Re: Coulomb's force lab.

[kowalskil <kowalskil@MAIL.MONTCLAIR.EDU>]

Ideally students are expected to discover (or be guided to
discover) the Coulomb's law. In part (c) it is the coefficient
of proportionality that is measured. I was speaking in terms
of epsilon_o because that Pasco manual  asked for it.

 From my recollection it used to be OK to say that epsilon-o
must be measured while mu_zero is "by definition." This was
before the new defintion of meter.
Ludwik Kowalski

David Bowman wrote:

> Regarding where Ludwik wrote:
>
> > ...
> > c) Measure epsilon_o. In this part F and Q must be
> > expressed in SI units.
>
> and
>
> > ...
> > Did you try to measure epsilon-o, Michael? How successful was it?
>
> Maybe it ought to be pointed out that [epsilon]_0 is not actually a
> measurable quantity.  Trying to measure [epsilon]_0 is like trying
> to measure one dozen.  The answer is determined by definition--not
> by measurement.  By the definition of the meter and the definition
> the ampere it is determined that
>
> [epsilon]_0 == (10^7)/(4*[pi]*299792458^2) (A^2 s^4)/(kg m^3) (exact)
>
> Rounding to 10 sig figs gives [epsilon]_0 = 8.854187818 x 10^-12 F/m.
> Any experiment that purports to measure [epsilon]_0 is not really a
> measurement of this quantity.  It is really a measurement of the
> accuracy of the calibration relative to fixed standards of one or
> more of the quantities that were measured in the experiment, or of
> any systematic error that prevented the apparatus from responding
> exactly in practice as it was modeled by theory, or both.
>
> This is sort of like an experiment that supposedly proports to
> measure the vacuum speed of light.  Any such experiment is
> really a measurement of the accuracy of the experimenter's standard
> of distance measurement, or standard of time measurement, or the
> systematic ability of the aparatus to actually have EM radiation
> actually travel through an actual vacuum, or some combinations of
> these.  Any deviation between the defined value of c and the
> "measured" value of c must be due to some combination of a
> miscalibration of some measured quantites or a systematic deviation
> of the experiment from actually doing what it is modeled to be
> doing.  In such an experiment one is not really measuring some
> constant of nature.  Rather one is testing the quality of one's
> apparatus.  It should be noted that this is *not* at all a bad thing
> to do.  It's just that we ought to recognize it for what it *really*
> is.
>
> David Bowman