# No subject

Sun Sep 21 16:24:00 EDT 2008

```electric constant of the cable can lead to significant errors based on the =
actual quality of the cable.
=A0
Even so called "high quality" cable can have large variations within a give=
n manufacturing run of cables.
=A0
Amateur radio operators often cut delay lines for antennas.=A0 Even on HF (=
below 30 MHz) I have seen errors up to 10% between calculated length and m=
easured length.
=A0
Jeff

Jeffrey K. Okamitsu, PhD, MBA
+1-609-638-5402

--- On Fri, 10/3/08, Hal Murray <hmurray at megapathdsl.net> wrote:

From: Hal Murray <hmurray at megapathdsl.net>
Subject: Re: [time-nuts] AC Connector On HP 5061B
To: "Discussion of precise time and frequency measurement" <time-nuts at febo.=
com>
Date: Friday, October 3, 2008, 1:51 PM

> On the DA you will want to ensure all your output cables are of
> identical length as this will ensure that all signals are in phase at
> the timebase input(s).   It's a small thing really a fraction of a
> picosecond but it's there and easily compensated for.=20

If you are after picoseconds, you have to match cable lengths very carefull=
y.

The rule of thumb is a foot per nanosecond.  Coax is slower: roughly 1/2 fo=
r=20
junk but faster for good (low loss) coax.

The way I remember the number for fibers (and good coax) is that 1 ft/ns=20
would be 5 microseconds per mile.  Fiber is 5 microseconds per kilometer.

All you need to calculate the speed of light is the dielectric constant. =
=20
(I'd have to dig out a book to find the formula.  It's probably
k/squt(e)) =20
Good coax is mostly air which reduces the dielectric constant and speeds up=
=20
light.  Similarly, traces on the outside layers of a PCB (half air if you=
=20
neglect the solder mask) are faster than traces on inside layers.

Good coax uses foam rather than solid dielectric.  It's mostly air which=20
reduces the dielectric constant.

--=20
These are my opinions, not necessarily my employer's.  I hate spam.

_______________________________________________
time-nuts mailing list -- time-nuts at febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nu=
ts