[time-nuts] Ultra low phase noise floor measurement system for RF devices.

[Magnus Danielson]

From: Dr Bruce Griffiths <bruce.griffiths at xtra.co.nz>
Subject: Re: [time-nuts] Ultra low phase noise floor measurement system for RF devices.
Date: Sun, 01 Apr 2007 13:49:03 +1200
Message-ID: <460F0F8F.2020902 at xtra.co.nz>

> Magnus

Hej Bruce,

> Balanced low frequency signal transmission with shielded cables is 
> easier with XLR connectors, provided one uses 1 connector per channel.

I'm a long time fan of XLRs from my audio-days. Neutrik is the favorite brand
naturally. As long as one wire up properly (1 - GND, 2 - Hot/+, 3 - Cold/-) it
is good.

> Enrico's paper on the use of counters to measure frequency stability
> 
> http://www.femto-st.fr/~rubiola/journal-articles/rubiola2005rsi-hi-res-freq-counters.pdf 
> <http://www.femto-st.fr/%7Erubiola/journal-articles/rubiola2005rsi-hi-res-freq-counters.pdf>
> 
> Shows why the SR620 is preferable to the Agilent 53332A/53331A

I assumed you mean 53131A/53132A.

> and the Pendulum counters for this application.

Depends on the model. CNT-90 has a mode which is most suitable, time-stamping
and everything.

> Using the HP5370A/B or other time interval counters in the picket fence 
> mode is also OK.
> 
> The cost of the SR620 highlights the need for a relatively inexpensive 
> reasonably high resolution timer counter or time stamp device for such 
> applications.

I like properly time-stamping counters, such as my HP 5372A. I need to look at
the details of the SR620, but as I recall things it should be able to pull
about the same stunts.

I haven't dwelved into the details of my SIA-3000P formats and how that would
be useable, but it got a 200 fs singel shot resolution which is rather good.
The software really helps alot when analyzing jitter.

Cheers,
Magnus