GeorgeM
Senior Heliman
Location: Malta, Europe
| Hi..
I think this is going to be a long post from me, but perhaps it will enable me to explain with some physical examples.
The round ball with the hole in front is a representation of the camera covering bubble ala wescam. The three purple pipes and the greenish disks are 3 identical tail booms + tail rotors respectively from a helicopter kit. They each have a brushless motor turning the tail rotor through a belt or torque tube arraingement. Whatever you like !!!
If one has ever experimented with heading hold gyros, even if there is no torque from the main rotor, one can observe that they will stay where they are unless a commanded force acts upon it....Either you from the TX or else wind/physical force. If the latter is exerted, the HH gyro will try to correct it and go back to where it was.
To see this, if you have a small electric heli with a HH gyro, take the complete rotorhead off, and suspend the model from a string so that the model balances horizontally. Run it up. You will see that the tail will remain pointing to where it was.
Now multiply your mini heli x3 and stick eaxh one 90deg away from each other... XYZ axis. each tail will control the movement of the gimbal init's axis. If wind or any other force acts on the thing, the gyros will compensate. I have tried this while repairing on of my logo20's. I ran it up without the head.. only the mainshaft and tail assy. I can tell you that if you try to change the direction to what the heli is pointing at, you'll have to fight against a fierce tail rotor 
That's how this idea started creeping in my mind !
The structure should have a 3 axis gymbal so that the ball is left free to rotate in any axis. But to understand this method without going into the real world of how to actually build it and let those tailpipes free to go in all axis, that's another matter to solve.
Having the sort of performance of a Gy401 and digi fast servo is an asset, and you'll eliminate the drift problems encountered with previous HH gyros.
Why complicate it so much....you might ask ??? There are systems from JohnC of CARVEC , Rotomotion, and others that have developed and almost perfected this stabilization in 3 axis in the model scale. I would like to purchase both systems to play with, but I have budget restrictions. Therefore I am still striving to solve this with as little outlay as possible, using off the shelf components.
A GY401 was painstraikingly designed to function so well in a model helicopter with the dynamics of the tail rotor working in air.
If you try to mount the gyro directly on a camera gymbal platform, you will notice fast oscillations. If you add delay, you will have somewhat lower frequency oscillations ( time untill the delayed signal passes to the servo ). However a gyro system works well where it's intended...on a helicopter !!!!!
What is needed is therfore for a more practical system ( to take out those tailrotors) is a normal 3 axis gymbal. A gyro sitting direcly on the chassis (gymbal ) it is moving / controling , a fast servo ( to react instantly and correct) and a little circuit in between !!!
This circuit...being a PIC or any other microcontroller or discrete components, will emulate the dynamics of a tail rotor......... ( wow ..... what a study in itself this is.... try going to http://www.unicopter.com/ and look for tail dynamics article....even do a search on google !! )
If we have this circuit emulating the tail rotor PID or characteristics, the servo can be used to directly control the same axis that is carrying the respective gyro ( trough a reduction ratio or directly if you have very powerful and fast servos ).
And this can be repeated for the 3 axis of a gimbal, thus eliminating the cumbersome tailbooms and tail rotors in the diagram link above.
Is there anywhere I have gone wrong ???
What I intend to do... or need help in , is the tail emulating circuitry. This will enable me/us to take those imaginary 3 tailbooms off the gymbal, and replace with the more conventional servo and gearbox....BUTTT with the gyro sitting on the respective actuated axis itself...thereby having tight close loop performance...( just like on a Vigor or Fury doing Fast 3D backwards and the tail remains solid !! )
I would like the Programming Gurus and those who have done stabilization experimentation to comment on this and point where I could be wrong.
As a sideplate argument, one can always have large mechanical gyros like Airfoilhelicam's, but to carry 3 of those, and the ascociated control of them, you'll need an R-Max by yamaha !!!
I am happy to have shared this with all of you. think it carefully. I would like to receive feedback.
I know this little essay will help clear things up a bit !!!
hope to hear comments !!
Best Wishes
George |
