It's effectively setting the angle of the wing with reference to the fuselage. This value is not represented in the Cl vs AoA table at all.

That’s 'wing incidence', No? :) ...a constant with no relation to AOA or drag...and set elsewhere.

It seems to muddle up wing incidence and aoa...and should be read from table 404CLvsAlpha IMO.

Induced drag is determined by table 404, it’s a product of lift.The AOA of min induced drag is found where the CL curve crosses the x-axis.Tip for 'Fuselage AoA at Min Induced Drag' From AAM...

Important for correct Induced Drag!Set to (AoAw@Cl=0)[TBL404] - (AoI+Twist/2)[TBL1204] typCd=Cdo + [(AoA+AoAmd)*(dCl/dAoA)Cl=0]^2 * Induced_Drag_ConstantdCl/dAoA is 'average Cl vs AoA slope' in TBL 404. Gritsevsky/Pierson

The above mentions [TBL1204].Table 1204 has been omitted from all but one FSX aircraft.This data is now in the config.Including 'Induced Drag Constant' which I think is now 'induced_drag_scalar'It's not clear to me Offset 80 has much to do with Induced Drag now.I've tested values -3,-1,3...and there is no effect on Cruise AOA, Cruise Pitch, Or Top speed.Maybe Offset 80 doesn't work anymore (?)

it’s a product of lift.

Ok... let's dispense with the 'lectures'. That's offensive and condescending. I'm well versed enough on the theories of aerodynamics for this discussion and I've assumed you are as well.Wing incidence is the angle of the root of the wing versus the fuselage, not the entire wing. Wing twist can change the incidence along the length of the wing.Having made commercial FDEs, and having worked in both FS9 and FSX I can assure you that the value is indeed used, indeed does have an impact on how the aircraft behaves in flight. Since all my work is commercial in nature, my own knowledge regarding air file contents won't extend past what is already publicly available. Hopefully that's understandable.

:( With respect...no offense is intended...I’m being sincere.I’m not interested in changing your point of view.And your insight is appreciated.It’s not a lecture, the detail is there to push forward the discussion of tbl1101 and offest80 specifically.Here’s my real agenda :)Tom Goodrick writing about FS98

There are conventional aeronautical engineering terms that could be applied to thisprogram - side force coefficient, roll moment coefficient, yaw moment coefficientand various damping derivatives. For some reason the authors of FS chose toinvent their own terms to patch up program performance.

The airfile has changed from a game programmer’s interpretation of aeronautics (FS98) to a very conventional textbook description of aeronautics (ESP). It’s really important to identify any legacy stuff we can.

Lifting line theory says the induced drag coefficient for a twisted wing is given by:CD_i = [CL²*(1+K₁) - K₂*CL*CL_a*T + K₃ (CL_a*T)²]/(A*pi)where CL is lift-coefficient, K_1, K₂, and K₃ are constants, CL_a is lift-curve slope, T is twist, A is aspect ratio, and pi is 3.142...The first term is the usual one, the other two only apply to a twisted wing:CD_i is minimised wheh dCD_i/dCL = 0 because all the other variables are fixed for a given aircraft .dCD_i/dCL = [2*CL*(1+K₁) - K₂*CL_a*T](/A*pi)This equals zero when:CL = K₂*CL_a*T/(2*(1+K₁)as CL ~ CL_a*a then angle of attack at minmum CD_i is:a = K₂ *T/(2*(1+K₁) and is directly proportional to twist. The formulae relate to the wing so if this is the basis of Offset 80 - Fuselage AoA at Min Induced Drag it would have to be corrected by the wing_incidence from the .cfg file. the values of a the second term in the expression can be calculated from this. The third term however seems to have been ignored.Interesting though this may be, the only way I know to determine K₁ and K₂ is by calculating the coeeficients of a Fourier series and summing them. I don't kow a way to determine them directly and simply from the aircrafty parameters. Anyone any ideas? EDITThe default Grumman Goose has Tables 1539-1545. The only non-zero enries in TRable 1101 are Airfarme Type and Airspeed Weight Factor - whatever they may be.

