Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Aerospace Flight simulator modelling

  1. May 6, 2009 #1
    Hello all.
    I'm modelling several new aircraft as mods for a flight sim.
    These mods are all WW2 birds like the Kittyhawk and Ta152.

    The generic computer models have a series of terms I am unsure about. So far I've been using existing aircraft modelling as a guideline but as we are expanding our mods to include types which are distinctly different (a short tailed Kittyhawk and a long tailed one for example, which should have distinctive handling differences)...

    critical Cy
    Cy max
    Vy factor
    Cx curvature
    Cx straightness
    K max
    Spin Cx loss
    Spin Cy loss

    As aeronautical engineers were involved in development of the game-engine it naturally follows some or all of these terms might be generic engineering terms for the field.
    Any light which could be shed on them, or opinions as to what they might represent would be most welcomed.
    I'm more of an engines man myself (the engine models I had no problem figuring out).

    edit, got more questions now. What is moment arm?
    Last edited: May 6, 2009
  2. jcsd
  3. May 6, 2009 #2


    User Avatar
    Science Advisor

    I'm not real sure about those, but are you just arbitrarily defining how these different models perform? Is there any sort of quantitative data that you'll be basing it upon?
  4. May 6, 2009 #3
    Well it's more of a revision of existing models in new slots and expanded for more selectable choices. For example there's a P-40E (short tail) and a P-40M (long tail) already modelled. I cloned these as new selectable choices marked Kittyhawk MkIa and Kittyhawk MkIII and then went about some revision based on RAAF, USAAF and Allison documentation.

    The weights and measures were off. The throttle calibration was entirely out, throttle heights were way too high, there was no pilot control for the mixture, loadouts were wrong, liquids capacities off and engine management needed to be enhanced. And I tried to demonstrate sharply the difference between the F3R and F20R motors.

    Well the new problem currently is all the values for things like Cy and Cx are identical between the two stock models...one being a short tail Kittyhawk and the other a long tail.
    A few figures ought to be really different. So I need some guidence to figure out which ones, since trial and error is very long winded and not entirely reliable. But you can use trial and error to a very limited degree to figure out how much to alter a given value (so long as you know what it is you're altering)...because you can use flight test data as a guideline (ie. play with the value singularly until you match performance data, assuming other values are pretty close to spot on).

    Of course I could just leave such things to aeronautical engineers. But that's funny actually, because if you go to one of the mod sites, they want to you leave it all to the site mod team (invariably bored middle aged bartenders in between home handyman jobs for the wife), whilst if you go to an aviation enthusiast site well they'd just rather you leave it all to professional pilot enthusiasts to model aircraft for sims. Basically wherever you go someone would rather someone else be better than someone and leave it at that, which is weird because it's sort of like holding a head underwater whilst you suggest, breathe, breathe. But aside from the fact I do indeed believe humanity is merely to justify vicious predators and call them pretty I would really like to do these mods.

    Any help would be greatly appreciated.
  5. May 6, 2009 #4
    Mmmm I do flight simulation and these don't seem right at all.... basically, I think this sim is a bunch of BS (sorry, it is). What your flying in it isnt a real flight dynamics model, its nothing more than a video game. That being said, I wouldn't even worry about trying to change how the airplane flies because I promise you none of the airplanes you have in that sim fly right to begin with based on what you just showed me.
    Last edited: May 6, 2009
  6. May 6, 2009 #5
    You know I have this engine modelling software, Engine Analyzer Pro. Real expensive but a couple of part time race engineers I know swear by it (by race engineer I'm talking about building open wheelers though more than say a member of the NASA administration).
    Well it's a good sim, models each individual stroke, every kind of coefficient you can think of, valve timing, tuned exhaust lengths, carby type, supercharger details, it's really a very good sim and undoubtedly saves dollars in trial and error for component matching in club racing and that sort of thing. Saved me several hundred dollars on my street car and correctly modelled conservative, but significant improvements for my brother in laws 5 litre and 2.8 litre Mercedes.

    I think in another thread you've been talking about some very expensive aerospace engineering software. Programmable laser cutters? I'm sure airframe and aerofoil modelling programs in addition to those working with tools.

    So I do feel obliged to point out this sim retails for around $35 and unfortunately is not the combination between my $700 engine modelling software and your $10K+ aerospace development tools.

    I probably wouldn't trust it to plot a backup landing window for the Space Shuttle, no :D

    Do you have any idea what the terms described above could relate to? please I do assure you I'd include a disclaimer for the accuracy of its physics far before I'd make a ridiculous claim, based upon your post Cyrus, since I'm not a liar. But our team would like to do the best we can, recognising it is a pastime and may be silly. Like women and anything written by hippies.
    Last edited: May 6, 2009
  7. May 6, 2009 #6
    This is great. It also has nothing to do with a nonlinear flight simulation routine - sorry.

    That thread is a totally different beast from what you are concerned with. Just ignore it, becuase it does not apply to you.

    Again, this price point you have made up really has nothing to do with what I'm talking about. I'm not paying for the simulator code, I'm writing it (for NASA). But I do have some code (from NASA) which is also NOT 10k, not even close.

