Advanced Flight Model for other aircrafts!
-
From my opinion, every other aircraft can be made to fly realistically close to the real one even if using the F-16’s flight model and modifying it accordingly, because after all what could we need to improve their behavior?
1. Realistic moments of inertia values (at least for Ixx, Iyy and Izz), taken either from a form of calculus/calculation or from experimental data;
2. Realistic aero data;
3. Realistic propulsion data.
If these are available, then they can be used into the modified F-16 OFM/AFM flight model data tables and there’s nothing more needed to provide more realism.
Lets compare the F-16 simulated by BMS, with a theoretical F-16 that has the same airframe and engine, but lacks a flight computer and has only a direct mechanical linkage to the flight controls.
One is flyable and the other is not. One has a high performance while the other is very under performing - although it could in theory have the same high performance as the FLCS model, it does not as no pilot can control it quickly enough.
If your flight model is limited to a specific set of flight control laws, you cannot use that to simulate other aircraft automatically. For instance, the Hornet is capable of high alpha maneuvering - the Viper is not, as its control laws prevent sustained high alpha.
This is not an issue for BMS because as mentioned above by the man who coded it, the BMS flight model is not limited to simulating aircraft that share the F-16s control laws.
Your three important things are all essential for getting correct performance - but you also need No. 4 - realistic control laws.
-
From my opinion, every other aircraft can be made to fly realistically close to the real one even if using the F-16’s flight model and modifying it accordingly, because after all what could we need to improve their behavior?
1. Realistic moments of inertia values (at least for Ixx, Iyy and Izz), taken either from a form of calculus/calculation or from experimental data;
2. Realistic aero data;
3. Realistic propulsion data.
If these are available, then they can be used into the modified F-16 OFM/AFM flight model data tables and there’s nothing more needed to provide more realism.
As it seems, the F-16’s AFM/OFM flight model is robust enough to carry a wide range of what’s needed for a good simulation and might be able to support even further improvements (this is not up to me though), so it doesn’t seem like every aircraft would need an individual flight model from scratch. The AFM 2 tested on the A-10 (if I’m not wrong), on the other hand, could be another interesting model for flight simulation.Speaking of witch…, I’ll try to get better aero data for the F-18C (Hornet) and E (Super Hornet) also, as I already have engine data chart for the F-404, as discussed earlier (so the C model would promise the most)!
Cheers, good day!
I really wonder how you manage to do all this work, gathering data should be a pain.
Is there a chance to “compute” them with x-plane and its engine to simulate aerodynamism ? (SORRY if this question is stupid, don’t jump down my throat guys).
-
Is there a chance to “compute” them with x-plane and its engine to simulate aerodynamism ? (SORRY if this question is stupid, don’t jump down my throat guys).
For the aero data, its possible to use a virtual ‘wind tunnel’ to determine how an aircraft is affected by wind. You want a very good model of the aircraft though, and it will take quite some time (or a fast fast computer). Thrust data, not so much, because its not determined by the shape of the aircraft (well, generally speaking - it is affected by the airflow into the inlet).
-
For the aero data, its possible to use a virtual ‘wind tunnel’ to determine how an aircraft is affected by wind. You want a very good model of the aircraft though, and it will take quite some time (or a fast fast computer). Thrust data, not so much, because its not determined by the shape of the aircraft (well, generally speaking - it is affected by the airflow into the inlet).
that would result in very poor modelization
the only right way to do it is retro engineering of em charts and accel charts
-
Far be it from me to contradict the expert. You would know more on the topic than I.
-
that would result in very poor modelization
the only right way to do it is retro engineering of em charts and accel charts
Those charts aren’t that easy to get I presume…
-
Hello again molnibalage,
I’ve already told you how I obtained those thrust curves, in a very rudimentary manner starting from those of the F-16’s Block 50 GE turbofan engine! I should now take Mav’s advice and try using an inverse-engineering technique to get the right curves (and correct values in the most critical parts) by knowing the exact plane’s forward accelerations at a certain drag coef. and plane mass, but only if I can get the correct plane’s acceleration charts at first, or from the plane’s e-m chart using the sustained turn rate, G-load, mass and true airspeed, do the job to estimate the CD (drag. coef) and then immediately find the Drag force = engine thrust for one particular point on the chart.
The e-m charts are much easier to get than acceleration charts (anyway they lead to the same thing), and so I’ll do my best to do what Mav said and painstakingly calculate the necessary thrust in as many points as possible (to obtain a more accurate curve for X altitude) and that’s as accurate as I can estimate thrust tables also from now on.
-
Hi Yzangard,
The true pain is with the engine thrust estimation now. Let’s say that I can more or less be on the right track with the aero coefficients of lift and drag, yet I can’t be 100% sure even here that at some point the error margin wouldn’t get higher than 5% (especially at Mach numbers higher than the Mach drag rise and shock stall appearance Mach).
Took me long enough indeed until I was able to obtain more correct data for the MIG-23ML (that I started with), by correcting and re-correcting my mistakes and other errors, but now as I can better trust the aero results, it doesn’t take very long to get on with them.
-
Those charts aren’t that easy to get I presume…
…no, they are not. But if you did have them, you could back out the thrust required to model engine behavior to match, and interpolate to fill the gaps. In fact, that is what you would have to do, I think. And outside of that, you’d also have to know something about how the subject aircraft flight control system works in order to model it’s maneuvering behavior properly…particularly for any fly-by-wire sort of aircraft. The F-16 FCLS and it’s control philosophy are a bit unique…and somewhat outdated by today’s engineering. It really doesn’t apply to any jet other than an F-16…unless that jet is also built by the same manufacturer. One can probably assume that for any aircraft.
-
FYI: Topolo has made the Viggen FM based on actual performance charts. It’s not an AFM like the F-16 FM’s, but still quite close to the real thing.
-
That sounds cool - now all you need is hi-fi Viggen avionics…nice! I wish there was a handbook for doing complete, accurate aircraft models as add-ons to BMS. I’m sure that dedicated 3rd party teams could come up with really great stuff, and then the BMS core team could continue to concentrate on what it does best - F-16 and world-sim AI modeling.
-
FYI: Topolo has made the Viggen FM based on actual performance charts. It’s not an AFM like the F-16 FM’s, but still quite close to the real thing.
Still can’t get over how sluggish the Viggen is when you mount the centerline tank and AAMs. Definitely not what I expected from an aircraft that is supposed to get in and out of highway strips.
(Not the FM’s fault. I’m probably too used to the F-16.)
-
That’s my doubt too. Not about the translation of the charts, Topolo has done a great job, but if that sluggishness is accurate. At a Swedish forum I’ve asked for rl pilot input in this regard.
-
That’s my doubt too. Not about the translation of the charts, Topolo has done a great job, but if that sluggishness is accurate. At a Swedish forum I’ve asked for rl pilot input in this regard.
I’ve checked the charts from the docs and they at least look reasonable. Give me a shout if you hear back from that pilot.