NOS Aluminum Electrolytics

[Dave Ishmael]

I have a few (more) comments about NOS aluminum electrolytics and aluminum
electrolytics in general:
1.    For NOS, do NOT apply the cap's rated voltage without first reforming
the capacitor.  The excessive DC leakage (DCL) currents that might result
will generate hydrogen gas that will cause the cap to catastrophically fail
due to venting.  As a general rule, I verify every cap's condition before it
is installed.
2.    I use a Heath IP-32 HV power supply to reform or check the condition
of the caps.  I slowly increase the HP-32's voltage while keeping the DCL in
the range of 10-20 mA until I reach the cap's operating voltage.  Verify
that the cap's DCL is acceptable at its operating voltage.  You will know by
the time the cap has sat at its operating voltage and watching the current
meter whether or not the cap will be OK for use.  Occasionally, I have run
into NOS caps that had excessive DCL after reforming (I usually take these
back to the store where I bought them and get them replaced).
3.    Cornell Dubilier used to recommend that their caps be reformed "after
long storage (e.g., 5 yrs).....by applying rated voltage for one hour
through a 1000 ohm resistor".  This recommendation has recently been
increased for storage periods of 10 years.
4.    A historical note.  Years ago, boric acid and ethylene glycol was used
in a pasty electrolyte.  Even though the end-seals in these particular caps
failed, this pasty electrolyte did not evaporate.  This probably explains
why so many of the receivers that I own from the 50's and 60's still have
their original caps in place.  In the 70's, as ESR's (equivalent series
resistance) were being decreased, the electrolytes became as liquid as water
and their solvents would evaporate.  With electrolytes prone to evaporation,
the end-seals have improved, but shelf-life issues have arisen.
5.    If I haven't used one of my older receivers (or xmtrs) for a couple of
years, I bring them up on a variac just in case their filter caps need to be
reformed.  How many stories have you heard about filter caps failing at
power-on??
6.    The end-of-life for an aluminum electrolytic capacitor is dictated by
its application.  Caps used in xmtrs typically see significantly higher AC
ripple currents and ambient temps.  In general, elevated temperatures is the
number one killer of aluminum electrolytic capacitors.  Reducing the ambient
temp 10C will  DOUBLE  the life of the capacitor (20C will QUADRUPLE their
life).
7.    Newer, replacement-type caps are physically much smaller than the
original.  For critical applications, where AC ripple currents and ambient
operating temperatures are expected to be high (e.g., xmtr service), select
the physically largest cap you can lay your hands on.  The larger caps will
minimize self-heating effects from the AC ripple currents and will run
cooler.  The larger the surface area of the cap, the cooler it will run for
any given application.  Several years ago I did a survey of 26 aluminum
electrolytics in the range of 22-50uF @400-450 Vdc and their case sizes
varied from 0.63"dia x 1.26"L  to  1.125"dia. x 2.156"L, over a three to one
range of surface areas.

With proper "care and feeding", these things can be quite reliable over
time.
Dave - WA6VVL

--- --- --- --- --- --- --- --- --- --- --- --- --- --
To subscribe: listserv at listserv.tempe.gov
and in body: subscribe HEATH yourfirstname yourlastname
To unsubscribe:  listserv at listserv.tempe.gov
and in body: signoff HEATH
Archives for HEATH: http://www.tempe.gov/archives
--- --- --- --- --- --- --- --- --- --- --- --- --- --