reported Artificial Horizon Intermittent

[-0303-]

 If plane didn't have caging don't add one.

[ldnz]

Yeah this is hopeless. I'm loving the V for Victory campaign, but the AH is really starting to annoy me. Such a fundamental instrument, and that campaign has plenty of clouds/nights/need for the damn AH.

Please @NineLine can you give this a nudge?

[-0303-]

On 9/23/2021 at 11:08 AM, scoobie said:

Secondly how it works in a real AI (fast forward to about 1:43):

Quoting a minor bit of and adding a real Spitfire example to scoobies's excellent post. Watch first 44 sec of this video to see a real Spitfire AI self correct as the instrument-on-table example above. Not a new video, just emphasizing the first minute. 15-44 sec:

 

 

Edited March 24, 2023 by -0303-

[Bozon]

Was the AH secretly fixed? 
I flew Reflected’s V for Victory campaign mission 1 yesterday - again… because in the last two attempts I flew into the fog, vertigo’ed in the white soup and crashed or broke the airplane in a dive - this is because the AH was completely off and was useless.

However, yesterday was a new attempt after the recent patch and the AH was showing close to the real horizon during the entire mission - and I was able to fly completely blind, instruments only, incl. a prolonged instruments-only IFR to Manston using the radio DF, until I broke out of the clouds over the English coast. 
 

The AH was close enough to true to a few degrees. This allows correcting pitch by zeroing out the RoC indicator, and correcting bank angle by freezing the DI compass, to get the AH “image” that I needed to maintain for level flight.

Anyone else noticed a difference since last patch? Or was I just lucky?

Edited April 3, 2023 by Bozon

[Hog_driver]

I tested it today with cold'n'dark startup and in level flight, it was 1-2 degrees off (roll). So, not great, not terrible.

PS. It's Reflected Simulations, not Enigma

Edited April 3, 2023 by Hog_driver

[Bozon]

3 hours ago, Hog_driver said:

PS. It's Reflected Simulations, not Enigma

 

Of course 

Corrected.

[-0303-]

If flown straight & level for minutes, it should not be off even a few degrees. It constantly self correct, it's how it works.

An unambiguous test is simple with any cageable Artificial Horizon. TF-51D.

Bank steeply, cage, uncage, roll level. Fly straight & level 10 minutes. Can use speedup. 10 minute should be more than enough for Artificial Horizon to self correct.

Edited April 3, 2023 by -0303-

[ldnz]

On 4/3/2023 at 5:44 PM, Bozon said:

Was the AH secretly fixed? 
I flew Reflected’s V for Victory campaign mission 1 yesterday - again… because in the last two attempts I flew into the fog, vertigo’ed in the white soup and crashed or broke the airplane in a dive - this is because the AH was completely off and was useless.

However, yesterday was a new attempt after the recent patch and the AH was showing close to the real horizon during the entire mission - and I was able to fly completely blind, instruments only, incl. a prolonged instruments-only IFR to Manston using the radio DF, until I broke out of the clouds over the English coast. 
 

The AH was close enough to true to a few degrees. This allows correcting pitch by zeroing out the RoC indicator, and correcting bank angle by freezing the DI compass, to get the AH “image” that I needed to maintain for level flight.

Anyone else noticed a difference since last patch? Or was I just lucky?

 

I was wondering that too. I've had no issues since the last patch. Yes it will go inaccurate when subject to continuous acceleration, but always returned to level

[Proper Charlie]

I'm still getting this problem, certainly on cold starts; not sure about hot starts (Multithread 2.8.4.39259).

You can see the artificial horizon is already significantly wrong a few minutes in, while still on the ground.

Track file attached.

MosquitoADIColdStart.trk

Edited April 23, 2023 by Proper Charlie

[Terry Dactil]

Yep. Same problem here on occasions.

However, I have found that increasing the revs to 1500 during warm-up speeds up the self errection process considerably.

(Probably increased suction that speeds up the gyro or something similar is modelled).

After that it seems to work OK for the rest of the flight.  I have not been getting the previous huge errord developing in level flight.

[Jshek]

I flew V for Victory mission 2 twice since I flew it the first time in MT, had no cockpit lights but did have a working AH for the entire mission. 
 

