"Phil B" wrote in message
...

AM broadcast stations are required to cut off their high audio frequency
abruptly at 5kHz to prevent interference to adjacent channels spaced +
or - 10kHz.


I've read articles which claim there was no specific limit on AM audio
bandwidth, but the FCC required the stations to limit bandwidth to limit
interference. Given the normal minimum station spacing of 30 kHz in a given
area, this would imply a maximum audio bandwidth of 15 kHz. If sideband
splatter is any indication, I know the old rocker WCFL at 1000 kHz would
splat out the lower sideband of KDKA at 1020 kHz in the Chicago area. Now,
KDKA is an easy nightthime catch.

As stations were added to markets, the FCC limited the audio bandwidth to 10
kHz.

I've read textbooks which claim a maximum allowed 5 kHz audio bandwidth but
I don't trust the textbooks, so I searched the FCC website.

I came up with:

3. Sound Broadcasting

Sound broadcasting, double-sideband..

BINFn/INF=2M, M may vary between 4000 and 10000
depending on the quality desired

This defination was among a group above the FCC's formulas:

BINFn/INF = Necessary bandwidth in hertz

So, if I'm reading this correctly, the necessary bandwidth for standard AM
will be twice the audio bandwidth, which must be between a minimum audio
bandwidth of 4000 Hz and a maximum audio bandwidth of 10,000Hz.

This is from:

http://frwebgate.access.gpo.gov/cgi-bin/get-cfr.cgi?TITLE=47&PART=2&SECTION=202&YEAR=2001&TYPE =TEXT

Frank Dresser

So, if I'm reading this correctly, the necessary bandwidth for standard AM
will be twice the audio bandwidth, which must be between a minimum audio
bandwidth of 4000 Hz and a maximum audio bandwidth of 10,000Hz.



Contrary to popular belief, AM stations broadcast with a bandwidth of
20KHz, which
makes for audio up to 10KHz. This much bandwidth in a tuner works well
for local stations,
but for DX work you'd want to cut your bandwidth in your receiver to
+-5KHz to avoid
excessive splatter. Even then, you'll still get some "monkey chatter"
from an adjacent channel
station. "Monkey chatter" is a technical term for the modulated audio
that came from station
A showing up "upside down" when you are listening to station B, 10KHz
away on the dial.
"Upside down" meaning that an audio tone of 9KHz station A transmitted
gets demodulated
by your radio when it's tuned to station B, as a 1Khz tone. 9.5Khz -
500Hz, and so on.
Human speach "inverted" this way sounds like "monkey chatter". The only
way to reduce
monkey chatter is to null out station A with the loop antenna. But if
there's another station C
10KHz on the other side of the desired station, and not in the same
direction of the first
undesired station, you're sunk.

You'll also want a sharp 10KHz notch filter (9Khz in Europe and
Australia and elsewhere).
That's to get rid of the heterodyne from the adjacent stations' carriers.

If you can find a copy of RDH4 (Radio Designer's Handbook edition 4), it has
lots of info on radio receiver design. But it assumes that you have a
working
knowledge of electrical engineering. That is, not a beginner's book.

Robert Casey wrote:




So, if I'm reading this correctly, the necessary bandwidth for standard AM
will be twice the audio bandwidth, which must be between a minimum audio
bandwidth of 4000 Hz and a maximum audio bandwidth of 10,000Hz.



Contrary to popular belief, AM stations broadcast with a bandwidth of
20KHz, which
makes for audio up to 10KHz. This much bandwidth in a tuner works well
for local stations,
but for DX work you'd want to cut your bandwidth in your receiver to
+-5KHz to avoid
excessive splatter. Even then, you'll still get some "monkey chatter"
from an adjacent channel
station. "Monkey chatter" is a technical term for the modulated audio
that came from station
A showing up "upside down" when you are listening to station B, 10KHz
away on the dial.
"Upside down" meaning that an audio tone of 9KHz station A transmitted
gets demodulated
by your radio when it's tuned to station B, as a 1Khz tone. 9.5Khz -
500Hz, and so on.
Human speach "inverted" this way sounds like "monkey chatter". The only
way to reduce
monkey chatter is to null out station A with the loop antenna. But if
there's another station C
10KHz on the other side of the desired station, and not in the same
direction of the first
undesired station, you're sunk.

What you are saying here is true about monkey chatter,
and "inversion" of music and speach.
But usually it only applies to distance listening.

And a sharp bridged T LC notch notch filter won't reduce the monkey
chatter, just the 9/10kHz whistle from carriers interfering.



You'll also want a sharp 10KHz notch filter (9Khz in Europe and
Australia and elsewhere).
That's to get rid of the heterodyne from the adjacent stations' carriers.

If you can find a copy of RDH4 (Radio Designer's Handbook edition 4), it has
lots of info on radio receiver design. But it assumes that you have a
working
knowledge of electrical engineering. That is, not a beginner's book.

Well said, the RDH4 isn't easy to read. But each sentence is never a waste of
words,
so you must consider each word you read.

Building radios the way they did in 1955 from nothing upwards using a few tubes
takes a lot of effort, and the main thing apart from knowledge that makes 95% of
ppl
abandon projects is the building of coils and the discipline about
constructional quality.

Patrick Turner.