INU Alignment - Is accelerated enough?

[Tuuvas]

On 12/17/2022 at 7:08 AM, Rhinozherous said:

A fast special option for normal would be nice... If I get it right the manual says engines should be off during alignment. So sitting for 15mins doing nothing is boring

Dont get me wrong, I like realism, but there are things that need an (optional) shortcut. 

This is what the 3min accelerated alignment is for.  It loads the normal alignment that was already stored in the Shark's computer.  Much like stored heading INS alignments in the F-16, F/A-18C, A-10C, etc.

[rrasfly2]

I am now convinced there is a serious bug with the INS alignment. I have followed your video exactly and had a good INS alignment once. Then the very next time I started off to the exact same procedure and failed and started drifting at well over 4 km per minute. ill try to make a trk file i can attach but thy keep going over 5mb

Edited December 20, 2022 by rrasfly2

[jubuttib]

10 hours ago, rrasfly2 said:

I am now convinced there is a serious bug with the INS alignment. I have followed your video exactly and had a good INS alignment once. Then the very next time I started off to the exact same procedure and failed and started drifting at well over 4 km per minute. ill try to make a trk file i can attach but thy keep going over 5mb

 

Just put it on anonfiles or something.

12 hours ago, rrasfly2 said:

I’m not sure what the exact parameters are that cause the light to pop but if you do a normal or a precision alignment the light won’t pop even if you start pulling some hard turns because the The NAV system is using the accelerometer data. 

FWIW I get the warning light popping up even during shallow turns when using accelerated alignment, and as you say I never get it with precise alignment. Despite that though the accelerated alignment drifts WAY less.

I've also noticed that the backup ADI keeps going on the wonk even when letting the autopilot fly the helicopter, with fairly nonchalant turns. Sometimes during my testing it just went all over the place after only a couple of turns and never recovered, sometimes it actually managed to sort itself out after a full lap.

Gonna also test what happens when if I fly a figure of 8 instead of constant clockwise patterns.

Edited December 20, 2022 by jubuttib

[Harlikwin]

On 12/17/2022 at 8:08 AM, Rhinozherous said:

A fast special option for normal would be nice... If I get it right the manual says engines should be off during alignment. So sitting for 15mins doing nothing is boring

Dont get me wrong, I like realism, but there are things that need an (optional) shortcut. 

Yup 100% agreed

[rrasfly2]

For anyone else having this alignment issue I have found a working solution although it doesn’t make any sense to me why it works. Do the alignment as stated in the manual with the motors off till you see the flashing INU PERC or INU NORM button signaling the alignment is complete. Do not hit the button to finish the alignment, just start up and get everything else ready to fly like normal. Then once both motors are up to full speed and you have both gens on and producing power, now confirm the alignment buy hitting the flashing button. I’ve tried to three times in a row now and all three times ive had a good alignment throughout the same 10 minute flight test I’ve been doing. This procedure makes zero sense to me why it works but it does. 
For everyone that keeps complaining about how long the alignment takes, I’ve attached a short video explaining all the the steps in a INU alignment.

 

Edited December 21, 2022 by rrasfly2

[ZuluThreeZero]

Up for giving the @rrasfly2 technique a bash, but otherwise, based on what I've read in this thread and others, is the basic accelerated align currently giving the best nav performance?

[jubuttib]

8 hours ago, ZuluThreeZero said:

Up for giving the @rrasfly2 technique a bash, but otherwise, based on what I've read in this thread and others, is the basic accelerated align currently giving the best nav performance?

Certainly from what I've seen the quick and dirty accelerated alignment is giving great results with fairly low drift. The positions of waypoints don't seem to be entirely accurate vs. ABRIS, but then again the PVI-800 is working with tenths of a minute accuracy, vs. ABRIS at hundreths of a minute. Using coordinates of Novorossiysk as an example, 1° of longitude would be ~79km, so one minute would be 79 000m/60 = ~1 317m, one tenth of that = ~131.7 meters, one hundredth = ~13.17 meters. So an offset of somewhere around 100 meters between ABRIS and PVI-800 doesn't seem too bad.

