"Richard Knoppow" wrote in message
I am in the process of restoring a Hallicrafters S-40A and have
posted a bit about this already. First of all, I replaced all the paper
caps. This is generally recommended when rebuilding old electronic
equipment. Most of these were the original wax filled types. I measured
these for capacitance and dissipation factor. D is the inverse of Q and is
a way of measuring ESR but has a constant value. The result was:
One cap a dead short.
One cap open
The rest had C much higher than the marked value and rather high D,
anywhere from about 0.2 to about 0.5, I think one was even higher. There
were a couple of molded plastic caps, one a blue colored Cornell-Dubilier
the other an Astron. These were probably replacements. Both were close to
the marked C value with D of around 0.1, probably still good. The paper
caps were made by Industrial Condenser Co, of Chicago and Sprague.
For comparison, new film caps measure right on the nose for C and have
D too low to measure on this bridge.
The measurements were made on a General Radio Type 650A, an oldie but
goodie, which has been calibrated with precision resistors and caps. The
capacitance measurements are made at 1.0Khz. If I get curious enough I may
check the RF impedance of a couple of these caps on a Q meter but I
suspect they are not very good.
While the high C value may not seem to be a problem I suspect it may
indicate some deformation of the capacitor elements or some other problem.
These were not remotely precision caps when made but the values are so far
off the marked values that I suspect a change in the internals rather than
manufacturing tollerance.
So, the advice to shotgun paper caps seems to be right on.

Another note, probably should be a separate post.
I mentioned before that the bandspread cap went the opposite direction
of any other I've seen in a Hallicrafters RX. I got a couple of responses
that others also had similar S-40's. I found a couple of other oddities in
my receiver (for instance someone had connected the output of the BFO
directly to the detecto rather than by means of the wire "gimmic" cap, it
won't work this way). I am now pretty much convinced that the band spread
cap was modified to make it "set" at 100 rather than the usual zero used
in H receivers. I was able to make it go in the right direction by
crossing the dial drive cable but I don't like makeshifts. Also, despite
the nearly symmetrical construction of the band spread cap I suspect the
minimum capacitance is slightly different when its running in the correct
direction. That would affect the "normal" dial calibration. The RX
calibrates OK but I am still going to have a go at removing the stop peg.
I think I will have to drill it out but will try something less "invasive"
first.
BTW, I used to consider Hallicrafters stuff junk but after rebuilding
this set and an S-20R have changed my mind. Actually both are well made
and well designed. Hallicrafters definitely built their equipment to fit
certain price ranges, mostly economy. I think they did quite well in
delivering good performance value at the prices they aimed at. One proof
is that the stuff is still working after 60 or 70 or more years.
The above really applies to the components too. While paper caps have
a bad reputation keep in mind that most of the ones we deal with are very
old and have worked well for a long time. Plastic film was not available
when these guys were made and other types of caps, mostly tin-can
oil-filled types, were very expensive in comparison to the common wax
filled types. I am not sure they lasted any longer. One exception is the
oil-filled molded plastic Sprague Black Beauty cap. These were made and
sold as extra-high quality, long life caps but Sangamo evidenty had
manufacturing problems, one of which was the way the oil filling tube was
sealed. These began failing within a couple of years of manufacture where
the "less reliable" wax filled paper caps went along for many decades.

Anyone ever bother to check resistor values? During overhaul and repair
of lots of tube era marine electronics, many composition (and films
sometimes) resistors had changed value considerably. Especially those used
in voltage dropping circuits. High value resistors ( half megohm or more)
seemed to be pretty wild too.
Of course checking frequently meant disconnecting from associated
circuitry to check is a real
pain in the you know what, but the end result, (including the condenser
replacements) resulted in amazing results!
Anybody find similar resistor drift?

Old Chief Lynn