Topband: Summary: Collins R390 and 75A4 receivers

[Steve Ireland]

G'day,

Thanks to all those who sent me data and their views on the Collins 75A4
and the Collins R390.  It has been fascinating.  What it has highlighted
for me, as a side issue, is the strengths of modern crystal filtering and
how dangerous nostalgia can be, when it comes to receiver performance.

For those R390/75A4 enthusiasts, Bob Sherwood and Sherwood Engineering has
produced some interesting data on both receivers.  Firstly, the dynamic
range of the R390A version of the R390 looks better than the 75A4, based on
Sherwood measuremements.   

First the Sherwood table with the R390A info (thanks, Bill W4ZV): 

Rig                           IMDDR3 (@ 5 kHz) BDR

Ten-Tec Orion                  101 (claimed)    ?
Elecraft K2                     88             126 
Ten-Tec Omni 6+                 86             119
Yaesu FT-1000MP                 83             111
ICOM IC-756 Pro                 80             104
Collins R-390A                  79 (@ 2 kHz)    ?
Yaesu FT-1000MP Mark V          78             106
ICOM IC-775DSP                  77             104
ICOM IC-706 MkII G              74              86
Yaesu FT-1000MP Field           73             107
Kenwood TS-570D                 72              87
ICOM IC-756                     67              98

Now the Collins 74A4 info (thanks, Mike W4EF), which was apparently
presented in a talk at the
1979 W9DXCC Meeting and is attributed to Bob Sherwood.

 Measurements done at 500 hz bandwidth.

 Noise floor:  -143.5 dBm
 Intermodulation: @20 Khz
 dynamic range: 66 dB
 3rd order intercept:  -44.5 dBm

 @2 Khz
 Dynamic range:  tone
 3rd order intercept:  -- 

Blocking dynamic range:
 @ 20 Khz:  95 dB
 @ 2 Khz:  tone

As can be seen, the dynamic range of the R390A looks to be better than the
75A4 - but both are well behind the transceivers serious topbanders
habitually use today, such as the FT1000MP/Omni 6, etc.  

When it comes to filtering, it seems the 75A4 has the edge over the R390
series, with its sharper skirted filters and passband/notch tuning system.
Tom W8JI, who has both R390 and R390A variants, points out:
"The R390A has only a single pole crystal way back in the IF chain to do
the less than 2kHz stuff, so the filter skirts are horrid. The R390, with
only L/C filtering is even worse. Both have more filter blow by than my
IC706!"

My observation in comparing my R390 with my FT1000MP on CW is the skirts on
the former sound as though they slope gently uphill - steep is definitely
not a word you would use.  The FT1000MP is like a razor in comparison -
even with just the stock 500Hz filters.

For a topbander, the Collins 75A4 looks like better prospect for DX
operation than a R390 with its better skirt selectivity and notch/passband
facilities, but its strong signal handling capabilities look so far behind
the radios of today, you could ask: "why are 75A4s still popular among some
Big Gun topbanders?"

>From the feedback received at VK6VZ, the key lies in the 75A4's abilities
to pull a weak signal out of the noise, owing to its lack of phase
noise/simple superhetrodyne design, particularly when it is HEAVILY
MODIFIED with better filters, as per the articles in Electric Radio
magazine, and backed up with a good audio DSP CW filter.

It seems to me that it is also far easier to modify/optimise the RF
performance of a 75A4/Drake R4C with their wide open spaces and simple
design than it is to do the same with a multiple IF/surface mounted
component FT1000MP, Omni 6, etc.  When I open up my FT1000MP, my heart
sinks...

As W8JI and others have pointed out, an stock unmodified R4C is pretty
useless as a serious 160m receiver for weak signal work. Sounds as though
the same may be true for the 75A4.  However, both can be improved out of
sight by a reasonably technically competent radio amateur - by fitting
better, more modern crystal filters, some optimisation and a good audio DSP
filter, such as the Timewave 599.

Vy 73,

Steve, VK6VZ