> > If not, how would you define a
'volt'? In terms of the steps on a
> > microwave-excited Josephson Junction? It's reproducable, and
> > frequency/time is also easy to define.
> I was thinking of defining it in terms of a
bandgap, such as used in
> precision reference "diodes".
I think a Josephson Junction is easier to make,
actually. The only real
problem is you need something around liquid helium temperature (you need
a superconducting metal IIRC). Apart from that, all the rest of the stuff
is pretty easy to make.
Of course for something like describing the supply
voltages and logic
thresholds of a chip, you don't need that much accuracy. Describing how
to make a weston standard cell and its voltage (at a given temperature)
would be massive overkill. I would think that describing how to make
_any_ common cell and specifying its open-circuit voltage would be good
enough.
Excuse me if I interupt, but you are talking about measuring
a specific voltage. But thats completly independant from the
definition. Only if you want to define Volt based on some
specific reproduceable effect - but thats not necesaty. To
get all electrical units you just need the 7 basic SI units
(Length, Weight, Time, Current, Temperature, Brightnes (?)
and Mol (sorry, no idea how to translate Stoffmasse)) and
only one is an electric unit (the Ampere). And only these
7 units have to be defined on specific, reproducable effects.
1V is defined as 1 kg m^2 / A s^2.
Gruss
H.
--
Ich denke, also bin ich, also gut
HRK