wing levelers

"Dave Hyde" wrote in message
...
I wonder if we've crossed professional paths sometime over the years.
Dryden?

Yup. I was at Dryden from 1993 to 1997 (internship, then as a Research
Associate on a grant with USC, then as a contractor after waking up one day
and realizing that I didn't really want a PhD after all), then went to work
at Lockheed in Palmdale, where I've been ever since. At Dryden, I was in
that Integrated Controls group over in code R.

PSFCC ring a bell?

It tickles my memory. Is that the processor upgrade that was put in one of
the Dryden F-18's? I never worked much with that gang, but I know a lot of
those guys.

Nauga is sitting this one out, Pete's doing a fine job :-)

I feel like I'm trying to get a kid to put his dad's gun down. The guys on
this group know enough to get themselves is a whole lot of trouble.

Pete Schaefer wrote...

I was at Dryden from 1993 to 1997...

I was there a few times over the years, mostly
in that time range.

Is that the processor upgrade that was put in one of
the Dryden F-18's?

Well, 'upgrade' is debatable, but we're talking about the
same program. g

I never worked much with that gang, but I know a lot of
those guys.

OK, there was another 'Pete' on that program, but I've
since remembered his last name, and it wasn't you. :-)

I feel like I'm trying to get a kid to put his dad's gun down.

At least no one's said they've already tested it in 'X-Plane',
so it *has* to work :-)

Dave 'phase margin' Hyde

On Mon, 7 Mar 2005 18:41:35 -0800, "Pete Schaefer"
wrote:

into the system through a couple of springs, and if you're asking for
more force than what they deliver then you've already gone off the wrong
side of the page.

Putting springs also puts lag into your system. More tendency for
oscillatory behavior.

I am not sure what he has in mind, but I've found that using preloaded
(tight wound) springs, when the "inputs" are applied via them,
introduce only the additional mass. If they are not preloaded ... the
mathematics gets beyond me.

Example ... take an old style screen door spring (sounds redundant,
doesn't it g), cut it to a lenght of an inch (for example only), use
an anvil to pop out a turn, pull on it ... some reasonable force is
applied before the spring begins stretching (macro level coils
separating ...not counting the tiny bits of stretch in the ends)

I used a similar method with shear load cells to protect them; the
first 10 pounds was applied to the cell with "no" deflection. The
next 2 compressed a spring and applied load to the cell. After that,
I had metal-metal preventing further load on the cell. Keeps the 15
pound cell from being destroyed when a man stepped in the weigh
hopper.

"Pete Schaefer" wrote

Lots of primary research was going on.
These days, I talk to people out there, and it sounds like they've got
NOTHING going on of any significance anymore. Sad.

That's cause there is nothing significant to discover, anymore! ;-o
--
Jim (grinning and ducking and running) in NC

It's wierd. When I was out there, it was still a really happenin' kind of
place. Lots of stuff was flying. Lots of primary research was going on.
These days, I talk to people out there, and it sounds like they've got
NOTHING going on of any significance anymore. Sad.

"Dave Hyde" wrote in message
...
I was there a few times over the years, mostly
in that time range.

Ron Webb wrote:
How can you tell whether you are straight and level? If
you're in a balanced turn, your accelerometer (which actually measures net
force) will believe you are straight and level.

And a GPS won't help much at all, mostly because its response rate is too
slow.

Frank


As I see it, accelerometers and GPS together are all that's needed. If the
vectors from all 3 accelerometers are in the right direction, and the
average GPS heading is not moving much- you're straight and level.

My GPS updates about once per second. That's plenty fast enough.Once every
10 seconds would work.When flying, do you react to every bounce, or just
ride with the flow and provide general guidance? I don't know about you, but
I try to stay relaxed.

The GPS does too make the differance between a really sticky problem and a
slam dunk.
I'm betting your practical experience was before $50 GPS and $3
accelerometers? (say, 5 years ago). Am I right?

Nope... I finished on my MTech thesis the year before last, designing a
system to automatically measure ocean waves. I used the ADXL202
accelerometers you've mentioned here, plus 3 magnetometers to measure
attitude, plus some digital filtering and stuff. Ocean waves, being in
constant variable motion, is probably a tougher environment than an
aircraft. And, as someone else suggested, I guess that an aircraft
system could be designed as a pilot aid rather than pilot replacement.

Your GPS can't tell you whether you're straight and level, only whether
you're straight. Whilst we're talking GPS, some interesting work was
done using GPS to measure attitude using 3 antennas about a foot apart.
This sounded like it had real promise.

