[time-nuts] Different Thunderbolt versions

[Bruce Griffiths]

Tom Van Baak wrote:
>> For observations at 1420MHz, and short integration times (~ 100sec) the
>> phase error needs to be 100ps or less.
>> For longer integration times (10,000 sec is desirable) and higher
>> frequencies the timing constraints are far more stringent.
>> Whilst clock modeling can be used to improve performance it can only
>> correct systematic effects.
>>     
>
> Bruce,
>
> Approximately what range of tau and ADEV(tau) would this
> VLBI requirement correspond to? A couple of the TBolts I've
> tested are at or just under 1e-12 at 1 second and stay almost
> flat out to 100 or even 1000 seconds.
>
> /tvb
>
>
>   
Tom

Really should use something like TDEV but roughly (within a factor of 2
or so at 1420MHz):

ADEV < 1E-10 or better @ tau = 1sec for 1 second integration time.
(integration time to short for anything but very bright objects)
ADEV < 1E-11 or better @ tau = 1sec for 10 second integration time.
(integration time to short for anything but very bright objects)
ADEV < 1E-12 or better @ tau = 1sec for 100 second integration time.
(integration time somewhat marginal for most objects)
ADEV < 1E-13 or better @ tau = 1sec for 1000 second integration time.
ADEV < 1E-14 or better @ tau = 1sec for 10,000 second integration time.



Also need to take into account that a similar LO would presumably be
used at each antenna.

For integration times of up to a few hundred seconds VLBI has been done
with a hydrogen maser at one antenna and a PRS10A at the other.
However the performance isnt Earth shattering and there are some issues
with long term drift and transport induced frequency shifts of the PRS10A.

For very short integration periods a good OCXO is adequate (at 1420MHz).
When an integration time of 1000sec or more is required the LO
requirents are more stringent.

Bruce