Stability augmentation promises to give you even less control

a wrote:
On Aug 1, 12:48Â*pm, wrote:
a wrote:
On Aug 1, 12:05Â*am, wrote:
Mxsmanic wrote:
writes:

OK, then single pilot in real IMC.

That's what wing levelers are for.

Not while manuevering, which is when the system would be active.

Another would be inadvertent VFR flight into IMC.

Though I will admit that since there has to be an autopilot installed which
this thing is installed on top of, it does sound a little like having a
belt and suspenders.

--
Jim Pennino

Remove .spam.sux to reply.

The reasonable approach would be to have the autopilot engage in
straight and level automatically if the sensors detect an out of
control condition. Not sure if a conventional A/P knows how to recover
from a spin, but that would be a modest software patch.

The article mentions loss of control as a major factor in the accident rate
without going into any details of what that means.

I can see the utility of something that monitors angle of attack and nudges
the nose down when it determines a stall is emminent.

--
Jim Pennino

Remove .spam.sux to reply.

Jim,with respect, if some device decided I wanted back pressure
released when a stall is pending that device and I would have a
discussion -- my end of it would be with wire clippers or a hammer! --
because when I'm landing the AoA is in the stall range and I don't
want the nose wheel to touch down first, especially on a soft field.

Since the system as described is easily overriden with manual inputs, I
don't see that as a problem.

It would take somewhat better programming than simply AoA. On the
other hand, my airplane never sees pitch and bank close to the
statutory limits, Those might be worth considering. Wait a minute,
maybe not. I would not want something to intervene if I needed big
pitch or bank inputs if trying to avoid another airplane or the like.
I'm guessing optimal spin recovery would be ok though, optimal being
defined as minimal loss of altitude. And maybe something to avoid the
JFK Jr kind of pilot auguring into the ocean.

JFD Jr augured in in what appears to have been cooridinated flight, so such
a system would have made no difference.

I would think the system would be somewhat usefull to prevent things like
departure stalls and such.



--
Jim Pennino

Remove .spam.sux to reply.

In article ,
says...

a writes:

The reasonable approach would be to have the autopilot engage in
straight and level automatically if the sensors detect an out of
control condition.

What would count as an out-of-control condition?

I stall, a spin, a spiral dive (excessive pitch down and speed) - many
things - as long as they are also with NO PILOT INPUT - then is it not a
good idea for such a system to "do something".?

--
Duncan.

On Aug 1, 7:48*pm, wrote:
a wrote:
On Aug 1, 12:48*pm, wrote:
a wrote:
On Aug 1, 12:05*am, wrote:
Mxsmanic wrote:
writes:

OK, then single pilot in real IMC.

That's what wing levelers are for.

Not while manuevering, which is when the system would be active.

Another would be inadvertent VFR flight into IMC.

Though I will admit that since there has to be an autopilot installed which
this thing is installed on top of, it does sound a little like having a
belt and suspenders.

--
Jim Pennino

Remove .spam.sux to reply.

The reasonable approach would be to have the autopilot engage in
straight and level automatically if the sensors detect an out of
control condition. Not sure if a conventional A/P knows how to recover
from a spin, but that would be a modest software patch.

The article mentions loss of control as a major factor in the accident rate
without going into any details of what that means.

I can see the utility of something that monitors angle of attack and nudges
the nose down when it determines a stall is emminent.

--
Jim Pennino

Remove .spam.sux to reply.

Jim,with respect, if some device decided I wanted back pressure
released when a stall is pending that device and I would have a
discussion -- my end of it would be with wire clippers or a hammer! --
because when I'm landing the AoA is in the stall range and I don't
want the nose wheel to touch down first, especially on a soft field.

Since the system as described is easily overriden with manual inputs, I
don't see that as a problem.

It would take somewhat better programming than simply AoA. On the
other hand, my airplane never sees pitch and bank close to the
statutory limits, Those might be worth considering. Wait a minute,
maybe not. I would not want something to intervene if I needed big
pitch or bank inputs if trying to avoid another airplane or the like.
I'm guessing optimal spin recovery would be ok though, optimal being
defined as minimal loss of altitude. And maybe something to avoid the
JFK Jr kind of pilot auguring into the ocean.

