Antonio,
I suspected something like that.
I realized that I've never done a similar test so I'm correcting that
now. Both the oscillator and the counter have been off for some time -
days for the counter due to equipment rearrangements and months for the
oscillator. I replaced the 04E oscillator with the 04A oscillator to
see if the 04A's lower power drain would help a power problem I'm
having. It didn't.
I installed the 04E oscillator and set up my 1992 to measure the
frequency of my Efratom FRK rubidium which has been running continuously
for about 3 months. I'm using a gate time of 10 seconds.
The first reading after turn-on was about 370 Hz (yes, Hertz) low.
After 20 minutes the initial warm-up was complete and the reading was
238 counts (e.g. 0.238 Hz) above 10 MHz - well within the 1e-7
specification. Now, after about 13 hrs, the reading is 78 counts high.
It's now dropping at about 1 count per hour, but there was a recent drop
of 4 counts in one hour as it works out the kinks.
It'll take a while to match the drift rate of your graph. Then I'll
check the standby drift again.
Ed
On 12/19/2011 10:49 AM, [email protected] wrote:
Hi Ed,
my mistake, the test started from cold, but the curve I've posted starts at a
certain midnight. For completeness, the very initial warm-up is missing in the
curve. Checking my archive, I see that the test started the previous day around
02:22 pm and the reading was 3 mHz higher, and I might have lost a few minutes
since turning on the oscillators and the beginning of the record. The test was
aimed at seeing how many days the Racal takes to stabilize around the 1E-10
level, and the answer was one week. And this is the overall sense of my curve.
Hence a few minutes less or more were not important.
But indeed this test is not a good indication of the very first warm-up
minutes (intentionally not from stby).
Antonio I8IOV
Hi Antonio,
Between you and Charles, I'm getting the feeling that my 04E oscillator
is defective. Yours only moved 15 counts over a period of days from a
cold start. Mine moved 40 points in two hours from standby. I think I
have some testing to do.
Thanks,
Ed
On 12/18/2011 6:01 PM, [email protected] wrote:
As a contribution to this discussion, I attach a 17-day trace recorded with one
of my 1992s on Feb 2008. The counter was counting a LPRO Rb. The test started
from cold for both sets. Gate time was 10 sec.
The 1992 is the 02M/04E military.
Antonio I8IOV
_______________________________________________
time-nuts mailing list -- [email protected]
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
