[time-nuts] Standards for units

Magnus Danielson cfmd at bredband.net
Mon Apr 2 04:56:47 EDT 2007


From: "David Dameron" <ddameron at earthlink.net>
Subject: [time-nuts] Standards for units
Date: Sun, 1 Apr 2007 20:45:49 -0700
Message-ID: <410-2200741234549750 at earthlink.net>

> Hi all, 
> I just realized that a meter is defined by the speed of light., see
> http://physics.nist.gov/cuu/Units/meter.html
> It is only to 9 significant digits, so if the speed of light (in some
> controlled environment) is measured more precisely than this, the meter and
> all other derived length units will change?

Incorrect. You have as many significant digits as you want. It is defined to
that number which just happends to use only 9 digits in decimal system. This is
a convenience that was chosen, but the significant digits is as many as you
need.

> (I was taught that 1 meter was 39.37 inches, to define the inch
> , but now I see more of 1 inch = 2.54 cm, as someone just referred to.)

The US Metric law defined 1 meter to be 39,37 inches, but this is unpractical
since you often build precission inch measures (in the world of C E Johansson)
from meter buildingblocks and then 25,4 mm is much more convenient. The
difference is about 2 ppm. Except in a few cases you must use the 25,4 mm
definition (which is now part of the SI system) rather than the 1/39.37
definition.

> I find the standard for the Ampere, mentioned in the nist  pages above more
> difficult, as 2 infinite wires to measure the force between cannot be
> found!

They are stapleware! :-) No, it's indeed a strange definition but it can be
converted into a more realizeable thing. I beleive that one will be redefined
eventually. We can count the amount of electrons passing a cross-section per
second and having the defined the element charge we simply use the second
definition which is the most precise SI unit we can realize anyway.

Go to BIPM and you can download the SI document from them. Includes parts of
the history. They also have more info online.

Cheers,
Magnus



More information about the time-nuts mailing list