[time-nuts] 5061 Cs frequency shift with long time constant?

[Tom Van Baak]

> Following up on Brian's comment yesterday, I'm investigating sources for 
> very low leakage caps.  If I find some, I'll let the group know, and if 
> there's a minimum purchase quantity that's reasonable, I may just buy a 
> bunch and make them available to the group here.
> 
> John

John,

Before you think of replacing the special one that
HP chose you might want to measure the leakage
you have. Do you have a matching schematic for
your A9 board? There are a couple of handy test
points where you can introduce an artificial load
across the cap to confirm the amount of frequency
offset vs. the amount of leakage. One of mine shows
a 100k in series with the cap becoming 12M in LTC
mode. So shunt the cap with something like 5M or
10M and see what happens to your frequency.

/tvb