[time-nuts] Segal law: freq difference for GPSDO

Vlad time at patoka.org
Tue Mar 29 15:56:51 EDT 2016



Hello,

I am looking for for the advise regarding some strange difference in 
frequency between of two GPSDO: Trimble Thunderbolt and Datum Starloc II

Measuring by HP 5386A counter it shows something around 0.0000014 delta.
If I connect TB as REF source then I am getting numbers like this: 
10.000001396 Mhz
If I connect StarLoc II as REF source then I am getting following:  
9.999998612 Mhz

May be somebody could recognize, what 10.0000014 Mhz could be used for ?

TB Lady Heather screenshot
http://www.patoka.org/OCXO/TB-LH-Mar.png

Allan Deviation for several hours looks like this:
http://www.patoka.org/OCXO/TB-SL.png


I connected both GPSDO to Rigol DS1102E Oscilloscope to see the phase 
difference. And its shows me the ellipse which change its orientation 
left to right and vice versa.

Its classical case of Segal's law: "A man with a watch knows what time 
it is. A man with two watches is never sure". This is exact situation. 
Now I am not sure which GPSDO (if any) is more accurate.

I have no third GPSDO (or equivalent) to compare. TB and Starloc 
connected to the same GPS Antenna distribution amplifier. So basically 
both using the same antenna. Both of them was turned on at relatively 
the same time and both was warmed up for few days.

I have a theory, that may be StarLoc II was customized somehow to 
produce the freq. slightly different from 10Mhz. But its just a theory.
Is it worth to compare 1PPS outputs using something like AD8302 + DAC + 
MCU ?

-- 
WBW,

V.P.


More information about the time-nuts mailing list