Re: Backup gyros - which do you trust?

[Dan Luke]

"Steve House" wrote:
> I've been reading with interest the several threads where a number of
people
> have strongly pointed out the advantages of a backup electric AI to
supplant
> a vacum driven main AI. But I'm reminded of the saying "A man with a good
> watch always knows what time it is. A man with two watches is never
sure."

This is a very interesting issue, to me. Reading the records of IMC
loss-of-control accidents is very unsettling to this single pilot IFR flyer
because of the cases where there *was* backup attitude instrumentation
available. Even when there wasn't, the pilots usually had at least the turn
coordinator to help keep the aircraft upright. It is too simple to chalk up
all these accidents simply to lack of proficiency. There is something else
going on - some human factors issue that has not been properly identified. I
suspect it may be related to task saturation. If so, instrument panel
clutter could be a contributing factor.

> So I'm toodling along in IMC with no outside horizon reference and I see
my
> two AIs don't agree with each other. How do I determine which to trust?
If
> I had a third, I could go with a 2 of 3 voting strategy of course, but
with
> only two, what do you do to decide which is operating properly and which
one
> has faulted? Obviously I can look for consistency with other
instruments -
> does my DG or Turn indicator show I'm turning, does the VSI show a climb
or
> descent - but what would be the best strategy given the various ways
vacuum
> or electric driven instruments can fail?

My strategy is to include a yoke-mounted GPS displaying a synthetic HSI in
my scan. This works wonderfully well in training, but I am not sure how well
I would do in a real situation where my AI suffered a gradual failure.
--
Dan
C172RG at BFM

[Ray Andraka]

There are several things you can add to help with the identification, In my
plane I have a low vacuum warning light (part of the precise flight backup)
mounted between the AI and DG. The AI is one of the sigmatec ones with a vacuum
flag, so that if vacuum is lost in the instrument but not in the system I still
know about it right away. These warnings cover identification of the more
common cause of loss of the AI. The other failure mode would be failure of the
gyro, in which case I don't believe you get the insidious gradual spin-down like
you do with loss of vacuum. I also fly with the GPS on the HSI page to offer
yet another source of redundancy.

Personally, I think the instrument scan typically taught relies too heavily on
the AI given its relatively low reliability. Unfortunately, the alternative is
a scan that works a bit more like a partial panel scan using the AI as
supporting, not primary. Such a scan is much harder to master and requires
considerable finesse to keep from chasing the needles. It is not one I would
expect to be able to teach someone just learning to fly by instruments.


--
--Ray Andraka, P.E.
President, the Andraka Consulting Group, Inc.
401/884-7930 Fax 401/884-7950
email
http://www.andraka.com

"They that give up essential liberty to obtain a little
temporary safety deserve neither liberty nor safety."
-Benjamin Franklin, 1759

[Sydney Hoeltzli]

Dan Luke wrote:

> This is a very interesting issue, to me. Reading the records of IMC
> loss-of-control accidents is very unsettling to this single pilot IFR flyer
> because of the cases where there *was* backup attitude instrumentation
> available. Even when there wasn't, the pilots usually had at least the turn
> coordinator to help keep the aircraft upright. It is too simple to chalk up
> all these accidents simply to lack of proficiency. There is something else
> going on - some human factors issue that has not been properly identified.

Concur. I, too, don't think it's entirely lack of proficiency. I think
there are pilots who have training and proficiency, who, in the words
of my CFI, "ought to be able to do it", who don't. And clearly backup
AI is no panacea.

>> Obviously I can look for consistency with other instruments -
>>does my DG or Turn indicator show I'm turning, does the VSI show a climb
>>or descent - but what would be the best strategy given the various ways
>>vacuum or electric driven instruments can fail?

Having an inventive CFI who has little habits like mind-f***ing
me into doubting my AI while palming the TC fuse, I think the best
strategy is delimited above. *Instrument cross check is essential*

Rod Machado's "Instrument Pilot Survival Manual" delineates something I
haven't seen elsewhere:

Turn triangle of agreement: AI, TC, compass
Pitch triangle of agreement: AI, VSI, alt static on/off

The point is to deliberately cross-check instruments
which depend upon independent power sources.

The problem (for me anyway) in training is that my compass is
mounted on the windshield bow and it's impossible to keep it
in my scan in VMC under the hood w/out extensive "cheating".

I also think Machado's under-utilizes ASI and hearing. I
think the reasoning is that there are three sources of ASI
failure and only two for VSI, one of which alt static
eliminates. But when forced to fly instruments without static
instruments, I found hearing was a fairly precise means of
pitch control (at constant power for a fixed-pitch prop)

Interested to see what others say: this topic should elicit a
lot of opinions.

Cheers,
Sydney

[Bob Gardner]

You have three sources of bank information in a typical panel: the attitude
indicator, the turn coordinator, and the heading indicator. If two agree and
the third does not, it is faulty. Add a fourth source and it makes
elimination that much easier.

Bob Gardner

"Steve House" > wrote in message
...
> I've been reading with interest the several threads where a number of
people
> have strongly pointed out the advantages of a backup electric AI to
supplant
> a vacum driven main AI. But I'm reminded of the saying "A man with a good
> watch always knows what time it is. A man with two watches is never
sure."
> So I'm toodling along in IMC with no outside horizon reference and I see
my
> two AIs don't agree with each other. How do I determine which to trust?
If
> I had a third, I could go with a 2 of 3 voting strategy of course, but
with
> only two, what do you do to decide which is operating properly and which
one
> has faulted? Obviously I can look for consistency with other
instruments -
> does my DG or Turn indicator show I'm turning, does the VSI show a climb
or
> descent - but what would be the best strategy given the various ways
vacuum
> or electric driven instruments can fail?
>
>

[James Robinson]

Dan Luke wrote:
>
> It is too simple to chalk up all these accidents simply to lack of
> proficiency.

