On Sat, 12 Jan 2008 04:27:08 +0000, The Natural Philosopher wrote:

mick wrote:
On Fri, 11 Jan 2008 00:49:22 +0000, The Natural Philosopher wrote:

mick wrote:
On Thu, 10 Jan 2008 22:54:21 +0000, Andy Wade wrote:

mick wrote:

[...] and the power station has to generate 60W to light your 30W
CFL (although you are only charged for 30W).
That's a complete misunderstanding of the idea of power factor. The
[supply system] doesn't have to generate 60 W, nor burn fuel at a
rate equivalent to 60 watts worth of output. The (RMS) current
drawn by the lamp is the same as for 60 W resistive load, so
resistive losses in the cables are increased by a factor of four
without power factor correction. However the I^2*R losses due to
current drawn by your lighting load will pale into insignificance
compared to that caused by much heavier resistive loads (cookers,
heaters, showers).


You sure about that? I oversimplified.

For a poor PF load the voltage and current are out of phase with each
other (how far depends on the PF PF=1 is in phase, PF=0 is 90deg
out). The generator is producing (and consuming fuel for) VA (real
power). However, domestic consumers pay by W (apparent power), not
VA. So you see 30W of load at the meter and can measure the AC RMS
current into the lamp, but the V and A waveforms are out of phase so
the actual V*A is greater than the W value. (W=VA*PF so a 30W
(apparent power) lamp with a PF of 0.5 will require 30/0.5=60VA input
to power it)

Agreed that the distribution losses into poor PF loads also escalate
with I^2R.

Also agree that the % difference on your bill will be insignificant.
:-)


Substations have BANKS..ACRES of capacitors to correct for power
factor, so that the generators do NOT have to run widely differing VI
phase differences. Its not really clear what sort of PF a CFL is
anyway..A bridge rect and an electrolytic maybe? Or a half wave rect
and an electrolytic..I bet there is a lot of input ripple..its easy
enough to stabilise output ripple with an HF SMPS..anyway a bot of C
across the mains is good, as its in the reverse direction to all those
motors and things..there the current lags the voltage..with capacitors
it tends to lead a bit.

It doesn't work like that. You can't correct for distortion-induced
power factor problems by throwing caps at it. That usually results in
worsening the PF and setting up resonances in the system.


Resonaces at 50hz are what you want.

Actually.


But not on the harmonics (particularly the odd ones). That leads to silly
things like the neutral terminals burning off your transformers... :-)


A good combination of inductance and capacitance will do the trick.
Largely the load PF problem is an inductive on,. Having a few caps
around will make it all nice again..;-)

Ho yes.... but adding de-tuning inductors to your nice shiny PF caps (as
often becomes necessary on distribution systems) starts to get complex
and costly. Ever priced up for a multi-stage PF correction system? The
prices start to look like phone numbers. :-( AFAICT all non-PF
corrected CFLs are capacitive anyway, so as a minimum you need an
inductor in series with each to correct them. There are PF corrected ones
available now (I don't know about UK availability though), with this
built in. :-)


--
Mick (Working in a M$-free zone!)
Web: http://www.nascom.info http://mixpix.batcave.net