Re: Mains conditioners and power cables


"Arfa Daily" <arfa.daily@xxxxxxxxxxxx> wrote in message
news:cNyRe.5414$x4.4455@xxxxxxxxxxxxxxxxxxxxxxx
>
> "Nick Gorham" <nick@xxxxxxxxxxx> wrote in message
> news:43169ee2$0$12892$cc9e4d1f@xxxxxxxxxxxxxxxxxxxxxx
>> Arfa Daily wrote:
>>
>>>
>>> * I'm having a bit of trouble with that one. A larger capacitance will
>>> certainly try to increase the * initial charge current over the first
>>> few cycles as the power supply comes up, which if allowed, * may exceed
>>> the peak forward current capability of the rectifier diodes. However, in
>>> a
>>> * transformer derived supply, the effect will be mitigated by the
>>> natural current limiting effect of
>>> * the transformer's secondary and to a lesser extent, primary, winding
>>> inductances. Any
>>> * resistance in the windings, or in the form of surge limiter resistors,
>>> will also contribute to the
>>> * charge current limiting.
>>>
>>> * Once the capacitor is charged, the ripple will be reduced by the
>>> effectively lower source
>>> * impedance of the increased value of capacitance. This effectively
>>> means that the capacitor is a * bigger bucket, better able to supply the
>>> load's current requirements, which haven't changed
>>> * from when a smaller value was fitted. This, it only needs to do, of
>>> course, between cycle
>>> * pulses from the rectifier. As the bucket won't drain down as far
>>> during these pulse gaps, it will * take less energy to fill the bucket
>>> back up, not more.
>>>
>>
>> 1.It will take exactly the same enery to "fill up" as the same amount has
>> been removed by the amplifier running with the same current requirement.
>>
>> 2.However as the capacitance is higher, removing the same energy per
>> cycle will produce a smaller voltage drop.
>>
>> 3. But because the voltage drop is smaller (2), the angle that the diodes
>> conduct over will be smaller.
>>
>> Therefore as the diodes have to provide the same energy as before (1),
>> but over a smaller time (3) the peak current MUST be higher.
>>
>> --
>> Nick
>
> Hmmm. Thinking on this some more, that sounds kinda right ( sorry Tim
> !! ). Obviously as the same amount of energy has been taken out, the same
> amount will need to be put back in. I must have been having a brainstorm,
> but it had been a long hard day...
>
> Thanks for putting it a bit clearer for my poor addled old brain to get
> itself around, Nick.
>
> Arfa
>

OK. I had a bit more of a think about it, and here's what I did. I hooked a
low voltage mains transformer secondary, to a standard bridge rectifier, via
a 3R9 power resistor. I then hooked a 1000uF cap across the other side of
the bridge, and loaded the whole shebang with a 10R wirewound.

Actual voltage / current values are arbitrary - I'm just going to work in
'scope graticule divisions. The actual measured DC was about 6v, but this
can't be taken as a reliable figure because of the ripple being deliberately
created.

The ripple voltage, with the described setup, was 2.4 divs. The voltage
'scoped across the 3R9 feed resistor ( and thus representative of the
current pulse amplitude into the reccy ), was 5.4 divs peak to peak. The
total time for one rectifier input cycle was 8.6 divs, and of that, the
rectifier was conducting for 6.8 divs ( 3.4 pos and 3.4 neg )

I then connected a second, identical, brand new and from the same batch,
1000uF cap across the first. As expected, the ripple amplitude across the
load, dropped in half to 1.2 divs, but here's the rub. The waveform across
the feed resistor remained exactly the same in both amplitude AND rectifier
conduction time. I actually put a switch on the second cap, and observed the
feed waveform closely, and you could not see a jot of difference. I then
went again, but with a 6800uF cap, and again, you could not see ANY change
in the feed waveform, but of course, the ripple across the load, decreased
to almost nothing.

I then did some experiments with the value of the load, upping it first to
18R. The result of this on the feed waveform, was to reduce it's amplitude,
as you would expect, to 4.2 divs, and reduce the conduction time to 6 divs
( 3 pos and 3 neg ). A further increase to 75R took the amplitude down to
0.4 divs, and the conduction time to 2.8 divs ( 1.4 pos and 1.4 neg ). The
value of the cap, again had no effect at all on these figures.

So, what's the upshot of all this ? It would seem to me, that

1) The value of the smoothing cap has no effect at all on the peak current
value through the rectifier diodes, for a given value of load current on the
DC side of the rectifier.

2 ) The only effect of increasing the capacitor value, is to proportionately
reduce the amplitude of the ripple voltage for a given load.

3 ) The only thing that affects the conduction time of the diodes, is the
value of load current drawn from the DC side of the circuit.

Perhaps I'm missing something here, but these - admittedly fairly simple -
tests, would seem to support my original contention that the only potential
problem to increasing a smoothing or resevoir cap's value, is that of the
initial increased charging current during the first few cycles after power
up, which may indeed lead to rectifier failure, or occasional fuse blowing.

Arfa


.

Relevant Pages

  • Re: nighthawk 550 electrical problem
    ... >> Maybe you're not clear on the concept of the rectifier regulator unit. ... >> You test the voltage regulator operation by doing the "charging ... >> on the voltage regulator to the DC positive output pin. ... >> check the first three diodes, touch the black lead's probe to the AC ...
    (rec.motorcycles.tech)
  • Re: Selenium rectifier question
    ... 2.8X of the voltage for the peak to peak from the AC. ... diodes should be rated to at least 3X the required current at minimum. ... install the tube rectifier assembly. ... Depending on the voltage rating of the main filter caps, ...
    (sci.electronics.repair)
  • Re: Selenium rectifier question
    ... 2.8X of the voltage for the peak to peak from the AC. ... diodes should be rated to at least 3X the required current at minimum. ... install the tube rectifier assembly. ... Depending on the voltage rating of the main filter caps, ...
    (sci.electronics.repair)
  • Re: 13.8V high current power supply
    ... gutsy dummy load or antenna to dump 100W into, ... When the 60 HZ wave drops, the diodes drop out of conduction and the filter ... cap is now supplying *all* the load current...and the voltage drops in the ... Since this is a full wave rectifier if we had ideal diodes they would be ...
    (rec.radio.amateur.homebrew)
  • Re: schottky SB340 -equivalents?
    ... gave s/c both directions, I don't have one in my spares box...anyone ... sometimes helpful to check the peak reverse voltage using a simple peak ... detector consisting of nothing more than a rectifier and capacitor in series ... directly across the rectifier under test, the measured peak voltage must be ...
    (sci.electronics.repair)