I was just reading the NTSB report of the King Air that crashed in
Colorado, attributed to spatial disorientation after a partial panel
failure. It seems representative of the problem. The facts are
chilling:

- IMC, alt. 23,200 ft.
- Two person cockpit.
- Experienced pilot - 5117 hours total, 2520 in type.
- Partial panel loss due to AC power failure.
- Failure immediately indicated by flags on affected instruments.
- Remaining instruments, powered by vacuum:
Left - airspeed, turn/slip,
Right - airspeed, turn/slip, altimeter, attitude.
- Aircraft began gently increasing turn within one minute of failure.
- Time between instrument loss and impact - one minute, 33 seconds
- Flight path consistent with graveyard spiral

http://www.ntsb.gov/publictn/2003/AAR0301.pdf

> There is something else going on - some human factors issue that
> has not been properly identified. I suspect it may be related to
> task saturation. If so, instrument panel clutter could be a
> contributing factor.

One comment in the report was that the pilot might have had a tendency
to focus on a single problem, and mot paid attention to other things.
He could have been trying to troubleshoot the electrical problem, and
not handed control over to the copilot, who would have had a better view
of the remaining functional instruments.

In any event, it is amazing how quickly the pilot lost control of the
aircraft, considering how this should have been fairly routine: If an AC
inverter had failed, then the changeover to the remaining inverter is
accomplished with a simple flip of a switch, and should have been almost
a reflexive action. The failure would have been immediately obvious, so
it wasn't one of those insidious failures that people don't notice at
first. An experienced IFR pilot should have been aware of the need to
maintain attitude and yet lost control almost immediately. In reading
the report, it seems like such an avoidable accident, yet...

[Darrell]

The presence of two attitude indicators is especially valuable when they
disagree. That disagreement will direct your attention to the needle/ball
and basic flight instruments to help determine which one is correct. With a
single AI you could more easily follow a gyro error without noticing a
difference in the other basic instruments until it was too late.

--

Darrell R. Schmidt

B-58 Hustler History: http://members.cox.net/dschmidt1/


"Steve House" > wrote in message
...
> I've been reading with interest the several threads where a number of
people
> have strongly pointed out the advantages of a backup electric AI to
supplant
> a vacum driven main AI. But I'm reminded of the saying "A man with a good
> watch always knows what time it is. A man with two watches is never
sure."
> So I'm toodling along in IMC with no outside horizon reference and I see
my
> two AIs don't agree with each other. How do I determine which to trust?
If
> I had a third, I could go with a 2 of 3 voting strategy of course, but
with
> only two, what do you do to decide which is operating properly and which
one
> has faulted? Obviously I can look for consistency with other
instruments -
> does my DG or Turn indicator show I'm turning, does the VSI show a climb
or
> descent - but what would be the best strategy given the various ways
vacuum
> or electric driven instruments can fail?
>
>

[Richard Kaplan]

"Ray Andraka" > wrote in message
...

> Good reason not to depart in low IMC I guess. I agree, that the AI is a
weak
> link and carries with it some risk. While nice to have, cost and panel
space
> make the back up AI difficult in some aircraft.

I think it is a matter of priorities... in an airplane which is IFR
certified, it is hard to believe there is not some space or economic
compromise which could not be made if a pilot felt this were an important
enough issue. I know I will get diagreement on this as always, but I think
the electric AI comes first before an IFR GPS. Even C152s sometimes have
Garmin 430s/530s nowadays; an electric AI would make much more sense IMHO.

As an even better solution, Hal Sheevers of Sporty's has for quite some time
been lobbying the FAA to permit an electric AI to replace a turn
coordinator... it does not seem as if the issue is getting very far with the
FAA, but I do think that would be a very good compromise if we started to
see electric AIs installed in place of the turn coordinator on planes where
panel space is tight.


--
Richard Kaplan, CFII

www.flyimc.com

[Ray Andraka]

The problem isn't just finding a home for it, it is finding a home for it that
is in a spot where it will be in the pilot's primary scan. I don't think it
would do much good on the other side of the panel where it might only be
referenced once a minute or less. I also agree that an IFR GPS is not
necessarily a high priority. Frankly, that (the GPS) is a lot of money for very
little added capability. In the case of a GPS/Nav/Comm, it also introduces a
single point of failure for all of the electronic nav gear with no back-up other
than what might be in your flight bag.

It is a shame that many of the simulators do not fail the AI the in the gradual
way it fails for real. I think training with realistic failures is the best way
to be able to recognize the failure (some of the accident reports indicate even
with redundant AI's recognition is not guaranteed). Unfortunately, the failures
can't be realistically simulated in the aircraft (at least not without an
illegal mod to the vacuum system), so ground based simulators must fill in
there. I've hear that some of the high end simulators such as the Frasca's do a
realistic AI fail, but the PCATDs I've played with all just pop from working to
tipped over instantaneously. Any of the current breed of PCATD's do any
better? Mine is a really old version of Elite, from about 1995 or so.

