the method buildInfluenceMatrix() in the panelanalysis class, and implement your own output.

  • André André posted a comment on discussion Open Discussion

    No the influence matrix is not provided in the output. You will need to get into the code to access it.

  • André André posted a comment on discussion Open Discussion

    Probably means that QtScript is not in your system's default libraries Try using QtCreateor. There are some compilation instructions in the File section.

  • André André committed [r945]

    - gui things

  • André André posted a comment on discussion Open Discussion

    Did you go through the tutorials? http://www.xflr5.com/docs/Point_Out_Of_Flight_Envelope.pdf and https://www.youtube.com/watch?v=_8thw2CQCaI&list=PLtl5ylS6jdP6uOxzSJKPnUsvMbkmalfKg&index=13&t=3s If you can't figure it out, please share the project file.

  • André André committed [r944]

    - corrected some memory leaks in foil modification routines

  • André André committed [r943]

    tagging version 6.38

  • André André modified ticket #102

    bug: panel method with activated tilt

  • André André posted a comment on discussion Open Discussion

    install "Visual C++ Redistributable for Visual Studio 2015" from this link https://www.microsoft.com/en-us/download/details.aspx?id=48145 or donwload the VC runtime libraries from the files section Any PC & graphics card which supports OpenGL 2.1+ will do. I guess that they all do these days.

  • André André posted a comment on discussion Open Discussion

    Just released v6.38 which fixes the random display bug in the 3D views.

  • André André committed [r942]

    Minor corrections to the 3d display

  • André André committed [r941]

    - Cleaning up the 3d widget classes

  • André André committed [r940]

    - fixed the random behaviour bug in the OpenGL view

  • André André modified a comment on discussion Open Discussion

    Hi everyone, There is some random behaviour in the OpenGL view which causes the 3d display to go black, or to display some weird stuff. Just to let you know that I'm on it, but OpenGL is a difficult world, so it may take a little time to fix, and your feedback will be useful. One thing to try if you face the issue: try to disable the light in the "view/3D light options" menu. Let me know. André

  • André André modified a comment on discussion Open Discussion

    Hi everyone, There is some random behaviour in the OpenGL view which causes the 3d display to go black, or to display some weird stuff. Just to let you know that I'm on it, but OpenGL is a difficult world, so it may take a little time to fix, and your feedback will be useful. One thing to try if you face the issue: try to disable the light in the "view/3D view preferences menu". Let me know. André

  • André André posted a comment on discussion Open Discussion

    Hi everyone, There is some random behaviour in the OpenGL view which causes the 3d display to go black, or to display some weird stuff. Just to let you know that I'm on it, but OpenGL is a difficult world, so it may take a little time to fix, and your feedback will be useful. One thing to try if you face the issue: try to disable the light in the <view 3D="" view="" preferences=""> menu. Let me know. André

  • André André posted a comment on discussion Open Discussion

    install "Visual C++ Redistributable for Visual Studio 2015" from this link https://www.microsoft.com/en-us/download/details.aspx?id=48145 or donwload the VC runtime libraries from the files section

  • André André committed [r939]

    tagging v6.37

  • André André posted a comment on discussion Open Discussion

    Hi everyone, This is a notification that xflr5 v6.37 has been released and is available for download. It fixes two very bad crash bugs that occured when trying to derotate a foil or set a flap. Some of you may still experience some issues with the 3d View. I have a hard time finding a default setting which will work on all Windows configurations. If you see a void 3D view, please go to the ?/OpenGL_info menu, and request a 2.1 context, which is the minimum needed by xflr5. The version of OpenGL detected...

  • André André committed [r938]

    - restored a missing image

  • André André committed [r937]

    - removed the default app view

  • André André committed [r936]

    - updating version 6.37

  • André André committed [r935]

    fixed two bad crash bugs when derotating a foil and when defining a flap

  • André André modified ticket #101

    wrong orientation of Lift in 3d

  • André André posted a comment on ticket #101

    I don't understand the problem. Maybe the issue comes from the fact that you are displaying a lateral mode? In the case of a type 5 polar with sideslip, the sign looks good to me.

  • André André modified ticket #100

    bug: lift calculation compare elliptic vs. bell

  • André André posted a comment on ticket #100

    To obtain the wing's lift, the integrand should be (CL x strip_area) and not just CL. If the chord length is uniform, then the strip_area can be moved out of the integrand, and you should find the expected ratio. Also, the lift is calculated as a discrete sum of strip_areas x the strip's lift, so there will be numerical errors.

