• Which the release of FS2020 we see an explosition of activity on the forun and of course we are very happy to see this. But having all questions about FS2020 in one forum becomes a bit messy. So therefore we would like to ask you all to use the following guidelines when posting your questions:

    • Tag FS2020 specific questions with the MSFS2020 tag.
    • Questions about making 3D assets can be posted in the 3D asset design forum. Either post them in the subforum of the modelling tool you use or in the general forum if they are general.
    • Questions about aircraft design can be posted in the Aircraft design forum
    • Questions about airport design can be posted in the FS2020 airport design forum. Once airport development tools have been updated for FS2020 you can post tool speciifc questions in the subforums of those tools as well of course.
    • Questions about terrain design can be posted in the FS2020 terrain design forum.
    • Questions about SimConnect can be posted in the SimConnect forum.

    Any other question that is not specific to an aspect of development or tool can be posted in the General chat forum.

    By following these guidelines we make sure that the forums remain easy to read for everybody and also that the right people can find your post to answer it.

P3D v3 Dysfunctional ADI

Messages
218
Country
greece
I've been trying to get the F-4B's ADIs to work in P3Dv3 with partial (alas no) success.

When using the FSX native model in v3 the ADIs work right only when facing the true north. There are offsets when rolling the plane and so on, but generally it's OK. When going south, though, the ADI flips out. Pitching the plane up registers pitch down on the ADI and so on.

The exact same model behaves perfectly in FSX and P3Dv2.

The problem lied in how I set it up.
KF0-360: Roll
KF361-721: Pitch
KF722-1082: Heading (for the big ADI)

So I went for a little polygons, one controlling roll (grandpa), one for pitch (child) and the sphere itself the heading (grandchild).

The ADIs worked perfectly in v3 when part of the model is exported! When the full model is exported though:
  • ESP 1.0 checked: The main and rear cockpit ADIs don't work right. Only heading does. Pitch & roll work in the back but are inverted.
  • ESP 1.0 unchecked: Visibility conditions work, as well as a single animation. All other animations are frozen in their 0 keyframe position.
I suspect I've hit an animation or part limit, so I started removing decoratory animations and some parts to see if that was the case. No difference. And the model works perfectly in FSX and v2.

Any ideas?
 
Alexis,
My very old F-4M and A-7D have this type of ADI and it works in Version 3.
I used stock animations.
The hierarchy was
Top: ADI Frame, basically the non-moving structure.
Level 1: attitude_indicator_back_up_pitch. This was hidden at the center of the system
Level 2: attitude_indicator_back_up_bank. This was a sphere with no markings.
Level:3: needle_dg.
Level 4 consisted of a top and bottom cap and 12 segments that rotated with needle_dg and had the pitch and heading texture.
The animations were all 0 to 360.
I probably could have constructed levels 2 and 3 in a simpler manner but I did it this way so I could check out each animation as it was built.
When the wings are level the heading on the ADI and HSI are the same. When pulling g the heading on the ADI appears to lead the HSI but only by an amount equal to the AOA, which makes sense.
Roy
 
An animated object can only respond to one keyframed variable at a time. It makes no sense whatever to attempt to apply two variables to the same object at the same time, much less three...

All three variables should be assigned to three separate but linked objects.
 
An animated object can only respond to one keyframed variable at a time. It makes no sense whatever to attempt to apply two variables to the same object at the same time, much less three...

All three variables should be assigned to three separate but linked objects.

What is strange is that the three animation object works in FSX/P3Dv2 but not v3. When applying the method you and Roy described the ADIs go crazy in v3.
 
So I went for a little polygons, one controlling roll (grandpa), one for pitch (child) and the sphere itself the heading (grandchild).

I think this is part of the problem. IMHO the order should be as I said above, pitch first then roll and heading last. The main function of an ADI is to tell the pilot which way is up. Pitch is the prime reference since it directly relates to attitude relative to the world. Bank is a secondary input. Bank is an angle relative to the vertical component of pitch, so the bank indications should be relative to pitch. If pitch is 90 degrees, bank will be in the horizontal plane relative to Earth axes.
When applying the method you and Roy described the ADIs go crazy in v3.
Your method was not the same as I described.

I can think of no change to P3D V.3 that affected anything as fundamental as the aircraft attitude and bank variables.

I really do suggest you try the method I described since it has worked in FSXA and P3D versions to date.

