Identical levels of precision when comparing full alignment to ALCM

[Biggus]

I'm not sure that this is intended behavior, and I could very well be doing something wrong, so if someone could test this too I'd appreciate it.  I've attached the mission file.

 

A couple of days ago I set out to test the difference in levels of precision with the INS alignment types.  I'd assumed that a full alignment would enable me to fly within a relatively close distance of an INS waypoint, and that an ALCM or ALR alignment would degrade this precision significantly.

To test this, I selected a VOR station at a distance of greater than 200nm from my starting point and placed a waypoint on it in the mission editor.  I made sure the 'stored heading' setting was selected and proceeded to fly the route a few times with different alignment types.  No enroute navigation fixes, no additional waypoints.  Just take off and fly to the waypoint on the VOR station.

I found almost no difference in precision between the different types of alignment.  An ALCM alignment would navigate me directly over the VOR station just as precisely as a full alignment, provided I paid sufficient attention to the INS needle.  ALR had the same result, although I haven't saved a track for it.

Which brings me to the actual apparent difference between the alignment types:  Needle behavior.  A full alignment was very close to correlating with the VOR needle immediately.  Maybe a degree or three away from the VOR station at over 200nm.  An ALCM alignment initially had approximately a 30 degree difference from the VOR needle and this difference was reduced the closer I flew to the destination.  So long as I could maintain the CAP caret over the INS needle, in nearly every case I could directly overfly the VOR station.  I was able to replicate this behavior with a similar flight from Liwa to Lar on the PG map.

Is this the way that the INS system is supposed to work?  Should every alignment type allow me to navigate to the exact same place with better than 0.5nm precision?

F1EE full alignment.trk F1EE ALCM 720 alignment.trk F1EE INS accuracy test Georgia.miz

[fausete]

Okay, to shed some clarity on alignment types.

In all cases, the coordinates used are introduced by the user, that doesn't change with the type of alignment. The alignment type affects the gyrocompassing, the precision in which the real heading is established by the INS.

ALCM means alignment from memorised heading (Cap Memorise), so as long as the last heading of the INS was correct and the aircraft has not moved since turning off the INS, ALCM is going to give the same results as ALN.

ALR is fast alignment, which will have a heading error. Currently the error is too big in this mode and we're polishing the alignment to correct this. Nevertheless, a certain degree of error is expected in this mode, which results in the INS not guiding in the correct (shortest) route to the next waypoint.

Hope this clears things and explains what you're seeing.

[Biggus]

Very interesting,@fausete.

You are saying that every alignment type will take me to the exact same point, provided the initial alignment was sufficient and only the route to that point will change?  Now I'm questioning my other assumptions and this thread probably belongs in another subforum.  I think I might have conflated alignment quality with INS drift.

Should I have seen some drift though?  My flights took me to a single waypoint over 250nm away with eye-watering precision.  Granted there was only one period of acceleration (the take-off) and one left turn of around 90 degrees, I'm surprised that I was able to get so close each time.

Edit:  Apparently maneuvering violently doesn't induce any drift in the INS.  Several 4-8G turns and I'm still able to overly a VOR station 280nm away.  Is this expected behavior?  Under what circumstances should I see drift or need to take a fix?

F1EE ALM heavy gs alignment.trk

Edited February 4, 2023 by Biggus
Added a track following some further testing.

[Iron Sights]

Perhaps the test needs to be conducted at a longer range. 280 miles may be too short of a distance, just speculation. Ins drift may take longer to crop up than you think. I have flown the KA-50 and noticed the drift took much longer to become an issue over time than expected. Do you have data on this? Maybe real data would tell us when we may need to do a fix? Once again, just spitballing here.

[Biggus]

2 hours ago, Iron Sights said:

Perhaps the test needs to be conducted at a longer range. 280 miles may be too short of a distance, just speculation. Ins drift may take longer to crop up than you think. I have flown the KA-50 and noticed the drift took much longer to become an issue over time than expected. Do you have data on this? Maybe real data would tell us when we may need to do a fix? Once again, just spitballing here.

That's entirely possible too.  I've got no data at all, I was hoping that perhaps the Aerges guys might be able to share some information.

[fausete]

Regarding drift:

Absolutely, it's what will cause a coordinate error un the long term and it's a separate and distinct error from the alignment error. This part is still being worked on but accelerations should play a role on the magnitude of the coordinate drift.