WHISTLER RECEIVER NOTES
		John H. Davis, PO Box 367, Warm Springs GA 31830

	In the August (1994) LOWDOWN, I was delighted to report finally 
hearing whistlers while on vacation in Kansas, and said I'd try to dig up the 
diagram.  Well, for once, I was able to do something when I said I would, and 
not six months later.
	The chief difference between this and most other whistler designs is 
the use of integrated circuits as the exclusive amplifying devices.  Combined 
with two supply rails (+9 and -9 volts), this technique allows simple 
inclusion of passive or active filters for additional noise attenuation, if 
needed.  It also makes for a very strong front end, free of overload and 
intermodulation problems.  While purists might argue the noise advantages of 
a discrete FET front end, the chief noise source in most whistler receivers 
is not the active device.  Thermal noise of the 2.2M input resistor, common 
to many designs, more than swamps the amplifier noise.
	I found early on that the Southeast has strong enough static almost 
year 'round to drown out the most delicate whistlers, and practically no 
sites where a long antenna can be strung in a power-line-free environment.  
Therefore, I gave the front end enough gain to deal with a 9-foot whip on top 
of the car (not fender mounted!) and yet not be overloaded by broadcast, 
Loran-C, and the ambient static levels.  The output is high microphone level 
or low line level, and can be balanced or unbalanced as needed.  (I first 
used it with a borrowed R-DAT recorder with professional balanced inputs, but 
have to settle for a cassette deck now.)  It's not designed for direct head-
phone listening.  Use the recorder's monitor jack or Radio Shack's little 
speaker/amp.
	As I mentioned, the whip is mounted on top of the car, courtesy of 
the luggage rack and a suitable bolt-on mount.  The metal box for the receiver 
sits next to the base of the whip, and connects with a very short length of 
coax.  The car body acts as a counterpoise in this arrangement, although I 
have a lug on the receiver box for a connection to earth as well; none of the 
equipment is connected to ground at any other point.  Well-shielded audio 
cable runs through the window to the recorder inside the car, where I can 
monitor warm and dry.  Best rejection of unwanted noise occurs with balanced 
600 ohm cable.  Even when recording on the unbalanced cassette deck, I use 
the 600 ohm cable and connect to the input through a Radio Shack 600-ohm/hi-Z 
transformer that has an XLR connector on one end and a phone plug on the other.
	The receiver and recorder have their own battery power, so I don't 
depend on the car's electrical system for anything.  I'm not sure what noises 
might arise in modern computerized cars, but in my old Toyota wagon, I could 
do most anything electrical in the vehicle--short of starting the engine--with 
no observable noise.  Better to take precautions, though.
	One peculiar thing you'll find when recording near a highway... tire 
noise!  Tire wear involves rubber molecules gradually being torn loose, one 
at a time, and sticking to the asphalt.  The electrical fields resulting from 
that tearing are easily detected within 150 feet or so, and sound very much as 
if you were recording the passing car with a microphone!
	Variations on the design are possible, as long as one is reasonably 
careful with layout and shielding:
	(1. Gain can be increased for easier line-level recording, by raising 
the value of the 22K feedback resistor of IC2A.  I wouldn't go more than 
double, though.
	(2. Gain can be lowered for easier mic-level recording, by reducing 
the 51K resistor at IC1A.  I wouldn't go below 22K myself.
	(3. To accomodate longwire antennas, change the input network, or make
a second network that can be readily jumpered into service; reduce the 3.3M 
and 10M resistors in proportion to each other, and raise the 150 and 47 pf 
capacitors in inverse proportion to the resistance change.  As a first 
approximation, and assuming a very hum-free location, I'd halve these 
resistors and double the corresponding capacitors for a 150' wire, and do the 
same again for every doubling of antenna length.  Alternately, I'd leave the 
input values as they are, include a 10M pot across the input jack, and trim 
for longer antennas that way.
	Whatever you do, DON'T try listening to whistlers from within a 
thunderstorm, not even with the car-mounted arrangement.  Cars are relatively 
safe from lightning only when they do NOT have a 9-foot antenna and lead-in
furnishing an engraved invitation.  Whistlers do have an ethereal charm, but
I'm not eager to see if they're on the Top 20 charts in the afterlife.


		     (This document is accompanied by two 
		 illustrations:  WHISREC.PCX and WHISMOD.PCX)