Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Vampire_pilot]

... and in real life too ... 

That would be okay if Il-2 would simulate the take off scenario correctly, but it does not. Torque can't be countered as you can in real life, and it is particularly bad on take off before the aero effects take on.
Excessive torque by definition means the presence of too much of it to be okay.

An example I recently had to work over was the B-36. It is not possible to take it off in Il-2 - literally, no way - if no tweaks would have been used in the FM to counter the "simple calculus FM torque". The game behavior is just not realistically calculated during take off.

[Mick]

... yep, in WW2 real life, very few twin engine AC had them spinning in opposite direction, to counter torque ...

The P-38 was one of the few ...



This meant building LEFT ONLY engines and RIGHT ONLY engines ...

[Birdman]

This torque issue should be caused by _Clip hook positions in prop meshes. Average of those hook positions should be at prop shaft, or otherwise engine position gets calculated incorrectly. Here those hooks are at tips of two blades and that gives bad average position for three-bladed props. Mirroring _Clip to another blade in other prop instead of using the same position in both props would give better take-off, but even better is to have one _Clip hook in tip of one blade and another at opposite position between other two blades.

I hope someone can test with

Code: [Select]

[Hooks]
_Clip00 <BASE>
_Clip01 <BASE>

[HookLoc]
-1 0 0 0 -1 0 0 0 1 0.0054831 -0.0089902 -1.6317796
-1 0 0 0 -1 0 0 0 1 -0.0054831 -0.0089902 1.6317796in PropL_D0.msh and

Code: [Select]

[Hooks]
_Clip02 <BASE>
_Clip03 <BASE>

[HookLoc]
-1 0 0 0 -1 0 0 0 1 0.0054831 -0.0089902 -1.6317796
-1 0 0 0 -1 0 0 0 1 -0.0054831 -0.0089902 1.6317796in PropR_D0.msh

I'm quite sure this will give nice amount of torque for both engines even without testing in game. There's still some torque but it shouldn't be excessive anymore when position of thrust vector should be OK with those _Clip hook positions.

With two engines and four-bladed props all should be fine with both hooks at the tips of opposite blades. With single engine all four hooks are in one prop and in that case all three blades can get one hook at their tips and fourth hook can be at the end of prop shaft or spinner.

Hopefully this post helps with fixing similar issue in many other twin-engine planes also with three-bladed props. I just wanted to explain the issue as well as I can instead of just posting two mesh files as a fix.

[andrey65]

Excellent! It's okay now. Thank you!

[Vasya]

Removing extra space in this line
(before the last zero) didn't help:
[Prop1_D0]
Mesh PropL_D0
Parent Engine1_D0
Attaching -1.000000 0 0 0 -1.000000 0 0 0 1 0 -0.622178 0

Why in all the world would you try that?
Attachment coordinates must be in a certain format, 3 sets of 3 vectors and then the absolute coordinates. Removing the space would eliminate one coordinate (z) and therefore render the whole attachment invalid.

Two very powerful engines, both turning the same direction will create excessive torque in Il-2 all the time.
What you can do is to introduce a second engine that turns the other way (only in FM calculation, not visibly) and therefore basically eliminating the torque.

Removing the space did not affect the airframe assembly - checked in the mission.
----------------------
With fm from the Pe-2, the Ki-108 had the same problem - it was describing takeoff circles on the runway.
Despite the fact that the M-105RF engines have only 1210 hp each.
The game, although ancient, but very accurately reflects the behavior of the aircraft both in the air and on the ground.
Otherwise it wouldn't be a simulator.
But all is well that ends well.
I'll try a new fix.
I think that 3D up to the Ki-108 airframe can still be improved - so as not to put crutches in the FM, which correct the problems of export from 3D Max to Il.
Regards to modders!

[Vasya]

Thanks for the fix 3D Ki-108!
I also updated my FM for this wonderful aircraft - adjusted the trim.
 

[SAS~Storebror]

This torque issue should be caused by _Clip hook positions in prop meshes. Average of those hook positions should be at prop shaft, or otherwise engine position gets calculated incorrectly.

Correctly pointed out by Birdman.
There's a second way to counter incorrect engine position calculation: You can simply specify the location and direction of each engine and prop vector in the flight model explicitely, as shown here on the B-36B for instance:

Code: [Select]

[Engine]
  Engine0Family Wright_Cyclone4360_Series
  Engine0SubModel R-4360-41
  Engine1Family Wright_Cyclone4360_Series
  Engine1SubModel R-4360-41
  Engine2Family Wright_Cyclone4360_Series
  Engine2SubModel R-4360-41
  Engine3Family Wright_Cyclone4360_Series
  Engine3SubModel R-4360-41L
  Engine4Family Wright_Cyclone4360_Series
  Engine4SubModel R-4360-41L
  Engine5Family Wright_Cyclone4360_Series
  Engine5SubModel R-4360-41L

  Position0x -5.38
  Position0y 19.42
  Position0z 0.5
  PropPosition0x -8.38
  PropPosition0y 19.42
  PropPosition0z -3.2
  Vector0x 0.99863
  Vector0y -0.052336
  Vector0z 0.0

  Position1x -5.13
  Position1y 13.46
  Position1z 0.5
  PropPosition1x -8.13
  PropPosition1y 13.46
  PropPosition1z -3.2
  Vector1x 0.99863
  Vector1y -0.052336
  Vector1z 0.0

  Position2x -4.88
  Position2y 7.58
  Position2z 0.5
  PropPosition2x -7.88
  PropPosition2y 7.58
  PropPosition2z -3.2
  Vector2x 0.99863
  Vector2y -0.052336
  Vector2z 0.0

  Position3x -4.88
  Position3y -7.58
  Position3z 0.5
  PropPosition3x -7.88
  PropPosition3y -7.58
  PropPosition3z -3.2
  Vector3x 0.99863
  Vector3y 0.052336
  Vector3z 0.0

  Position4x -5.13
  Position4y -13.46
  Position4z 0.5
  PropPosition4x -8.13
  PropPosition4y -13.46
  PropPosition4z -3.2
  Vector4x 0.99863
  Vector4y 0.052336
  Vector4z 0.0

  Position5x -5.38
  Position5y -19.42
  Position5z 0.5
  PropPosition5x -8.38
  PropPosition5y -19.42
  PropPosition5z -3.2
  Vector5x 0.99863
  Vector5y 0.052336
  Vector5z 0.0
However, as Andy rightfully pointed out, the game still fails to calculate engine torque properly if you have just too much of it:

Torque can't be countered as you can in real life, and it is particularly bad on take off before the aero effects take on.
Excessive torque by definition means the presence of too much of it to be okay.

An example I recently had to work over was the B-36. It is not possible to take it off in Il-2 - literally, no way - if no tweaks would have been used in the FM to counter the "simple calculus FM torque". The game behavior is just not realistically calculated during take off.

We've really tried our best to solve this, but there was no way to do it with IL-2's physics without denying the presence of torque altogether.


Mike