Richard Kaplan wrote:

> "Ray Andraka" > wrote in message
> ...
>
> > Good reason not to depart in low IMC I guess. I agree, that the AI is a
> weak
> > link and carries with it some risk. While nice to have, cost and panel
> space
> > make the back up AI difficult in some aircraft.
>
> I think it is a matter of priorities... in an airplane which is IFR
> certified, it is hard to believe there is not some space or economic
> compromise which could not be made if a pilot felt this were an important
> enough issue. I know I will get diagreement on this as always, but I think
> the electric AI comes first before an IFR GPS. Even C152s sometimes have
> Garmin 430s/530s nowadays; an electric AI would make much more sense IMHO.
>
> As an even better solution, Hal Sheevers of Sporty's has for quite some time
> been lobbying the FAA to permit an electric AI to replace a turn
> coordinator... it does not seem as if the issue is getting very far with the
> FAA, but I do think that would be a very good compromise if we started to
> see electric AIs installed in place of the turn coordinator on planes where
> panel space is tight.
>
> --
> Richard Kaplan, CFII
>
> www.flyimc.com

--
--Ray Andraka, P.E.
President, the Andraka Consulting Group, Inc.
401/884-7930 Fax 401/884-7950
email
http://www.andraka.com

"They that give up essential liberty to obtain a little
temporary safety deserve neither liberty nor safety."
-Benjamin Franklin, 1759

[Richard Kaplan]

"Ray Andraka" > wrote in message
...

> The problem isn't just finding a home for it, it is finding a home for it
that
> is in a spot where it will be in the pilot's primary scan. I don't think
it

Put it where the turn coordinator is located and the put the turn
coordinator off to the side somewhere.. the regs say you must have a turn
coordinator but do not say where the turn coordinator has to be on your
panel.


> tipped over instantaneously. Any of the current breed of PCATD's do any
> better? Mine is a really old version of Elite, from about 1995 or so.


The current Elite software allows a choice between instant or gradual AI
failure. They have a reasonable upgrade program for their software as well.


--
Richard Kaplan, CFII

www.flyimc.com

[David Megginson]

"Gary L. Drescher" > writes:

> In MS FS2002, the AI fails abruptly, in just a few seconds. But I've heard
> that FS2004 (due late this month) has more-realistic gyro failures.

If you don't want to wait, FlightGear already has gradual gyro
failures. You can fail an individual gyro or an entire system
(i.e. vacuum or electrical):

http://www.flightgear.org/


All the best,


David

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

[Ray Andraka]

I thought one of the reasons for the turn coordinator is that it won't tumble in
unusual attitudes where the AI will. I once tumbled my AI in IMC, and it was
not a pretty thing. Gave me a real bad sense of vertigo. At the time I had a
needle and ball rather than a TC, but that was the instrument that let me sort
things out. I think I would still want either a T&B or a TC in my primary scan.

Richard Kaplan wrote:

> Put it where the turn coordinator is located and the put the turn
> coordinator off to the side somewhere.. the regs say you must have a turn
> coordinator but do not say where the turn coordinator has to be on your
> panel.
>

--
--Ray Andraka, P.E.
President, the Andraka Consulting Group, Inc.
401/884-7930 Fax 401/884-7950
email
http://www.andraka.com

"They that give up essential liberty to obtain a little
temporary safety deserve neither liberty nor safety."
-Benjamin Franklin, 1759

[C J Campbell]

Newer vacuum AIs have warning flags, dual vacuum pumps, and vacuum warning
lights on the annunciator panel. If I had an electric AI and a vacuum AI and
they disagreed markedly, I would be suspicious of the one having warning
lights and flags all over it.

Otherwise, comparing them to the other instruments, as you suggested, is the
best approach. AOPA Safety Foundation just ran some tests on how long it
takes a pilot to recognize that his vacuum system has failed, and how long
it takes pilots to recover from those failures. Many times it took over 90
seconds, which is way too long. Some never did recognize the failure.
However, none of them were convinced that the vacuum instruments were
working and that it was the other instruments that had failed. IIRC somewhat
less than half discovered the problem quickly and took quick action. Only
about 1/4 of them covered the failed instrument.

[[email protected]]

On 13-Jul-2003, Ray Andraka > wrote:

> The problem isn't just finding a home for it, it is finding a home for it
> that is in a spot where it will be in the pilot's primary scan. I don't
> think
> it would do much good on the other side of the panel where it might only
> be
> referenced once a minute or less.


Our Arrow has a backup electric AI on the right side of the panel. It is
turned on whenever I am in or near IMC. I do not include it in my primary
instrument scan, but check it periodically to make sure it is on and stable.
I have fortunately never had a failure of my vacuum AI in IMC, but I have
practiced under the hood with the vac AI and DG covered, using the electric
AI for guidance. For me, it only takes a few seconds to get comfortable
with looking across the panel, and sure beats sweating bullets trying to
hold attitude and a reasonable course with only the TC, particularly in
turbulence.

So, the only issue is whether I could detect a failed vac AI before it leads
me (or the autopilot) into an unusual attitude. If the problem is (as is
most likely) a vacuum pump failure, I have a vacuum warning light right in
front of my eyes that would be hard to miss. But a failure of the gyro
itself could be trickier. I tell myself that if I can't keep a reasonable
course (per the DG) with a wings-level attitude (per the AI), then something
is wrong and I need to immediately refer to the TC and the electric AI to
sort things out.

