1)    on occasion (not a real consistant prob.) d5 failed without reason.
when moved to d6 anode as paul has said
to get around this i left d5 where it is & just added another diode on the
anode side of d6 instead
hence leaving d5 to be a zener only.
never had a problem since, not once !!

my favourite diode is 1n4007(100v 1a) cheap & robust !! ideal for mains
applications.
2)    another way is to use back to back zeners ( vz +0.6) + a diode +cap
for your dc supply
occasionally the same happened as above so i added 2 diodes , one each in
parallel with each zener.
no prob.

item (1) is what i use for reliability & cost.
glen
********************

Paul B. Webster VK2BZC wrote:

> Harold Hallikainen wrote:
>
> >  Placing the SCR in a FW rectifier seems to be a worse solution (to
> > me) than using a triac.  You'll get two diode drops plus the SCR drop,
> > which I believe would be considerably higher than the drop on a triac.
>
>   Depends on the load.  For driving transformers or AC motors, this is
> the only way to do it in order to achieve symmetrical triggering, though
> the half-wave supply for the operating voltage would not be satisfactory
> - perhaps a secondary regulator.
>
> >  Also, the drive circuit floats a bit wildly since the negative
> > side of the bridge is now alternately connected to each of the AC
> > terminals of the bridge, depending on line polarity.
>
>   Yes.  It doesn't solve the isolation problem at all.  Best for small,
> encapsulated products such as stand-alone knob-operated dimmers/ power
> controllers.
>
> >>From: Peter van Hoof <pvh@VERTONET.COM>
>
> >> how about this solution
> >> http://www.vertonet.com/members/pvh/scr.pdf
> >> get power for the pic from the other side of the load with diode, cap
> >> and resistor in series.
>
>   Did *no-one* notice the blooper here?  Did anyone look?
>
>   D5 is on the *wrong* side of D6.
> --
>   Cheers,
>         Paul B.