Metallisation creep ?
 

I have hundreds of Germanium OC71's , now useless although perfectly
alright when bought perhaps 20 years ago, stored indoors in normal
human comfortable conditions.
Looking under a microscope and probing it would appear the
C-E shorts are due to metalisation creeping over the die.
Over the years I've found a few times that those, usually 3 lead, ceramic
resonators can fail by going resistive over the edge of the thin element as
though the metalisation has crept around the edge.
Is this a recognized failure mode and anyone know the proper term or where I
could research it ? Whether silver or Al or both have this effect ?

--
Diverse Devices, Southampton, England
electronic hints and repair briefs , schematics/manuals list on
http://home.graffiti.net/diverse:graffiti.net/

A slightly different kind of creep happens in the silver-mica
capacitors in old radios. Many old radios from the 1950's had a small
silver-mica sheet in the base of each IF transformer. Each sheet had
two small capacitors deposited on it. One is in the plate circuit of
one tube, with 100 to 150 volts on both its terminals. The other
capacitor, about 1 cm away, is in the grid circuit of the next tube, at
near zero volts. Over the years many of these have shorted out from
one capacitor to the other. You can see the shorting is due to metal
creeping across the gap.

The shorting in this case is often intermittent-- as the 150 volts is
quite capable of zapping the thin metal film. If you ever hear a radio
that emits a constant crash of static, as if in a monsoon, this is
likely the reason.

I've never seen this happen in transistor radios, and I've played
hundreds of old ones. Is this problem limited to certain transistor
designs? Most US transistors went from alloy-junction type directly
to silicon planar, circa 1962.

"Ancient_Hacker" wrote in message
oups.com...
A slightly different kind of creep happens in the silver-mica
capacitors in old radios. Many old radios from the 1950's had a small
silver-mica sheet in the base of each IF transformer. Each sheet had
two small capacitors deposited on it. One is in the plate circuit of
one tube, with 100 to 150 volts on both its terminals. The other
capacitor, about 1 cm away, is in the grid circuit of the next tube, at
near zero volts. Over the years many of these have shorted out from
one capacitor to the other. You can see the shorting is due to metal
creeping across the gap.

The shorting in this case is often intermittent-- as the 150 volts is
quite capable of zapping the thin metal film. If you ever hear a radio
that emits a constant crash of static, as if in a monsoon, this is
likely the reason.

I've never seen this happen in transistor radios, and I've played
hundreds of old ones. Is this problem limited to certain transistor
designs? Most US transistors went from alloy-junction type directly
to silicon planar, circa 1962.


I could understand where there is a heavy polarisation DC but there would be
minimal DC

disimilar mertal potential at most ) for stored trannies.
Usually for ceramic resonators they are in ac only/(dc only by
association) in filter circuits. I can only assume it is some weird chemical
process going on with no DC around.

--
Diverse Devices, Southampton, England
electronic hints and repair briefs , schematics/manuals list on
http://home.graffiti.net/diverse:graffiti.net/

Ancient_Hacker wrote:
I've never seen this happen in transistor radios, and I've played
hundreds of old ones. Is this problem limited to certain transistor
designs? Most US transistors went from alloy-junction type directly
to silicon planar, circa 1962.

I have heard that such phenomenon happens in old Philips AF116/AF117
transistors, commonly used in the sixties in the FM tuner.

---
Met vriendelijke groet,

Maarten Bakker.

"N Cook" wrote in message
...

I have hundreds of Germanium OC71's , now useless although perfectly
alright when bought perhaps 20 years ago, stored indoors in normal
human comfortable conditions.
Looking under a microscope and probing it would appear the
C-E shorts are due to metalisation creeping over the die.
Over the years I've found a few times that those, usually 3 lead, ceramic
resonators can fail by going resistive over the edge of the thin element
as
though the metalisation has crept around the edge.
Is this a recognized failure mode and anyone know the proper term or where
I
could research it ? Whether silver or Al or both have this effect ?

Interesting, but OC71's are surely just a collectible now?

N