[time-nuts] Follow-up on Z3801A high EFC reading

[John Ackermann N8UR]

Last week I posted about one of my Z3801As having a very high EFC 
reading and an "EFC ERR" message in the health status when powered up 
again after about 18 months of sleep and a 500 mile location change.

Over the last week, the error message has continued, but the EFC has 
consistently trended downward.  It started out at about 1015000 and is 
now down to 1012070.  During the first day or so, the EFC count was 
going down more than one unit per minute; in the last day or so, it's 
more like 4 or 5 units per hour.  But since it's moving in the right 
direction, I'm not going to try to rip open the outer oven at this point.

This sort of aging effect is a bit odd given that this unit had been 
running continuously for about 7 years before it was shut down for the 
move; I might expect to see this is in a brand new unit but not in one 
with so much running time.

I've been comparing the 10 MHz output against a cesium during this time, 
and an ADEV plot is attached.  It seems to indicate that the performance 
is just fine -- the short term floor is probably limited by the 5061B's 
noise, and we can see the 5061A long-term noise floor starting to show 
up at 200K seconds, but there's certainly nothing here to question the 
performance of the GPSDO.

Now that I've captured this baseline data, I'm going to power cycle the 
Z3801A to see if it lands in the same general vicinity.

John
-------------- next part --------------
A non-text attachment was scrubbed...
Name: z3801a_2_screenshot.png
Type: image/png
Size: 60627 bytes
Desc: not available
URL: <http://febo.com/pipermail/time-nuts_lists.febo.com/attachments/20110716/7336c071/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: z3801a_2-hp5061a_2-adevplot.jpg
Type: image/jpeg
Size: 54476 bytes
Desc: not available
URL: <http://febo.com/pipermail/time-nuts_lists.febo.com/attachments/20110716/7336c071/attachment.jpg>