Vic Rosenthal wrote:
> Steve Thompson wrote:
>
>> The high current 'gulps' I meant are the ones every line half cycle,
>> not at syllabic rate. The line voltage your meter measures tends to
>> reflect a cycle to cycle average, and doesn't indicate what happens
>> during those few ms whereas the psu feels the full effect.
>>
>> An example (imprecise, so please don't shoot me down): If your amp is
>> pulling an average of, say, 10A from the line for something
>> approaching a kW input, the charge goes into the capacitors in a
>> space of maybe 2ms during the half cycle of 8.33ms. The average
>> current during those 2ms has to be 40A, with the peak being 50, 60
>> or even higher. That sets the voltage drop that the psu suffers
>> whereas your voltmeter and lights don't react so much to the short
>> term disturbance.
>
> If you put a scope across the primary of the transformer, you might
> see a distorted waveform due to the large current drawn during part
> of the cycle. But the output voltage of the supply is related to the
> voltage that appears across the filter capacitors. This total charge
> is (I would guess) proportional to the area under the curve of the
> distorted waveform I mentioned. So the resultant output voltage drop
> due to the resistance in the circuit feeding the PS is an average (of
> some kind) of the voltage drop across the primary over the period of
> the cycle multiplied by the square of the turns ratio.
>
> Another way of putting this is that although the voltage drop due to
> the power line resistance varies greatly during the cycle, the
> capacitors even it out.
Yes, that's one reason why I labelled it 'imprecise'. The current flowing
into the caps has to fall to zero at the crest of the sinewave, so you can
argue that the peak voltage on the caps should match the peak line voltage
regardless of line impedance - but it doesn't seem to work that way. I
haven't done the maths, but one simple way I have of looking at it is to
consider the total line impedance and the caps as a low pass filter - then
you can imagine that the peak voltage and charge replenishment could fall
short. Every time I've measured a real psu, the PSU Designer prog. does a
good job of predicting things - and high peak amps really hurts.
Cheers,
Steve
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|