:( Don’t take my word for it, in my very limited experience I haven’t dealt with wing twist.My impression is induced drag in FSX is not very elaborate.As you know we don’t set the drag curve, FSX calculates it from lift. How I don't know.I suspect this stripped down version of the formula above might apply too.Cdi=Cl^2/pieAeCl=coefficient of LiftA=aspect ratioe= oswald efficiencyThe implication would be ‘oswald efficiency’ (Kd?) and 'induced_drag_scalar' are the parameters for tuning CDi.Aircraft.cfg does have a wing twist parameter. (among other things)Does it use it? Did it use it? Does it still use it? I don’t know.My impression is wing Lift and Drag are simplified in FS.And the effects of wing twist may be achieved by tuning the stability coefficients.Particularly those tables that factor in angle of attack. (451, 456, 460, 464, 1535, 1536, 1537, 1538)It’s interesting you mention the Grumman Goose, because it’s the only aircraft with TBL1204-wing stuff.All the other aircraft have moved this data to the config. (...Beatle’s post above)Maybe it uses an old approach for the wing and an ESP approach for basic Aerodynamics.(?)Using AirUpdate.exe it wouldn’t take long to add 1539-1545 and delete 1101…just to see.:( BTW I googled your formula...I'm not sure how these apply...they might help. κD = planform contribution to the induced drag factorκDL = lift-twist contribution to the induced drag factorκD_ = twist contribution to the induced drag factorκL = lift slope factor

FS doesn't go nearly that complex in calculating lift... let alone drag. The two of you are 'over-thinking' quite a bit.

I don’t think so.Cdi=Cl^2/pieAe seems pretty standard.I don't think FS needs to be more complex than that...Ed, does wing twist still work in FSX?Reading this thread it sounds like Offset 80 might somehow account for wing twist.I recall reading it doesn't...but I can't be sure.Do you know?By the way, I replaced my default Grumman Goose airfile with a custom one.That's the one I was looking at when I commented above.So no it doesn’t have table 1204, as I said it did, and it doesn’t have table 404 either.I think Grumman Goose could be made to the ESP conventions. :smile:Danny

Yes, the formula is standard... however, I don't believe FS uses it. In my own code I don't use that formula for drag nor do I use it's counterpart for calculating Cl. Oh, and just a side note... lift is for the aircraft, not the wing in FS, in other words lift isn't being calculated for the wing but rather for the entire aircraft. I don't recall which version of FS changed this, but if I recall correctly it was reported by Ron Freimuth.I believe that's why wing twist was reported as no longer 'functional'. However, I will warn you that many things have been reported as 'non-functional' with regards to flight dynamics in FS... and many of them were indeed functional.Just an FYI: ESP and FSX are one and the same. You are thinking they are somehow different. They are not.

Thanks, I did a quick diff on all ESP aircraft.asm samples, to confirm the aerodynamics sections were consistent across all three.They are, so these aircraft don’t pick and choose from a list of supported parameters.For aerodynamics aircraft all use a standard set of concise coefficients.It appears this does not include table 1101 anymore. (BTW, the only differences were in the choice of engine tables...as you'd expect)There’s an unlikely chance that Asm2Air.exe automatically generates 1101. Also FS may generate it too.I understand to improve aircraft compatibility…missing airfile data may be approximated by the sim on load.(AIRCRAFT SIMULATION TECHNIQUES… (FS2004), M Zyskowski ). ED, I think you misunderstood what I meant by ‘ESP conventions’ I’m referring to how ESP presents the airfile data in the SDK.The idea is it applies to FSX as well…because they are the same.

It's my understanding that what you're seeing in the ESP 'SDK' (and I'll use that term extremely loosely) is not complete. Much is missing. Whether the intent was to provide what was missing with the ESP 2.0 or not, I don't really know. Since ACES is gone as is ESP... we'll probably never know.