-Elliott Drucker

[Roy Smith]

Ray Andraka > wrote:
> Personally, I think the instrument scan typically taught relies too
> heavily on the AI given its relatively low reliability.
> Unfortunately, the alternative is a scan that works a bit more like a
> partial panel scan using the AI as supporting, not primary. Such a
> scan is much harder to master and requires considerable finesse to
> keep from chasing the needles. It is not one I would expect to be
> able to teach someone just learning to fly by instruments.

You are probably right. When I did my initial IFR training, my
instructor was very big on partial panel work. As a result, I learned
to not rely on the AI, and I find partial panel approaches (in training,
anyway) almost a non-event.

The downside, is that I suspect I don't use the information the AI gives
me as much as I should. I tend to fly pitch by airspeed, not by the AI.
This probably makes me not as smooth and precise as I might otherwise
be. But I do have a lot of confidence that I can fly an approach on the
TC and ASI alone.

[Sydney Hoeltzli]

James Robinson wrote:

> One comment in the report was that the pilot might have had a tendency
> to focus on a single problem, and mot paid attention to other things.
> He could have been trying to troubleshoot the electrical problem, and
> not handed control over to the copilot, who would have had a better view
> of the remaining functional instruments.

This is poor CRM if it is the case.

Did the report say anything about the training of the pilots? I
woulda thought they did regular sim stuff, where I assume the
instructors put you through the wringer on various failues.

> The failure would have been immediately obvious, so
> it wasn't one of those insidious failures that people don't notice at
> first. An experienced IFR pilot should have been aware of the need to
> maintain attitude and yet lost control almost immediately. In reading
> the report, it seems like such an avoidable accident, yet...

Yeah, that's what gets me about so many of these.

Sydney

[David Megginson]

Roy Smith > writes:

> The downside, is that I suspect I don't use the information the AI gives
> me as much as I should. I tend to fly pitch by airspeed, not by the
> AI.

Are you sure that using the ASI for pitch doesn't make you smoother?
I think that a couple of knots difference is more noticeable than a
fraction of a degree change in the AI pitch indication.

My problem is that managing pitch with the ASI gets hard in turbulence.


All the best,


David

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

[Sydney Hoeltzli]

C J Campbell wrote:
> Newer vacuum AIs have warning flags, dual vacuum pumps, and vacuum warning
> lights on the annunciator panel. If I had an electric AI and a vacuum AI and
> they disagreed markedly, I would be suspicious of the one having warning
> lights and flags all over it.

CJ,

How do newer vacuum AIs come with dual vacuum pumps? :)

We have a "warning flag" on our newer AI. I note that it is really
a low vacuum flag. It doesn't say a thing about how reliably the
instrument itself is operating.

Sydney

[Sydney Hoeltzli]

Ray Andraka wrote:
> There are several things you can add to help with the identification, In my
> plane I have a low vacuum warning light (part of the precise flight backup)
> mounted between the AI and DG. The AI is one of the sigmatec ones with a vacuum
> flag, so that if vacuum is lost in the instrument but not in the system I still
> know about it right away. These warnings cover identification of the more
> common cause of loss of the AI. The other failure mode would be failure of the
> gyro, in which case I don't believe you get the insidious gradual spin-down like
> you do with loss of vacuum.

Ray,

I'll speak to the latter.

A failing horizon gyro may not "spin down". But it can still
be insidious. Example: our AI had a period where, in level flight,
it would jump up and indicate a rather nose-high attitude. Fail
to catch it and you'd be in a rather steep dive. Then it would
go back to normal. Then jump up again....finally it broke and
unmistakably started spinning in a nauseating fashion, but the
"breaking" process could easily have caused a loss of control for
a pilot w/out a good cross-check (our failure happened VMC)

Cheers,
Sydney

[Sydney Hoeltzli]

Richard Kaplan wrote:

> Put it where the turn coordinator is located and the put the turn
> coordinator off to the side somewhere..

N. F. W.

I think it's time I had a really, really, good hunt for
that post about the Grumman getting flipped upside down
and dumped into IMC after (what was probably) a collision
with an RC plane.

Those Electric and Vacuum AIs come with a get-your-life-back
guarantee they honestly, really truly, won't tumble, never ever,
no matter what, even if I do? How do I test it, in a non-aerobatic
plane not approved for spins?

Cheers,
Sydney

[Newps]

C J Campbell wrote:
> Newer vacuum AIs have warning flags, dual vacuum pumps, and vacuum warning
> lights on the annunciator panel. If I had an electric AI and a vacuum AI and
> they disagreed markedly, I would be suspicious of the one having warning
> lights and flags all over it.

I just bought a new AI this spring. For an extra $50 or so you can get
the one with the flag. Since I already have the EI volts/amps
instrument that has warning lights for both high and low voltage, but no
warning light for the vacuum pump, I spent a little extra for the flag.
Now I don't need a suction warning lamp.

