>>| R = V/I was called "resistance",
>>| and r = dv/di was called "dynamic resistance".
Bernard Cleyet wrote:
> those are the
definitions I absorbed also. Definitions change.
Yes, definitions changes.
Also: It's never good to get into a wrangle about
"right" versus "wrong" definitions.
But let me try to sell people on the disadvantages of
R := V/I [1?]
as compared to the advantages of
R := dV/dI [2]
which is the limiting case of
R := (delta V)/(delta I) [3a]
= (V2 - V1)/(I2 - I1) [3b]
First of all, when talking about _internal_ resistances
and load lines etc., definition [1] is a non-starter.
Starting from equation [3b] we can simplify it to obtain
R = (V2 - V1)/I = (delta V)/I [4]
in the case where I1=0 and we set I:=I2. Alternatively
we can simplify [3b] to obtain
R = V/(I2 - I1) = V/(delta I) [5]
in the case where V1=0 and we set V:=V2. But there is
just no way we can obtain the extreme simplification
R = V/I
because we cannot simultaneously set V1=0 and I1=0 since
the point (0,0) is not on the load line.
=========
To pound on this point (if any additional pounding is
needed), note that it is easy to construct cases where
V/I is essentially infinite while dV/dI is almost zero.
Example: a nearly-ideal constant-voltage source. We
see in this example that V/I doesn't tell us anything
about the _internal_ or _intrinsic_ properties of the
source in question ... but dV/dI does.
(V/I might tell us something about the resistance of
the external load ... but it's not gonna tell us
much about the Thévenin resistance ["internal"
resistance] of the source).
=========
OK, just a little more pounding:
For any circuit, knowing the resistance R:=dV/dI at a
given operating point (V,I) tells us something useful
because it tells us how to extrapolate to other nearby
operating points. But knowing the ratio V/I at a
given operating point (V,I) doesn't tell us anything we
didn't already know. That is, knowing V and I and V/I
is only knowing two things, whereas knowing V and I and
dV/dI is knowing three things ... except in the much-less-
than-general case where the (V,I) curve is a straight
line through the origin.