-=> Mike Ross said to Greg Mayman
 -=> about "VEHICLE LED'S" on 09-11-03  23:16.....

 MR> You're manipulating what I wrote to mean something else since the
 MR> first case was a real case and the other an ideal case. Peaches and

Of course. But you keep jumping between the real and the ideal
and making statements about one that should only be applied to
the other.

 MR> The lamp is obviously not equal to the dynamo's internal resistance.

No it isn't. But who wants maximum power transfer anyway? At that
point the efficiency is at best only 50%. Surely you'd want
better than that with your cycle lighting generator!

 MR> What we're trying to achieve is to cancel the effect of the dynamo
 MR> internal reactance on the load we have. We need an ac power transfer.

Exactly.

The best way to cancel the series inducive reactance is by adding
_series_ capacitive reactance. When XL=XC, the output impedance
of the dynamo will be equal to the resistance of the windings,
plus other effects that appear as resistance.

But this will only occur at one particular frequency and speed.
That wouldn't be a lot of use, would it?

 MR> If we look at the reactive power and the true power we find that the

REACTIVE POWER???

 MR> when Q=1/2 there is no resonance and the voltage rise broadens to
 MR> cover a wider range of frequencies.

If the Q is any figure greater than zero, there is resonance.
If the voltage rises over a range of frequencies, rather than
being equal at all frequencies, that is resonance.

 MR> However, you are right in your
 MR> objection to maximum power transfer but your above statement is only
 MR> partly right because it would only apply when no reactive components
 MR> are present in the circuit.

No, it is 100% right for ALL cases, whether reactive components
are present or not.

Check your textbooks on the methods of matching a radio
transmitter to the antenna -- a case where almost all the factors
are reactive.

 MR>  My real world dynamo has an inductance in it!

As I said before, the series inductance can be cancelled at one
particular frequency by adding _series_ capacitance.

 MR> Doesn't the capacitor supply current back to the circuit when the
 MR> dynamo voltage drops to zero? Doesn't the inductive reactance of the
 MR> windings maintain that current flow? Is the lamp lit?

Lets look at those three questions.

The first two are contradictory. If the inductive reactance of
the windings maintains the current flow, the dynamo output would
NOT fall to zero, so what need would there be for the capacitor
to "back up the circuit"?

And of course the lamp is lit, but if it didn't have thermal
mass, it would flicker on and off at twice the AC frequency.

Where do you think the capacitor gets the current from to "back
up the circuit"? It takes it from the dynamo! Which means that it
"steals" it from the lamp!

Consider the currents over the parts of the cycle, over steps of
45 degrees, where unity = peak current from the dynamo.

Degrees  dynamo current  capacitor current   lamp current
  0           0             -1                   -1
 +45        +0.7            -0.7                  0
 +90        +1.0             0                   +1
 +135       +0.7            +0.7                 +1.4
 +180        0              +1                   +1
 +225       -0.7            +0.7                  0
 +270       -1               0                   -1
 +315       -0.7            -0.7                 -1.4
 +360        0              -1                   -1

Notice how the capacitor supplies current to the lamp as the
dynamo current goes through 0 (at 0 and 180 degrees) but takes
it from the dynamo, effectively stealing it from the lamp, at 45
and 225 degrees.

The lamp does NOT get current supplied over a greater part of the
cycle, although the amplitude of the current is increased.
 
 MR> And if the amplitude of the voltage is greater doesn't more current
 MR> flow over part of the output cycle?

You are confusing two separate factors. More current (or voltage)
- greater amplitude - is not the same as the current flowing
during a greater part of the cycle than before.

What you are saying is the equiovalent of saying that if the sun
shines hotter the days must be longer.

Hey, maybe that's right! The sun is hotter in summer and the days
ARE longer!!!! 

 MR> I'm not debating Sherlock with a connoisseur!   What is
"imprable"?

Doesn't your computer make typos too? 

 MR> It is accurate for a portion of the frequency region of interest with
 MR> respect to reactive effects. I wrote it was obviously wrong for very
 MR> low frequencies down to DC. This doesn't mean its results are garbage.

No, but what it does mean is that the simulation will NOT be the
same as the practical case. Therefore you have to allow for this
and not assume that what it shows must be correct.

It may not even be approximately right if you've missed a key
factor. And key factors are not always obvious.

 MR> One needs to asses what is being simulated in the knowledge that no
 MR> simulation is ever perfect. Are you just a contrarian, man?! 

What have these asses got to do with it? Do you mean donkeys or
rectums? 

Of course no simulation is perfect. But you must not assume that
just because you have _some_ of the factors right the results will
necessarily be meaningful.

From Greg Mayman, in beautiful Adelaide, South Australia
   "Queen City of The South"    34:55 S  138:36 E

... Wisdom is knowing what to do next.
___ Blue Wave/386 v2.30

--- FLAME v2.0/b