[time-nuts] Measuring speed of light or reproducing a metre

[Poul-Henning Kamp]

In message <CALH-g5boAbvTcgtbtdr8uoAv0Mwwhg5vfBrK8K-gDJ7NdYC9Dw at mail.gmail.com>
, Jim Palfreyman writes:

>With a 3325B, a 5370B, and other time-nut miscellany, what's the quickest
>way you can come up with to measure the speed of light OR reproduce the
>metre.

Run a couple of meters bare wire across your table, terminate one
end in 50 Ohm and feed the other end with a sharp square-wave.

Hook the 5370B up with two oscilloscope probes and set it to TI mode,
AVG=1000 or so.

Start with both probes the same place, SET REF on 5370B to cancel
out the differential delay of the probes.

Now slide the probes along the wire, record distance between them and
measurements, have student plot them on {white|black}board and you
have a pretty good approximation of speed of light.

Even the most battle-scarred physics-teachers will look astonished,
when you nail the speed of light down on their table like that...

However, the meter you get is not long enough, due to the
lack of vacuum and excess of copper atoms.

-- 
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.