  • André André modified ticket #98

    3D Modell can't show out

  • André André modified ticket #99

    Units problem in 6.32

  • André André modified ticket #99

    Units problem in 6.32

  • André André posted a comment on discussion Open Discussion

    Usually when you build a wing from balsa ribs for instance, ypou position them parallel to the fuselage axis. So I guess this convention makes sense.

  • André André committed [r934]

    Minor improvements

  • André André posted a comment on discussion Open Discussion

    Is it linked to one project file specifically? If so can you share it?

  • André André committed [r933]

  • André André posted a comment on discussion Open Discussion

    Good to know, Still, I'll implement a verification of the OpenGL version to prevent similar issues in the future. xflr5 requires OpenGL 2.1 or greater.

  • André André posted a comment on discussion Open Discussion

    Hi Jim, Just sent you a private message André

  • André André posted a comment on discussion Open Discussion

    just in 3D view globe/sphere is very dark (well, maybe it was so dark in earlier versions...nevermind) Turn the light on?

  • André André posted a comment on discussion Open Discussion

    Doesn't look good. I'd like to fix this, but I'll need a little more information: - Is the problem linked to a specific project file? If so can you share it? - Is it the same thing if you don't load a project file at all? - Is it any better if you restore the default settings in the options menu? Depending on the answers I may ask your help to run a debugging version on your machine.

  • André André posted a comment on discussion Open Discussion

    Hi everyone, Just released v6.35 which fixes a number of small issues with the GUI. As always, feel free to post back your issues, I'll do my best to fix them in a timely manner. André

  • André André posted a comment on discussion Open Discussion

    Just released v6.35. It's very similar to v6.34, so it may or may not change things. If it doesn't let me know, I may need your help to fix this issue. More specifically, is it only the 3d view which is blank or also the polar and operating point views ? If it's only the 3d view, can you try different contexts in the ?/OpenGl info menu, and see if it any of the configurations produces a display of the flying banner. Try also context 2.1, and please post back the content of the information window....

  • André André posted a comment on discussion Open Discussion

    Even if you define the elevator and fin, in the case of an LLT analysis, these will be ignored and the results wll be related to the wing only. In the case of LLT, the slideslip is ignored. Yes, the viscous part is included. It is calculated by interpolating the foil data using the local lift coefficient. Very clearly an approximation, at the limit of what can be asked of xflr5's simple model.

  • André André committed [r932]

    Minor gui corrections to the body edition interface

  • André André posted a comment on discussion Open Discussion

    Sent you a private message. In your spams maybe ?

  • André André committed [r931]

    - gui things

  • André André posted a comment on discussion Open Discussion

    Well isn't the stall point always for high aoa anyway? Bear in mind that xflr5 is less and less reliable as the aoa approaches the stall region. This is because the VLM is a linear/inviscid method, and does not predict stall. The stall behaviour is deduced from foil 2d data, and its transposition to 2D analysis is an approximation.

  • André André posted a comment on discussion Open Discussion

    Hi Israel, Unfortuantely, this kind of airfoil is probably beyond the capability of xflr5 to analyze correctly.

  • André André posted a comment on discussion Open Discussion

    Could you try something: there should be a file XFLR5.ini in C:\users\myuser\appdata\roaming Please make a copy of this file just in case it's the source of the problem, then delete it and try again to open a file. If it works please share the .ini file or paste it's content in this thread. If not I'll send you a PM to try something else. André

  • André André posted a comment on discussion Open Discussion

    If it's linked to a specific file, could you share it please.

  • André André posted a comment on discussion Open Discussion

    In such a case, I'd recommend to start with a conventional configuration, fore CoG, positive wing angle vs. elevator, negative slope of Cm=f(aoa), which gives the usual modes. Then try sensitivity analysis and change the values towards those of your configuration. I have encountered those strange modes too, but I don't if they have any physical meaning.

  • André André posted a comment on discussion Open Discussion

    Difficult to say.This could mean that the CoG is too much backwards.

  • André André committed [r930]

    Fixed a minor display issue.

  • André André posted a comment on discussion Open Discussion

    This is the implementation. void modeProperties(std::complex<double> lambda, double &omegaN, double &omega1, double &zeta) { omega1 = fabs(lambda.imag()); if(omega1 > PRECISION) { omegaN = sqrt(lambda.real()*lambda.real()+omega1*omega1); zeta = -lambda.real()/omegaN; } else { omegaN = 0.0; zeta = 0.0; } } I don't know how far it's reasonable to trust the results when they don't produce the traditional phugoid and short period modes.