Today flew the mission in regular OB, had working cockpit lights but the AH was completely messed up.  Stayed straight and level from Manston to WP1 at Beachy head, the AH was a distaster. Reading at the bottom of the gauge and 30 degrees off. 
 

Flew around the convoy etc (don’t want to spoil it for anyone). Never levelled out. Flew straight and level back to Manston at the end of the mission, was 90 degrees to direction of flight…. 
 

So no cockpit lights in MT, non functioning AH in regular OB. 

[Proper Charlie]

I had a quick look in the Mosquito lua scripts, and found where you can increase the power of the suction pump.

\Mods\aircraft\MosquitoFBMkVI\Cockpit\Scripts\suctionpump_device.lua

suctionFactor = 70.0 -- Was originally 35.0

This worked for me - just played through V for Victory mission 2 myself, and had no problems with the horizon.

I also played through the track I shared above with this setting, and the problem went away there too.

Edited April 23, 2023 by Proper Charlie

[Hog_driver]

I just tried the Day Ranger mission in the campaign (latest OB, MT) and the AH worked well. I opened the mission in ME, didn't spend much time on the ground and taxied to the runway avoiding sharp turns and swerving (I think that may be a factor). I've also been told to "Fly straight, hold Boost 16+   2-3mins"

The AH worked well up to St. Omer, then I hit a tree like an idiot. Screenshots attached as a proof.

[Jshek]

39 minutes ago, Proper Charlie said:

I had a quick look in the Mosquito lua scripts, and found where you can increase the power of the suction pump.

\Mods\aircraft\MosquitoFBMkVI\Cockpit\Scripts\suctionpump_device.lua

suctionFactor = 70.0 -- Was originally 35.0

This worked for me - just played through V for Victory mission 2 myself, and had no problems with the horizon.

I also played through the track I shared above with this setting, and the problem went away there too.

 

Awesome! I’ll try this! Thanks!!

[Jshek]

7 hours ago, Proper Charlie said:

I had a quick look in the Mosquito lua scripts, and found where you can increase the power of the suction pump.

\Mods\aircraft\MosquitoFBMkVI\Cockpit\Scripts\suctionpump_device.lua

suctionFactor = 70.0 -- Was originally 35.0

This worked for me - just played through V for Victory mission 2 myself, and had no problems with the horizon.

I also played through the track I shared above with this setting, and the problem went away there too.

 

This seems to have worked to be honest. Was only able to test cold start, take off then flew around for 5 mins. 
 

Did some serious aerobatics just to see if I could get it to skew. Everything seemed good. 
 

Thanks a lot for the suggestion. I’m assuming it wouldn’t be a bad idea to adjust the Spit as well. Unless ED decides to make the adjustments so we won’t have to after each update. 
 

Thanks again!

[-0303-]

Video. Sound on. Real P-51 gyro horizon spinning up and self correcting within 12 seconds.

https://www.reddit.com/r/WWIIplanes/comments/1375rj4/a_cool_little_artifact/

Note. He states erroneously "AN5726". In every drawing I've ever seen it's military designation "AN5736".

Edited November 5, 2023 by -0303-

[RodBorza]

It seems to be working after the update from June 8th. I flew with it a little bit and the Artificial Horizon seems to be working. Can someone confirm that?

Enviado de meu SM-A127M usando o Tapatalk

[Jshek]

On 6/9/2023 at 1:40 AM, RodBorza said:

It seems to be working after the update from June 8th. I flew with it a little bit and the Artificial Horizon seems to be working. Can someone confirm that?

Enviado de meu SM-A127M usando o Tapatalk
 

Flew 2 V for Victory missions in a row in 2.9 MT and it’s still an issue. I’m not sure how this hasn’t been fixed yet… 

@NineLine 

Both missions flown today were from full cold start. AH is fine for 30 seconds then the rest of the mission it’s no good. I was flying the V for Victory mission 4 & 6 (high power settings the entire flight at 30 000ft for high alt photo recon) nothing I did was able to fix the AH. 

As @Proper Charlie mentioned, maybe it’s an issue with the in game suction model? I will adjust mine from 35 to 70 but why do this everytime the simulator is updated. Why not fix the model for the Spit and Mossie which both don’t have manual pull to center artificial horizons. 
 

