"Greg Mayman" bravely wrote to "Mike Ross" (18 Sep 03  08:50:00)
 --- on the heady topic of "VEHICLE LED'S"

 -=> Mike Ross said to Greg Mayman
 -=> about "VEHICLE LED'S" on 09-15-03  23:10.....

 MR> No it isn't of use because that isn't how to achieve a maximum AC
 MR> power transfer.

 GM> It works well at one frequency ;-)

Sure, line power factor correction is one example, or like you mentioned
SWR at a fixed radio carrier frequency.


 MR> "conjugate" of the AC source impedance

 GM> There's notyhing about that in any of my books. Maybe they did it
 GM> all the oldfashioned way, just because it worked in practice.

 GM> But that "conjugate loading" is a lot different from adding a
 GM> shunt capacitor to cancel out the series inductance.

 GM> That way, the output would start to increase as the frequency
 GM> increased towards resonance, reach a peak at resonance, then fall
 GM> off fast at higher frequencies due to the low pass filter action
 GM> of series L + shunt C.

This method of making XL=XC=RL gives a good result when adding the XC
across the load. As far as the output voltage is concerned the XC makes
the network behave as a 1:1 transformer over a wide range of
frequencies.

However, you are right in that the XC makes the voltage drop faster at
higher frequencies. Normally with only the XL, the voltage would
gradually drop at -6dB per octave as an LR filter would.

By adding the capacitance the bandpass becomes flat by lifting the XL
droop but the price of this is that above the cutoff frequency the load
voltage will drop a lot faster at -12db/octave.

In this application though if the resonant point is made quite high,
then the lamp will not extinguish. In fact the capacitance I used was
double what I calculated. This results in a slightly higher average
voltage and a slightly larger voltage rise at resonance but since the
bike won't go that fast there is no problem.


 MR>  When you
 MR> use a series capacitance it will only be possible to have maximum power
 MR> transfer at a single frequency.

 GM> Yes, that is exactly what I said.

You are right of course but we are not talking about an unloaded tank
circuit, nor a single frequency. The lamp draws power from the tank
resulting in the load voltage remaining very close to constant
throughout a range of frequencies up to and including resonance. At that
point the reactive voltage rise instead of being extreme, becomes
quenched and rises just a tiny bit. The tank behaves very much like a
1:1 voltage transformer when it is critically loaded. It works for me!

BTW what method would you suggest to get more power to the lamp?

 Mike
 ****

... A couple of volts below threshold.
--- Blue Wave/DOS v2.30