[Clay]

Stefan <"stefan"@mus. INVALID .ch> wrote in message >...
> Darrell wrote:
> >
> > The presence of two attitude indicators is especially valuable when they
> > disagree. That disagreement will direct your attention to the needle/ball
> > and basic flight instruments to help determine which one is correct. With a
> > single AI you could more easily follow a gyro error without noticing a
> > difference in the other basic instruments until it was too late.
>
> I don't understand this. Needle/ball is always included in my basic
> scan. As a starting point on toubleshooting, they are more reliable than
> the AI. Am I the only who knows this?
>
> Stefan

Stefan, You have hit the nail on the head. BRAVO!!!!!!!
Many of the real good instrument pilots know that needle, ball,
airspeed are the three essentials to flying. These are the most
reliable instruments on the panel. Once these are mastered then the
rest of the instruments will just make things easy.
Too many of us have forgotten the basics or have just gotten lazy.
When is the last time you flew a true NDB approach? With loran, GPS,
LOC & glide slopes, and vectoring, insturment flying is not that
difficult.
During an ICC, have your instructor give you a real good partial panel
workout.
It is rare to loose a gyro but it happens. Many pilots have lost
their battle with gravity during IMC because they forgot the basics.
This is especially true after loosing a vac pump or electrical system.
Clay

[Richard Kaplan]

"Sydney Hoeltzli" > wrote in message
...

> Those Electric and Vacuum AIs come with a get-your-life-back
> guarantee they honestly, really truly, won't tumble, never ever,
> no matter what, even if I do? How do I test it, in a non-aerobatic
> plane not approved for spins?

Sure they can tumble. And I would agree they would tumble before the turn
coordinator. But certainly they won't tumble at anything less than 60
degrees bank; the whole idea of the two AIs is to stop the spatial
disorientation profile way before you approach anything like an aerobatic
maneuver.

Also I am not saying you have to put the turn coordinator somewhere where
you will get vertigo... just not necessarily in the immediate 6-pack area,
perhaps instead closer to where most airplanes have their CDIs. It will
still be readable if you somehow get rolled nearly inverted by wake
turbulence. And in the far more likely situation of a vacuum failure, you
will be very glad your electric AI is right up there in your primary scan
area.

--
Richard Kaplan, CFII

www.flyimc.com

[Richard Kaplan]

> wrote in message
...>


> Our Arrow has a backup electric AI on the right side of the panel. It is
> turned on whenever I am in or near IMC. I do not include it in my primary
> instrument scan, but check it periodically to make sure it is on and
stable.

I think that is a really worrisome location. I just flew last month with a
very experienced instrument student who had a similar arrangement in his
182RG and developed vertigo and nausea while flying partial panel with his
primary AI covered up.. he literally could not manage the plane and asked me
to take over for a while to let him take off the hood and relax.

If you do have the electric AI in that location, it would be helpful for you
to occasionally fly a fairly long cross-country flight with a safety pilot
while you use a hood and cover up the primary AI... that would be helpful to
ensure you are not prone to vertigo from flying with the backup AI alone.


--
Richard Kaplan, CFII

www.flyimc.com

[Sydney Hoeltzli]

Julian Scarfe wrote:

> http://www.dft.gov.uk/stellent/groups/dft_control/documents/contentservertemplate/dft_index.hcst?n=5232&l=4
>
> tells a similarly chilling story of a Bandeirante that lost one of its two
> AIs resulting in a loss of control.

What do you think of the conclusions? They seem to be:
1) prevent AIs from failing
2) since 2 AIs weren't enough to keep the plane upright (combined
with 2 turn and banks, 2 of every other instrument), require
passenger planes to have 3

(agree, chilling)
Sydney

[Sydney Hoeltzli]

Richard Kaplan wrote:

> A vacuum faliure with a gradual spooldown of the AI and DG at vary rates is
> a subtle process, very different from having an instructor suddenly cover up
> a gyro. You are correct that there is enough redundancy in a typical GA
> plane to FLY partial panel; the problem is that there is not enough
> information for most pilots to IDENTIFY a partial panel situation before the
> situation has become critical.

Richard,

There seem to be a number of instances where the pilot was aware
of the problem -- had described it to ATC and possibly requested
some form of assistance -- had been flying the plane for some time
more or less under control, and then lost it. So the pilot did
identify a partial panel situation before it became critical,
he simply couldn't FLY partial panel.

The accident Julian posted the link for more-or-less seems to
fall into this case, as did a couple of local accidents.

Cheers,
Sydney

[Richard Kaplan]

"Sydney Hoeltzli" > wrote in message
...

> There seem to be a number of instances where the pilot was aware
> of the problem -- had described it to ATC and possibly requested
> some form of assistance -- had been flying the plane for some time

No doubt there will always be crashes we cannot avoid. I am sure there are
people with tip tanks who run out of fuel; that does not mean tip tanks fail
to increase an airplane's range.

The fact is that any sim instructor will tell you many pilots have a
difficult time identifying gradual loss of the vacuum system and that
redundant gyro equipment helps to identify this failure earlier.

--
Richard Kaplan, CFII

www.flyimc.com

[[email protected]]

On 13-Jul-2003, "Richard Kaplan" > wrote:

> I think that is a really worrisome location. I just flew last month with
> a very experienced instrument student who had a similar arrangement in his
> 182RG and developed vertigo and nausea while flying partial panel with his
> primary AI covered up

Our electric AI isn't THAT far to the right. Actually, it's mounted in the
right-hand radio stack, right above the DME. (Arrows and other Cherokees
from the late 1960s on have a double radio stack. With modern avionics
typically taking up less space, there is often room left over. It is a
simple matter to fashion a "blank" with a 3" instrument hole to fit in this
space, and that places the backup AI within reasonable scan distance for the
pilot.)

-Elliott Drucker

[C J Campbell]

"Sydney Hoeltzli" > wrote in message
...
| C J Campbell wrote:
| > Newer vacuum AIs have warning flags, dual vacuum pumps, and vacuum
warning
| > lights on the annunciator panel. If I had an electric AI and a vacuum AI
and
| > they disagreed markedly, I would be suspicious of the one having warning
| > lights and flags all over it.
|
| CJ,
|
| How do newer vacuum AIs come with dual vacuum pumps? :)
|

They come with a new airplane attached to them. Not unlike the $350,000
Cessna ball cap I bought.

