[Milsurplus] Command Set Transmitter Keying

[Mike Morrow]

>Interesting theory, but I'm not so sure a spring would weaken over time.

Perhaps not, John.  

But, in a different arena, it's fairly well documented that springs in
firearm magazines that have been left loaded for long periods of time
often show degradation.  Also, I had to replace a 60-year old operating
rod spring on a M1 rifle due to it exerting below normal compression force,
causing cycling problems.  If properly designed, these critical springs
should not have been exposed to stress past the elastic limit.

I was also part of a incident investigation for loss of a 1100 megawatt
electric nuclear plant's output when a voltage regulator swap from auto to
manual mode failed.  The opening spring force in the mode switching relay
was found to be insufficient to make the normally closed contacts wipe
and conduct upon de-energization.  It had been OK earlier in its life.

So, who knows?  I don't rule out much when I first examine a condition.
I would *initially* consider everything that *could* delay K52 drop out.
One could even look at other mechanical issues, such as corrosion effects on
the armature pivot bearings or coil varnish adhesion to the armature.  All
are very unlikely for K52...all are effects that couldn't cause the consistent
timing problem reported.  But still, if there isn't something pathological
going on, then this characteristic of the transmitter HV being applied
for 30 mS after the antenna has been disconnected was occuring 70 years ago
as a simple consequence of the design.  That extra 30 mS of transmitter
operation after each key opening is about half the length of a dot sent at
20 wpm, the command sets were extremely rarely and irregularly used on CW,
and a station well removed from the transmitter would not hear that extra
30 mS bonus warble due to the opening of the antenna relay.  Perhaps this has
occurred since command set day one.  It's just hard to accept.

So, assume that the old K52 is functioning today exactly as when it was new.
An interesting experiment would be to install an appropriate resistor across
K52 coil terminals (thus across through some wiring resistance the coils 
of K51, K55, K1, K2, and K3).  I'd try something around the 400 ohm K52 coil
resistance.  Only a 2 watt rating is required.  If it would dissipate the
inductive energy of the six parallel coils quickly enough to let K52 (HV)
drop out before K55 disconnects the antenna, then this would be a cheap and 
simple corrective action that doesn't require blocking diodes.

Mike / KK5F