Wonderful attention to detail!!!

I decided to see what the high-alt performance was like. It's pretty amazing - 1940s equipment going up to 41,000 feet and still showing a few hundred fpm.

Most of the time up there was spent in a lazy right-hand turn. After levelling out along the way, I saw something I have never yet seen modelled in any sim. See if you can spot it in the attached screenshot!

its been sometime since i played " i spy with my little eye". do tell !

Wow, that is detailed alright. Nice spot.

Its in the spoiler Ali. Took me a sec as I haven't heard the effect called "self rectifying" before, in my parts we call it gyro precession.

Hehe, yeah, and it's pretty annoying. :) Especially if you don't trim perfectly.

If I was a coder writing this stuff, I'd be thinking "is anyone actually going to notice this?". I am glad people do, it must be rewarding for the coders.

Kaiza,

it is not precession per se, even if the mechanism makes use of precession.

You apply torque to the gyro assembly in proportion to the local field of gravity, thus causing it to precess towards vertical and keeping it useful as a reference in spite of unwanted precession due to e g gimbal friction and other disturbance sources.

It can be done in rather ingenious ways, e g through ball bearings on a rotating plate with slots they fall into causing an imbalance 90 degrees off from the tilt of the assembly from the local perceived gravity.

The downside is, of course, that in a prolonged coordinated turn the AI will align itself with the vertical axis of the aircraft - as seen above. That's one reason we hold in racetrack patterns rather than in orbits.

(Edit: Deleted hogwash about this happening in climbs.)

@Sobek: It's also pretty annoying if you blunder into a cloud with your AI still out of whack to an unknown degree, as I did soon after taking the above screenshot. It can be noted that

a) It is possible to fly ball and paddle in DCS: P-51D

b) I suck at it. :D

Cheers,

Fred

Kaiza,

it is not precession per se, even if the mechanism makes use of precession.

 

You apply torque to the gyro assembly in proportion to the local field of gravity, thus causing it to precess towards vertical and keeping it useful as a reference in spite of unwanted precession due to e g gimbal friction and other disturbance sources.

 

It can be done in rather ingenious ways, e g through ball bearings on a rotating plate with slots they fall into causing an imbalance 90 degrees off from the tilt of the assembly from the local perceived gravity.

 

The downside is, of course, that in a prolonged coordinated turn the AI will align itself with the vertical axis of the aircraft - as seen above. That's one reason we hold in racetrack patterns rather than in orbits.

 

You will more commonly see this after a prolonged climb, where the AI will move towards indicating level during the climb and indicate a nose-down attitude once you level off.

 

 

@Sobek: It's also pretty annoying if you blunder into a cloud with your AI still out of whack to an unknown degree, as I did soon after taking the above screenshot. It can be noted that

 

a) It is possible to fly ball and paddle in DCS: P-51D

b) I suck at it. :D

 

Cheers,

Fred

Well you learn something new every day - thanks for that.

Nate

Heh and I thought it was a bug :D

Just wasn't sure what excactly was causing it.

Thanks for the explanation :thumbup:

What?

What?

+1

TOTALLY at sea as regards the above. Could you please repeat in English :D

Haha, allright.

The attitude indicators in aircraft will, like any gyro, drift over time. To counter this, mechanisms are included which turn the attitude indicators to align with local gravity. Usually aircraft fly level most of the time, so this works well. The exception is when you remain in a turn for a long period of time, when the gyro will start aligning itself with the aircraft rather than the horizon. Stay in the turn long enough, and the gyro will indicate level - only to indicate a turn the other way when you return to flying straight ahead.

This is usually not modelled in simulators, as it is an effect rarely seen.

Ok, got it, thanks for the (re-)explanation Effte. Rep inbound...

Oh. :D

I'm rally looking forward when all motor settings/features will become alive!

e.g. not to be able to operate the merlin at max for long time.

I bet there will be more reply like this:

Heh and I thought it was a bug

Just wasn't sure what excactly was causing it.

Thanks for the explanation

The OP has explained this effect so well that I hesitate to doubt him. I can only say that I have never noticed this behavior in gyro equipped plane before. Having said that, the oldest instrument equipped airplane I have ever flown was a 1965 Beechcraft Baron. It had a black-on-black AI and so LOOKED like the horizon in the P-51 but perhaps it functioned differently. What it definitely did not do was the behavior described above. We commonly taught steep turns with 45-60 degrees of bank and a hard reversal on the original heading after 360 degrees of turn. The AI was always solid. And the reason we do racetrack patterns is because there is no way to circle along a fixed course.

Edit: I think, Fred, that you are stating that one must be in a sustained constant turn for a very long time before the gyro "corrects" itself in this manner. Something that a pilot flying on instruments would never do since there is never a need to turn past 270 degrees. If true that is quite something to learn after holding a CFII for 25 years.

Smokin,

affirm. You will never notice this in normal flight operations. You'd have to spend quite some time circling. It is very slow, as it only has to be rapid enough to prevent the drift from friction in bearings, gimbals etc from creating a significant error.

You have had bad students, as apparently not one of them has questioned how the AI figures out which way is down in 25 years. This in spite of the space-fixed direction of 'down' changing constantly due to the movement of the aircraft and the rotation of the earth. ;)

Well my instrument/ME instructor career ended many years back. But I still hold the tickets to remind myself that life is safer now. No my students were rarely bad. They just shared my lack of interest in the magic behind some of the systems. To us (use your Tarzan accent here): Suction = good, vacuum failure = bad. I was a kid then. Now I am always happy to learn more.

Me not have Tarzan accent! Me fluent! :D

I'm so jealous of the cost of tickets stateside. Here, once you have a couple of ratings just 'hanging on to the tickets' will set you back thousands of dollars every year. The administration has to 'cover the cost', they say. Expensive ink in those stamps.

At the same time, a colleague of mine decided to renew his FAA CPL/ME/IR while we were at FAA HQ in OKC. Two bucks. Two friggin' bucks! The postage to send him the paperwork was more than two bucks!

For the latest in Euro side lunacy, check out what it'd take for you to be allowed to fly here. Un-be-fricking-lievable, and I deeply apologize on behalf of our policy-breakers.

A pleasure to read!

Rep all around :D

This was already modelled in warthog and black shark wasn't it? At least gyro drift itself was, if the EGI ever failed then the AI would start getting errors very quickly in both pitch and roll.

i wonder if this is why my SAI in blackshark is always off level.