[Milsurplus] Rethinking Re-Capping

Sat Dec 2 20:57:25 EST 2017

DaveI, like most members of this list, admire your ability to troubleshoot and repair this ancient equipment.  Your goal of not disturbing things any more then necessary to keep it working as designed and built is noble but it is "mission: impossible", IMHO.
Books like "Reliability Factors for Ground Electronic Equipment" https://archive.org/details/ReliabilityFactorsForGroundElectronicEquipment
clearly points this out.  On page 1-1: ..."during WWII, up to 60% of airborne equipment shipped to the Far East was damaged on arrival.  Furthermore, as much as 50% of the equipment and spares in storage became unserviceable before it was ever put into use."  With stats like this, chances are pretty good that your repaired radios will again fail on the shelf, let alone while actually operating.  
Just after WWII, a great effort was put into developing reliable components.  Tons of research was done and specifications written with the goal of increasing electronic equipment reliability.  And the effort was, for the most part, pretty successful.  But as the equipment complexity increased by factors of 10 and more, the reliability of the systems again suffered.  When it was all said and done, some complex avionics systems of the mid to late 1960s had a MTTF measured in the 10's of hours, or even less.
The only technical museum that has operating displays, that I am aware of, is at InfoAge in New Jersey, see: http://infoage.org/index/radio-technology-museum/I am sure that there are others but I don't know of them.  Pretty sure that even with modern components replacing the originals, it is a constant effort to keep the displays operating.   Al Klase and Ray Chase would know much more about this then I do since they have the "hands on" experience of the day-to-day operation.
The only answer I can see to this problem is to have two items, one "shelf queen" to display that is pristine and one to operate which is up to date component wise.  Pretty sure that 99.99% of the visitors to the museum could not see the difference between the two pieces of equipment.
Jim

      From: David Stinson <arc5 at ix.netcom.com>
 To: ARC-5 <arc5 at mailman.qth.net>; milsurplus at mailman.qth.net 
 Sent: Saturday, December 2, 2017 4:07 PM
 Subject: [Milsurplus] Rethinking Re-Capping

Rethinking Re-capping.

Concerning the restoration of
historically-significant radios: my views and
goals have evolved over the years.  My goal-
personal fulfillment from answering the challenge
of these sets aside- is to give the radios the
best chance to be preserved beyond my time as
their temporary care-taker.  To do so, I believe
the radios must play, and should do so with the
minimum disturbance of the original assembly
possible.

There was a time when I would say "replace very
electrolytic and every paper and every Micamold"
as a matter of course.  I no longer do that.  Been
working on a set of three RAX receivers that have
been awaiting their turn in my barn and storage,
for at least 12-15 years.  All have been playing
12-24 hours now without failure.  One needed no
cap replacements at all.  The second needed one
and the third needed two.  I changed no resistors.
Did repair some bad trimmer caps in two receivers,
but not counting them for this post; different
issue.  To find those that actually need te be
changed requires a reliable schematic, a good
meter, a variable B+ supply, patience and a finger
tip.

In a tube type receiver such as the TCS, SCR-274N
or RAX, there are several paths that B+ travels to
ground.  Most of them go through tubes.  When the
tubes have no filament voltage, the tube plate or
screen pin is a "dead end;" an open circuit.  No
current should flow through the tube's plate or
screen circuits.
But there are usually other paths- like
screen-voltage dividers- that provide a path for
some current flow.

If you spend some time with the schematic, you can
trace all the B+ distribution busses, looking for
paths to ground.  For instance- in the -274N
receiver, two 7000-Ohm resistors are connected
between B+ and ground as a screen-voltage divider.
With the tube filaments off, the receiver out of
the rack and with no local control box installed,
the only path for B+ to ground is through those
two resistors, unless there are leaky bypass caps.
So, with no filament voltage, if we connect our
variable V+ supply and increase to 100V, we should
observe no more than  100/14,000= 7 milliamps of
current flowing.  If we have more, we have leaky
capacitors somewhere.

So how do we find the leaky cap(s)?  Did this test
on one of the RAX receivers.  After allowing
"reforming" time for the electrolytic caps (and no
filament voltage), at 100V on the buss, it was
drawing 20mA.  100V was dropping somewhere and at
20mA, the leak was turning that current into 2
Watts of heat.  That doesn't sound like much, but
it builds-up quickly.  After a few minutes, a
finger-tip on one of the .05 uFd screen bypass
caps felt the heat and the ZM-11 proved it was the
villain.  After replacing that cap, the current on
the B+ buss dropped to 6 mA; just 2 mA above that
expected and close enough to operate.  Left the
100V on the buss for four hours and no increase in
the current.  That's not a promise no other cap
will fail- they are 70 years old.  But if the goal
is to do the least possible and given I'll be
running these at reduced B+, the risk of damage is
small and acceptable.  If one can't "feel the
heat," a study of the B+ distribution on the
schematic will reveal places one can lift a wire
or part to isolate the branch with the bad actor.

In many radios (not in RAX but in others), there
are coupling caps that go to grids which can be
leaky.  Apply B+ with no filament voltage as
before.  Measure the grid of the tube fed by the
capacitor coupling.  If you have a positive
voltage on that grid with no filament voltage,
change the coupling cap.  NOTE: Use a VOM for
this, not a VTVM.  The very-high input impedance
of the VTVM may be "spoofed" by an electro-static
charge on the grid side of the capacitor plate.  A
good VOM will load it just enough to give you a
"real" reading.  Exception:  Leaky cathode
bypasses, especially the one at the Audio PA, need
to be checked individually, since there's no
current path to find them without filament
voltage.  Quick check:  Calculate the current that
should flow through the cathode resistor if you
put +10V on the tube cathode.  i.e. if the cathode
resistor is 330 Ohms and you connect +10V to
ground to the cathode, you should see 30 mA of
current flow.  If (after a reasonable "reform"
time) it's pulling more than 30 mA, your cathode
bypass is leaking.

But what about OPEN bypass or coupling caps?  With
the danger of frying things from bad bypasses and
coupling caps past, one can power the set and
check performance.  If you have open bypasses or
coupling caps, *the circuits will tell you.*
Oscillation in IF amps or Converters usually
equals open bypasses.  A screwdriver touched at
tube bases will often change the "motorboating"
and tell you which stage is oscillating, or use a
scope.  If the oscillation changes with tuning,
the Mixer/Converter stage is oscillating (this
happened with one of the RAX receivers- open
cathode bypass on the Converter).  If the audio is
distorted, look for open bypasses in the 1st Audio
and Audio PA.  If you have no gain through a
stage, check any coupling caps that might be
involved.  But we shouldn't assume in these cases
that capacitors are the culprit; resistance and
voltage chart checks are your friends.  For
instance- I've revived three TCS receivers lately
and I find wonky resistors to be an even bigger
problem than capacitors.

With these techniques, I have three RAX receivers
on the bench playing nicely for hours.  I've
changed three capacitors, no resistors and one
tube.  Did also repair some trimmers as noted
before.

GL OM ES 73 DE Dave AB5S

______________________________________________________________
Milsurplus mailing list
Home: http://mailman.qth.net/mailman/listinfo/milsurplus
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:Milsurplus at mailman.qth.net

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qth.net/pipermail/milsurplus/attachments/20171203/88f7084c/attachment-0001.html>