I cannot confirm whether it happens during any other scenario besides a cold start but all of my missions are always cold start which is exactly what I like. 

Also can’t keep the engines at 1200rpm on the ground while taxiing, that’s much too fast 900-1000 is a perfect rpm for taxi which seems to be an issue for the suction on the aircraft and ruining the rest of the mission with a bad AH. 
 

Thank you. 
 

 

 

 

[kablamoman]

I don't have the mosquito, but the same problem certainly seems to be pervasive throughout the sim in all of ED's modeling of traditional (steam gauge) artificial horizons.

Here's a post I made with regards to the one in the mustang that I think outlines the issue fairly clearly: 

Hopefully ED is able to revisit how they've modelled these instruments.

[Jshek]

On 11/4/2023 at 5:53 PM, Jshek said:

Flew 2 V for Victory missions in a row in 2.9 MT and it’s still an issue. I’m not sure how this hasn’t been fixed yet… 

@NineLine 

Both missions flown today were from full cold start. AH is fine for 30 seconds then the rest of the mission it’s no good. I was flying the V for Victory mission 4 & 6 (high power settings the entire flight at 30 000ft for high alt photo recon) nothing I did was able to fix the AH. 

As @Proper Charlie mentioned, maybe it’s an issue with the in game suction model? I will adjust mine from 35 to 70 but why do this everytime the simulator is updated. Why not fix the model for the Spit and Mossie which both don’t have manual pull to center artificial horizons. 
 

I cannot confirm whether it happens during any other scenario besides a cold start but all of my missions are always cold start which is exactly what I like. 

Also can’t keep the engines at 1200rpm on the ground while taxiing, that’s much too fast 900-1000 is a perfect rpm for taxi which seems to be an issue for the suction on the aircraft and ruining the rest of the mission with a bad AH. 
 

Thank you. 
 

 

 

 

@NineLine Track attached, freshly made today. Cold and Dark start, V for Victory Mission 6. Taxi to runway, take off, right hand turn, climb, low cruise. Does not align. Did a few maneuvers to break it loose, it still goes back to its left wing low position. Fly with AH aligned, in a left hand turn. Go to a cruise attitude and power setting, still holds left wing low attitude. Watch the AH as I taxi, especially pre run up at the end of the runway and as I taxi into position for takeoff. 

Thank you,

Mosquito Artificial Horizon.trk

[-0303-]

On 11/5/2023 at 12:53 AM, Jshek said:

Also can’t keep the engines at 1200rpm on the ground while taxiing, that’s much too fast 900-1000 is a perfect rpm for taxi which seems to be an issue for the suction on the aircraft and ruining the rest of the mission with a bad AH.

Obviously it's ridiculous to have to maintain a high rpm on the ground in order for the Artificial Horizon (AH) to work correctly in the air. If it works, it's just a band aid  to prevent AH going wrong even before the wheels leave the ground...

~~~

~~~

Ye olde classic or "steam gauge" AH have two functional attributes, the lack of either makes it useless. DCS lacks the second. I've run tests for hours without self correction (Spitfire and/or P-51).

1) A spinning gyro to keep it rigid in space.

2) A self correcting mechanism that continuously corrects it towards the gravity vector (pendulous vanes).

A misconception I struggled a bit with: Centrifugal force of continuous turn will bias the AH. No it won't. A 180 turn will bias it. An additional 180 to full 360 will bias in the opposite direction, canceling the bias out.

~~~

~~~

How to test

A proper test is simple. Whenever you notice the AH is misaligned just fly straight and level for 10 minutes (enough for a real Spitfire). If the AH hasn't  substantially or completely corrected itself, it's wrongly modeled. Using speedup (CTRL Z) is fine.

A plane with cageable AH is quick and easy to deliberately misalign for testing purposes: Bank, cage, uncage, roll level. An uncageable AH is more work. Either fly around for 15 - 60 min until it's misaligned or ...

