A simple, cheap bipolar power supply

[Magnus Danielson]

>>>>> "ME" == Mariano Elizalde <mariano-elizalde at worldnet.att.net> writes:

 ME> At 10:53 PM 2/23/97 +0000, you wrote:
 >> Here is a simple bipolar linear supply.  Using good heat sinking, over
 >> 1 ampere is available from each pole.  Other regulator ICs are
 ME> ...
 >> Here is an ascii-matic:


 ME>   ... just a couple of suggestions;

 ME> ... use a transformer rated to 18VAC instead of the one you suggest because;

 ME> at 12VAC <secondary> you will not have enough raw dc to hold regulation into
 ME> the LM7815/LM7915
 
 ME> at 12VAC as the secondary voltage, the rectified DC will be about
 ME> (12(vac)*1.414)-0.6 or, 16.38 vdc.

 ME> at 18vac <secondary> the raw dc will be around 25 volts... you stress a bit
 ME> the regulators but if they have a heatsink, they will survive..

 ME>  the LM7815/7915 are rated for an input of 3 volts above the regulated
 ME> output or about 18 volts...

There's a diffrent view on this... having an transformer giving higher
secondary voltage will allow for the PSU to survive a wider window of
primary voltages (downwards in this case). One cannot assume that you
get exactly 117V, 120V, 220V, 230V or 240V (or any other nominal
value) but must consider that the voltage may very well vary over at
least +/- 10% of the *nominal* value. This is what the powercomanies
have standardized as acceptable (they have actually lower, nominal and
upper voltages specified) and what you get on some places may very
well vary even further down...

If you intend to have your gear at a gig you migth find that the line
will certainly go down below standardized lowest voltage... especially
when there is much ligth on the stage and the bass kick pumps the heck
out of the power feed.... I have seen the ligths on the mixer fade due
to such downs... it's there for real!

Also, there is nothing which is more anying than gear failing during
the gig but works "just fine" back home... cutting the corners too
hard in the PSU *will* eventually get back to you... and redesigning
an PSU can in worst case become a kludge job... this is one of the
things that should be done rigth from the begining in my
book. Conservative design (in the sense of large margins) is paying
out here I think.

A well designed switching PSU is also a plus... I have seen 230V
(nominal) amps which uses switched PSU even survive 100V nicely... ok,
I did not load it, but it did not shut down...

 ME> and last; suggest you multiply times 10 the value of your capacitors  in
 ME> order to reduce the ripple and to hold max. output power (I*V) since you are
 ME> using half wave rectifiers there...

 ME>  ... just a suggestion...  :)

Not a bad one...

Cheers,
Magnus