I agree with Carl that you don't want spikes to get into your electrolytics
but any semiconductor as well, which of course includes the rectifier
string.
You might ask how they get through the mains transformer and I'm told the
answer is simply the capacitive coupling from primary to secondary. If you
don't have a grounded inter-winding layer (which used to be standard fit as
a safety barrier but is often not because many use either split bobbins on
one limb or use primary and secondary on separate limbs) then the fast
rise-time spikes have an easier path to the secondary. Topaz specialise in
high isolation transformers for this purpose.
I fitted mains filters with large chokes, X and Y capacitors and a vdr in
vending machines I used to make. In my first trial batch I had 4 vdrs that
exploded because I put them at the mains input as you see in many circuit
diagrams. I did a better analysis (that I should have done to start with)
and realised that to work properly a vdr is like a zener: it needs something
to pull against, so I moved the vdr downstream of the filter, ie next to the
transformer. Everything ok for the next 4000 units, no failures due to
mains spikes.
I agree with the practice of a ceramic from supply to ground, but I'm not
sure what noise it is suppressing, I don't think it's mains "spikes" but HF
noise of some kind.
For low frequency rectification, I wouldn't use high speed rectifiers, the
hard switch over can be a source of oscillation with the wiring which can
radiate. I would only use "soft" rectifiers, ie vanilla flavour that we all
know and love.
David
G3UNA
> If there's a problem with spikes on the line, isn't it better to
> get rid of them with filtering at the line input so they don't
> reach the transformer in the first place?
>
> Steve> >
> > You dont want intense line spikes going thru electrolytics.
> >
> > Carl
> > KM1H
> >
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|