[sdiy] Mysterious ground noise

[Antti Huovilainen]

Hi

I fired up my oscilloscope yesterday again after a long pause and
ocntinued debugging the mysterious ground noise problem with my Midi2CV
unit (on protoboard). Turned out, that the scope shows the same noise
(about 4mV p-p) even if the probe tip is connected to probe ground lead
and I touch the tip/ground (doesn't matter which one, since they're
connected) to circuit ground. The noise goes away if I remove power from
the digital section (PIC, UART & 3 latches). It can't be just the case of
grounding the circuit (which uses wall-mart supply and is thus floating
with respect to mains ground) to the scope as I can ground it with other
channel and the noise appears only on the channel that is connected to the
circuit ground (It's not a defective channel on the scope. I've tried
switching them). The noise also goes away if I have a resistor (x kohm,
didn't measure. 1k < x < 50k) between the circuit ground and the
probe-tip/probe-ground.
Does anyone have any ideas how this can happen? A defective scope?

  /----\
--+    +-+15V ----+-----[100ohm]----78L05---digital +5V
  |7815|          |
  \-+--/          \-------- Analogue +15V
    |
----+- gnd --+--------- Digital ground
             |
             \--------- Analogue ground

Antti

"If you don't eat yer meat, you can't have any pudding!
 How can you have any pudding if you don't eat yer meat!?"
   - Roger Waters