[78-L] AM radio--WAS: acoustic recording

[Michael Biel]

> . . . if a 1000KHz signal is modulated with a 14KHz sound,
> the sideband signals generated would be 1014 and 986 KHz,
> both well outside the station's assigned bandwidth...and
> interfering with broadcasters assigned frequencies of 1020 and 980 KHz...?!
> Steven C. Barr

AM stations do not have an assigned bandwidth.  They only have an 
assigned center frequency.  The station at the "frist adjacent 
frequencies" to the 1000 KHz station, 990 and 1010, are assigned enough 
distance and power to not be able to interfere with their first 
adjacents.  They are allowed to completely overlap each other because 
during the day they will not be heard in the territory of the other 
station. At nighttime many of the stations sign off, reduce power, or go 
to directional antennas.  It was the second adjacent stations, 980 and 
1020 that were the problems.  In the late 60s the FCC started a policy 
of "short-spacing" which meant that some of these second adjacents were 
too close together or too powerful.  Audio above 10K would cause 
splatter over the 2nd adjacents IF the radios had a bandwidth that could 
hear those high frequencies. So the radio manufacturers started to dumb 
down their AM sections and reduce the bandwidth, which filtered out all 
the high audio frequencies.   The  reviewers at the magazines would give 
the audio response for FM sections but ignored this on AM sections where 
it was more important.  Before anybody noticed it, most AM radios and 
tuners had top audio response to only 3.5 K.  So radio stations started 
to emphasize the highs even more to try to push the sound through, and 
this made the interference worse. 

In the 1980s the broadcasters requested the FCC to require the receiver 
manufacturers to widen their bandwidth AND require the reduction of the 
transmitter frequency response to match.  Because of the misguided 
"marketplace" policy of the Reagan appointed FCC Chair, Mark Fowler, 
they said it could be voluntary.  Of course nobody did anything because 
both sides had to happen in order for it to work.  In the late 90s the 
FCC enacted NRSC, National Radio Systems Committee, which did restrict 
the frequency response of stations to 10KHZ, and the broadcasters 
encouraged radio manufacturers to improve their bandwith.  A few radios 
are good at that, such as the GE SuperRadio III and the Grundig A-350.  
But stations still overlap their first adjacent stations but no longer 
overlap the second adjacent.  There are some papers about this at the 
National Radio Systems Committee website.

http://www.nrscstandards.org/SG/NRSC-1-A.pdf

http://www.nrscstandards.org/SG/NRSC-2-A.pdf

The first two show the history and the transmitter rules.  The nest 
paper gives the test results of about 50 radios.  If you take a look at 
page 57 you will see that even the famous Bose Wave Radio stinks, with a 
frequency response down to -22 db at 5 KHz.  The problem of AM sound is 
in the radios. 

http://www.nrscstandards.org/SG/NRSC-G100.pdf

Mike Biel  mbiel at mbiel.com 

> ----- Original Message ----- 
> From: "Michael Biel" <mbiel at mbiel.com>
>   
>> I do take exception to David's contention that broadcasting has limited 
>> frequency response.  FM mono extends to 20 K and FM stereo extends to 15 
>> K.    I don't know about Canadian law, but I do know that in the U.S. 
>> there was no upper limit on frequency response on AM until the late 
>> 1990s.  Before then it was routine for AM stations to have response 
>> beyond 10K, usually uip to 13 or 14 K.  The limitations in AM sound is 
>> due to the lousy radios made since the mid-70s.  Radios from the 50s and 
>> 60s are much better before manufacturers purposefully reduced their 
>> bandwidth because of increasing interference from new short-spaced 
>> stations.  Top response now is allowed to 10 K and must be filtered 
>> beyond that.  There are only a couple of radios made today that can meet 
>> that response, such as the GE SuperRadio 3 and the Grundig S-350, but 
>> the stations are broadcasting to 10K, at least here in the U.S.  
>>
>>     
> How is that wide response possible for AM broadcast? An AM signal
> has "sidebands" equal to Fc (the centre frequency) + AND - the modulating
> frequency...am I right here? Therefore, if a 1000KHz signal is modulated
> with a 14KHz sound, the sideband signals generated would be 1014
> and 986 KHz...both well outside the station's assigned bandwidth...and
> interfering with broadcasters assigned frequencies of 1020 and 980 KHz...?!
>
> Steven C. Barr