[time-nuts] Cs stability

SAIDJACK at aol.com SAIDJACK at aol.com
Tue Jul 17 01:13:58 EDT 2007

In a message dated 7/16/2007 12:10:54 Pacific Daylight Time,  
SAIDJACK at aol.com writes:

>For  example tilting the Cs unit by 90 degrees will typically give an error  
>of about 1E-09 or so.

>It would take some time for the  control loops to compensate for this  error.

Hi guys,
I did tilt my 4050 when I first installed the unit, that number is from my  
memory. Notice that Magnus is right of course, the Cs will compensate the tilt, 
 given enough time to do so. I did not mean that the 1E-09 error will stay  
permanent of course, as I said in the follow-on sentence.
Also the 4050 is not very well temperature compensated compared to modern  
units, I was not impressed when I saw the FTS tempco specifications. A good  
GPSDO can outperform the 4050.
I have a military OCXO sample (new product) that is supposed to be  
"g-insensitive" and even that one has about E-09 frequency shift per 90 degrees  tilt.
One of the only ways to get around that is to use three crystals in  series, 
with the three crystals oriented in the the XYZ axis, so their  errors 
compensate out.
Vibration is also a big enemy of crystal oscillators of course.

************************************** Get a sneak peek of the all-new AOL at 

More information about the time-nuts mailing list