YASim users wanted

[jsb]

Dear all,

I am working on some little features to extend yasim functionality, basically to support aircraft developers. I believe yasim CLI tools could give us some more information about the model that has been calculated from our XML input.
E.g. I have added some code to calculate the MAC of a wing and have the CLI tool print it as well as the CG_x as % of MAC.
I also added some XML attributes that allow you to input mass, altitude, speed in kg,meters,kmh or lbs,ft,knots as it can improve the readability and you do not have to wonder what units are used (of course the old attibutes are still available and unchanged).

I also discussed some ideas with Buckaroo (the guy who wrote the yasim guide) to add some more XML attributes to configure some details, e.g. for flaps.

Any ideas or whishes regarding this from your side?

Please keep in mind YASim is YASim and not jsbsim and it should stay this way, so your "feature requests" shouldn't be too complex. It is more about those little details that could make life easier or add flexibility by making things configurable.

[Octal450]

Fix EGT in the jet engines
Simulate proper startup/shutdown, so users don't have to work around it and fake it.

J

[StuartC]

Proper Turbine support for choppers using engine function would be good. It would save the tons of extra nasal.

[LesterBoffo]

P.S. Thanks for taking on improving the YAsim coding.

Some sort of data table look up for airfoils and propellers, honest propeller gyroscopic precession forces, and number of propeller blades, off the top of my head.

Fuselage drag with Alpha and Beta angles, maybe as simple as categorizing them as '0' for least draggy to '4' for most draggy?

Icing on the cake, Pie in the Sky stuff: A real fix for turbine helicopters lack of native fuel consumption, exhaust gas section and pressure ratio to power turbine power levels, tapping the cold air section for reactive forces.

Reciprocating engine categories such as Cetane (Diesel or Compression ignition) and Octane (spark ignited engines), bore and stroke, power and torque curves, icing effects on carbed engines, gasoline engine misfire with too rich of fuel mixture at altitude.

[DFaber]

Many thanks for your efforts to improve Yasim. It's really apreciated!

I really miss an option to "subtract" a wing element from the simulation. Use cases are Failure simulation, which I currently do by adding lots of drag. Also carrier aircraft with foldable wings would benefit from this feature when they have to fit through a narrow elevator.

It would be nice to have more control over the control surfaces on wing/mstab elements, e.g. deploy slats or spoilers on one wing only. The F-111 and P-61 use wing spoilers for roll control. The slats on the bf109 might open only on one side in tight turns.

Speaking of the bf109, a variable pitch propeller would be fantastic.

Greetings

Detlef

[Michat]

Hi

Two things that I want to point but have no idea it's is totally related to the YASIM paradigma.

A .Quite notice in small models but in general happen too. I note how difficult is to get an speed reduction of the aerodynamic model. Of course this should be related with the xml e.g flaps you talk about.

B. Just after Auto throttle Set Off. It's pretty noticiable. How much throttle user must to apply to the engine throttle control, In my experience with a joystick lever. Don't know if happens with keyboard or mouse, since I never use them.
I have to move up or down the thottle lever to some point where to get the throttle sensibility back and then make quick adjustments and fine tunning to get again balance and the desire speed set previously by the APP Speed. HUD helps. I'm not sure if is related to YASIM or happen in JSB ? (Good nick for a YASIM modeler) Maybe is a problem on the FG side.

I'm gonna say that all my favorites aircraft are YASIM because I have a favor for those russian models performance and design.The programmer have to be a SuperShamman with special tactic logic- anti-logic agaisnt YASYM logic paradigm in order to satisfy the output we get , vivid alive and kicking models.


At the other side as the basic geometrical YASIM reference I have our wonderful http://helijah.free.fr/flightgear/hangaren.htm.

I know Marc Kraus from The Lake of Constance Hangar had to deal with a lot of unsolvable problems of YASIM especially notice in our dear I-16 Polikarpov Bombable for Helijah Hangar. We cannot decrease speed easily. Happens only with small YASIM aircraft ? Related with A.

What I find flying Herbyw's russian aircraft YASIM models exceeded my expectations of what is possible or not in YASIM. Those big models were silly floaty once. But no more since there is some magic on it. Dont't know if this magic is also applicable to small aircraft, since others have declared unsolvable.

Your project could help thousand of aircrafts and few developers.

.
29.92

[LesterBoffo]

I'd just like to add for Michat's comments about certain aircraft coasting with no engine/propeller aerobraking with the throttle closed. This might be because of, at least with piston engine planes with a governed, variable pitch, constant speed propeller, the default coding is for a propeller to be controlled through the 'advance' channel which is more of an engine governor function. I've found that adding this line into the engine section

Code: Select all

min-throttle="0.05"

will force the engine to close it's throttle to a slower idle speed, it also improves taxiing. This entry will need slight adjustment, up or down depending on performance and whether you have supercharging.

