Working on getting our L-39 into DCS using the mod structure given to us by Alex, with good results so far! Thought you would like to see our progress. At this point it has its own slot, FM, cockpit (still early WIP), collision model, normal maps (early WIP) and all animations and lights are working properly.

Originally Posted by Krebs20

Great work. Is this going to be a public release? Payware or freeware? It would be a good deal of fun. To play with.

Too early to say. We would like to model all aircraft systems when the resources are available to do so. The ZA (combat trainer) and C (trainer) versions, that is.

http://forums.eagle....ead.php?t=92197

Figured I'd start a new thread for this little nugget of information.

After a whole 3 days of head-scratching and bug-fixing on both sides of the ED dev fence, I'm proud to announce work has begun on the L-39 ADVANCED FLIGHT MODEL.

As I type, the sim is flying straight and level, and we now have the unenviable task of breaking the aircraft into multiple parts, performing aerodynamic analysis of each part, and computing the actual performance of each component, in order to create a very high-accuracy, high-fidelity aerodynamic model of the aircraft.

Development time of this is unknown, but as soon as we have something worth sharing, you'll know as soon as we calm down enough to write.

Best regards,
Tango.
-----------
The AFM is in large part our own. We feed back the results of our computations to the sim, the sim does a small amount of additional processing to complete the model, then it is applied to the sim.

All that is being added by ED is the gravitational model, collision detection, and some physics related to the weapons.

Best regards,
Tango.
------
I must give credit where credit is due. A big thanks to Alex and anyone else on the ED team that is facilitating AFM integration and 3rd party development in general.

In addition to what Tango described, we are feeding vector summed moments and forces which we must developed from scratch and the sim will run these through the 6-DOF equations of motion. It goes without saying that this will be a long and arduous process of development for not only our AFM module, but also for the development of the AFM API for 3rd party Devs.

If anyone out there is experienced in flight dynamics simulation or flight model creation and would like to volunteer their knowledge or resources to help develop this process, just shoot us a PM. C++ knowledge is not needed, but helpful.

http://forums.eagle....14&postcount=39

Well the process for doing it isn't a secret, but our own individual code may or may not be shared. Currently we are working with an ED dev to get us the correct coordinate system transformation data we need to make local body frame force and moment calculations.

The API is still WIP, so we haven't really started to make a ton of progress until we have nailed the first step.

http://forums.eagle....74&postcount=53

We have 12 L-39 rated pilots on the Patriots Jet Team who we will be looking to when it comes time to test flight model fidelity, as well as access to around 8-10 of the actual jets. Rest assured we will be making it as near-perfect as we humanly can.

http://forums.eagle....87&postcount=54

but how does the ED team know at the end, if the 3rd party flight model corresponds correctly to the actual aircraft and DCS standards?

Short of ED owning/testing/flying the real aircraft - the only way is to see if the FM matches available real world data.

Nate
_____

http://forums.eagle....10&postcount=55

My guess is you need to know math, physics etc first, ie fluid mechanics, multidimensional vectoring, aerodynamics etc, and then you need to know C++ etc to translate this into a function a computer can understand I presume...


This. I would add that you should probably understand time/frequency domain dynamic analysis for accurate phugoid characteristics and controls for accurate autopilots and flight control systems (should your aircraft have them). While we all wish we were all experts in these areas, we are doing just fine learning as we go. We currently have 2 engineers and 1 software professional writing and optimizing the code, but it takes a lot of work just to set the FM up.

Over 1500 C++ lines of code are written and the aircraft only has 3 source of lift added and flies nothing like an aircraft, yet. It takes a really fine touch to balance all 6 of the forces and moments to create something that resembles and FEELS like a real aircraft. It's one thing to match numbers and a completely different thing to model proper dynamics as ED and Yo-Yo have done in their products.

http://forums.eagle....69&postcount=64

**UPDATE**

The following lift and drag forces have been added to the AFM. These are by no means tweaked to their final values, but their effects can be seen in game.

LIFT

Lift due to left wing
Lift due to right wing
Lift due to left wing flap
Lift due to right wing flap
Lift due to alpha rate
Lift due to elevator deflection

DRAG

Drag (zero lift) due to fuselage
Drag (zero lift) due to left wing
Drag (zero lift) due to right wing
Drag (zero lift) due to vertical tail
Drag (zero lift) due to horizontal tail

Drag (profile) due to beta
Drag (profile) due to alpha
Drag (profile) due to elevator deflection
Drag (profile) due to gear
Drag (profile) due to left flap
Drag (profile) due to right flap
Drag (profile) due to speed brake

Drag (induced) due to left wing lift
Drag (induced) due to right wing lift
Drag (induced) due to left wing flap lift
Drag (induced) due to right wing flap lift

L-39 news

 

http://forums.eagle.ru/showpost.php?p=1532548&postcount=103

And here is what you've all been waiting for. Here's a first look at our WIP AFM!

