Today's Message Index:
----------------------
1. 06:36 AM - Re: Looking for a used PC680 battery for testing (hotwheels)
2. 11:50 AM - autopilot servos (David)
3. 01:10 PM - Re: autopilot servos (Robert L. Nuckolls, III)
4. 01:30 PM - Re: autopilot servos (Peter Pengilly)
5. 04:02 PM - Exciting News from Aircrafters (Timothy Farrell)
Message 1
INDEX | Back to Main INDEX |
NEXT | Skip to NEXT Message |
LIST | Reply to LIST Regarding this Message |
SENDER | Reply to SENDER Regarding this Message |
|
Subject: | Re: Looking for a used PC680 battery for testing |
Thanks to all that replied to my query. I'm going to find something local that's
close enough.
Regards,
Jay
Read this topic online here:
http://forums.matronics.com/viewtopic.php?p=366467#366467
Message 2
INDEX | Back to Main INDEX |
PREVIOUS | Skip to PREVIOUS Message |
NEXT | Skip to NEXT Message |
LIST | Reply to LIST Regarding this Message |
SENDER | Reply to SENDER Regarding this Message |
|
Subject: | autopilot servos |
Guys, can you help guide me towards what the specs are for autopilot
motors? My idea is to use high-torque steppers so that they can be
computer controlled by software that I write in a computer that I design
(maybe an Arduino.) This is being done mostly for fun but also to be
safe and practical in my OBAM aircraft. How much torque do I need? The
Mustang II takes very, very small stick movement once airborne and the
faster you go, the less torque required. Basically, once airborne, you
weld your hand to your leg and just think about which direction you want
to go and that is enough to turn or climb.
I think it would be easiest, in programming, to have feedback from the
servo on it's current position and there are other methods but the
encoded type sounds easiest to implement. Thoughts?
Lastly, AIUI, most steppers become free-wheeling once power is removed
so that gives me control back instantly in event that is necessary
simply by turning them off, leaving no drag on the stick or ruddders.
I would like to control all 3 axes.
So which steppers and corresponding controllers do you guys recommend?
Or am I barking up the wrong tree?
Once fully debugged, I may play with automated take-off and landing but
that will be very, very closely watched if I ever get to that phase.
Too easy to break something important (like me!)
Thanks,
David M.
--
Tell the truth. Be honest. Be responsible to and for yourself.
We want our freedoms back. Strike the illegal legislations called: 1) obamacare,
2) "Patriot (HA!) Act", and 3) 'presidential orders' that affect anyone besides
gubmnt worker bees.
Hate crime laws? Really? Thought police? Orwell would be proud.
Every gram of cocaine you buy from elsewhere contributes to an innocent being murdered
in Central and South America. Grow your own or Stop taking it.
Message 3
INDEX | Back to Main INDEX |
PREVIOUS | Skip to PREVIOUS Message |
NEXT | Skip to NEXT Message |
LIST | Reply to LIST Regarding this Message |
SENDER | Reply to SENDER Regarding this Message |
|
Subject: | Re: autopilot servos |
At 01:21 PM 2/17/2012, you wrote:
>
>Guys, can you help guide me towards what the specs are for autopilot
>motors? My idea is to use high-torque steppers so that they can be
>computer controlled by software that I write in a computer that I
>design (maybe an Arduino.) This is being done mostly for fun but
>also to be safe and practical in my OBAM aircraft. How much torque
>do I need? The Mustang II takes very, very small stick movement
>once airborne and the faster you go, the less torque
>required. Basically, once airborne, you weld your hand to your leg
>and just think about which direction you want to go and that is
>enough to turn or climb.
That's been the fondest wish of probably every a/p
designer since WWII. Guys like Ed King, Honeywell,
Bill Lear, Jim Younkin, et. als. can point to many
decades of development spanning dozens of recipes for
success.
>I think it would be easiest, in programming, to have feedback from
>the servo on it's current position and there are other methods but
>the encoded type sounds easiest to implement. Thoughts?
What have you read about autopilot history? Have you
studied any of the legacy design goals for failure
mode effects?
>Lastly, AIUI, most steppers become free-wheeling once power is
>removed so that gives me control back instantly in event that is
>necessary simply by turning them off, leaving no drag on the stick or ruddders.
Sort of. Steppers have a small but noticeable
"cogging" in back drive torque. Jim Younkin
went to stepper motors with a step down gearing
on the order of 10:1. Typical torque needed at
the aileron bell crank on an RV is 30 oz inches.
