[time-nuts] GPS referencing of broadcast stations...

Burt I. Weiner biwa at att.net
Thu Oct 13 14:33:24 EDT 2011

Hi Mike,

Several years ago there was some discussion about using a FM carrier 
as a calibrating signal by driving close to a tower supporting a FM 
broadcast antenna then adjusting the counter's time base so the 
display agrees with the station's assigned frequency.  The fellow 
suggesting this soon found out that this was not such a great idea.

One of the most annoying problems I have in measuring FM stations is 
the time involved.  I need to catch the carrier, as the FCC refers to 
it, at rest.  This can take quite a while and I have to stare at my 
equipment and not blink.  In the olden days I could call the station 
at 2 or 3 o:clock on Monday mornings during transmitter maintenance 
or get them to remove modulation for 2 or 3 seconds by having all 
faders down at the end of a song in order to get a quiet 
carrier.  Now-a-days, if the PD thinks there's even 1 second of 
silence at 2 AM, all the listeners will all tune out never to come 
back, they will not show up in the ratings, the station will go 
bankrupt and they'll all lost their jobs and of course it will be his 
fault for allowing it.  It's not the FM modulation that's so much the 
problem as it is the exciter's AFC loop chasing modulation.  A real 
good demonstration of this is to beat the incoming FM carrier against 
a stable signal generator.  If you were to suddenly remove the 
modulation, say pull the BNC off of the composite input, you would 
hear the beat flop back and forth until it settles.  How fast and how 
much it settles depends on the particular exciter and its AFC loop 
constants.  Some settle almost instantly, other's may take 2 to 3 
seconds regardless of the reference source.

As you know, with FM exciters there is a problem with truly locking 
the carrier (FMO) to any reference.  Definitely you can tie the 
exciter's internal reference to a GPS reference and lock that part of 
the exciter.  The real problem comes when you try and "hard lock" the 
FMO portion of the exciter to that reference.  The FMO has to be able 
to be "springy" enough to be modulated by the lowest modulating 
frequency; that was the whole point of moving away from crystal 
controlled FMO's or multiplying a bazillion times back in the 40's 
and 50's.  If you were to hard lock the FMO to the reference it would 
almost be the same as simply replacing the FMO with a crystal 
controlled oscillator.  I know there are some folks who are field 
modifying the time constants and going to a lower impedance to drive 
the varactor modulator diodes to reduce modulation peak overshoot, 
but you can only take that so far before you lock the modulation 
right out of the beast.

As far as synchronizing the audio between the digital and analog 
signals goes...  Well, I have a Radio Shack Accurian, a Radioosphphy 
something or other, and a Boston acoustics receiver. None of them 
agree in the delay through them in spite of using iBiquitiy's 
software - and the difference can be quite noticeable.  I suppose the 
best you can do is to go by the delay in your IBOC modulation monitor 
and hope for the best.  A buddy of mine goes by his PD's radio which 
results in the best sleep at night.

I realize my two diatribes have been sort of off topic for the group, 
but since the subject of using broadcast signals as references had 
come up I though I'd squeak up from my normal lurking.

Burt, K6OQK

At 07:39 AM 10/13/2011, time-nuts-request at febo.com wrote
>We don't have any AM stations, but the big reason for GPS locking of
>our FM stations for IBOC is to stabilize diversity delay.
>The Analog component of the audio for HD-1 is delayed to synchronize
>with the digital components so that if the digital carriers are lost, an
>IBOC receiver will "blend" back to Analog. If it's not synchronized with
>in a few samples, you hear a jump in the audio.
>Mike - Vermont Public Raido Engineering

Burt I. Weiner Associates
Broadcast Technical Services
Glendale, California  U.S.A.
biwa at att.net

More information about the time-nuts mailing list