 

 

Suggestions and constructive criticisms are welcome, as this is a very early build of the AFM and has only been in development for three weeks. This video demonstrates wing stalls, spins, tail slides, rolls, loops, rudder kicks, dutch rolls, attempts at landing with no gear (not implemented by ED, yet) and even displays the precision of the flight model with some formation flying.

 

Current known issues:

 

Wrong wing rises during spin

All moments and forces need to be adjusted to reduce alpha and beta rates

Ground handling not added to the 3rd party API, yet

Fuel weight not a factor, yet (this aircraft is LIGHT and FAST!)

http://forums.eagle.ru/showpost.php?p=1607830&postcount=156

**UPDATE**

 

Thermodynamic Gas Turbine Model Implemented

 

Inputs:

 

Atmospheric Conditions

Static Temperature

Static Pressure

Local Speed of Sound

Velocity/Mach of Inlet Air

 

Geometric

Core Area

Nozzle Area

Bypass Ratio

RPM

 

Properties

Burner Pressure Ratio

Nozzle Pressure Ratio

Max Combustion Temperature

Nozzle Temperature Ratio

Inlet Efficiency

Fan Efficiency

Compressor Efficiency

Burner Efficiency

Turbine Efficiency

Nozzle Efficiency

 

Thermodynamic Outputs:

 

Solved Properties

Inlet Pressure Ratio

Fan Pressure Ratio

Compressor Pressure Ratio

Turbine Pressure Ratio

Inlet Temperature Ratio

Fan Temperature Ratio

Compressor Temperature Ratio

Burner Temperature Ratio

Turbine Temperature Ratio

 

Solved Conditions

Inlet Total Temperature

Compressor Total Temperature

Burner Total Temperature

Turbine Total Temperature

Fan Total Temperature

Exhaust Gas Temperature (EGT)

Inlet Total Pressure

Fan Total Pressure

Compressor Total Pressure

Burner Total Pressure

Turbine Total Pressure

Nozzle Total Pressure

 

Performance Outputs:

Net Thrust (Gross Thrust - Ram Drag)

Mass Flow Ratio

Fuel Air Ratio

Specific Fuel Consumption (SFC)

Fuel Flow

Overall Engine Efficiency

 

Example Performance Curves

This example features the peformance of the GE CF6 Turbofan engine

http://forums.eagle.ru/showpost.php?p=1609415&postcount=161

Landing dynamics added to the flight model, along with new custom mechanical animations for gear, flaps, and airbrake.

 

http://forums.eagle.ru/showpost.php?p=1497736&postcount=73

AIR DATA

 

* Determination of static and dynamic pressure from first principles

* Air density calculation

* Mach calculation

* Airspeed derived from static and dynamic pressures

* Airspeed error due to density (altitude and/or temperature)

* Compressibility effects and "ram rise" by modelling adiabatic compression using the first law of thermodynamics

* Airspeed error due to AoA and slip angle

* IAS position error precisely matches the L-39 profile

* Complete modelling of static and pitot system with all consumers connected to the relevant system(s)

 

Best regards,

Tango.

 

http://forums.eagle.ru/showpost.php?p=1502507&postcount=75

ALTIMETER

 

* Connected to static system

* Correct failure behavior

 

 

VSI

 

* Connected to static system

* Correct failure behavior

 

 

ASI

 

* Connected to static system

* Correct failure behavior

 

Best regards,

Tango.

 

http://forums.eagle.ru/showpost.php?p=1613796&postcount=79

New external mesh for the C model, thanks to DGambo.

http://forums.eagle.ru/showpost.php?p=1617134&postcount=80

More testing of the new external model. Pilot is currently a place-holder, and even more detail is being added to the airframe almost every day. Bare-metal textures for testing purposes only.

Thanks for the updates,Minoslas!

 

Regards.

That's looking fantastic! Thanks for all the updates.

http://forums.eagle.ru/showpost.php?p=1619189&postcount=86

While the L-39ZA is the primary focus of our module, we are also working on a C variant that will focus more on the initial training of aircraft systems (and maybe someday dual-seat training). We are also considering including civilian aerobatic and racing versions as extras. Here's a popular L-39 that visits my hometown airshow every year, N139PJ "Angel 7". Also, a WIP of a start to increasing engine exhaust detail.

 

Screens of aircraft are all work-in-progress and do not represent the final aircraft or content.

