Conference vmszoo::rc

    
    Yes, I'm pretty sure you can use two channels - in fact some
    radio manufacturers do that now for redundancy as protection
    against radio failure or interference.  I don't think you'll
    have too much trouble with the spacing - of course 3IM is something
    to think about but _any_ three equally spaced frequencies can give you
    3IM problems.  'Course Tycho can probably take a lot more radio
    hits than our planes can without a catastrophy!
    
    Bill
    

    I can see it now, the HUD in your scale Eagle lights up and
    says
    
    %%%%%%%%%%  OPCOM 17:45:20 %%%%%%%%%%
    UPLINK-F-PARITYERR, control frequency corrupted, switching to
    alternate.
    
    %%%%%%%%%%  OPCOM 17:45:21 %%%%%%%%%%
    UPLINK-I-NEWCHAN, control established on channel !AS
                                                                      
    -----
    
    Wonder how much it would cost to have "frequency-agile" RC radios?
    Change the data frame to be 16 bits of ident, then 16 bits of command,
    then 32 bits of ECC.  Frames repeat every 64/1600 or 25'th of a
    second.  Reciever hunts around till it finds (after ECC) a transmitter
    with the correct ident, then loiters there, reading commands.
    
    If interference gets heavy, the frames won't pass ECC and so the ship
    fails to respond.  Pilot hits "new freq",  transmitter picks a new freq
    (after a short listen), transmits a a few frames of "goto new freq N"
    on the old freq, and then goes back to sending data on the new
    frequency. 
    
    Meanwhile, the reciever may or may not be able to recieve the "new
    freq" command.  
                                                       
        If it can recieve the "new freq" message, it just jumps there.
    
    	If it can't recieve the message, the reciever has more work
    	to do. It sees total loss of signal, but it still remembers 
    	the old frequency, and it has a table in ROM describing the 
    	search path that the transmitter will look for an open freq.  
    	So, it starts searching for a transmission freq. that has 
    	a clear signal, and the proper ident code.  It takes a full
    	frame time for the reciever to get enough info to decide that
    	it has (or hasn't) found the right freq.  Plus, it's likely
    	that the time alignment of various transmitters will not be
    	correct, so add another half-frame time per frequency searched.
    	
    	So, it looks like it will take 60 milliseconds/frequency on
    	the average to find the transmitter again.  Figure that there
    	might be six frequencies that will be probed, that's just
    	.36 seconds to reconfigure.
    
    -----
    
    And, if you've got the bucks for this system, try:
    	
    	Reciever has a low-duty cycle on-board transmitter.  When it
    	gets .25 second (ten frames) of errors in a row, it pulses out the 
    	ident with a message that says "Hey, there's a lot of junk here, 
    	let's go find another channel.  Try channel N".  Transmitter hears
    	this burst, and goes for a new channel.
    
    	Note that this system could detect interference and change channels
    	slightly faster than a human could react to being hit in the
    	shins with a leaf-rake.
    
    -----
    
    Anybody got $500,000 in venture capital?
             
    		:-)
    	Bill Yerazunis, FCC commercial+radar lic. # PG-2-4483

    Bill, some great ideas there, but you forgot spread spectrum!  In
    the not too distant future (on a later model) I may look into some
    ham packet radio gear, which would allow reasonable amounts of digital
    data transfer.
    
    For what it's worth, I got an encoder going, driven from my new
    analog I/O board, so sometime tonite I ought to have Tycho controlled
    by computer!
    
    Willie

    Interference can also occur at the Rx end of this "loop".  If
    there is broad-band RF interference at the receiver, it is
    unlikely, unless alternate channels are widely separated, that
    hunting frequencies would be of any benefit.  If channels were
    separated enough to make such a scheme workable, I think that you
    will find that a VERY large increase in available frequencies
    would be necessary.

