[time-nuts] Shopping list

Poul-Henning Kamp phk at phk.freebsd.dk
Wed Aug 23 18:44:00 EDT 2006



I've been playing with my HP5370B and HP J06-59992A tonight ;-)

I am surprised by how consistent the calibration results
are.

When using 10K sample averages, 10MHz/1.5Vpp square from a HP33120A
they vary only a few picoseconds:

     TI++     TI--     TI+-     TI-+      TI++--   TI+-+-
     -----------------------------------------------------
     281.150  215.050  206.850  292.200     0.000   -2.175
     280.750  215.350  207.050  288.550     0.000    0.100
     281.800  215.950  207.650  288.750    -0.225    0.100
     281.900  216.000  208.300  293.100    -0.150   -2.300

1MHz/1.5vpp square, also from HP33120A:

     TI++     TI--     TI+-     TI-+      TI++--   TI+-+-
     -----------------------------------------------------
     107.300   30.000    2.200  139.800     0.250    2.600
     107.000   30.750    3.000  140.850     0.075    1.275
     106.650   30.050    2.600  139.600     0.300    2.100
     106.850   30.800    2.650  141.600     0.325    2.075

10MHz via TADD-1 from PRS10:

     TI++     TI--     TI+-     TI-+      TI++--   TI+-+-
     -----------------------------------------------------
     275.450  183.900  177.700  278.700     0.825    0.500
     277.100  182.500  178.050  282.800     0.800   -2.275
     276.150  182.250  178.250  285.700     0.650   -2.425
     278.400  180.350  179.000  281.300     1.175    0.400

The interesting thing is that it is also possible to compare coax
cables to each other using this method: not measuring the absolute
propagation delay of course, only the relative propagation delay
between two cables...

But still...  light takes 3ps to travel a millimeter...

Enjoy!

-- 
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