Page 2
IL-2's Oleg Maddox
By Bob "Groucho" Marks
Groucho: What can you tell us about the flight modeling in IL-2?
Maddox: The FM of the IL-2 sim includes several sub-modules, almost each with hundreds parameters:
- Standard atmospheric model, including altitude effects of air pressure, temperature, and density.
- Equations of motion include standard non-linear six degree-of-freedom equations of motion used for aircraft flight dynamics analysis.
- Aerodynamic model consists of several sub-modules for determining forces and movements produced by various parts of the plane. We use various graphs for WWII planes to determine aerodynamic forces. This model also computes aerodynamic effects due to Mach number, angle of attack, and angle of sideslip including compressibility, temperature, and density of air at altitude, stalls, and spins. Also included are aerodynamic effects of configuration changes such as flaps, landing gear, external stores, airframe damages and ground effect at low altitude (landing and take off) are taken into consideration.
- Mass properties of the model take into consideration the plane construction model, its "empty mass", fuel mass, ammo load, mass movements and products of inertia.
- Propulsion model includes engine and propeller models and computes the current power of engine that depends on controls state, temperature and altitude. Temperature mode of engine depends on the state of water and oil radiators. Propeller model includes airspeed and altitude effects and propeller movements.
- Landing gear model computes forces and movements for takeoff/landing.
We tune our common FM for each type of plane with single or multi engine(s) and in this case each FM is comparable to the original plane characteristics. Really our flight model already has so many parameters in each module that it is impossible to describe them outside of technical documentation. :)
If we speak about the real IL-2: ORIGINAL manual of IL-2 (1940) and trial docs tell us that this aircraft wasn't able to get into a spin in any situation. When we did the FM (THOUSANDS parameters, including such as mass distribution, aerodynamics, loadout, etc) of this plane we got similar results - it is almost impossible to get a spin for this plane. We now understand why the pilot with only 10 hours (!) of the wartime flight school was able to fly this bird independently!
At the same time, in a version that you have, there is a flyable model of La-5FN.... (Really this plane will be AI controlled in the release, so you now can understand that even AI planes of our sim have similar to flyable planes FMs with all the parameters). Just try to fly at high altitude on low speed or fly vertically and you'll get the stalls and spins. Not always - depends of situation such as altitude, maneuvers, common speed, speed of propeller rotation, etc. All depends of real aerodynamics and physics laws of the flight at DIFFERENT altitudes and parameters of the plane model itself. Just try it, you'll get it to react just as it is stated in REAL TRIAL DOCS OF NII VVS (NII VVS is a Soviet military aviation research institute, where were in trials Russian WWII planes of all types, all captured German WWII planes---98% of all types!---all WWII lend lease UK and USA planes) and like stated it real pilots... So, ALL PLANES, WHICH PRESENT IN OUR SIM, will be modeled according to these documents and consulting.
Just for example, when our common FM model was already almost finished, we begun to tune it for different planes and compare with all known for us trial docs. The result was excellent! We got so good and corrected so many parameters, such as sharp turn time and common maneuverability, climb time, maximum speed at diving attack, flatter, real speed at different altitudes (including indicator speed differences on different altitudes), etc., that comparable to real planes data at 95-98%!
In addition, to be sure that we do it all correctly, each of our FM and physics programmers took many real flights with an instructor with a chance to take the flight controls and performed many intense maneuvers including those with a high degree of overloading. Most flights were in the cockpit of a Yak-52, a two-seat prop sports plane. Personal experience of real flight control is one of the most important requirements for any developer that is developing a flight simulation. It is necessary to get a feeling of the real flight and know what you are developing . . . :) Also several our developers previously attended the Moscow Aviation Institute as well as I did (then I worked as a lead engineer for one of the largest military research institutes for 11 years).