[Kyler Laird]

Sydney Hoeltzli > writes:

>We have a "warning flag" on our newer AI. I note that it is really
>a low vacuum flag. It doesn't say a thing about how reliably the
>instrument itself is operating.

Yup, I had one freeze but the vacuum was fine so it didn't complain.

--kyler

[[email protected]]

On 14-Jul-2003, "Richard Kaplan" > wrote:

> > Our electric AI isn't THAT far to the right. Actually, it's mounted in
> > the right-hand radio stack, right above the DME. (Arrows and other
> > Cherokees from the late 1960s on have a double radio stack. With modern
> > avionics
>
> OK, that sounds fine... in fact perhaps a good location for others
> searching for a location for an electric AI.

Yes, it works for us. Also, it should be an easy and relatively inexpensive
installation (unlike swapping instruments with resultant requirements for
changes in wiring harnesses).

In our plane, this installation also allowed plenty of space to install an
ON/OFF switch next to the electric AI. We leave it off when flying in solid
VFR conditions, with the understanding that this will extend its life. So
far, so good. It's required no maintenance in over 7 yrs/1400 hrs (knock on
wood).

-Elliott Drucker

[[email protected]]

On 14-Jul-2003, Roy Smith > wrote:

> Not really. When I'm doing partial panel work, I find I miss the DG
> much more than I miss the AI.


A yoke-mounted GPS in either HSI or moving map mode is a good substitute for
a failed DG. Strictly for "situational awareness" of course! You have to
be careful in that the update of the GPS display is not instantaneous like
the DG's response to heading changes, so it's easy to find yourself "chasing
the needle". That's where the backup AI comes in.

[Jon Woellhaf]

Why do you think leaving it off most of the time will extend its life?

> wrote in message
...
> ... We leave [our electric AI] off when flying in solid
> VFR conditions, with the understanding that this will extend its life. So
> far, so good. It's required no maintenance in over 7 yrs/1400 hrs (knock
on
> wood).
>
> -Elliott Drucker

[Ray Andraka]

I thought that gyro bearings were very susceptible to damage if subjected to
mechanical shock while not turning. If that is true, then turning it off in
flight may actually reduce the service life.

wrote:

> In our plane, this installation also allowed plenty of space to install an
> ON/OFF switch next to the electric AI. We leave it off when flying in solid
> VFR conditions, with the understanding that this will extend its life. So
> far, so good. It's required no maintenance in over 7 yrs/1400 hrs (knock on
> wood).
>
> -Elliott Drucker

--
--Ray Andraka, P.E.
President, the Andraka Consulting Group, Inc.
401/884-7930 Fax 401/884-7950
email
http://www.andraka.com

"They that give up essential liberty to obtain a little
temporary safety deserve neither liberty nor safety."
-Benjamin Franklin, 1759

[[email protected]]

On 14-Jul-2003, "Jon Woellhaf" > wrote:

> Why do you think leaving it off most of the time will extend its life?

Seems to me that the two wear items in the electric AI are the brushes in
the DC motor and the gyro bearings. Also seems logical to me that both will
wear a lot less when not turning. As noted in previous post, this strategy
seems to be working for us. I would certainly entertain contrary wisdom
from anyone with more specific knowledge.

-Elliott Drucker

[Michael]

Sydney Hoeltzli > wrote
> What do you think of the conclusions? They seem to be:
> 1) prevent AIs from failing

Well, there's a lot to be said for that. For one thing, it's far from
unlikely that BOTH of the AI's failed, not just one.

Did you miss this: (All quotes from the referenced report)

"These 21 artificial horizons had an MTBUR of 257 hours." That's mean
time before unscheduled replacement, but... "The artificial horizon
fitted to the EMB-110had no specified overhaul life and was treated as
an 'on condition'item" and thus all replacements were unscheduled.

Why was such a shockingly high failure rate tolerated? Well, "The
BCAR Section under which the aircraftwas certificated did not
stipulate the reliability requirementsthat the artificial horizon
should meet in order to ensure thatthe occurrence of a double failure
was a statistically remoteevent." Gotta love the way those regs
protect us...

And sure enough it was not statistically remote - it had happened
before! "An EMB-110 operated by another UK company suffered two
double artificial horizon failures in 1995. The first,on 4 June 1995,
involved a double instrument failure" There were only a handful of
EMB-110's in the UK...

> 2) since 2 AIs weren't enough to keep the plane upright (combined
> with 2 turn and banks, 2 of every other instrument), require
> passenger planes to have 3

Yes, that's the recommendation. IMO it's unmitigated crap. First
off, AI's should not be failing at an average of less than 300 hours.
Second, there were still two good PNI's (basically HSI's) and
turn&slip indicators. But could the pilots use them? Probably not
because "This technique, commonly referred to as 'limited panel' (see
paragraph 1.5.3.2) does not form part of a professional pilot's
recurrency training and testing."

So the most likely causes of the crash are AI failure (quite possibly
double AI failure), and the inability of the flight crew to fly
partial panel because SURPRISE they get no recurrent training in
partial panel flying. Exactly what kind of outcome could one expect
when you fit proven failure-prone AI's to an airplane and don't give
the flight crew any recurrent partial panel training?

