[sdiy] Problem with 723 PSU

[jh.]

I can only guess here - I never noticed such a problem, and I may be
completely wrong. Anyway:

Maybe you're creating a big loop that can pick up radio frequency, or
act as a resonator. With the dual voltage psu in that configuration,
you don't have a common ground from the transformer to DC output.
Instead, the Regulators are floating, and the single ground connection
is made between the output of the (positive) regulator of the negative
branch, and the reference (gnd) pin of the "real" positive regulator.
That would be one end of the loop. The second end could be the capacitive
coupling between your transformer windings, or simply the capacitance
between the wires. Ok, there's the bridge diodes and large electrolytic,
but this might be more transparent to high frequency than we might suspect.
It's just a vague idea, but if there is such a loop, one of the regulators
sits right inside of it.

I would try ferrites around the long wire, and use twisted pairs from
transformer
to the main unit - just try if it makes any difference. And / or small
ceramic caps
across the AC (secondary) to see if the frequency of ringing changes. Best
to start with that last one, to see if this is the right clue at all.

JH.


-----Ursprüngliche Nachricht-----
Von: <jorgen.bergfors at idg.se>
An: <synth-diy at node12b53.a2000.nl>
Gesendet: Mittwoch, 18. April 2001 10:37
Betreff: RE: [sdiy] Problem with 723 PSU


> >maybe if you put the whole PSU external and run DC into the
> >case?
>
> I definitely would not want to run the regulated DC through long wires.
That's begging for trouble.
> One solution could be to put the rectifier and some filter caps in the
floor box and the regulators in the main unit. But that could just as well
cause similar trouble.
> I had hoped that it could be fixed by changing some component values, but
if not, the best bet is to cram everything into the main unit. It's doable,
just a bit of additional, no fun work.
> On the other hand, you would expect that 0,1 ohms in the wires from the
transformer should not be able to break the design.
>
> /Jorgen