Help me understand an oscilloscope display

[veracohr]

I'm not entirely sure this is the right forum, sorry if it's not.

I've been working on some circuit experiments, seeing exactly what I expected on the o-scope out of my breadboarded oscillator, fuzz & filter circuits. Today I hooked the output of my keyboard (Dave Smith Instruments' PEK) to the o-scope to get an idea of what voltage swing I should expect from a synth output.

What I saw was a funky waveform of 60Hz (so clearly related to power supply, right?) that was about 160Vpp. This was with no detectable output level, metered or audible, when connected to a mixer. When I brought the output level of the synth up to some nonzero level, nothing on the o-scope changed. Nor did anything change when I changed the oscillator shape. This remained when I turned the synth off, and only changed when I unplugged the cord from the output or unplugged the synth's (external) power supply.

Obviously I'm missing something I should know. Can anyone enlighten me?

 

[star882]

Make sure the ground is connected properly. You were probably picking up capacitive leakage.

 

[infinia]

What I saw was a funky waveform of 60Hz (so clearly related to power supply, right?) that was about 160Vpp. Obviously I'm missing something I should know. Can anyone enlighten me?

yikes that's a lot of voltage.
60Hz usually indicates a measurement system ground loop. PS stuff is usually fullwave so 120 Hz is the norm. When looking at power supply stuff change the scope trig. source to <Line>.

Always use a earth / safety 3 prong plug at the scope, never 'float it' or defeat the 3rd wire. When you connect the probe gnd., that DUT point then becomes the only gnd connection.

 

[veracohr]

Er...what ground? The synth's power supply is ungrounded, and the o-scope is grounded to the best of my ability (I don't know how it is beyond the wall outlet).

Edit: this was in response to star882's post.

 

[infinia]

Er...what ground? The synth's power supply is ungrounded, and the o-scope is grounded to the best of my ability (I don't know how it is beyond the wall outlet).

good one gnd, keep searching, the 160V is coming from somewhere. Can you get a stable display and trigger off it?

 

[veracohr]

yikes that's a lot of voltage.

Exactly what I thought! But that's what it said.

60Hz usually indicates a measurement system ground loop. PS stuff is usually fullwave so 120 Hz is the norm. When looking at power supply stuff change the scope trig. source to <Line>.

Always use a earth / safety 3 prong plug at the scope, never 'float it' or defeat the 3rd wire. When you connect the probe gnd., that DUT point then becomes the only gnd connection.

I triggered off the channel that was connected. I'll try Line next time. I did, however, try both the tip and sleeve connections of the TS connector, and both were the same.

The scope is grounded, through a grounded power strip. As far as the probe ground, perhaps I'm missing that. The only time I've connected a ground from the probe is when doing the scope's self-calibration, as described in the manual. Should I be connecting the probe ground to the ground connection of the synth output?

 

[veracohr]

good one gnd, keep searching, the 160V is coming from somewhere. Can you get a stable display and trigger off it?

It's triggered off the channel that's connected to the signal. The display is completely stable.

 

[infinia]

The only time I've connected a ground from the probe is when doing the scope's self-calibration, as described in the manual. Should I be connecting the probe ground to the ground connection of the synth output?

Reverse of that

Yes Every connection needs a return, so always use a gnd from the probe. Except maybe at the scope itself, ie at the cal. port.

Think about this for a second
So without using a probe gnd the scopes return path is only thru a long convoluted path > the pwr line.

 

[infinia]

yikes that's a lot of voltage.
60Hz usually indicates a measurement system ground loop.


(...)

When you connect the probe gnd., that DUT point then becomes the only gnd connection.

 

[veracohr]

Awesome, thank you! I get it now - the reason I didn't have a problem not using a probe ground on the breadboarded circuits is that both the scope and the circuits' power supply are grounded.

Now all makes sense in the world.