So the stepper would need to supply about 3 oz
inches of torque. At the same time, gearing magnifies
the backdriving forces for the de-energized motor.
10:1 is probably a golden goal for trading off physical
size of motor and tolerable back-drive forces.
Gene Brown and I discussed a stepper motor driven
a/p design probably 25 years ago at Electro-Mech.
It's the easiest to implement both mechanically
and electrically. No disconnect clutch needed.
Have you considered an RC servo driving a tab
on the surface to be controlled? RC servos are
cheap and easy to integrate. They need some 'help'
in the EMC department but that's doable.
Driving a tab keeps you totally disconnected
from primary flight surfaces and in absolute
control for over-ride forces needed to deal
with a tab that's jammed to the stop.
Most AUTOPILOTS are designed to do pretty much
everything a MANPILOT would do. But my idea of
an small-letters autopilot is to keep the airplane
pointed in the right direction with a minimum
of maneuvering . . . very light fingered touch
from the electronics.
This suggests that you don't need fat servos
capable of 2G loops and 360 degree per second
aileron roles.
>So which steppers and corresponding controllers do you guys recommend?
>Or am I barking up the wrong tree?
There's LOTS of trees to contemplate before you
start barking. Exactly what do you want this
product to do for you?
>Once fully debugged, I may play with automated take-off and landing
>but that will be very, very closely watched if I ever get to that phase.
>Too easy to break something important (like me!)
I can tell you that we had some VERY talented folks
working that problem on the AGATE program. There is
a Bonanza fitted with standard controls and 14v
hardware for the right seat and fly-by-wire, glass-
cockpit, highway-in-the-sky hardware in the left
seat that ran from a totally independent 28v
system. We installed our own SAAS (small aera
augmentation system) right on Beech Field to give
us real-time, centimeter correction for GPS signals.
I wont suggest that you can't do it . . . but I've
seen it done and the task is . . . shall we say
. . . monumental?
Your probability of success for any phase of
your endeavors begins with small goals with
stopping places for evaluation of results. Suggest
you start with a simple mod to one aileron to
add a tab, drive it with a servo that you can fit
into your aileron, then go fly it with manual
control knob on the servo. Can you achieve a servo/
tab combination that gives you the control authority
you need to be a precision wing-leveler/heading-hold
system in moderate turbulence? If the servo drives
to either stop, is the airplane manageable without
breaking a sweat?
Tame that dragon to the extent he's not trying to
eat you, then let's talk about ways to make him
heel on a leash . . . and ultimately tap-dance.
Bob . . .
Message 4
INDEX | Back to Main INDEX |
PREVIOUS | Skip to PREVIOUS Message |
NEXT | Skip to NEXT Message |
LIST | Reply to LIST Regarding this Message |
SENDER | Reply to SENDER Regarding this Message |
|
Subject: | Re: autopilot servos |
David,
I admire your courage in taking on one of the more difficult challenges
in light aircraft avionics. An autopilot is primarily a flight
control/flight dynamics problem - if you know little of this subject
(for example if you are unfamiliar with stability derivatives) I would
urge you to read up on the subject or get help from someone who does.
There are also several non-intuitive gotchas that can sneak up on you.
Could I suggest you read the relevant chapter of FAR23 - I think it is
para 1329. I would suggest you ignore yaw until you have the other two
axes sorted (or perhaps for ever).
Servo torque must the the minimum that will do the job - as the pilot
must be able to over ride the servos and the autopilot must not be able
to break the aeroplane if it decides to inject full deflection elevator
(especially at max level speed). If possible I would measure the stick
force that is required to pull 2 g throughout the flight envelope, and
also the stick displacement. Then measure how much force and stick
displacement to roll at 30* per second. Now calculate the mechanical
advantage of the stick, and so the force the servo will have to exert at
the place you decide to connect it to the control system (make sure the
servo cannot under any circumstance go 'over centre'). I suspect you
will have to use some rate feedback, as well as position feedback, to
accurately apply the desired control surface angle. If you do not
include autotrim in pitch how will you tell the pilot that the servo is
about to become saturated so he can trim manually? How will you ensure
there are no large out of trim transients when the autopilot is disengaged?
You will find quite a change of displacement required from circuit speed
to fast cruise. CG position will also affect the stick force required.
The logic is also non-trivial, at what bank angles will you transition
from heading hold to bank angle hold? 5* is usual. You will also have to
decide which parameter to close on for altitude control. You only have
control over aircraft pitch attitude, so how do you use that to control
altitude? What will you control in the pitch axis if you aren't holding
altitude? Speed (IAS, TAS or ground speed) or rate of climb/descent?
Will you have a minimum speed that would won't let the airplane
decelerate below? For example, you are holding 5000ft and the engine
power slowly decreases (for whatever reason), at what point do you give
up trying to hold altitude and let the airplane descend at a minimum
airspeed - or will you hold altitude come what may and allow the
airplane to stall? If you are trying to follow a route, and for some
reason your are off the track line how will the autopilot regain track
and what information do you require from the nav system (GPS?) - draw
some pictures it will take me too long to type! For that matter, where
will get heading information for for heading hold, if from GPS what do
you do if the GPS quits? How will you tell the pilot that the autopilot
is no longer in control and he should start flying the airplane again?
Over powerful autopilots have the ability to break airplanes and kill
the crew, under powered servos will be no use at all. Best of luck with
your project, please keep us informed of how you progress.
Regards, Peter
As an example a military autopilot I worked on only controlled one
aileron through 2* (two degrees) and the elevator through about 15% of
max travel - you will need more but will not need to operate over the
full control surface travel.
On 17/02/2012 19:21, David wrote:
>
> Guys, can you help guide me towards what the specs are for autopilot
> motors? My idea is to use high-torque steppers so that they can be
> computer controlled by software that I write in a computer that I
> design (maybe an Arduino.) This is being done mostly for fun but also
> to be safe and practical in my OBAM aircraft. How much torque do I
> need? The Mustang II takes very, very small stick movement once
> airborne and the faster you go, the less torque required. Basically,
> once airborne, you weld your hand to your leg and just think about
> which direction you want to go and that is enough to turn or climb.
>
> I think it would be easiest, in programming, to have feedback from the
> servo on it's current position and there are other methods but the
> encoded type sounds easiest to implement. Thoughts?
>
> Lastly, AIUI, most steppers become free-wheeling once power is removed
> so that gives me control back instantly in event that is necessary
> simply by turning them off, leaving no drag on the stick or ruddders.
>
> I would like to control all 3 axes.
>
> So which steppers and corresponding controllers do you guys
> recommend? Or am I barking up the wrong tree?
>
> Once fully debugged, I may play with automated take-off and landing
> but that will be very, very closely watched if I ever get to that
> phase. Too easy to break something important (like me!)
>
> Thanks,
> David M.
>
Message 5
INDEX | Back to Main INDEX |
PREVIOUS | Skip to PREVIOUS Message |
NEXT | Skip to NEXT Message |
LIST | Reply to LIST Regarding this Message |
SENDER | Reply to SENDER Regarding this Message |
|
Subject: | Exciting News from Aircrafters |
To all our dedicated customers and collegues,
Aircrafters has some significant and exciting news to announce. We have a
new partial owner! Tim Farrell has become a part of Aircrafters and will b
e assuming all of the administrative, sales, and marketing responsibilities
=2E Dave Saylor will be focusing his time and energy on the shop floor and w
orking on your airplanes.
I am originally from Oregon and have recently settled in the Monterrey Bay
area. I 'm an instrument rated private pilot with years of experience buil
ding Lancairs. I love experimental aviation and aviation in general. I h
ave experience teaching and I look forward to helping Aircrafters spread it
's knowledge about building and buying all kinds of experimental aircraft.
We look forward to having you back in our hanger. This year we will celebr
ate our 15 year anniversary, that makes us one of the oldest and most exper
ienced builder's assistance and experimental aircraft maintenance shops in
the world. Today we are still staffed 100% by A&P mechanics and Dave Saylo
r is our in house IA. We have a great relationship with the Sacramento FSD
O and always assure compliance with the FAA's 50% rule. We make sure that
every aspect of your build is quality, detailed and lasting.
As a new service, we are offering maintenance on Light Sport Aircraft, both
experimental and certified, along with Rotax service. Today's Light Sport
Aircraft utilize technology that is new to many shops, but old hand for us
=2E We have been working with composites, advanced avionics and glass panels
for over a decade, one reason that makes us the best shop around for LSAs.
Considering an RV-12? Give us a call.
Feel free to drop by or drop a line anytime, I can't wait to meet you in pe
rson,
Tim Farrell
Manager/Owner
Aircrafters
(831) 722-9141
motoracer@gmail.com
Other Matronics Email List Services
These Email List Services are sponsored solely by Matronics and through the generous Contributions of its members.
-- Please support this service by making your Contribution today! --
|