[Collins] Re: S Line series 3, 2nd m8ixer mods

[Dr. Gerald N. Johnson]

On Tue, 2007-09-04 at 08:10 -0500, w9hak wrote:
> Gerald,
> 
> You peaked my interest when you mentioned modifications to the 2nd mixer 
> to reduce noise. I use my 75S3 daily to chase dx with the signals many 
> times 339 or lower. The unit is stock Collins, including the great 200 
> cps filter, and I have done all the Dennis Brothers mods per his CD. The 
> sensitivity is as good as it gets for this receiver. Therefore 
> improvement is the s/n ratio caught my attention. Any information would 
> be greatly appreciated.
> 
> Smith Bradford
> W9HAK

I spent months investigating noise sources in the receiver. What I found
was a bit of hum modulated noise coming from the mechanical filter. I
cured that with the mu-metal over the plastic cased filter. The major
noise was from the second mixer. A tube with a gm of 4000 doesn't make a
low noise mixer. So I made a plug in adapter that holds a tube with a gm
of 50,000 as I recall an Amperex 6688. I didn't have a 32S, so I didn't
bother keeping the cathode follower for transceive operation. I changed
the bias, I'm sure and then I tapped the grid down to the capacitive tap
on the first IF transformer. The same transformer is used in the
transmitter and the taps are connected for mixer neutralization. For HF
service this is a hair too far down on signal, but I was using the
receiver mostly with a 2m converter that had enough RF gain to overcome
the loss in sensitivity though the converter had just a single 6CW4
neutralized RF stage and a 6CW4 mixer. I put an attenuator between the
2m converter and the 75S-3B. I adjusted the attenuation to the maximum
value that didn't increase the system noise figure.

At 2m before the mod, W5WXV, who lived about 20 miles away running a
good 2m signal was found 20 or 30 times in 200 KHz and after mod was
found three places in that spectrum, full scale where he was
transmitting and about even with the noise the other two places. Now
this was a tough test with his quad of Telrex yagis probably 17 or 18 dB
gain with 650 watts RF to the coax and I had a pair of HyGain yagis
probably 13 dB gain pointed towards each other. For years Al kept a
regular schedule running 2m CW with a station at Quincy IL. Al was in
Garland TX and he had golden ears that could extract a signal 20 or 30
dB down from noise.

I'm sure I have circuit details somewhere, but the Amperex tube is
critical to the low noise mixer with dynamic range and I don't know if
it can be found. I believe I have a spare but there's no bench space to
put the receiver on so it may be a while before it gets attention. Its
been a while since its been turned on and 43 years old electrolytics
will probably object to having power applied today. While I've not tried
it, I don't think a double gate MOSFET will handle as much signal as the
tube and double gate MOSFETs are nearly as hard to find as tubes.

I've not seen the Dennis Brother's CD. Its probably something I should
study to see if I agree with its mods.

Ii know W0MLY sold a double gate MOSFET replacement for the first mixer
and thought I was daft fiddling with the second mixer.

Speaking of audio S/N, it would be nice to have a SSB filter right
before the product detector to minimize IF stage noise from the other
side band. I have a 2m transceiver kit begun near the workbench that
does just that and uses VCXO for the local oscillator and a single
conversion receiver. It should be superb for recovered S/N with very low
LO phase noise and filters at both ends of the IF strip. Called
"Hohentwiel."
-- 
73, Jerry, K0CQ, Technical Advisor to the CRA
All content copyright Dr. Gerald N. Johnson, electrical engineer