[R-390] R-390A vs RA 17

[David Wise]

> From: Ed Tanton [mailto:n4xy@earthlink.net]
> 
> I have both... and both are terrific examples of the
> then-state-of-the-art. You have it exactly correct about repairing the
> RA-17... I also have an HRO-500 that I need to work of since 
> it doesn't
> lock above 20MHz... it is similar to the RA-17 with its early 
> PLL design
> (the '17 being implemented using tubes and the '500 with all germanium
> transistors.) My R-390-A is the best one can be, having been given the

I can't speak for the HRO-500, but there is no
PLL in the RA-17 et al.  The key to the PLL idea
is feedback: the phase difference between the
desired signal and the oscillator is fed back
to the oscillator as a frequency control voltage,
in such a way that it tends to reduce the phase
difference.

The Wadley Drift-Cancellation Loop employed by
the Racals, the Yaesu FRG-7, and several others
is a feedforward design involving filtering.
(IMO "Loop" is a misnomer.  Also IMO, it's harder
to understand than a PLL.)  When you turn the
MHz dial you're basically selecting one particular
harmonic of a 1MHz crystal to use in the conversion.
The drift cancellation happens because,
mathematically speaking, one conversion adds the
HFO while another conversion subtracts it.
Result: no HFO drift.  In spite of having an
HFO, it will be as stable as your R-390A, and
if you ignore the way the "first oscillator"
and "second oscillator" waveforms are generated,
it's simply another triple-conversion receiver.

Caveat: I know the FRG-7.  If you want to know
more about its rendition of this neat circuit,
ask.  If I've glossed or mixed up some points on
the Racal, please enlighten me.  And if you have
an RA-17 that's looking for a home, look no further :-)

Regards,
Dave Wise