Paul:

 Reading all your posts on EFI systems and GPS intefacing has kept me
 thinking about what might be the best operating system to use if one
 wanted to create an entire home-brewed glass-panel display.

 I ran the idea past my wife's cousin, who is now retired from a position
 as a programmer for some Silicon Valley hardware developer.  He
 suggested a look at TCL (pronouced: Tickle)

 see:
 http://en.labs.wikimedia.org/wiki/Tcl_Programming/Introduction#So_what_is_Tcl.3F

 He sparked my interest so I've been hacking around with it for the last
 couple of day.  The weather has been snowy/rainy and we occasionally
 have up to 3 dogs and 8 grandchildren under 8 years of age ramming
 around the house, so hiding in my office is often prudent behavior :)

 I first explored methods using TCL GUI tools to draw the panel
 instrument (which I've done with the vertical gauges on the bottom of my
 panel).  I  then decided that displaying images of real instruments was
 a faster/prettier way to do it. I can now simulate the appearance of
 most anything on the screen.

 I now need to figure out how TCL can be used to read an external
 data-source.
 I will probably test it by having it read test-numbers from a simple
 external data file, but according to claims from online sites, I should
 be able to read data from most any data port as well.  That is my next
 project.

 This is one of my most recent panel displays.

 Robin Bergland


 Robin,
 This is very impressive and very nice appearance.

 SUGGESTION: include an attitude indicator, and layout the "clocks" in the
 standard T layout.

 Philippe CHENEVIER

 Can TCL be used to program a Honeycomb tablet?
 Tablets have build in gyro's.


 Paul Lamar

 I believe JAVA is the only programming language supported except
 native mode. So basically no. But we can use TCL with Ubuntu or Linux.

 Henry

 Unfortunately the notebooks that support Linux do not have build in GPS
 or gyro's. The GPS is easy to fix as there are several USB GPS's. I have
 it working
 in Ubuntu 10.04 but the gyro is more of a problem.
 There must be  merge bewtween Android Honeycomb and Linux sometime soon.

 "A microelectromechanical system (MEMS) is an embedded system that
 integrates
 electronic and mechanical components at a very small scale.  A basic MEMS
 device
 consists of an ASIC and a micro-machined silicon sensor.  The AGD1 2022
 FP6AQ chip
 found in the iPhone 4 is a MEMS gyroscope designed by STMicroelectronics."

 Apparently these come in two version. One is designed to be embedded into
 the memory
 map of a computer and the O.S. (Android Honeycomb) must be written to
 access it... I think.
 My guess Android Honeycomb does this as does the I-pad.

 The second one is a standalone 3 axis analog output probably used in toy
 helicopters.
 Here is a pdf. The AVRXplain $30 PC board could be used to convert these
 three analog outputs
 to a USB signal.... Perhaps..... If we are lucky. See gyro the data sheet
 pdf.

 BTW there could be a market for a stand alone attitude gyro using this PC
 board and a gyro
 chip and an OLED display. Piece of cake after you learn AVR asm language.
 Cost would be well under $100 while stand alone electro mechanical gyro's
 are $1000 to $3000.


 Paul Lamar


You could be on to something with the small gyro- I cant believe I didnt
think of it. My model helicopter has a single axis gyro for yaw. My first
one was a real spinning gyro, the second iteration is a little more
sophisticated. It uses a fiber optic cable wrapped around a tube oriented
in the direction of the axis. The chip reads the "red shift" of the light
frequency to know when the helo is rotating around its rotor.

There should be no reason this $100 component cant be used (or 3 of them)
in an A/C to provide real time gyroscopic input to the control or
monitoring unit. The unit on my toy helo uses a simple 3 prong plug- 7.2
volt +, 7.2 volt - and a signal out. The signal out increases or decreases
in voltage depending on which way I forgot to add rudder when I am flying.

Kevin Alderman

Got any references for that fiber optic cable wrapped around a tube?
I am ignorant of that device.

The solid state attitude gyro  uses the microcomputer to integrate a rate
gyro such as a MEMS device. About ten years ago, I as I recall, Goodyear
had a both at OSH with a demo unit. At first I was impressed. Then I picked
it up and shook it. It lost horizontal reference. The engineer in the
booth was
a bit upset that I did that. He claimed one did not see that level of
turbulence in
a real aircraft.

IMHO combining the GPS with the rate gyro info can produce a good attitude
gyro and is not out of the question for an experimenter on here (once one
learns AVR asm :)) to build using a cheap $30 microcomputer board
with A to D's.

Assuming for the moment the throttle is fixed and the
GPS speed is increasing the aircraft is diving. If the GPS is telling you
there is a heading change and the readings for the yaw rate gyro and roll
rate
gyro are significant you know the roll angle assuming a coordinated turn.
One can then get data from the ball in a turn and bank in an uncoordinated
turn. If the GPS says there is no speed change and and no turn you are
obviously
flying straight and level. Other parameters that could be monitored is the
manifold
pressure and RPM to detect a change in power.

The basic idea  here is take data in from multiple sensors (not just the
rate
gyro's) and use logic and reason to determine what the pitch and roll
angles are.

The classic mechanical attitude gyro uses only one sensor, namely the
spinning flywheel.


Paul Lamar

http://www.rchelicopterfun.com/rc-helicopter-gyro.html

This site gives a run down of the latest gyros. The FO unit isnt listed,
and like the mechanical unit may be considered a dinosaur.
Note the "heading hold" function. This lets the RC pilot flip a switch on
the transmitter and the helo will stay pointed in the same direction. It
works quite well. When I get airborne, I can flip the HH switch and no
matter what I do other than move the rudder stick, it stays pointed in the
same direction. Quite handy for a helicopter.

I haven't tried it, but I would assume for fixed wing RC it would work like
a single axis autopilot, i.e. wings level.

http://www.hobbypartz.com/gyro.html

This is a site I found that has more info- note that since my helo is
several years old (which is a good thing it means I haven't destroyed it in
a crash) I didnt know that the prices were as low as they are. Some of
these are as low as $12.70.

I am not quite to the point of making a test unit, but if any of you are it
might be beneficial to get a cheap unit and test it.
3 of these in the 3 axes of an A/C could make a really inexpensive
autopilot.

Also, in RC we have servos just like the big planes for control. And we
have some planes that are 1/3 scale and larger. This means BIG servos with
lots of power. I will keep looking.............

Kevin Alderman