[time-nuts] Vectron Type 229 Crystal Oscillator

[ed breya]

Whatever you do, don't apply power to the pins indicated as "VCXO 
Supply." I'm pretty sure that is the internal regulated supply that's 
accessible only to connect a pot so that you can remotely trim the 
frequency - if it has the feature, including a varicap diode in the 
oscillator. One end of the pot goes there, the other end to "Return," 
and the wiper to "VCXO Input." So, the first thing to do is to see 
with an ohmmeter whether that "VCXO Supply" pin shows any resistance 
to the case or ground/return, indicating that it's hooked up 
internally. If you then bring up the + main power connections slowly, 
you should see that pin voltage rise and stop at some point like 6.2 
V. You could also apply a test voltage through a resistor and see what it does.

The reason I know this is that I had several Vectron 10 MHz ovenized 
oscillators, but one didn't work right, so I took it apart to see 
what I could do. I deciphered part of the circuit and found that the 
oscillator supply was supposed to be +6.2V, set by a shunt regulator 
temperature compensated zener like a 1N823.

But, the zener had a small choke in series with it - I don't mean as 
a noise filter, I mean actually in series with it, so it affected the 
supply voltage. I could only surmise that they used the small copper 
winding resistance of the choke as a strongly-positive tempco (~3000 
ppm) resistor to provide another degree of compensation.

The problem with this unit was that the tiny choke was burnt open, 
and the body was blackened, so I could not read the color codes, but 
could tell that it was similar to other chokes in the circuit - the 
small axial epoxy molded types. So, the shunt zener did not regulate 
the supply - it ran instead at about 9V, and way off in frequency. I 
tacked in a similar small choke, and the voltage then was about 6.2V.

So, it appears that somehow before, power was applied to that pin at 
over 6.2V, and forced enough current to burn the choke open. I don't 
know if other damage was done, but the circuit appeared to work after 
patching. I assume that the value of that copper resistance was 
important, and possibly unique to each zener or oscillator, so unless 
I get the same amount, it won't quite be up to snuff. Since the 
original choke was destroyed, I don't know exactly what it should be, 
and I didn't want to take apart any other oscillators to find out.

Ed