That Albatross is a beauty

Blue Angels,Sweet!!

http://forums.eagle.ru/showpost.php?p=1625986&postcount=106

3DS Max line render of the cockpit in it's current state. More details and renders to follow Q1 of next year, and nothing here is finalized, but for now this should get your attention!

 

Screens of aircraft are all work-in-progress and do not represent the final aircraft or content.

http://forums.eagle.ru/showpost.php?p=1644640&postcount=203

Are you going to develop different cockpits for the versions L39C and L39ZA?? The real ones have little differences between them.

There actually are some pretty significant differences (mostly layout of the main panel) between the C and ZA cockpits, and yes we plan to model both (front and back seats).

L-39 Albatros News

Quick Update:

 

Tonight I started working on a massive flight data export function (150+ values per time step) to assist in debugging and flight model tweaking. In the process of writing this function, I got the idea to write a raw input recorder and reader for replaying flights from the cockpit, similar to Phantom Control in FC1 and FC2. Within an hour, I had a fully functioning recorder and within four hours, I had a fully robust reader/input replicator. Whether skimming the hilltops, performing stalls/spins, or cycling gear/flaps/etc., the playback was perfect (even at 50X simulation rate), allowing me to fly a 20 minute flight landing at another airbase, only to see the aircraft smoothly touch down on the exact spot I had just previously marked.

 

The reason for the robustness of the playback is due to the fact that inputs are recorded AND read at 166.667 Hz (0.006 s), as this is the rate at which the external flight routine model is called. The playback is independent of controller setup/curves/etc., as the playback disables controller input and reads solely the values from the recorded file and uses them to set internal variables I have defined in my FM dll. The playback is pausable and you are free to click buttons in the pit and do whatever it is you would like while the program is flying your aircraft.

 

After all the data has been read, the inputs cease to be read from the recorded file and the control is given back to the pilot (if he is even at his computer )

 

The only disadvantage of this method over a more general Phantom Control program is that it will only function on a aircraft that has a 3rd party developed external flight model that has implemented a method similar to this. I'm writing this here to see if people would be interested in this sort of feature staying in the module in a final release? In my opinion, the potential for abuse using this program is minimal to none, but I wonder what others think.

 

Regardless, this is big news for any aerobatic teams out there who are missing Phantom Control in DCS! I figure another use for this may be pre-recorded flights, such that a "AI" backseat instructor could fly the aircraft during some instruction or during early introduction flights.

http://forums.eagle.ru/showpost.php?p=1756566&postcount=233

 

**UPDATE**

 

Systems Ported From Lua to C++

Air data system

Electrical

ADI

Altimeter

Radio Altimeter

Accelerometer

Mechanical (gear/canopy/flaps/speedbrakes)

Fuel

Engine Control Unit

Environmental

Flight Control/Trim

Warning Indicator Lights

 

System Updates

 

Altimeter:

Reference Barometric Pressure can now be set

 

Radio Altimeter:

80 degree lobe added. Above 40 degrees away from pointing the belly straight down, slant range is taken into account

 

ACS-1M Clock and Stopwatch:

Main 12 Hour clock

12 Hour Stopwatch

30 Minute Stopwatch

 

Engine Control Unit:

Engine Vibration, Fuel Pressure, Oil Temperature, and Oil Pressue gauges added

EGT is now dynamic with atmospheric and engine conditions. Heating and cooling are modeled during extended high or low throttle settings.

 

Flight Data Recorder:

When turned on, it records specific flight data/pilot inputs to time based csv files in Saved Games folder. (Does not record by default like the real aircraft, as recording all data can become quite large on a user's system over time)

 

Caution/Warning Indicators:

Too many changes to list, but these are being tied to systems as work is completed

 

Misc:

LCtrl-M key command prints METAR conditions to a message on the screen (Only time, wind, temp, dewpoint, and pressure are correct at the moment)

Smoke system has been added for aerobatic use using a temporary method. The C version will likely feature a smoke system equippable similar to a smoke pod on final release.

 

And many many more small changes...

 

Work In Progress

Landing gear physics (still a huge mystery and very very confusing)

Flight model tweaking (induced drag tweaking, pitch torque balancing, side force tweaking)

Tango's expertise on navigation and radio communication systems will be utilized for RSBN and other systems in the near future

http://forums.eagle.ru/showpost.php?p=1760767&postcount=236

 

**UPDATE**

 

External Pilot Head Movement

6DOF head movement from TrackIR in the cockpit is translated to the external pilot's head. Now you can see everywhere your teammate is looking. Video demonstrates pitch, roll and yaw, however x, y and z translation TrackIR axis's are working but not implemented. They will be animated as the pilot leaning about in the seat.

 

Pilot model is temporary for testing purposes. This video does not represent the final product, and everything is subject to change.

 

Enjoy!