Fw: Re: [Collins] Identifying BBs

[Dr. Gerald N. Johnson, electrical engineer]

Carbon composition resistors change value from age, heat, and humidity.
They are somewhat temperature sensitive but don't return to the same
value when cooled. They take a random walk, every time they are used.
The antifungus coating, if it protects them from humidity may slow the
average wander in value at least it should retard the humidity effects.

Its not only the high value carbon composition resistors that will
change value, all will change value, generally rising if they aren't
IRC. IRC tend to go lower in value. IRC are distinguished by having
rounded ends and a visible mold line the length of the resistor. Inside
they are actually a carbon film on a glass or quartz tube which
concentrates the heat in a much smaller volume than that of the true
carbon composition resistor which is a lump of resistive material with a
covering, and that concentration of heat tends to char the molded case
lowering the effective resistance. In the 60s, Collins component
specifications for carbon composition resistors accepted ALL makers
except IRC, probably for that reason and for the fact that IRC don't
fail open as reliably as carbon composition resistors (sometimes chosen
to be used as wired in fuses), and the RF characteristics of the IRC are
different than the standard carbon composition resistor. IRC made a
resistor without the outside molding and claimed it had better RF
characteristics than carbon comp.

Some of the black beauty failures come from moisture intrusion, but also
come from the oil leaking out. The bad black beauties used kraft paper
(as in brown wrapping paper) saturated with oil as the dielectric. Its a
toss up whether the oil preserves the paper or the paper retains the
oil. As the capacitor ages, the oil sneaks away (most rapidly when the
case cracks, other wise slowly along the leads), putting more work on
the paper. Worse, the foil/paper/foil stack develops air gaps between
the foil and the paper (that were initially filled with oil). These air
gaps, no matter how tiny have a much lower dielectric constant than the
oil and paper, and as the result the electric field is higher on the
air. That leads to dielectric breakdown of the air and that arc heats
and chars the paper. Raises the leakage level. Doesn't require loosing
all the oil, just enough to let some place on the capacitor develop that
air gap and that air gap will develop in the capacitor. Then the kraft
paper wasn't perfect to begin with as it included imperfections and
inclusions of conductive material. If the foil is really thin, the
makers hoped that the foil would burn away from the bad places in the
paper the first time voltage was applied.

Its just been my experience that only one or two out of the first 40 or
50 oiled paper capacitors (whether molded or waxed) that I've tested
have had leakage small enough to not change the bias on an audio output
stage, while the Orange Drop leakage was always far better than I could
measure (using a 11 Megohm input R VTVM as a current meter on the 1.5
volt scale, passing criteria being under 1 volt of leakage indicated,
e.g. less than 0.1 microamp with at least 400 volts applied to the
capacitor). 

As any leakage adds heat to the capacitor its effective at pushing oil
out to be replaced by air to lead to further damage to the capacitor
dielectric. And so the marginally leaky capacitor gets worse with
operation. Some weren't good while still in the factory box.

As for changing them in a receiver, my opinion is this: If you don't
replace ALL the black beauties, don't come complaining about having to
trouble shoot low gain (leaky screen bypass capacitors cut the screen
voltage and the stage gain), short output tube life (leaky capacitors
put positive bias on the grids causing higher plate current), and poor
AGC operation (low screen voltage on an AGC amplifier leads to low AGC
voltage, while leaky AGC bypass and time constant capacitors limit the
AGC voltage developed).

Its only half the trouble to chop out a black beauty and stuff in a new
Orange Drop than to work out the old capacitor with enough leads left
for it to go back, then to test it for leakage, and then to stuff in a
new Orange drop. And its a far better use of workbench time to replace
all the black beauties than it is to search for all the reasons the gain
is a little low, the AGC performance a little poor and the audio quality
a little distorted.

73, Jerry, K0CQ, Technical Advisor to the CRA
-- 
Entire content copyright Dr. Gerald N. Johnson, electrical engineer.
Reproduction by permission only.