[time-nuts] Measuring ADEV using TBolt-Tic tester

WarrenS warrensjmail-one at yahoo.com
Fri Oct 7 17:00:42 UTC 2011


Few TIC instruments can be used to measure 1 to 10 sec ADEV noise of a good 
oscillator because of the TICs high noise floor due to it's limited 
resolution.
A 1 second noise floor of 1e-11, requires a time interval resolution of 
10ps.
The good news is that many time nuts have a great TIC counter with that 
capability and do not even know it.

A stand alone Tbolt, IN THE MANUAL DISABLED MODE, WITH A GOOD ANTENNA SETUP,
makes a respectable TIC counter with resolution near 10 ps on it's PPT Freq 
data output and near 100 ps using it's Phase data.
This gives a noise floor near e-11 at 1 sec and e-12 at 100 seconds, which 
is low enough to test many of the good 10 MHz oscillators.
I find the Tbolt is the Only thing needed to do high precision mid and long 
term performance testing of 10811s and LPRO Rbs
A Tbolt can give very useful data for making ADEV plots past 10 to 100 
seconds for either an externally connected Osc (or for it's own internal 
osc) in both the disciplined and disabled modes.

Attached are some 'TimeLab' ADEV plots comparing a stand alone TBolt-Tic 
tester to a TPLL2.0
The  data shows that a Tbolt can make accurate ADEV measurements on it's 
own, down to 2e-12  past 100 sec when using the logged FREQ data.
The TBolt_Tic can also give very accurate long term data to 1e-13 in one day 
by using it's phase output data when the Tbolt is in the DISABLE mode.

This allows the Freq nut an alternative to needing a special TIC counter (by 
using an external osc Tbolt) or needing a separate reference Osc (by using 
the GPS signal), when doing high performance testing.

The Tbolt plotted has it's internal Osc removed and uses an external 10 MHz 
osc.
I made test plots with the TBolt driven from both a 10811 and a LPRO using 
the GPS as the reference signal.

Like all testers, the noise measurement is limited by the reference used,
so it is very Important to get the Best GPS signal possible to the TBolt, 
which includes but is not limited to,
Having a good sky view, getting the fixed location accurate and taking the 
short term data when the satellites are not switching.

Very Important note, The above is NOT available directly from LadyHeather's 
ADEV plots  (at least not yet - Mark is a revision coming?),
What one needs to do is to log the Tbolt's Freq and Phase data at the 1 sec 
rate and then use that data with an external ADEV program such as Ulrich's 
Plotter or John's TimeLab.
Note for TimeLab users. LH adds some extra comment data lines in it's log 
file that need to be manually removed before using the data file with some 
versions of TimeLab.

Another very useful thing that can be done with LadyHeather and a Tbolt that 
has been set up to use an external osc (and has a good antenna system),
is to measuring or set the frequency of any external 10 MHz Osc.
Short summery of how;
Put the TBolt in the Manual disable mode, (D D)
The Oscillator's EFC input and the Tbolt's Dac output are NOT connected or 
used,
Use the (J J) command to rezero the Phase plot (plus the cable delay if 
desired)
and then read or adjust the Osc freq on the ppt plot.
With LadyHeather's filter off (F D 0) useful resolution is about 1e-10 in 
one second,
With the filter set to 10 sec (F D 10)  useful resolution is near 1e-11 in 
well under one minute
With the filter set to 100 sec, resolution is around 1e-11 in a few minutes. 
adding a trend line helps (G O L)
For higher accuracy and resolution, watch the Phase change trend over a 
couple hours using (G P L),
This can provide useful resolution as low as 1e-12. (delta 3.6 ns per hr = a 
1e-12 freq offset)
If you want even more accuracy, watch the PHASE change over a few days. This 
can check your best Primary Cs standard's frequency.

Have Fun
ws 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TBolt_Tic Noise Floor2.gif
Type: image/gif
Size: 57584 bytes
Desc: not available
URL: <http://www.febo.com/pipermail/time-nuts/attachments/20111007/48d1ab68/attachment-0001.gif>


More information about the time-nuts mailing list