Most FM radios with integer PLLs will have a loop filter bandwidth of
maybe 150 Hz, - enough to deal with most microphonics.
In order to modulate below about 250Hz, LF audio, like subaudible tones
, are injected into the reference oscillator- the PLL will follow that
because it is within the loop bandwidth. The voiceband audio is split
with HP and LP filter. ABove 250 Hz goes to the varactor modulator on
the VCO, and below 250 Hz goes to the reference frequency modulator.
Modern 2 way radios have fractional divider PLLs and loop bandwidths
of maybe 60 kHz. If applied to the varactor VCO modulator, the
modulation would be corrected- these radios use IQ modulators to
directly modulate the (clean) carrier frequency produced by the
fractional PLL.
So: for the majority of FM gear, do not bank on any decent fidelity
below about 300Hz. Your mileage may vary- it will depend exactly where
the loop bandwidth was set.
The newer 2 way PMR radios, like Tait TM8000 etc, they offer external FM
down to DC modulation on their external IO connectors, - they have an IQ
modulator and the sky is the limit
-glen
On 12/04/2020 4:47 am, Steve wrote:
oops, I keep typing Symbols instead of Samples
/* Slow Data + Voice Bits + Sync Bits = 747 Bits = 5976 Samples */
/* Full Data + Sync Bits = 747 Bits = 5976 Samples */
Without the bit-stuffing the spectrum cleaned up and it actually
sounds more random, but this is just a function of your test program,
as long runs of 1's would make a mess without scrambling. I think most
FM direct modulators have a filter for below 100 Hz.
_______________________________________________
Freetel-codec2 mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/freetel-codec2
