I was scoping the input of each stage in an amp build,look what I found hitchhiking on my signal

[SoulFetish]

Setup:
1kHz/100mV sine wave into the input of the amp. Fully bypassed, center biased, with a 100k load resistor. 1st stage drives a FMV tone stack from the plate with a volume control connected to the second stage.
Obviously the volume controls the voltage at the grid, so I took a 622mV measurement at the junction of the treble wiper and input of the volume pot. (TMB controls set at max).
now, here is something interesting; I included an adjustable resistor in series with the 20uF cathode bypass stage because this stage has a real susceptibility to hiss and noise. So, when I attached the scope probe to the grid of V2, here is the shape of the sine wave:




So, I tried attaching the probe ground clip to a bunch of different ground nodes around that stage to see if I was making an measurement error. Oddly enough, the distortion component of the wave form here did decrease when I floated the probe ground lead unconnected.
What do you guys make of this?
HF instability? I will say, I would make serious changes to how I would lay this out if I could do it again. Even though the grid wire is shielded, there is a lot of wire length along the signal path here due to the tone stack and running the volume wiper lead from the front of the chassis to the back of the chassis.

[xtian]

Looks like "ringing". I'm trying to remember when I found this in one of my builds. I suspect it was a microphonic ceramic capacitor. Turn off your sine wave, but continue watching the scope as you tap your caps with a chopstick.

A complete schematic will be helpful.

[SoulFetish]

Looks like "ringing". I'm trying to remember when I found this in one of my builds. I suspect it was a microphonic ceramic capacitor. Turn off your sine wave, but continue watching the scope as you tap your caps with a chopstick.

A complete schematic will be helpful.

Ahhh I use a ceramic cap as the treble cap in my tone stack. (It’s an friggin’ NP0 though, and they are supposed to stay on their best behavior)
although.. i do you have 100pF ceramic bright cap across the volume pot

[Helmholtz]

Is your input signal perfect?
Something like crossover distortion would be differentiated and emphasized by the treble boost in your amp.

How does a square wave signal look?

[Helmholtz]

So, I tried attaching the probe ground clip to a bunch of different ground nodes around that stage to see if I was making an measurement error. Oddly enough, the distortion component of the wave form here did decrease when I floated the probe ground lead unconnected.
What do you guys make of this?
.

Not connecting the probe's ground clip tends to give false results especially with HF components in the signal.

The scope input needs a ground reference (you can't measure a voltage with a single wire).
So if you don't connect the ground clip, the scope uses its own ground as reference.
Scope ground and amp ground are not identical, there's always some impedance (mainly inductance) between.

[pdf64]

My take is that sort of thing is usually the tonestack / bright cap emphasising distortion of an imperfect sine from the sig gen.
if the treble and bright are turned down / off, generally it disappears.

Any unbypassed cathode resistance on an early stage will tend to increase hiss, and increase liability to heater hum contamination (unless heaters are elevated etc).

[Helmholtz]

My take is that sort of thing is usually the tonestack / bright cap emphasising distortion of an imperfect sine from the sig gen.
if the treble and bright are turned down / off, generally it disappears.

Yes, that's what I meant above.

If there's no square wave ringing at different signal levels, I'd exclude instability.

[stokes]

I got a similar thing with a Vibro Champ.There is a 330pf cap from grid to cathode that was missing.Putting it back cleared the problem.In this case the amp was bought to me for a loud crackling sound in the output, when I put the scope on it I found that "distortion" on the signal.

[SoulFetish]

Is your input signal perfect?
Something like crossover distortion would be differentiated and emphasized by the treble boost in your amp.

How does a square wave signal look?

my signal generator sine wave is extremely low distortion. extremely.

[SoulFetish]

My take is that sort of thing is usually the tonestack / bright cap emphasising distortion of an imperfect sine from the sig gen.
if the treble and bright are turned down / off, generally it disappears.

Any unbypassed cathode resistance on an early stage will tend to increase hiss, and increase liability to heater hum contamination (unless heaters are elevated etc).

In this particular amp, I find the opposite to be true. Plus, as I said, the sine wave is clean

[Helmholtz]

Try a square wave test signal. You can use the (attenuated) calibration signal of your scope.

[SoulFetish]

Not connecting the probe's ground clip tends to give false results especially with HF components in the signal.

The scope input needs a ground reference (you can't measure a voltage with a single wire).
So if you don't connect the ground clip, the scope uses its own ground as reference.
Scope ground and amp ground are not identical, there's always some impedance (mainly inductance) between.

The inductance is probably the reason the distortion component was attenuated when the ground lead was lifted, which I figured was a measurement error. The probe ground was connected to the stage's signal ground for this photo.

[SoulFetish]

Try a square wave test signal. You can use the (attenuated) calibration signal of your scope.

I actually have a pretty good square wave signal generator from 1kHz into the hundreds of kHz. I'll check it out.

[Greg_L]

I have no answer for you. I find it interesting that you have a pretty symmetrical hiccup on both sides of the sine wave. What does that say to us? It's happening at a certain pretty specific amplitude range?

[pdf64]

How about if the input stage anode is scoped?
Has a different valve been tried?