> (agree, chilling)

There are plenty of chilling accidents out there. This isn't one of
them. This was inevitable.

Michael

[Richard Kaplan]

> wrote in message
...

> seems to be working for us. I would certainly entertain contrary wisdom
> from anyone with more specific knowledge.


Well my understanding from the last shop which overhauled my AI was that if
an AI can be caged then it should be shipped in the caged position because
it can be damaged if transported uncaged.

Perhaps the same logic applies to flyingwith an uncaged and unpowered gyro,
but I cannot give any other source except the above.

--
Richard Kaplan, CFII

www.flyimc.com

[[email protected]]

On Sun, 13 Jul 2003 22:10:08 -0400, Roy Smith > wrote:

> David Megginson > wrote:
>
> > Roy Smith > writes:
> >
> > > The downside, is that I suspect I don't use the information the AI gives
> > > me as much as I should. I tend to fly pitch by airspeed, not by the
> > > AI.
> >
> > Are you sure that using the ASI for pitch doesn't make you smoother?
>
> Well, for example, on my initial climb, I'm looking for Vy on the ASI.
> Since I'm approaching Vy from below, it means I have to just sort of
> guess at the right pitch and watch the ASI to see what the airspeed
> does. The "right" way is to know what pitch attitude I want, peg the AI
> there, and then watch the ASI to confirm I'm getting the performance I
> expected.

I guess I don't agree with the "right" way then. The pitch attitude for Vy
(or any other airspeed) will be somewhat dependent on aircraft weight,
density altitude, etc. I do exactly what you do, use the ASI as my climb
attitude reference, not the AI. Same for descents -- I peg whatever
airspeed i want, keeping the rpm's where I want them, and then make sure
that I have a reasonable descent rate using the VSI. Adust pitch as
necessary to keep everything where it should be.

>
> I know I do it wrong, and I make a concious effort to retrain my scan,
> but it's very difficult to unlearn the way you first learned to do it.

[Aaron Coolidge]

In rec.aviation.owning Roy Smith > wrote:
: David Megginson > wrote:
:> But once you're established, do you find it easier to hold pitch using
:> the ASI?

: Not really. When I'm doing partial panel work, I find I miss the DG
: much more than I miss the AI.

I totally agree with Roy here. I also don't use the AI as much as some. I too
fly pitch using the ASI, although I use the AI as a reference to set pitch
during level-offs. I too miss the DG more than the AI in partial panel
work. I do fly a fixed-gear Cherokee, though.

As an aside, I had an AI failure in IMC, on the 3rd or 4th flight after
I bought my plane. I didn't realize the AI was a little slow to align at
startup- I had only owned the plane a couple hours!
In flight, I noticed the AI leaning a couple of degrees and thought "Oh,
the AI isn't installed in the panel straight- the DG isn't turning."
About 30 seconds later, the AI started spinning in the roll axis. At
perhaps 600 degrees a second! This was pretty easy to diagnose.
The next day (VFR) the AI worked fine. It was very sensitive to temperature,
and having cold-soaked overnight at Will Rogers World, it just wasn't
happy.

PS, modern AI's won't tumble in roll, and should not tumble until
more than +/- 85 degrees of pitch. (REF the R.C. Allen install guide.)
--
Aaron Coolidge (N9376J)

[Sydney Hoeltzli]

Bob Gardner wrote:
> You have three sources of bank information in a typical panel: the attitude
> indicator, the turn coordinator, and the heading indicator.

Um, isn't this leaving out something fundamental (and pretty
reliable, if difficult to interpret)?

Sydney (Be Expert With Map and Compass)

[Julian Scarfe]

"Roy Smith" > wrote in message
...

> > Are you sure that using the ASI for pitch doesn't make you smoother?
>
> Well, for example, on my initial climb, I'm looking for Vy on the ASI.
> Since I'm approaching Vy from below, it means I have to just sort of
> guess at the right pitch and watch the ASI to see what the airspeed
> does. The "right" way is to know what pitch attitude I want, peg the AI
> there, and then watch the ASI to confirm I'm getting the performance I
> expected.

I think it depends on the aircraft. Flying a simple trainer on trends in
the secondary instruments is not too hard. The more slippery the aircraft,
the more difficult it becomes: by the time you've picked up the trend in the
ASI needle, your attitude is already in a mess. I noticed this
transitioning to the Mooney, and I think it's much more marked in faster
aircraft -- one of the reasons why jet pilots don't spend a great deal of
time practising partial panel but rely on multiple AIs for redundancy.

Julian Scarfe

[Dennis O'Connor]

There is one IFR flight school that gives the student the introduction to
the AI during their first briefing in the first hour...
Instructor: "See, that's the AI... See how it gives you both pitch and
bank."
Student: "Yes, that's really nice."
Plonk goes the sink stopper...
"That's the last time you will see it until your check ride, son.!"

Denny

"Aaron Coolidge" > wrote in message
...
> In rec.aviation.owning Roy Smith > wrote:
> : David Megginson > wrote:
> :> But once you're established, do you find it easier to hold pitch using
> :> the ASI?
>

[Dennis O'Connor]

My big, ugly, AN horizon has been working reliably for decades... Makes you
kinda wonder...

Denny

"Michael" > wrote in message
om...
> Sydney Hoeltzli > wrote
> > What do you think of the conclusions?

