Vasile,

Your series capacitance voltage divider circuit has certainly received
criticism.
So I hope that I'm not kicking a dead horse.
While a transformer would definitely be the safest approach for the typical
hobbyist, I'm glad to see that there is someone who is willing to be more
intimate with the electrons.
The concept and configuration of your series capacitance voltage divider
circuit is fine.
I have designed industrial applications which have provided several amps of
current at low voltages from 220 Vac and 277 Vac mains using a series
capacitor. Industrial equipment and machines frequently use a series
capacitor instead of a transformer for providing lower voltages.
A few points:
I would not add a resistor in series with the capacitor (C9) as this will
vary the effective capacitance reactance which is providing the desirable
voltage drop and will also dissipate heat.
I would not add a resistor in parallel to the capacitor (C9) as this will
vary the effective capacitance reactance which is providing the desirable
voltage drop. This resistor would also dissipate heat.
I would not be too concerned with the capacitor (C9) remaining charged as
the approximate 6 mA current consumption from the voltage regulator will
quickly discharge the capacitor. UL and TUV stipulate that the voltage
potential on an unplugged power cord must diminish to 1/3 of the mains
voltage within 10 seconds. So, If I wanted to increase the rate at which
the capacitor discharges, I would place a resistor in parallel to (C5) on
the low voltage side of the circuit.
I would want a 1/4 amp fuse between the 220 Vac main and the series
capacitor (C9).
I would also want the series capacitor (C9) rated for 800 V to 1000 V.
If the series capacitor (C9) is getting warm at all, then the voltage
rating should be increased.
I would want the diode rectifiers voltage rating increased to 800 V to 1000
V (1N4006, 1N4007) to assure that there is not any brief reverse conduction
during the delay of the capacitors charge cycle.
Otherwise, looks good to me.
Just a few comments from an old engineer.

Sincerely,
Ned Seith
Nedtronics
59 3rd Street
Gilroy, CA 95020
(408) 842-0858
G O T    G A R L I C ?
GILROY GARLIC FESTIVAL 2001 !
 >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

At 12:10 PM 7/26/01 +0300, you wrote:
>I've done one, some time ago, see
>http://www.geocities.com/vsurducan/c520.htm
>at the bottom of the page, up to 150mA, but I don't recommend such power
>supplies for production devices ( maybe on 120V mains, only )
>Note that in my design, the supply don't like to remain without load...
>Vasile
>
>On Thu, 26 Jul 2001, Roman Black wrote:
>
> > Alexandre Domingos F. Souza wrote:
> > >
> > >         I designed a circuit with a PIC (12C508) that turns on and off a
> > >relay, controlled by an external source. The power to this circuit (around
> > >120mA) is a small transformer. I'd like to build a transformerless PSU
> into
> > >this circuit. But EVERY scheme I found on the internet was unusable. I saw
> > >something on the Embeeded Control Handbook from Microchip, but it only
> goes
> > >to (if I'm not mistaken) 45 mA. Maybe someone can help me to develop
> (or at
> > >least learn to) a simple transformerless PSU that 120mA.
> >
> >
> > Hi Alexandre! This can be done if you REALLY want
> > to do it. The Mchip app note shows 45mA, and that
> > is with a half-wave rectifier (two diodes).
> >
> > Just change to a proper rectifier bridge, this will
> > double the current to 90mA. Then using a 30% larger
> > value of the main Xc cap will give you 120mA.
> > This web page has Traffic.PDF sent by Richard Ottosen
> > which shows a larger current Xc supply with a
> > full-wave bridge:
> > http://centauri.ezy.net.au/~fastvid/tube4w.htm
> >
> > IMPORTANT! using a full wave rectifier will give
> > lethal voltages at ALL points of your 24v circuit.
> > You no longer have a reference ground! I know you
> > said you were using an optocoupler but remember at
> > all times this circuit is dangerous...
> >
> > Some handy points, you can get reliable relay
> > operation at half the hold current usng the old RC
> > trick, which gives it full current for pull-in
> > but only half current under hold. Costs one resistor
> > and one 470uF electro. Maybe this would get you
> > down to 60mA circuit current. I have used this trick
> > down to about 25% hold current.
> >
> > Remember the Xc supply ALWAYS runs full current,
> > so if your 120mA relay is OFF, something else
> > (large zener?) must be drawing the 120mA. Have a
> > serious think about this. Xc supply is best for
> > circuits that always draw the same current.
> > :o)
> > -Roman
> >
> > --
> > http://www.piclist.com hint: PICList Posts must start with ONE topic:
> > [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads
> >
> >
>
>--
>http://www.piclist.com hint: PICList Posts must start with ONE topic:
>[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics