You are in IMC, when *everything* fails! Now what?

"Julian Scarfe" wrote:
"As for rudders - I teach - rudder to STOP yaw. Not to pick the wing up...."

We're starting to diverge a bit here in what we're talking about. The
original thread had to do with what you do in IMC with no gyros. The
ONLY information you have is what you can get from thw wet compass and
(as the argument evolved) a moving map GPS giving you a course line.

All you can deduce from a GPS moving map is heading, and even that, only
if you assume that heading more or less equals course (certainly
d(heading)/dt is very close to d(course)/dt for small values of heading
change. So, as I said, you use the rudders to correct deviations off
the GPS course line. This corresponds exactly to stopping yaw. In IMC
with no direct roll or bank information, picking a wing up isn't even
possible, except as a by-product of holding heading.

Perhaps we're just arguing different arguments. I'm assuming that you
just lost your entire panel in IMC, in which case it's not unlikely that
you've entered an unusual attitude by the time you figured out you had a
problem. Maybe you're in a steep spiral. Maybe you're close to stall,
or have stalled already. Maybe you're in a spin. Using rudder to
counter course change indicated on the GPS is the only thing I can think
of which will work in all those cases.

As for "I resign myself to the fact that at best we will make constant
small s-turns all the way", If I've lost all my instrumnts in IMC, I'll
be very happy if the worst I do is make small s-turns until I get into
visual conditions.

I certainly agree that once the plane is on an even keel and stable,
using coordinated aileron and rudder is superior to just using rudder.
It would be foolish to argue otherwise. But if you're not sure if
you're upside down or not, I'd stay away from the ailerons.

"Gary L. Drescher" writes:

Sorry, I wasn't very clear as to what I was proposing. I wasn't
suggesting to let the pendulum hang down to get bank information.
Rather, the idea is to let it swing from side to side to get yaw
information. As the plane changes its heading, the pendulum's
direction of motion changes relative to the plane.

It would have a very significant lag, since pendulums tend to keep
going for a long time with a single push. Were you looking for
something bigger than the ball in the TC?


All the best,


David

--
David Megginson, , http://www.megginson.com/

Sydney Hoeltzli wrote
Julian Scarfe wrote:
Well yes, but solving the roll problem and getting the wings level will
solve the pitch problem. Solving the pitch problem doesn't solve the roll
problem.

I'll have to try this again next time I'm out playing, but I guess I'm
trying to say (at least in my plane) I don't think they're separable.
That is to say it's not enough to *get* the wings level, if the pitch
is oscillating a wing will drop. (this is a practical observation)

I have another practical observation. If an unusual attitude has
developed, then even regaining perfect control over roll may not
salvage the situation. I tried this experiment once (link and text at
end of post).

Pitch stability is one of those terms that doesn't necessarily have
any meaning outside of control theory. Inside of control theory it's
too dull to think

No, seriously - positive dynamic pitch stability, as it's generally
understood, basically means that if you have the plane trimmed out to
a certain speed, and then make a step change to that speed, the plane
will undergo some oscillations but eventually return to the trim
speed, within a certain tolerance. However, it's pretty obvious that
it won't do that for arbitrarily large step changes. The magnitude of
the step change is defined by certification rules - and if you've
allowed the speed to diverge outside those limits, then even if you
bring roll under control you're still a test pilot.

Michael

http://groups.google.com/groups?hl=e...3D30%26hl%3Den

One piece of advice that non-instrument pilots are always given should
they blunder into IMC inadvertently is to turn on the autopilot if
they have one. The reason is obvious - an autopilot, assuming it
works properly, will maintain wings level in the clouds whereas an
untrained person may not. Makes sense, right?

I've never owned a plane with an autopilot, and when I did
inadvertently blunder into IMC I kept the plane under control and flew
RADAR vectors based on the training all private pilots get in basic
attitude instrument flying. But what if I had an autopilot?

Recently I installed an Autocontrol III in my Twin Comanche. I tested
it in both wing level mode and heading hold mode (it has no altitude
hold) and I was satisfied that it would be a good tool for reducing my
workload at night or in IMC. But would it help the untrained pilot in
IMC?