    No, becuase those values are crap-o-la. In other words, if you want a real simulator. Meaning, a real deal flight simulator, you need lots of wind tunnel data on your aircraft of interest. Not a few values that you listed, which again are bogus (I have a vauge idea what they are, but if I am right they are definately NOT what you need).

    To give you an idea of what your simulator needs to do, you need to understand these notes and implement them. If you dont, you dont have a simulator, you have a video game.

    http://www.princeton.edu/~stengel/MAE331Lecture8-9.pdf [Broken]
    Last edited by a moderator: May 4, 2017
  8. May 6, 2009 #7
    Sorry mate, honestly not trying to offend your sensibilities.

    Let me start again. What we're trying to do is simply introduce an enhanced atmosphere to a video game.

    In trying to work with the game-engine I came across a series of values and I'm trying to work with those to change the handling and feel of the arcade models in the game.

    I've read somewhere once before that Cy is a term used to describe a coefficient of drift? Cx drag? Cz lift?

    In this terribly simplistic game modelling, I was simply wondering how to use these figures to change the feel for the game player.

    I apolegise that I'm simply not in a position to reprogram the game engine to become a more effective simulator, or to develop a flight simulator from scratch.

    You have already been most helpful towards a much improved personal understanding and for this I sincerely thank you. But it still doesn't help me with our modelling project for our video game.

    To try a new tact, perhaps I can adapt real world physics into unrealistic game modelling terms which in no way reflect real world physics but may nevertheless help us provide an improved model for the Kittyhawks in the game.

    For the Kittyhawk MkI a short tailed fuselage was used.
    In the Kittyhawk MkIII a long tail fuselage was implemented (from the F-5 and K-5 series production variants onwards), in order to deal with higher torque ratings due to constant revision of the take off settings (beginning around 46"Hg MAP and winding up anything between 60-70"Hg in the field, ie. leaping from ca. 1150hp to 1780hp at sea level, though 56"Hg wep is an average returning around 1470hp...still quite a jump).

    So for the real deal, what sort of terms are used to describe the increase of a fuselage tail area in order to deal with engine torque at take off?
    I should start there perhaps, I am coming in cold but can pick up things quickly (am self taught in most interests). I will get lost in complicated formula which would assume an industry background however.

    I guess I should just forget the game for the moment and concentrate on understanding the real thing first.
    Am downloading the pdf links as I write, any other links or info to help is welcomed.

    Thanks again, and for the replies and positive assistance.
  9. May 6, 2009 #8
    Here is an example of part of the airframe modelling in the game (values are metric). The engine models are separate (I have no problem dealing with those).

    Type 1
    Crew 1
    Wingspan 11.36
    Length 9.5
    Seaplane 0
    Canard 0
    Jet 0
    JetHiV 0
    Empty 2821.0
    TakeOff 3936.0
    Oil 55.0
    Fuel 515.0
    Wing 21.92
    Aileron 1.45
    Flap 2.76
    Stabilizer 2.80
    Elevator 1.60
    Keel 0.9
    Rudder 1.4
    Wing_In 4.96
    Wing_Mid 4.0
    Wing_Out 2.0
    AirbrakeCxS 0.0
    Aileron 4.16
    Flap 1.67
    Stabilizer 4.95
    Keel 4.99
    Elevator 5.70
    Rudder 5.85
    Wing_In 1.65
    Wing_Mid 4.16
    Wing_Out 5.41
    Wing_V 2.5
    GCenter 0.12
    GCenterZ 0.00
    GC_AOA_Shift 0.45
    GC_Flaps_Shift 0.15
    GC_Gear_Shift 0.10
    CriticalAOA 16.00
    CriticalCy 01.52
    CxCurvature 00.70
    CxStraightness 00.15
    CriticalAOAFlap 18.00
    CriticalCyFlap 01.80
    SpinTailAlpha 17.00
    SpinCxLoss 0.03
    SpinCyLoss 0.015
    Vyfac 1.0
    Tfac 1.03
    Vmin 165.0
    Vmax 525.0
    VmaxAllowed 760.0
    VmaxH 595.0
    HofVmax 3800.0
    VminFLAPS 130.0
    VmaxFLAPS 250.0
    Vz_climb 13.0
    V_climb 280.0
    T_turn 19.5
    V_turn 340.0
    K_max 13.5
    Cy0_max 0.15
    FlapsMult 1.0
    FlapsAngSh 10.0
    Range 1080.0
    CruiseSpeed 310.0
    SensYaw 0.45
    SensPitch 0.67
    SensRoll 0.35
    lineCyCoeff 0.097
    AOAMinCx_Shift 0.0
    Cy0_0 0.14
    AOACritH_0 16.0
    AOACritL_0 -15.0
    CyCritH_0 1.31
    CyCritL_0 -0.7
    CxMin_0 0.0240
    parabCxCoeff_0 4.2E-4
    Cy0_1 0.87
    AOACritH_1 15.0
    AOACritL_1 -22.0
    CyCritH_1 1.82
    CyCritL_1 -0.7
    CxMin_1 0.072393276
    parabCxCoeff_1 5.8E-4
    parabAngle 5.0
    Decline 0.006
    maxDistAng 40.0
    draw_graphs 0
  10. May 7, 2009 #9


    User Avatar
    Science Advisor

    Perhaps the best bet is to get with the developer of the game engine and see if you can get some documentation.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook