[time-nuts] What is the real source of the TBOLT's PPS output?

Ulrich Bangert df6jb at ulrich-bangert.de
Fri Jan 23 15:21:10 UTC 2009


Gents,

considered an OCXO with an OAVAR of or better say 1.0E-11 @ 1 s (as used
in the TBOLT) and a divider chain to generate an 1PPS from the 10 Mhz
oscillator signal. What OAVAR @ 1 s would we expect if we compare the 10
MHz to the PPS derived from it? 

I know, the answer is not given easily, but: If the divider chain is
engineered correctly (i.e. a synchronous and not a ripple divider) then
we are basically comparing the 10 Mhz with a delayed version of itself.
The delay will be the typical clock-to-output propagation delay of a
d-flipflop of the dividers chain's semiconductor family. 

Since we are basically measuring the propagtion delay of a semiconductor
I would expect to measure an OAVAR that is almost independend from the
OAVAR of the 10 MHz clock and which is dominated by the jitter that is
typical for this semiconductor family. In other words: I would rather
expect to measure a LOWER OAVAR than to measure an OAVAR that is higher
than that of the 10 MHz clock. 

Would I measure a SIGNIFICANT HIGHER OAVAR than that of the 10 MHz
clock, this were an indication that the divider chain (beneath the delay
introduced by it) would contribute a significant amount of instability
to the PPS. I encourage you to perform such an comparing measurement on
your TBOLT. Use the PPS output to start a TIC and the 10 MHz output to
stop the TIC and record the results. 

I did so and received the results to be seen in TBOLT_External.PDF. @ 1
s the OAVAR is 3.3E-10 which is an almost unbelieveable bad value when
one takes the 1.0E-11 @ 1 s of the 10 MHz clock into account. If THIS
PPS were derived from the 10 MHz then Trimble's engineers had done a
very bad job on the TBOLT. Which is not what I believe! But then: If it
is NOT derived from the OCXO, where does it come from then?

Is it perhaps the PPS coming from the receiver ??????? Well, considered
that a high-grade implementation of an GPSDO with a M12+ may be able to
produce a GPS PPS with an stability of 2.0E-9 @ 1 s (inclusive
SAW-correction), then 3.3E-10 @ 1s were an incredible GOOD value for the
TBOLT's receiver. Can that be????? This encouraged me to have a look
again to the measurements that have been discussed in another thead:
TBOLT-internal measurments as reported with the TBOLT monitor program.
TBOLT_Internal.PDF shows the results of such an internal measurement
while the TBOLT was in the disciplined state. @ 1 s we notice an OAVAR
of 2.1E-10 which is even a tick better than the external measured
3.3E-10. 

This similarity makes me believe that what comes from the PPS output of
the TBOLT is basically the GPS PPS and NOT ONE DERIVED FROM THE 10 MHZ.
Unfortunately the manual makes no absolute clear statement about this
question but there is one clue of which I believe it seconds my opinion.

The manual says about the PPS:

BNC Connector TTL levels into 50 ohm 10 microseconds-wide 
pulse with the leading edge synchronized to GPS or UTC within 
20 nanoseconds (one sigma) in static, time-only mode.

That is: The synchronisation of the PPS to GPS/UTS is dependend ONLY
from the receiver's time-only working mode, nothing else, especially NOT
from the state of the PLL controlling loop. If the PPS were derived from
the 10 MHz, then everything described about the regulation loop, i.e.
the whole PPS pulse shifting action to start with an low offset of +/-
50 ns with and the disciplining starting after that would lead IMHO at
least SOMETIMES to effects that the internally generated PPS may be far
without the window specified. If Trimble can specify this, it MUST be
the true GPS PPS!

On the other hand, if this were true we could turn off all our GPS
receivers and instead supply every GPS-PPS that we need from the TBOLT,
which should be better an factor of almost 10! 

Your opinions on that topic are welcome as always!

Best regards          

Ulrich Bangert
www.ulrich-bangert.de
Ortholzer Weg 1
27243 Gross Ippener 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TBOLT_External.pdf
Type: application/pdf
Size: 22562 bytes
Desc: not available
Url : http://www.febo.com/pipermail/time-nuts/attachments/20090123/5726ab0a/attachment-0002.pdf 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TBOLT_Internal.pdf
Type: application/pdf
Size: 23058 bytes
Desc: not available
Url : http://www.febo.com/pipermail/time-nuts/attachments/20090123/5726ab0a/attachment-0003.pdf 


More information about the time-nuts mailing list