  • André André posted a comment on discussion Open Discussion

    Current operating point / export Yes

  • André André committed [r929]

    - GUI improvements for foil modification in the xfoil analysis module

  • André André posted a comment on discussion Developers

    Sure new features are always interesting. Feel free to try to make it work, and if sucessful we can introduce it in the program for good. André

  • André André posted a comment on discussion Open Discussion

    I don't think xflr5 can answer these questions directly. It can just give you the lift and drag, from which you can deduce the power of level horizontal flight. Climb and descent will depend on how much extra power you add through engine throttle.

  • André André committed [r928]

    Fixed some font issues for MacOs

  • André André committed [r927]

    Minor improvements for macOS

  • André André posted a comment on discussion Open Discussion

    Is it just the program's window or the whole screen ? Is it the same issue as this one https://sourceforge.net/p/xflr5/tickets/98/ ? If so let me know if this version improves things http://www.xflr5.com/docs/xflr5_v6.34_win64.zip

  • André André committed [r926]

    Final mods for v6.15

  • André André modified a comment on discussion Open Discussion

    The main issue seems to be that neither the main wing nor the CoG are in the z=0 plane. When defining an analysis with custom CoG position, it would be better to set it in the plane z=55mm. In which case the curve Cm=f(aoa) is almost flat for NP=100.7mm. But I'd rather trust the result of a stability analysis (Type 7) which gives NP = 104 or this formula https://sourceforge.net/p/xflr5/tickets/49/ which gives NP 101 - displayed in the bottom left corner I don't think you can expect more precision...

  • André André posted a comment on discussion Open Discussion

    The main issue seems to be that neither the main wing nor the CoG are in the z=0 plane. When defining an analysis with custom CoG position, it would be better to set it is the plane z=55mm. In which case the curve Cm=f(aoa) is almost flat for NP=100.7mm. But I'd rather trust the result of a stability analysis (Type 7) which gives NP = 104 or this formula https://sourceforge.net/p/xflr5/tickets/49/ which gives NP 101 - displayed in the bottom left corner I don't think you can expect more precision...

  • André André posted a comment on discussion Open Discussion

    Well first, since the Cm value for the airfoil is calculated at the NP, the values are low, and strongly nonlinear vs aoa. When it comes to the wing, each span position of the wing sees a different total angle = aoa + twist + downwash. I don't think you should expect too strong a relationship between the two. What's more, in your model the NP is roughly at 0.059m and the CoG is at 0.05m from the leading edge. Since MAC = 0.2, this is still (0.059-0.05)/0.2 = +4.5% static margin.

  • André André posted a comment on discussion Open Discussion

    Submenu Current Polar / Export analysis to xml

  • André André posted a comment on discussion Open Discussion

    In the foil analysis, the Cm coefficient is calculated at the NP which is at 25% fo the coord's length. In a linear analysis it would be constant and independant of a.o.a. Using XFoil, the variations of Cm are due to viscous effects. In a wing analysis, Cm is calculated at the CoG.

  • André André committed [r925]

    - added an option to adjust the number of worke...

  • André André committed [r924]

    - Fixed memory leak and memory overflow bugs in...

  • André André committed [r923]

    Fixed some minor bugs

  • André André posted a comment on discussion Open Discussion

    There may be a problem with the batch analysis. If this is the case, please use single...

  • André André posted a comment on discussion Open Discussion

    There is no way to invert a foil in xflr5; you'll need to modify manually and reload...

  • André André posted a comment on discussion Open Discussion

    https://sourceforge.net/p/xflr5/tickets/98/

  • André André committed [r922]

    Tagging version 6.33

  • André André committed [r921]

  • André André committed [r920]

    Tagging version 6.33

  • André André posted a comment on discussion Open Discussion

    I'm not sure I understand the concern, but here are a few thoughts. The downwash...

  • André André posted a comment on discussion Open Discussion

    OK

  • André André posted a comment on discussion Open Discussion

    You can access the force values by exporting the graph data to a text file: context...

  • André André modified ticket #99

    Units problem in 6.32

  • André André posted a comment on ticket #99

    The calculation is performed at 60kt, it's the graph label which is wrong. It has...

  • André André modified a comment on discussion Open Discussion

    Well, the program stops at the first zero-moment angle it finds. I didn't imagine...

  • André André posted a comment on discussion Open Discussion

    Well, the program stops at the first zero-moment angle it finds. I didn't imagine...

  • André André posted a comment on discussion Open Discussion

    There is no reason for type 1 and 2 polars to be different with a zero deflection...

  • André André modified a comment on discussion Open Discussion

    There are two issues with the olw340 plane 1. the NP is at x= 0.365 and the CoG is...

1 >

Get latest updates about Open Source Projects, Conferences and News.

Sign up for the SourceForge newsletter:

JavaScript is required for this form.





No, thanks