hey /diy/ i'm trying to build an RC airplane, what's the best software for wing design, air flow and various other physics related parameters.
thank you
Microsoft Paint is a good starting point
>>999551
>>999547
anyone ?
>>999557
There has been a lot of "I want to build plane but know nothing." lately.
Got any previous experience in building model aircraft?
If you need to ask, I doubt you have the skills.
>>999569
Hopefully OP does know how to build a plane and is just looking for some modeling software.
You're probably right though. Designing an RC plane is much more complicated than a lot of people think, since it's, ya know, designing an aircraft.
>>999566
yes, but never used any software.
>>999569
why are you fags so crusty ?.
i don't need your idiotic opinion.
be helpful or fuck off.
>>999551
Stfu
>>999547
Inventor
>>999551
this
Ansys is good for fluid dynamics and aero. But it's about 30-40 grand for a licence.
>>999547
If you want to start performing some simple wing analysis for long wings, you can avoid the full-3D route and just perform some wing-level (i.e. a global view that ignores some quasy-2D phenomena) and aerofoil-level (i.e. wing sections) analysis and design.
>for dem wingies
Just use Prandtl's long wing theory (if the wing is straight) or perhaps a discrete Weissinger's method (roughly a vortex lattice with just one panel in the chordwise direction). That would give you reassonable estimations of lift and lift-related magnitudes, and induced drag and induced-drag-related magnitudes. As we are not in the 1940s anymore and you won't want to spend half an hour doing a simple computation by hand, you can code-it-yerself any of the methods in a breeze.
>for dem sections
You can use published aerofoil data, thin aerofoil theory (which gives you the local pitch moment, but no drag information), code your own panel method or conformal mapping method for full 2D potential analysis (for the same data as obtained through thing aerofoil theory), couple that potential analysis witjh boundary layer theory (plenty of methods here) for section drag data and even corrections to the potential lift curve slope and moment results, be a faggit and waste your time with a full Navier-Stokes CFD, or just l2use XFOIL for that interactive potential-boundary layer results you just want.
For more complex aerodynamic analysis, like including the fuselage and nacelles, you can code yourself a simple panel method (just go with a vortex lattice) or be a cool guy and get yourself a Hoerner to dig the sweet empirical drag data.
>>999547
>other physics
Your model plane is just a collection of slender beams and thing plates. Just compute it by hand or quickly code your own throwaway programs.
>>999547
>moar other physics: flight dynamics
Just to make sure your model is stable and won't fall on your head, assume small perturbations around a steady state (like flying horizontal) and linearise the shit out of the rigid body dynamics equations. Yes, rigid body dynamics: discard all aeroelastic phenomena for now. After that, just assume the state is a complex exponential law on the time. You will get a system of algebraic equations which is an eigenvalue problem on the complex time constant. Negative real parts are stable and possitive real parts are unstable and will kill you.
>>999547
>best software
As you just saw, you can code everything yourself. It's not like the maths about aircraft desig are a secret or something. The best software is the software you code to suit your specific needs.
If you're lazy as shit. Just use xflr5 or X-Plane
>>999911
lol
>>999578
>Designing an RC plane is much more complicated than a lot of people think, since it's, ya know, designing an aircraft.
Lolno
https://www.youtube.com/watch?v=-75v2G1o40U
>>999547
Ansys Fluent is fairly decent and comes with a geometry and mesh package as well.
Open foam works too but i think you need a separate meshing program for that one.
>>999547
If it's your first build, then use an existing design.
Also, with RC planes, the thrust:weight ratio is so high you could build a plane blind and it would still fly.
Plane with more holes then airframe yet still flies :
https://www.youtube.com/watch?v=DksEy8eWwno
>>999919
>xflr5
Pretty much made for this exact purpose, OP.
Granted, you're seriously overengineering this thing by model airplane standards if you're actually running a full-blown simulation of it. Model airplanes aren't exactly difficult to get working; TLAR principles and a very basic knowledge of aircraft controls, weight and balance is usually enough to get the job done. And as cheap as they are to build, design by trial and error/iteration is often a no-brainer as well.
THIS THREAD IS GOLD!!!
-Ansys
-OpenFOAM
-xflr5
thanks anons, now i know when i make a plane
https://www.youtube.com/watch?v=OBae-GV9KEA
>>999905
no shit all this was already scienced out like 75 years ago by NACA. go through their catalog of aerofoil profiles and pick the one you like.
>>1000623
I was talking about the strength of materials/structural mechanics part.
For aerodynamic performance of aerofoils at low Reynolds numbers, yes, OP might as well grab a copy of the Abbott or the original NACA Airfoil Data.
>>1000609
Good luck OP. The next step is getting the boundary layer right, so i suggest you search for the most appropriate turbulence model that suits your case.
Unless OP is building OPs photo, all this modeling won't mean a thing. You can make the most aerodynamically impaired aircraft out of foam, throw a powerful engine on it and it will fly fine.
>>1000627
>I was talking about the strength of materials/structural mechanics part.
Which is in some ways more important to the flight characteristics of a model than the aerodynamics itself. Heavy airplanes generally don't fly well. Light ones can, even if they're draggy.
>>1000831
Except that stability is a pretty big deal. If he doesn't know what he's doing it will crash with no survivors
>>1000843
Stability isn't hard to achieve, especially if you're going with a fairly conventional design.
Anyone who's spent time folding paper airplanes should have a good enough feel for stability to make it work.
Kerbal Space Program with ferram aerospace.
>>1000414
just checked it out, i didn't know softwares could be this expensive.
i could literally buy a fucking Mercedes for that price. or at least hire a team of fags do do the job for me.
>>1000610
so whats the conclusion of these two ?
>>1001375
Well yeah, if you fucking pay for it.
>>1001375
You shouldn't use a NS/RANS CFD for this. If the flow configuration is easy enough, and assuming you can build a correct mesh for capturing the relevant boundary layer phenomena (protip: you can't), simplified methods (even lifting line combined with 2D aerofoil data) will give you comparable or even better results with a tiny fraction of the effort. For complex flow configurations (separated flow), your CFD results will be so worthless you might as well check tea leaves or read your horoscope for better accuracy.
>sauce: I've seen my share of disastrous amateur and not so amateur CFD computations
>>1001375
Ansys is like a few $100/year for an educational license, but yeah a commercial one is terribly expensive
But OpenFoam is free no? I've only worked with Ansys so i'm not an expert on that program.
>>1001624
Pretty much this. It takes a very long time and devotion to obtain a plausible solution if you're using CFD.
>mfw trying to get a mesh that doesn't crash
If OP really wants to get into this i suggest he starts with the flow around a cilinder. There's plenty of documentation about that and getting a decent mesh is trivial, or can even be found online.
>>999773
No bingo
thread allowed.