    Well, just on the off chance that anyone cares, I can now control Tycho
    thru my S-100 machine!  I had a little trouble with the analog output
    section of the board, it has 0.2 volt spikes at around 33 KHz [they use
    a single converter and scan 8 track/hold amplifiers] which was
    asynchronous with the frame rate, and there was so much jitter on all 7
    channels that the poor Tyke just sat there and shivered! A little
    filtering and I managed to drive him around the apartment, but there's
    still some RFI getting into my analog inputs, so I have to change over
    to shielded wire.  Not terribly portable, and it'll be worse when
    we get the Amiga hooked up and running as a Head Up Display....
    
    Definate reccomendations for the Signetics NE5044 R/C encoder chip,
    with the exception of the noisy voltages I was feeding it, it worked
    just fine first try!
    
    Willie 

    True, broadband jamming near the Rx would knock it down (i.e. servo
    with no bypass caps, electric motor unshielded without bypassing,
    etc.).
    
    Maybe spread-spectrum is the way to go.  What are the "unlimited"
    frequencies now?
    
    	-Bill
    

    I finally got the software written to do the 3 second delay on the
    controls, and it's rather difficult to control.  I'm going to have
    to wait for daylight, as the Tyke is chewing up the furniture here...
    Those who saw Tycho at the car race remember how slowly he moves,
    it's actually dangerous (to the furniture) to run him at full throttle,
    3 seconds is a long time!
    
    I had a rather interesting time when I first fired up my program,
    I had swapped the input and output pointers in my ring buffer, so
    instead of having a 3 second delay, I had a 38 second delay!  Move
    the joysticks carefully, nothing happens.  Move them to the limits
    of their travel, nothing happens.  Start checking to make sure the
    transmitter is turned on and all the wires are hooked up right and
    he suddenly wakes up and starts moving!
    
    I did learn one thing, the Futaba transmitter modules are not meant
    to run at 12 volts for long periods of time, as they get quite hot.
    I've cut a hole in one and soldered a heat-sink to the final output
    transistor which helps a lot, but it still gets warm...           
    
    Willie

    As far as I know, there are no legal telemetry frequencies available
    to us RC'ers.  I doubt that you would run into legal problems if
    you used an RC transmitter on a surface vehicle this way, but you
    may have overload problems on the local receiver in the vehicle.
    To minimize this, you may want to lower the output power of the
    transmitter on the vehicle and separate the frequencies as far as
    possible.  This will allow the filters in the vehicle's receiver
    to reject the strong off frequency signal right near it.  You may
    also be able to help things by playing with antenna loacations.
    I have seen receiver overload at the flying field many times.
    If you put an interfering transmitter near a plane (three feet),
    it will overload the receiver and glitch the servos.
    
    CHarlie
    

re: .8

    A long, long time ago I remember seeing an Estes catalog with a
    transmitter module which would let you run a downlink. As best as I
    can remember, it was on a 27mhz freq. This would be disaster for
    a control function but if you're just monitoring stuff, and you're
    running either checksums or CRCs on you data blocks, then it shouldn't
    be that much of a problem if you pick a reasonably clean channel to
    start with.

    There's supposed to be a model rocketry notes file around someplace
    and i'll bet that those guys know how to do it. Might be worth checking
    it out....

                     !
                   --+--
G. Schrader     o___<0>___o
                  *  *  *

    Kind of short notice, but here it is FWIW:
    
    Thursday, June 6, I'll have my WALDO vehicle at LTN1, and will be
    taking him for a drive after lunch (around 1:00-ish).  Waldo is a
    rather large scale (3 feet square) radio controlled vehicle, which I'm
    using for lunar teleoperations research [and just plain fun].  It uses
    lawn tractor tires, cordless drills, and has a 25 amp-hour (that's
    right, 25,000 mAH) 8-volt battery pack.  It's currently controlled by
    touch-tones over a 2-meter ham radio, but will be upgraded soon to
    packet-radio control.  Some of you who were into R/C a couple of years
    ago may remember Tycho, my heavily modified Clod Buster R/C truck with
    the TV camera, Waldo is his big brother...
    
    I don't follow this notes file any more, so if you have any questions
    or comments, send mail to SNDPIT::SMITH.
    
    Willie