[time-nuts] Difference between HP 5370A and 5370B

Poul-Henning Kamp phk at phk.freebsd.dk
Sat Apr 16 07:02:51 EDT 2005


In message <42604A91.60407 at pacific.net>, Brooke Clarke writes:
>Hi David:
>
>There's an appendix in the Stanford Research manual for the PRS10 Rb 
>oscillator that explains how to make 1,000 measurements each second and 
>thus get 1 ps accuracy every second.  It involves using a precision 
>external clock into the SR620 and connecting the Ref Out (1 kHz derived 
>from the internal or Ref oscillator) into the External Gate input.  The 
>Start and Stop signals need to be something like 10 MHz, or probably any 
>frequency at or above 1 kHz.  And the gate is used to enable a 
>measurement 1,000 times per second. 

I do this all the time with my HP5370B.

Recently, after I wrote a GPIO driver for FreeBSD, I have managed
to get it to spit out "binary format" timestamps, so now I can make
it take 5000 measurements per second (with 20ps resolution and about
50ps noise) and get all the readings onto my computer, not just the
average/stddev.

I've been playing with doing FFT's of the samples and I belive I
could clearly see the modulation frequency on my old Rb unit,
but I never really found time to play a lot with it.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.




More information about the time-nuts mailing list