(Please someone check my math and tell me if I made a mistake. I used LAT 44.712993418734854 and LONG 37.76625581947498, and this website to get the distance between the points: https://gps-coordinates.org/distance-between-coordinates.php)

Edited December 21, 2022 by jubuttib

[jubuttib]

11 hours ago, rrasfly2 said:

For anyone else having this alignment issue I have found a working solution although it doesn’t make any sense to me why it works. Do the alignment as stated in the manual with the motors off till you see the flashing INU PERC or INU NORM button signaling the alignment is complete. Do not hit the button to finish the alignment, just start up and get everything else ready to fly like normal. Then once both motors are up to full speed and you have both gens on and producing power, now confirm the alignment buy hitting the flashing button. I’ve tried to three times in a row now and all three times ive had a good alignment throughout the same 10 minute flight test I’ve been doing. This procedure makes zero sense to me why it works but it does.

Trying to test this right now, and am seeing a fair bit of drift within the first 10-20 minutes (around 1.3 km when comparing PVI-800 vs. ABRIS self-coordinates). Could you upload a video of how exactly you do your startup?

EDIT: 2.3 km after 30 minutes.

Edited December 21, 2022 by jubuttib

[jubuttib]

After testing rrasfly2's new method, which didn't work for me, I ran tests with a "start from parking, hot" helicopter with the "Normal Alignment with Gyrocompassing" selected in the ME, and wanted to compare that to the accelerated alignment. I also tested this on a figure of 8 pattern this time.

Short story, didn't really see any real differences in accuracy between the two, both were much better than trying to do precision alignment yourself from what I could see. The figure of 8 pattern seems to make the INU drift more and faster vs. just doing a square pattern (the waypoints were the same, just the order was different). I recorded the PVI-800 self coords vs. the ABRIS self coords after each turn, and sometimes before as well, to try and gauge whether the INU drifts more when turning or when flying straight. Used the same website I mentioned a couple of posts before for measurements. Right now it seems that the worst thing for accuracy are counterclockwise turns, but honestly the whole thing is such a mess that it's hard to make any solid conclusions...

Here are my notes. First is whether measurement was taken after/before however many turns I'd done, then how long I'd flown by then (approximate, not exact but close enough), what the heading was AFTER the turn, what the drift was at that point, and what direction the previous turn had been in (CW = clockwise, CCW = counterclockwise):

Hot helicopter using "Normal Alignment with Gyrocompassing":
After 1 turn, 5 minutes,   E	= ~100 meters	CW
After 2 turns, 10 minutes, SW	= 333 meters	CW
After 3 turns, 15 minutes, E	= 543 meters	CCW
After 4 turns, 20 minutes, NW	= 841 meters	CCW
After 5 turns, 25 minutes, E	= 1 021 meters	CW
After 6 turns, 30 minutes, SW	= 1 218 meters	CW
After 7 turns, 35 minutes, E	= 1 368 meters	CCW
After 8 turns, 40 minutes, NW	= 1 540 meters	CCW
Before 9 turns, 45 minutes 	= 1 660 meters
After 9 turns, 45 minutes, E	= 1 563 meters	CW

Cold helicopter using accelerated alignment done using APU power:
After 1 turn, 5 minutes,   E	= 732 meters	CW
After 2 turns, 10 minutes, SW	= 623 meters	CW
After 3 turns, 15 minutes, E	= 834 meters	CCW
After 4 turns, 20 minutes, NW	= 1 116 meters	CCW
After 5 turns, 25 minutes, E	= 1 273 meters	CW
After 6 turns, 30 minutes, SW	= 190 meters	CW
After 7 turns, 35 minutes, E	= 1 230 meters	CCW
After 8 turns, 40 minutes, NW	= 1 229 meters	CCW
Before 9 turns, 45 minutes	= 1 160 meters
After 9 turns, 45 minutes, E	= 1 529 meters	CW
After 10 turns, 50 minutes, SW	= 403 meters	CW
Before 11 turns, 55 minutes	= 414 meters
After 11 turns, 55 minutes, E	= 1 871 meters	CCW

And no, the drift going down from over 1 km to just a couple of hundred meters after turns 6 and 10 in the second test are not typos, I checked the numbers multiple times and they're correct as given by the game. (EDIT2: The difference was smaller at first, but every time the helicopter turned to SW, the error got smaller. Interesting.)

