[sdiy] Magnetic Bug squashed !

[Magnus Danielson]

From: harry <harrybissell at prodigy.net>
Subject: [sdiy] Magnetic Bug squashed !
Date: Sat, 26 May 2001 01:38:58 -0400

> Thanks all for the tips. The magnetic bug is now dead. As some pointed
> out... it could be a grounding bug. It was BOTH... the magnetic hum was
> masking a more subtle ground loop problem.

I think I hinted you towards that. In my experience it is typical, you
have a full suite of problems and you need to figure out the worst
ones before you go ahead and learn about the others...

I've gotten a little bit out of practice, but I used to be pretty good
at it once... (naturally in my own view).

> This was a rack mounted "360 Systems Slavedriver"... a semi-useless
> Guitar/CV interface. I use it for the string select logic, with mods...
> to get
> a low note priority input to a custom octave dropper / bass synth. As
> practice
> starts in 10 hours (morning)... I had to fix it right away.
> 
> I removed one power supply and used a wall-wart. Then I opened the 360
> box... it had a 723 based dual tracking supply... with a shitty little
> 24VCT
> E-I core tranny... that was the major source of mag field... much more
> than
> the potted block supply. So I used the potted block (+/-15) to supply
> the
> 723... eliminating half the AC in the box.
> 
> The I isolated the PCB from the box... I use nylon shoulder washers to
> insulate
> the 1/4" jacks... XLR connectors with the shield floating. The only
> ground to the system is now at the output lead. Hum gone.

Depending on your transformer this might or might not be
good. Transformers have capacitive leakage which will make the
secondary coils to experience a signal insertion relative to primary
coil 0/chassi. This could be a source of trouble. Fortunately, this is
at a very high impedance so tossing in a 250 kOhm resistor between
chassi and signal ground can be sufficient to reduce this noise alot.
I've seen boxes where this noise would reprogram the DSP from FIR to
IIR and have the whole thing selfoscillate ;o)

It took some searching to find, but was obvious when I figured it out.
I really like all transformers to have double individual layers of
conductive isolation foil since then, since that will give a very
effective tool on handling these problems.

> This also eliminated an isolation transformer in the bass output...
> which SCREAMS
> "I've got a ground loop problem" don't it. Should have listened
> earlier....

You only learn just too late what your actual problems where, don't you?

> Still, its nice to kill that bug after 20 years of buzzing...

Right.

> Thanks again all....

You are most wellcome!

Cheers,
Magnus