[David Megginson]

Aaron Coolidge > writes:

> In rec.aviation.owning Roy Smith > wrote:

> : David Megginson > wrote:
> :> But once you're established, do you find it easier to hold pitch using
> :> the ASI?
>
> : Not really. When I'm doing partial panel work, I find I miss the DG
> : much more than I miss the AI.
>
> I totally agree with Roy here. I also don't use the AI as much as some. I too
> fly pitch using the ASI, although I use the AI as a reference to set pitch
> during level-offs. I too miss the DG more than the AI in partial panel
> work. I do fly a fixed-gear Cherokee, though.

Note that I wrote "ASI", not "AI", so it might be that all three of us
agree.


All the best,


David

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

[Julian Scarfe]

"Sydney Hoeltzli" > wrote in message
...
> Michael wrote:
>
> >>2) since 2 AIs weren't enough to keep the plane upright (combined
> >> with 2 turn and banks, 2 of every other instrument), require
> >> passenger planes to have 3
>
> > Yes, that's the recommendation. IMO it's unmitigated crap. First
> > off, AI's should not be failing at an average of less than 300 hours.
> > Second, there were still two good PNI's (basically HSI's) and
> > turn&slip indicators. But could the pilots use them? Probably not
> > because "This technique, commonly referred to as 'limited panel' (see
> > paragraph 1.5.3.2) does not form part of a professional pilot's
> > recurrency training and testing."
>
> You Have Got To Be Kidding.
>
> Are you serious? Yes, I missed that. Are they asserting this
> shocking hole in proficiency training is widespread?

I think you're barking up the wrong tree here. Classic teaching of
partial/limited panel involves covering an instrument and then continuing to
fly without it. In the case of the Bandeirante accident, that wasn't the
issue. There was still a perfectly serviceable AI in the panel, and a pilot
sitting in front of it. The issue was identifying the failed instrument in a
complex cockpit environment.

The chances of being left with no working AI in the panel of a transport
aircraft (which starts with 3 AIs) but still having the instrumentation to
fly partial panel are so remote that it's not worth the time to train on it.
That time is better spent on other exercises, one of them *recognition* of
instrument failure.

For GA aircraft the situation is different. The probability of ending up
with a TC but no AI is much higher, and controlling the aircraft
successfully without it is easier. That makes it well worth the practice.

Julian Scarfe

[Sydney Hoeltzli]

Julian Scarfe wrote:

> I think you're barking up the wrong tree here. Classic teaching of
> partial/limited panel involves covering an instrument and then continuing to
> fly without it. In the case of the Bandeirante accident, that wasn't the
> issue. There was still a perfectly serviceable AI in the panel, and a pilot
> sitting in front of it. The issue was identifying the failed instrument in a
> complex cockpit environment.

I'm not sure I'm barking up the wrong tree.

Possibly practicing flying partial panel makes little sense. OTOH,
practicing partial panel *does* teach which combinations of instruments
can be used to provide the same information as the missing AI.

Surely this is relevant to obtaining and maintaining a good
crosscheck -- and wouldn't good crosscheck be the key to identifying
the failed instrument in a "complex cockpit environment"?

BTW, my reading of the accident report is that they weren't
certain but what both AIs had failed -- something that was certainly
within statistical likelihood given the low MTBUR

Cheers,
Sydney

[Roger Tracy]

I have a Garmin 196. It has a simulated instrument
panel page on it that is pretty accurate. It would be
the tie breaker in the event of confusion over what
the gyro instruments were saying.


"Sydney Hoeltzli" > wrote in message
...
> Julian Scarfe wrote:
>
> > I think you're barking up the wrong tree here. Classic teaching of
> > partial/limited panel involves covering an instrument and then
continuing to
> > fly without it. In the case of the Bandeirante accident, that wasn't
the
> > issue. There was still a perfectly serviceable AI in the panel, and a
pilot
> > sitting in front of it. The issue was identifying the failed instrument
in a
> > complex cockpit environment.
>
> I'm not sure I'm barking up the wrong tree.
>
> Possibly practicing flying partial panel makes little sense. OTOH,
> practicing partial panel *does* teach which combinations of instruments
> can be used to provide the same information as the missing AI.
>
> Surely this is relevant to obtaining and maintaining a good
> crosscheck -- and wouldn't good crosscheck be the key to identifying
> the failed instrument in a "complex cockpit environment"?
>
> BTW, my reading of the accident report is that they weren't
> certain but what both AIs had failed -- something that was certainly
> within statistical likelihood given the low MTBUR
>
> Cheers,
> Sydney
>

[Dennis O'Connor]

I played with that page on the 196 yesterday in moderately gusty
conditions... You have to be smooth on the controls, make your correction
and then wait for the unit to update... Flown that way it is useable and I
suspect that I could fly an approach with it in real conditions (an ILS to
minimums would be hairy).. The work load is significantly higher than with
the gyros..

But, if you do not wait for the screen to update you get into a world of
hurt... Since it was nice and bumpy I put the hood on and then flew it like
I was panicky - rapid, big,. inputs - It only took about 30 seconds to get
out of sync, with the ship laid over on it's side, whereupon I had the fun
of recovering from an unusual attitude... Other than the speed having
gotten further into the yellow arc than I like it was good exercise... About
this time the controller came on and asked me to say intentions - uh, oh,
busted!

Denny

"Roger Tracy" > wrote in message
...
> I have a Garmin 196. It has a simulated instrument
> panel page on it that is pretty accurate. It would be
> the tie breaker in the event of confusion over what
> the gyro instruments were saying.