Maybe we have bumpier bumps here, or maybe you fly something with a high
wing loading, or whatever, but in the circumstances where an active wing
leveler would be useful, I think it would need to react accurately and
correctly within (say) 0.1 seconds to be useful. And would need to not
need to be reset more than (say) once in 20 minutes. I just can't see
that with a 1-second GPS update of position and track (not airspeed or
heading). If there's no bumps, then manual aileron trim will do the
trick just as well.

Frank

Your GPS can't tell you whether you're straight and level, only whether
you're straight.

Agreed.

But if the GPS says I'm straight (over a 10 second average) and the
accelerometer says the roll axis vector is pointed toward the floor of the
aircraft, then it seems to me I'm straight (otherwise the heading would be
changing) and level - since if there is no heading change the only vector
remaining is G.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ron,

Try adding in some cross controll input & acelerometer error Bzzzzt. The
deal is that over time, any acelerometer, even 3 axix will accumulate
error, and ignoring your idea of non-realtime GPS directional input for
the moment, what you need is a leaky integrator. To accomplish this, you
need to be able to sense the horizon in an accelerated frame of
reference. You can accomplish this in a number of ways. The two that
come readily to mind are magnetometers & solid state or otherwise steam
gage type gyros. Sadly, this last requirement for instrumentation is
often where the expense for your system will come in to play.

Evan Carew

Ron Webb wrote:
Your GPS can't tell you whether you're straight and level, only whether
you're straight.


Agreed.

But if the GPS says I'm straight (over a 10 second average) and the
accelerometer says the roll axis vector is pointed toward the floor of the
aircraft, then it seems to me I'm straight (otherwise the heading would be
changing) and level - since if there is no heading change the only vector
remaining is G.



-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFCLdNmpxCQXwV2bJARAo9QAJ9IMxPQ/igwTJZ+enML69Su6SRd9QCfbHYp
VCxBa5Nc6LdrmsK+D3W6/Tc=
=JQDQ
-----END PGP SIGNATURE-----

On Tue, 08 Mar 2005 00:33:00 GMT, Ernest Christley
wrote:

Frank van der Hulst wrote:
Ron Webb wrote:

"Predictor" wrote in message
oups.com...

Ernest Christley wrote:
"Dan, did you ever get a chance to work with fuzzy logic?"



Why overcomplicate things. This is just too easy an application with a
BasicX board and an accerleometer.


Actually, its not that simple. Trust me, I worked on a similar problem
for 2 years. How can you tell whether you are straight and level? If
you're in a balanced turn, your accelerometer (which actually measures
net force) will believe you are straight and level.

And a GPS won't help much at all, mostly because its response rate is
too slow.

Frank

And the cruise control on your motorhome doesn't allow you to go to the
back and get a cup of coffee while tooling down the highway.

Why does it have to be all or nothing, people? Why can't it be an
electonic 'assistant' that can take the edge off of a bumpy ride and
keep the course drift to a minimum, without being asked to save the
pilot from an IFR death spiral? How can it tell your straight and
level? It WON'T, and no one need ask it to. Just pretty much maintain
the status quo, thereby making flight more enjoyable.

It's that "bumpy ride" that takes a quick response.

You need response (and sensing) times in milliseconds.
By far the simplest I can think of would be to tie into a gyro for
both pitch and roll. Then if you have the time and inclination, add
the output from the blind altitude encoder for altitude hold.

For an experiment, feed the gyro signals into a lap-top and process
them. You could even use an old AI for a test. Just let the
calculated outputs drive bar graphs, or position indicators and record
the functions. It'd be nice to be able to record the actual control
inputs as well for comparison. When the outputs look good then move
on to driving servos for trim tabs.

All failure modes should be neutral. Failure of any signal should
drop the system off line and that would put the servos at neutral.

A while back a friend had the elevator trim tab break. This
immediately put them into a 6-G pull up. Only a good pilot (who's
hand was on the stick) saved their bacon. BTW that was in IMC.

There is no need for an amateur built AP to be a dangerous exercise,
but the risk depends solely on how much foresight is put into the
system. Never leave a control system with a single point failure
mode.

Roger Halstead (K8RI & ARRL life member)
(N833R, S# CD-2 Worlds oldest Debonair)
www.rogerhalstead.com

As I see it, accelerometers and GPS together are all that's needed. If the
vectors from all 3 accelerometers are in the right direction, and the
average GPS heading is not moving much- you're straight and level.

Ron-Using accelerometers in 3d space will soon become inaccurate due to
small errors in the accelerometers, math rounding errors and drift. How do
you propose to 'erect' such a device to keep the shiny side up?