Roy
 
I do what exactly Roy and Bill mention. I use P3D v3 and I have no issue for unusual behavior
do structure like Roy said and make sure you applying one variable at one animation object like Bill said.
linked it from one to another. make sure you check direction X, Y, Z each object linked have same direction
if you have difficulty to set animation key frame when already linked, then do one by one of the animation for each object. after all object have their own animation, align it to one point object as parent-child (one to another).
 
The problem is solved, so let me sum it up with the history, solution and observations for the new guys that want to replicate it.

Purpose: Make an ADI that can show Pitch, Bank and Heading
Method: I created the sphere and applied all three animations to it in consecutive keyframes.
  • 0-360 for roll
  • 361-720 for pitch
  • 721-1080 for heading
Result & Problem: worked perfectly in FSX:A, FSX:SE and P3Dv2. In Prepar3Dv3 though I had the following issue:
  • When heading North, ADI worked perfectly
  • When veering off North, the ADI started behaving erratically, with the behavior worsening the further I veered off.
  • When heading South, pitching the plane up resulted in pitch down on the ADI, and roll was reversed as well
  • The backup ADIs that didn't show heading showed the same behaviour as well.
Apparently P3Dv3.4 cared about something that FSX and v2 didn't.

Solution & second problem: I decided to resort to the traditional method that requires that you animate three parts, one for pitch, one for roll, one for heading and link them appropriately.
The problem was that when this was applied, I was getting the following issues in FSX and P3D:
  • Exporting only the ADI worked fine
  • Exporting the entire cockpit made only the heading work properly, while pitch was frozen to -180 (inverted) and roll was obviously indicated for inverted flight.
Reason for 2nd problem:
In the end, some parts of the model were corrupt at various levels of hierarchy:
  • for the front ADIs
    • The Main Panel ("grandfather")
    • An obsolete part where the ADI frame was linked. The sphere was child to the ADI.
  • for the rear ADI
    • The Main Panel ("grandfather")
    • The "EJECT" light -parallel in hierarchy to the ADI itself
  • in the main scene
    • The "interior" polygon
Solution:
  1. The offending parts were cloned.
  2. Children were linked to the clones and hierarchy was restored.
  3. The clones were linked to the parents of the corrupt parts
  4. The corrupt parts were deleted from the scene
After this was done for all five parts, the ADIs started behaving normally.

Useful observation:
The standard procedure for making an ADI, which Roy justified involves the following hierarchy:
  • Pitch polygon
    • Bank polygon
      • Heading polygon
I observed however that in P3Dv3.4 ONLY all the ADI spheres showed signs of sideslipping. Alas, when flying straight and level and doing a roll, the sphere moved off-center slightly indicating different pitch than the actual value, that could even be 3-5 degrees off. This is observable in the UH-60 that comes with P3Dv3.
Here's a screenshot of the UH-60 ADI at a bank. You will notice that the cross is evidently off-center which shouldn't be the case. This is exactly what I got in my Phantom too, but smaller in scale.
ADI.jpg

68zdq5l2p


I found out that the hierarchy that keeps the ball centered is:
  • Bank polygon
    • Pitch polygon
      • Heading polygon
So this was the problem and solution. Thanks to everyone for your input!
 
Last edited:
Alexis,
As they say in commercials "your experience may be different" Glad you got it working.
I have just flown my F-4 in P3Dv3.4. It works just like it did in FSXA, P3D 1.4, V2(various) V3 (various).
The model is native FSXA and has just been copied over and left there as updates took place.
This suggests to me that it is not the simulator per se that is creating your difficulties, must be something to do with the compilation.
The only deviations between the ADI and HSI headings I have seen are when turning and the ADI sees a component of AOA in its heading whereas this does not affect the HSI.
I am, for some reason, using heading DG, should have been True or Magnetic, but I probably deferred to the standard modeldef coding. However magnetic variation in the area is only just over 2° West wrt grid heading.
Regards
Roy
 
The only deviations between the ADI and HSI headings I have seen are when turning and the ADI sees a component of AOA in its heading whereas this does not affect the HSI.

The "AoA component" in the ADI is part of the problem which started this research. P3Dv3 seems to be the only one with this issue and this is why I experimented with the hierarchy, leading me to the solution mentioned above.
The problem was there even with only the ADI exported, so it doesn't have to do with model complexity -probably. There is something P3D doesn't seem to like in the hierarchy, which leads to "AoA" or what's seen in the screenie above.

The point is that it works, at long last!
 
Back
Top