[D-ECHO]

Hi jsb and thanks for your effort
It would be really great to have a drag by extension parameter for gears so the ac dev can set how much drag is created by the gear when extended

[LesterBoffo]

...
It would be really great to have a drag by extension parameter for gears so the ac dev can set how much drag is created by the gear when extended

I've found that you can add little vstab entries that do the same thing.

IIRC, YAsim aircraft with retracts supposedly are able to mimic gear drag by adding an

Code: Select all

extension="x.xx"

entry to the gear. But I've found it's minimal if at all.

Here's a typical vstab entry I use..

Code: Select all

 <!-- Nose gear drag -->

   <vstab x="3.95" y="0" z="-1.049"
             taper="0"
             incidence="0"
             twist="0"
             chord="0.294"
             sweep="0"
             camber="0.0"
             dihedral="0"
             length="1.093">
                   <stall aoa="30"
             width="18"
             peak="5.5"/>
      <flap0 start="0.00"
                end="1"
                lift="0.00"
                drag="1.386"/>
      <control-input axis="/controls/gear/gear-down"
                            control="FLAP0"/>
      <control-output control="FLAP0"     
                                                 prop="/gear/gear[0]/position-norm"/>
   <control-speed control="FLAP0"
                            transition-time="5" />
   </vstab>

Admittedly it's a fudge at best, but it does allow you to scale your gear length and size, just make certain it doesn't extend past the gear's ground contact point, match the gear's transition timing, and you might take into account whether the gear doors are left open when extended because their being open will add to the scaling of the drag.

[enrogue]

I'm working on a patch for decoupling gear drag from compression - the drag area of each gear unit is currently tied to 3x the compression number, which is definitely not ideal

I also have an updated version of the (now ancient) yasim ground friction fix - just in the middle of updating it to be turned on & off (and not be the default)

[fmg]

Great attempt!
Just a few wishes/thoughts:

Crosswind sensibility seemed to be to strong during start and landing (taildraggers).

Possibility to have several gears (fixed and retractable) to simulate planes with skid and start trolley so that you don't crash when trolley is retracted/away.

Strange behavior of piston engines. Rotation gets up in a climb and down in a dive. Shouldn't it be the over way round?

Rocket engine support.

More engine parameters. E. g. to tune fuel consumption.

No automilling propellers.

No sliding on ground while parking.

Supersonic effects.

[D-ECHO]

@enrogue: Great to hear
@Lester: Yep, definetely possible, but quite dirty, would be great to have some yasim built-in thing

Some other thoughts from my side:
support of electric propeller engines
spin
more realistic stall characteristics (if somehow possible)
possibility to slip

[jsb]

Thanks for all the input and sorry for my delayed answer. I was a little busy but my latest work just made it into the FG repo ("next") and I updated the http://wiki.flightgear.org/YASim
The CLI tool will output more information about the model that YASim created from the XML file.

Regarding your answers and ideas:
Very interesting input, i really appreciate it
Some of them might be quite easy to implement, most of them are beyond my current knowledge so I definitly need more details.

You may want edit your posts and add
1) how you'd like something to be configured in the XML e.g. define new attibutes that should be added to the XML parser
2) specifiy what behaviour you would like to see e.g. give math model for values to be calculated.

Just to pick the first as an example:

Fix EGT in the jet engines
Simulate proper startup/shutdown, so users don't have to work around it and fake it.

What does "fix" or "proper" mean in this case? Should EGT be a fixed value? Should it be calculated from other values and if so what formular and parameters are needed?
I guess we could search for nasal code in different aircrafts and look how they do things but it would be better to have someone with knowledge who could explain and give the equations we need

Merry x-mas

[Octal450]

Well, in Yasim, EGT starts at 20F or something stupid like that with idle engines, it's very unrealistic. JSBs isn't much better, so I'm working on a custom equation for that. Will get back to you.

Kind Regards,
Josh

[LesterBoffo]

The problem with automilling propellers can be fixed at the animation .xml level by adding this to the propeller spin animation group.

Code: Select all

    <condition>
      <greater-than>
        <property>engines/engine[0]/rpm</property>
        <value> 12 </value>
      </greater-than>
    </condition>

You'll want to increase the value if the engine has a large, geared propeller.

Of course to fix this at the YASim coding level would have to account for how many cylinders a recip engine has, their displacement in bore and stroke, and throttling effects, which would require a ton of internal lookup tables and and some complex maths needing to be run continuously while the recip engine runs.

The upshot of this is a very realistic fuel consumption and other operational features like engine misfire/damage from running outside of fuel mixture stoichiometric, or air to fuel ratios for the altitude being flown at.