Phase or frequency modulation?

I have been checking the performance of a variable-reactance type of
frequency modulator which 'pulls' a crystal oscillator.  After
multiplication and mixing, the signal appears at 145 Mc/s.

Listening to this signal on an Icom 706 MkII transceiver I found it was
barely intelligible, with severe high frequency cut.  At first I
suspected my modulator but I checked the audio output of the Icom with a
good-quality signal generator and found the response was:

200c/s : -3dB
400c/s :  0dB
750c/s : -3dB
1 Kc/s : -6dB
1k5    : -10dB
2k0    : -13dB
2k5    : -16dB
3k0    : -18dB

(Using the wideband FM setting of the Icom produced similar results, so
the limitation was in the detector/A.F. stages, not in the I.F. filter)
This looks as though EITHER a 6dB per octave response is being imposed
on the output of the FM detector OR the detector is expecting phase
modulation.

The handbook for the Icom refers throughout to frequency modulation and
does not mention phase modulation.  Most references to modulation in the
2-metre band (144-146 Mc/s in the U.K.) mention frequency modulation and
the use of phase modulation would cause 'splash' into adjacent channels
at higher audio frequencies because of the rising characteristic.

Has my Icom been designed for a market where phase modulation is the
norm or is there another explantion?

--
~ Liz Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk