"Jasen Betts" bravely wrote to "MIKE ROSS" (13 Sep 03  06:52:51)
 --- on the heady topic of "VEHICLE LED'S"

 JB> 11-Sep-03 23:16:28, MIKE ROSS wrote to Greg Mayman
 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
 MR> the windings maintain that current flow? Is the lamp lit?

 JB> it does that, but it doesn't cause current to flow for a greater time.

 JB> if either way you're getting a sine wave there only two instants when
 JB> there's no ccurrent flowing.

The lamp will always see a (our assumed) sine wave. One difference is
that without the capacitor the voltage across the lamp lags the dynamo
source. The other is that when the cap is added the lamp signal is in
phase or close to it at the frequency where XL=XC=RL and the voltage
across the lamp will rise by about +3dB.


 MR> It's the load voltage amplitude which
 MR> rises.

 JB> Mean or RMS voltage?  - does it actually make the lamp brighter ?

The lamp doesn't care about the capacitor or the inductor. It only
responds as its resistive nature permits and if the voltage at its
terminals rises from 5.77vac to 6.24vac then the lamp will certainly
glow a bit brighter.

The real bonus, I've noticed in the field tests, is it will glow with a
more stable brightness over a wider range of rpm's. But as noted, except
below the dynamo's natural low frequency cut-off. Realize that below a
different, though higher, frequency point the capacitor has no effect
but the inductance then has less effect on the output voltage so the
voltage tends to remain roughly the same.


 MR> And if the amplitude of the voltage is greater doesn't more
 MR> current flow over part of the output cycle?

 JB> measuriong mean or RMS voltage will give self-consistant results for
 JB> current, but, unless you've boosted the RMS voltage you've not
 JB> increased the power output to the lamp.

A lamp's filament tends to behave as a constant power load. Thus as the
source emf drops the lamp resistance drops as well but they don't quite
follow the same transfer function so the lamp eventually goes out before
the source emf does. However, at the other end this holds roughly true
in it's normal brightness range and if the voltage rises beyond a
certain point the power used by the lamp stays roughly the same. As I've
noticed there seems to be a critical region where the filament power and
light output ratio or rather the efficiency is best. This is true also
of different lamp filament types and their rated lifetimes.


 JB> the phrase "greater part" seems meaningless. Curent is flowing for
 JB> essentially all of the cycle, it crosses zero twice but never
 JB> settles there.

We should say instead that power flows twice each period and if there is
less of it lost internally and more of it reaching the lamp then the
"greater part" isn't meaningless as is being made seem but it may have
been a poor choice of words. ;-)

 Mike
 ****

... This message transmitted on 100% recycled photons.
--- Blue Wave/DOS v2.30