JFD Jr augured in in what appears to have been cooridinated flight, so such
a system would have made no difference.

I would think the system would be somewhat usefull to prevent things like
departure stalls and such.

--
Jim Pennino

Remove .spam.sux to reply.

It's been a while since I read the NTSB on the JFK Jr crash, but I
seem to remember he went from 5500 feet to impact in much less than a
minute in a classic death spiral. Seems to me sensing a tight
overbanked rapidly descending turn (5000 fpm down should get a
sensor's attention!), very rapid rate of turn could be a trigger for a
control system to take over. Critical point, if the PIC is as
disoriented as one might have expected JFK Jr to have been, is that
pilots should NOT be allowed to over ride the system. I would bet
serious money he had the the yoke in his belly and would not be
surprised if the controls were otherwise neutral, and he was wondering
why, with so much up elevator, the damn thing was still going down.,

Dave Doe writes:

I stall, a spin, a spiral dive (excessive pitch down and speed) - many
things - as long as they are also with NO PILOT INPUT - then is it not a
good idea for such a system to "do something".?

I suppose that's a good start, but when you write computer software, you have
to be way more specific ... and if you don't want to kill anyone, you also
have to be absolutely right.

A perfectly programmed computer might be able to do it, although I should
think it would also need mind-reading ability in order to determine the
pilot's intentions. But in any case, people write the software, so it's never
perfect--and that's the problem.

Stephen! writes:

You haven't lived until you've piloted an SEL over the mountains in solid
IMC without an autopilot. I logged 2 hours IMC in a 182 over Montana a few
weeks ago. That was a LOT of fun!

But a standard autopilot does what you tell it to. You turn it on and off
yourself, and you tell it what heading to hold, what altitude to maintain,
when to level the wings, and so on. That's very different from a computer
that independently decides when you need "help" and acts on its own initiative
because it thinks you are "out of control."

In solid IMC over mountains, you need precise situational awareness. You don't
have that if a computer is making decisions behind your back.

It's a bit like having a silent copilot who does things whenever he deems them
appropriate without any interaction or instructions from you first. Does that
really enhance safety?

Mxsmanic wrote:
Stephen! writes:

You haven't lived until you've piloted an SEL over the mountains in solid
IMC without an autopilot. I logged 2 hours IMC in a 182 over Montana a few
weeks ago. That was a LOT of fun!

But a standard autopilot does what you tell it to. You turn it on and off
yourself, and you tell it what heading to hold, what altitude to maintain,
when to level the wings, and so on. That's very different from a computer
that independently decides when you need "help" and acts on its own initiative
because it thinks you are "out of control."

In solid IMC over mountains, you need precise situational awareness. You don't
have that if a computer is making decisions behind your back.

True in general but this has nothing whatsoever to do with the system under
discussion.

It's a bit like having a silent copilot who does things whenever he deems them
appropriate without any interaction or instructions from you first. Does that
really enhance safety?

It depends on how smart that silent copilot is.

The world is not black/white, one/zero, yes/no, all/nothing and once that
fact sinks in you might have a better understanding of it.


--
Jim Pennino

Remove .spam.sux to reply.

On Aug 3, 10:45*am, Mxsmanic wrote:

It's a bit like having a silent copilot who does things whenever he deems them
appropriate without any interaction or instructions from you first.

YOU OBVIOUSLY ARE CLUELESS and never flew a real plane.

LOOK UP CRM.

"Mxsmanic" wrote in message
...
It's a bit like having a silent copilot who does things whenever he deems
them
appropriate without any interaction or instructions from you first.

Then you probably won't be buying an aiplane with this feature.

--
Geoff
The Sea Hawk at Wow Way d0t Com
remove spaces and make the obvious substitutions to reply by mail
When immigration is outlawed, only outlaws will immigrate.

writes:

True in general but this has nothing whatsoever to do with the system under
discussion.

It has a lot to do with it. A system that decides to "stabilize" the aircraft
because it thinks you have "lost control" is indisputably related to this.

It depends on how smart that silent copilot is.

When the silent copilot is a computer, that depends on the barely-tested,
bug-laden software that it is running.

writes:

LOOK UP CRM.

There is no CRM with a computer.