Reverting to the original analysis with a = Offset 80 - Fuselage AoA at Min Induced Drag K₂ *T = a *(2*(1+K₁) on substituting this back into the second term of the original expression givesCL*CL_a*a *(2*(1+K₁)This is eliminates and K₂ and only requires CL*CL_a (both of which are known in the sim), and K₁ which represents the effect of twist on lift curve slope. It's small, and, within Flight Simulator's accuracy, could be omitted - just look at a typcal CL vs AoA curve. So Offset 80 could have easily been used to model the effect of twist on induced drag. I assume that was Microsoft's intention but, as they say, the road to hell is paved with good intentions.Another relevant question. What does CL_a ractually represent? The ESP piston and jet samples give values of 6.16 and 6.48 respectively The theoretical value for an "infinite" wing is 6.283 (2*pi). The FSX default 747-400 is 5.199.. Correcting the theoretical value to allow for aspect ratio gives 4.98 which marches the FS value closely enough. Microsoft seem confused betweeen section lift curve slope (which is for an "infinite" wing ) which is always significantly greater than the wing lift curve slope . I must ad, . mit I 've always supposed that the table was section lift curve slope and that FS corrected it for aspect ratio!

:( If someone can demonstrate with a flight test what offset 80 does I’m still open.Otherwise I’m very skeptical :) It no longer has any effect in my own tests.From the name and description, it sounds like it was a quick way to adjust deck angle.I’m convinced it simply offsets CL vs Alpha, not a good thing….but no matter.It’s not a conventional aerodynamic coefficient (so I call it FS programmer art).If I'm wrong…it would be very helpful to show me :) :( So as I see it...Any effect twist has on CL should be described in CL vs Alpha.Any effect twist has on Induced drag will be calculated by FSX if CL and Oswald are correct.Any effect wing twist has on stability should be described in body roll and yaw.As I understand, and you pointed out too, Cla is 'lift variation due to alpha'…the slope of CL vs Alpha (tbl404 typ).It does fluctuate over 360 deg, but it’s constant from ~inverted stall to ~upright stall.So I guess it is constant much of the time.In my mind’s eye I can’t quite picture what it does…but I can imagine while it is constant pitch changes will be more smooth.I know, from test flights, flying a constant path and G load, for say a 1G stall, is very difficult while CLalpha is shifting.Several online 'textbook' sources put B747’s Clalpha at 5.500 so FSX looks reasonable :)CL vs. Alpha doesn’t refer to a simple airfoil profile, like an XFoil output(2D-infinite)...as far as I know.It’s better to imagine it as the entire 3D body…which it less as you stated.AVL(free) might be a way to calculate these body coefficients.Otherwise flight testing works very well too.The empirical method is the only way to get the stall speed and cruise speed within very tight limits.While I rely mostly on flight testing. I do use formulas to get a sense of the landscape of CL and CDi.I have two questions about these calculations...1) To determine lift from a foil you need to know the right Reynolds number…what speed and altitude should I use to calculate Reynolds(e)?2) FS needs the correct Oswald efficiency to calculate induced drag...how do I estimate that? (I found a value for my aircraft…but is it right?)A word of warning...for folks not already accounting for this...There are two definitions of AOA.The Angle of attach is measured from the wing’s geometric cord angleThe Angle of attach is measured from the wing’s CL=0 angle FS uses the former...but some sources will use the latter.

http://www.grc.nasa.gov/WWW/BGH/reynolds.htmlI meant to post this link... I assume altitude should be 0, and speed should be cruise(?) or maybe stall(?)

Having not once ever needing to use the Reynolds number of a wing, I have to ask... why are you even going there?

I agree it’s not really necessary. It’s to use Xfoil to calculate a lift curve for the airfoil.To get the best results, Xfoil needs a reasonable idea of Reynolds number.The airfoil’s CLvsA is only for reference...and curiosity too:)