So I tested it under nearly ideal conditions. I had the plane loaded
about 400 lbs under gross, and with the cg toward the forward part of
the envelope for good stability. I slowed the plane to about 120 kts,
trimmed it up, and released the controls at a safe altitude. I wanted
to see what would happen. For a minute or so, the plane wandered in
heading but continued to hold altitude (plus or minus a bit - there
were some bumps) and wings mostly level. Eventually, of course, it
got a wing lifted by a bump, and off it went into a graveyard spiral.
At 50 degrees of bank, the speed had built to about 160 kts and the
descent rate was over 1000 fpm. I decided that this was a classic
nose-low unusual attitude. So I engaged the autopilot in wing leveler
mode.

Immediately and smoothly, the autopilot leveled the wings. So far, so
good. The airplane immediately began to climb and slow down. "Wow,
this actually works!" I thought. Not so fast. The airplane blew
right through its trim speed, and the nose kept coming up. The wings
did stay approximately level, and the heading stayed pretty constant,
but airspeed bled off. I put my hand on the throttles, sensing what
was coming. Sure enough, I was slowing right through red line. The
power setting I used was low enough for a power-on stall, and the
altitude was high enough, so I kept my feet lightly on the rudders,
ready to recover. The plane slowed through 65 mph (power-off stall is
over 70) and then nosed down and dove. At the bottom of the dive, I
think it may have been doing even more than 160 kts - but then it
began to climb again. At the top, I felt the bite of the stall, but
before I could take action the nose dropped again. Still the wings
were level.

As the nose began to drop again, I lost my nerve. As the nose crossed
the horizon, I held the yoke for a few seconds to hold that attitude.
Then I let go. The plane flew, holding wings level and maintaining
altitude reasonably well in the bumps.

Maybe it's a good thing I had the instrument rating and the confidence
to use it before I had an autopilot...

Michael

"Michael" wrote in message
m...
"Gary L. Drescher" wrote
Sorry, I wasn't very clear as to what I was proposing. I wasn't
suggesting
to let the pendulum hang down to get bank information. Rather, the idea
is
to let it swing from side to side to get yaw information. As the plane
changes its heading, the pendulum's direction of motion changes relative
to
the plane.

Sounds great. How is that different from the ball (inclinometer)?

Two differences. First, the ball is not given a shove to keep it
oscillating back and forth even when the plane is going straight (any such
oscillation damps out almost instantly). Second, the ball is constrained to
move in only one plane. The idea of the pendulum is to start it swinging
side to side (say, east-west if you're heading north) but let it move
freely; then, if the plane yaws, the pendulum is still swinging east-west,
so the yaw is detectable. At least in principle.

--Gary


Michael

There shouldn't be

Many people feel there is no need to "pick up a wing". Just lower the
AOA and then return to level flight using coordinated controls.

This is what I teach, and is what is in the FAA's Airplane Flying
Handbook.

In a spin you use the rudder to stop the rotation

We were talking about stalls.

Greg Esres wrote:
In a spin you use the rudder to stop the rotation

We were talking about stalls.


You may have noticed that often a wing drops in a stall; beginning of
the incipit phase of a spin. Use rudder to stop the rotation (wing
dropping), elevator to lower the AOA...

I was basiclly agreeing with the earlier post.

Use rudder to stop the rotation (wing dropping),

Isn't necessary. No stall, no spin.

You should read the Airplane Flying Handbook for the proper recovery
procedure, rather than what some instructor taught you.

Greg Esres wrote:
Use rudder to stop the rotation (wing dropping),

Isn't necessary. No stall, no spin.

You should read the Airplane Flying Handbook for the proper recovery
procedure, rather than what some instructor taught you.

Surely you're not saying that an FAA document is the be-all and
end-all when it comes to understanding aerodynamics, are you?

(Teacherjh) wrote
One piece of advice that non-instrument pilots are always given should
they blunder into IMC inadvertently is to turn on the autopilot if
they have one. The reason is obvious ...

I'm not sure I agree.

With what? That it's a good idea? Obviously it's not always a good
idea, as my experiment showed. That it's common advice? I think it's
common enough that most people have heard it.

Michael