[time-nuts] Thunderbolt behaviour with long time constants.

[ws at Yahoo]

The couple decades comment ONLY applies at your fastest tau.

You'll should be able to get good results with the tester two to one better 
than the DUT.
At your Minimum tau which was 0.25 sec, your Noise floor measurements of the 
5370b would suggest it is about 1x10-10.
A good TBolt will do near 1x10-12 there, which is where the two decade 
comment came from.
If you only care about noise of  >2e-12 at tau's  >25 seconds, No problem.

BTW setting a typical Tbolt's TC up above around 500 will increase its noise 
at mid taus, unless everything is in a very controlled situation.

ws
*********************************
Thanks for the comments.    I did a quick check of the noise floor of the 
5370b while feeding the 10 mhz output of the tbolt to the start and stop 
inputs of the 5370b with a t connector. At a tau of 10 (approx 2.5 seconds) 
the Allan deviation is 1x10-11, at a tau of 100 (approx 25 seconds) it is 1 
x 10-12.   Optimizing the trigger settings might lower this some what but I 
used the same settings as before.   I was not aware you need an instrument 
with a noise floor a couple of decades below the signals you are trying to 
measure.
Sent from my iPad

On 2011-02-09, at 9:19 AM, "WarrenS" <warrensjmail-one at yahoo.com> wrote
> Looks to be mostly a plot of the 5370B's noise floor at low taus.
> You may need something a couple of decades better if you want to measure 
> the Tbolt at that tau setting.
>
> BTW, You should NOT set the TBolt's TC to >999 sec. I  have found that the 
> control can become somewhat unpredictable, AKA S/W bug.
>
> ws
>
> *********************
> ----- Original Message ----- From: "Mark Spencer" <mspencer12345 at 
> yahoo.ca>
> To: <time-nuts at febo.com>
> Sent: Tuesday, February 08, 2011 7:48 PM
> Subject: [time-nuts] Thunderbolt behaviour with long time constants.
>
>
> First apologies for the zipped attachement.
>
> I thought some members of the mailing list might find the attached plot of 
> the
> performance of my Thunderbolt to be interesting. After looking at the
> performance of my Thunderbolt with a time constant of 225 seconds which 
> seems to
>
> be about right for my Thunderbolt, I noticed that the Allan deviation 
> appeared
> to be be reduced at high values of Tau (approx 2,000 seconds..)
>
> I then changed the time constant of my Thunderbolt to 2,000 seconds and 
> compared
>
> the two plots. I'm quite surprised at the result and the similarity 
> between
> the peformance at low values of tau with the two different time constants.
> I took 4 readings per second with a 5370B, hence each unit of tau on the 
> graph
> is approx .25 of a second.
>
> I'll likley try other time constants in the future
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to 
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.