[time-nuts] Tbolt instability

[David C. Partridge]

How large a self survey did you take with the TB?  The default survey size
is relatively small.  If you increase the size of the self survey to be
(say) 24 hours of data (24*60*60 = 86400) you may find you get a much better
self survey position which should result in better stability (based on Mark
Sims' recent research).

Mark will soon be releasing a new version of Lady Heather that will collect
position readings over 48 hours and then do some fancy stats on this data to
set the position accurately.

Side question to Mark: Any thoughts on release dates?

I'm not sure how long the TB needs to be on to reach a stable state, but do
know that it suffers from a "heisenberg problem" in that the temperature as
reported by the onboard sensor chip starts to climb quite fast if you start
monitoring the TB using the serial port.

I'm sure the Z3801 will be more stable, as IIRC it uses a very long time
constant for disciplining the 10811A.

I suspect that if you had an accurate source like a maser to check against,
you could come up with a better time constant and damping factor for the
TBolt than the factory default settings, and this would allow it to give
much better results, but what do I know? 

I only have one clock that's reasonably trustworthy (the TB), so I'm happy -
I'm sure if I had more I would start to worry ...

Dave 

-----Original Message-----
From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On
Behalf Of John Green
Sent: 08 October 2009 14:51
To: time-nuts at febo.com
Subject: [time-nuts] Tbolt instability

Tbolts seem to have a very good reputation for accuracy and stability. Most
comments I have seen compare them favorably to the HP/Symmetricom Z3801.
Mine may be OK in the very long term but doesn't look too good in the short
term.

 <snip>