A quick (somewhat) and dirty test. Spawn runway, shut engine off / on until AH is misaligned "enough". Four cyclings is often enough for 30 or even 60 degrees off. Now, lock brakes, open all cooling flaps, set RPM enough for good suction, use speedup, run and watch for 10 minutes or hours.  Any properly functioning AH will correct itself within minutes. See any number of real life videos.

~~~

~~~

As Kablamoman says, this seemingly is a problem with every DCS "steam gauge" plane. Just re- tested the Spitfire. Engine cyclings took AH to 45 degrees off. Running two (2) hours, it did not self correct, it got worse, nearly 90 degrees off (see pic). RPM held at 2000, but engine broke and quit at two hours, hence zero RPM in pic. Test start at 8:36, note clock left 10:36 'ish.

I got the Spitfire reported 2 years ago.

spitfire AH test 2 hours from 45 to near 90.trk

Edited November 10, 2023 by -0303-

[Jshek]

41 minutes ago, -0303- said:

Obviously it's ridiculous to have to maintain a high rpm on the ground in order for the Artificial Horizon (AH) to work correctly in the air. If it works, it's just a band aid  to prevent AH going wrong even before the wheels leave the ground...

~~~

~~~

Ye olde classic or "steam gauge" AH have two functional attributes, the lack of either makes it useless. DCS lacks the second. I've run tests for hours without self correction (Spitfire and/or P-51).

1) A spinning gyro to keep it rigid in space.

2) A self correcting mechanism that continuously corrects it towards the gravity vector (pendulous vanes).

A misconception I struggled a bit with: Centrifugal force of continuous turn will bias the AH. No it won't. A 180 turn will bias it. An additional 180 to full 360 will bias in the opposite direction, canceling the bias out.

~~~

~~~

How to test

A proper test is simple. Whenever you notice the AH is misaligned just fly straight and level for 10 minutes (enough for a real Spitfire). If the AH hasn't hasn't substantially or completely corrected itself, it's wrongly modeled. Using speedup (CTRL Z) is fine.

A plane with cageable AH is quick and easy to deliberately misalign for testing purposes: Bank, cage, uncage, roll level. An uncageable AH is more work. Either fly around for 15 - 60 min until it's misaligned or ...

A quick (somewhat) and dirty test. Spawn runway, shut engine off / on until AH is misaligned "enough". Four cyclings is often enough for 30 or even 60 degrees off. Now, lock brakes, open all cooling flaps, set RPM enough for good suction, use speedup, run and watch for 10 minutes or hours.  Any properly functioning AH will correct itself within minutes. See any number of real life videos.

~~~

~~~

As Kablamoman says, this seemingly is a problem with every DCS "steam gauge" plane. Just re- tested the Spitfire. Engine cyclings took AH to 45 degrees off. Running two (2) hours, it did not self correct, it got worse, nearly 90 degrees off (see pic). RPM held at 2000, but engine broke and quit at two hours, hence zero RPM in pic. Test start at 8:36, note clock left 10:36 'ish.

I got the Spitfire reported 2 years ago.

spitfire AH test 2 hours from 45 to near 90.trk 6.2 MB · 0 downloads

 

Tried high rpm on the ground, 1200rpm minimum. Doesn’t work. 
 

If you try the Mossie, note that the AH will bank left or right as you turn on the ground…definitely not working correctly. 

[Lau]

So, what’s the feeling of flying the Mossie in pitch black and clouds (Full IMC)?

Scary !

Currently and as simulated the attitude indicator is nothing but a useless piece of equipment that is never erected properly. Oh you can get a hang of it by taking precise mental pictures of its position at the exact moment when the gyro compass stops turning, but that would be the best you will get out of it and that of course will have to be a process that you will have to repeat over and over again for the duration of the flight. The pleasure of experiencing the above is only true if you make it after the take off with a tilted horizon, the landing lights being a better indication of what is going on in front of the aircraft, if you see anything just PULL!

Was it that bad IRL?

Best,

Lau

[Art-J]

6 hours ago, Lau said:

Was it that bad IRL?

It wasn't. Oldschool gyro AHs have been modelled incorrectly in DCS for years, however, and that issue carries over to module after module after module after module etc.

There have been many threads reporting the problem, these ones for example provide more info:

 

[Lau]

Hi @Art-J,

Thanks for the feedback, looking forward to read what @NineLine and @BIGNEWY have to say about this one.

Best,

Lau

Edited April 2, 2024 by Lau