EDIT: Also noticed that the autopilot has SEVERE problems making left turns vs. right turns. After it has turned left, it then tries to straighten the helicopter out, but uses so much anti-torque to the right that the nose turns hard to the right while the helicopter is still banking left. This doesn't seem to happen after right hand turns.

Oh, and here's an image of what the flown path on the "accelerated alignment using APU power" flight looked like after about 50 minutes of flying:

Edited December 21, 2022 by jubuttib

[rrasfly2]

I’ll try to make a video showing the precision alignment technique I got to work for me. 2.3 km drift in 30 min is at the high end but still close to what the ED advertise drift rate is. It is interesting that some alignment techniques work for some people but not others though. As for the drift of 1.563km in 45 min in jubuttib’s second test that seems pretty good based on EDsh claimed 4km drift rate per hour which seems correct for a 1980 INU. What Direction I’m flying when the drift rate is decreased seems to be dependent on your initial alignment heading. 

[Er.tz]

I am having the same problems. I am new to dcs so i thought i was doing something wrong.Perhaps iam, but it is driving me mad. The PVI coords  and the Abris ones  no matter what procedure i chose or editing waypoints,  dont align on flight. Route mode  has a drift that sometimes  the abris cant recognice when I have passed a waypoint so doesnt update to the next one (even if the message sterpoint passed is showed).  Sometimes there is a message in the abris "CTD limit" that i have no idea what means and it seems is undocumented in the manual. DT is hell. If i use route mode before waypoint 1, the helicopter stays above the airfield trying to hover and so on. Is the HSI or the abris working as intended?

That drift is intended or a bug? Are the developers answering those questions?

I smell there is something wrong/bugged with the magnetic deviation that is impacting the INU aligment.

Edited January 10, 2023 by Er.tz

[Schlomo1933]

If you use the route mode to wp1 u really need to setup your heading with the compass. The magnetic setting is only recommended if no metallic objects are in the near of the shark or u do a inflight alligment. 
 

after inu alligment in fast or norm, - 1. put in the true heading into the Inu scale

2. Flip the gyro switch to the lower position 

3. flip it back to gyro

Done. 
 

3.1 Because U will need allign the heading again after a while ( mostly after 25 km— is buggy ) in flight,  turn in the magnetic variation into the inu scale. 
When a Realligning is needed:  you switch in flight the gyro switch into magnetic mode and then back .

i can recommend to make a fixpoint between every waypoint and make a hover  fix—- NOT A SHKVAL FIX (it’s also buggy). Till the unrealistic drift is fixed. 
(the last fixpoint should be short before the target area)

and fixpoint 3 and 4 are changed if u set them up in the missioneditor

-so u need to push fp4 for 3 and fp3 for 4

Edited January 10, 2023 by Schlomo1933

[Er.tz]

hace 19 horas, Schlomo1933 dijo:

If you use the route mode to wp1 u really need to setup your heading with the compass. The magnetic setting is only recommended if no metallic objects are in the near of the shark or u do a inflight alligment. 
 

after inu alligment in fast or norm, - 1. put in the true heading into the Inu scale

2. Flip the gyro switch to the lower position 

3. flip it back to gyro

Done. 
 

3.1 Because U will need allign the heading again after a while ( mostly after 25 km— is buggy ) in flight,  turn in the magnetic variation into the inu scale. 
When a Realligning is needed:  you switch in flight the gyro switch into magnetic mode and then back .

i can recommend to make a fixpoint between every waypoint and make a hover  fix—- NOT A SHKVAL FIX (it’s also buggy). Till the unrealistic drift is fixed. 
(the last fixpoint should be short before the target area)

and fixpoint 3 and 4 are changed if u set them up in the missioneditor

-so u need to push fp4 for 3 and fp3 for 4

 

Thank you for your answer.

"true heading" is the magnetic compass + 7? (in Caucasus map) and also the "track (T)" of the abris? What is the best form to know true north heading?

The HUD and HSI degrees are from inertial calculations? For example, at the start of the mission if i have done a precise INU aligment, the HUD caret mark  and HSI white  marks will mark true north degrees?

Edited January 11, 2023 by Er.tz

[Schlomo1933]

F10 map. Click on your aircraft , then it will show u you true heading in the down left box

[Tuuvas]

Or alternatively, LCtrl+Y while in the cockpit to get the info bar at the bottom to tell you your true heading