[Drake] More on the R4-C sensitivity problem

[Martin Sole]

Firstly I would like to thank all who responded to my initial question
about a perceived lack of sensitivity. It seems all the usual
possibilities were candidates, tubes, R's and C's in the AGC circuit,
even perhaps a procedure in the book that rarely works exactly.

I have spent a little more time working on the problem and have
discovered that it may be a little more deep rooted. Before going any
further I think it would be useful to 'set the scene' a little. This
R4-C is a 21xxx series and came to me with 1.5KHz and 500Hz filters
installed. Though I had little opportunity to fully appreciate what I
was getting I am pleased that it is a radio I can 'play' with rather
than being a pristine show piece. It does seem though that this poor
unfortunate has been the victim of some seriously errant butchery and
might well have become someones ugly duckling.

Issues identified so far:

AF gain knob incorrect. Has 1/4" hole but AF gain has 3/16 shaft. PTO
knob has a crank handle installed instead of the normal finger disc. PTO
is stiff, I have re-set the end float. Cleaned and re-greased the
bearing and lubed the brass dial collar. No burrs on the shaft but the
thing is still stiff. 3rd mixer injection mod. 
C100 changed to 0.47uF tantalum, R75 (150R) removed (now replaced).
AM filter bracket missing.
4-NB mounting stand-offs missing.
T7C core very loose.
3rd mixer mod. Seems to be as per the mod listed in drakemod V6 except
that R132 was lifted and C199 retained.

Digging around with a scope showed all sorts of noise on the supply
lines, this was being caused by failure/weakening of both multi-section
electrolytics. Replacement has cleaned this up no end and I am able to
get decent sensitivity, <0.1uV detectable. I removed C199 and fitted
R132 then found that I could not set T6, all sorts of howling,
oscillation etc. It was touchy before but now impossible to contain.
Traced this problem to T5 having been aligned to the wrong side of the
passband. So I now have a half decent stock R-4C but still have a couple
of questions. The S meter alignment now works as per the book with lots
of adjustment range on the bias, zero, and sensitivity pots but having
done the set up in SSB I see that the zero moves upscale when selecting
the 1.5KHz and 500Hz filters and down below zero in the AM position. Is
this normal? Second point I noted was that my 500Hz filter (Shinwa
Drake) appeared to have a very high (20dB) insertion loss. A second
filter (Heath Dynamics) was found to be good. With careful optimisation
of T6 and C42 I can get the Drake filter to produce the same output
level as both wider filters. Do I have a bad filter or is this standard
practice?

73's
Martin, HS0ZED G4UQF