[Elecraft] PLL reference oscillator failure

[Ron D'Eau Claire]

The most sensitive part is almost always the problem part! Try gently
tapping other parts. I've often used the rubber eraser end of a pencil to
make gentle pokes to determine which part is the most sensitive. Sometimes
it's hard to tell! 

One thing that can cause this behavior is marginal feedback. Crystals
oscillate by vibrating mechanically. A small change by bumping a marginal
crystal can start it, or the change in capacitance caused by touching the
can. Make sure C84 and C85 are the right value parts. They control the
feedback. Also be sure the diodes are in the right way around: D16, D17 and
D18. C86 is important to keep the source of Q19 at RF ground. And, of
course, you should confirm that you have 8 volts between ground and one side
of C86 that doesn't disappear if you tap things! 

If it's your crystal, it could be an intermittent crystal but that's really,
really rare! One problem people run into installing crystals is letting too
much solder flow around the pins. The solder joint look fine on the bottom,
but the solder flows through the hole (they are plated all the way through)
and 'puddles' around the lead under the crystal where you can't see it! If
the puddle is just tiny bit too big, it'll short the crystal to the case.
It's often a 'cold', intermittent joint because the case wasn't hot enough
to take the solder. The blob of solder around the pin is simply resting
against the bottom of the crystal case. 

Try checking continuity from the pins to the crystal case while
tapping/wiggling it to see if there's an intermittent. Keep in mind that
most DMM's are useless for this purpose because they take a while to detect
the short. If it's intermittent the DMM may not detect it. Many DMM's have a
"continuity" check that sounds an alarm if even a brief contact is detected.
The option is to remove the crystal and look! You'll need a good solder
sucker (at least a spring-loaded unit like the Soldapullet Elecraft
recommends) and take care not to put more heat on  the leads than you need
to. 

Ron AC7AC 

-----Original Message-----

Hi everyone,

Thank you all so much for all your suggestions and advice!  It looks  
like I have some kind of intermittent fault in my PLL Reference  
Oscillator.  I've retouched all the solder points in that area incase  
it was a cold joint, and after poking around at it for a bit it  
suddenly seemed to be working.  Then after a few minutes it went dead  
again, no oscillation.

So I poked the crystal X1, just kind of tapped it, and suddenly it's  
oscilating again.  I've verified that there are no cold solder  
joints, the crystal is securely soldered and the case is grounded.   
After working on it for a while, it seems like the slightest touch or  
movement to the rig can cause the reference oscillator to start and  
stop.
Does this problem suggest a failure within the crystal itself?

Thanks also for the advice and feedback about the case bottom, I've  
got the feet parallel now and everything seems good there.

73 de
Stephanie
va3uxb