Snubber caps on preamp tubes...

[Bob M.]

Is there a characteristic noise a snubber cap might make when it breaks down? These are the small, high voltage caps usually seen on preamp tubes going from plate to cathode and are very typically seen on Rivera-era Fenders. I've added these to an unruly SF Princeton Reverb a while ago and now I'm getting a splattering noise on top of the note when I hit a transient, a hard driven note, on my guitar. The amp worked fine before for a long time. This snubber cap(s) is/are my most recent addition. I think I'm might have also added a thermistor as well, but I'm not completely sure, I'll check my notes.

So many typical amp faults have characteristic noises that help to identify the culprit and location and I was just wondering if that might be true with preamp snubber caps. I stayed away from these little buggers for a long time but I've realized that they have their place and function when other less invasive solutions aren't getting it done.

Thanks in adv.,
Bob M.

[Helmholtz]

Is there a characteristic noise a snubber cap might make when it breaks down?

These caps are not really snubbers as there are no switching spikes to be snubbered and the caps are not supposed to "break down".
Rather they produce a low pass filter effect by reducing high frequency gain - either to shape frequency response or to avoid parasitic oscillation. They don't need to be ceramic types.

While ceramic caps are known to sometimes produce acoustic noise, they should not cause signal distortion - provided they are not cracked.
Why did you add them?

[Enzo]

MAy I suggest: instead of hoping to find a list, try finding out. Just disconnect one end of the cap. Now did the issue stop? Or does the amp still make the noise?

The key to troubleshooting is to isolate the problem.

[Jon Snell]

A lot of manufacturers use a low value capacitor in series with a selected value resistor and place the combination in parallel with the anode load resistor as Helmholtz correctly states.
Using the formulae F (3db cut off point) = R (resistor value) x C (capacitor value) a rough HF cut can be introduced to help stop parasitic take off in the stage it is fitted to. (This obviously ignores anode impedance etc).
It can simply be used as a treble cut as well.

[SoulFetish]

Is there a characteristic noise a snubber cap might make when it breaks down? These are the small, high voltage caps usually seen on preamp tubes going from plate to cathode and are very typically seen on Rivera-era Fenders. I've added these to an unruly SF Princeton Reverb a while ago and now I'm getting a splattering noise on top of the note when I hit a transient, a hard driven note, on my guitar. The amp worked fine before for a long time. This snubber cap(s) is/are my most recent addition. I think I'm might have also added a thermistor as well, but I'm not completely sure, I'll check my notes.

So many typical amp faults have characteristic noises that help to identify the culprit and location and I was just wondering if that might be true with preamp snubber caps. I stayed away from these little buggers for a long time but I've realized that they have their place and function when other less invasive solutions aren't getting it done.

Thanks in adv.,
Bob M.

There are two things that come to mind reading the description of the noises you heard. Okay, 3 things:

1st - Bad Tube. Tubes, can exhibit all kinds of noises that can be fixed by replacing it.. Sometimes its a noise you've never heard before, other times it sounds just like something else.
Tubes are the easiest thing to cross off the list.
2nd - Noisey 100k plate resistor. Sometimes a noisey grid leak, but this less common than noisey plate resistors.
3rd - Noise voltages coupling into the signal through board leakage. CBS decided to mark this era by waxing the shit out of the eyelet boards. Over time, the boards can become resistive/conductive. The fix is find where the issue is coupling into the circuit, and fly the leads or install an isolated terminal.

If you can hear the sputtering at idle, one quick way to isolate where in the circuit the noise is to remove each preamp tube, one at a time until the noise stops. Start at V!, and see if the noise goes away. If the noise is still there, pull V2, and so on.
If the nose does go away, then swap in a know good tube. If that doesn't fix it, you know to start looking at the input stage, tone stack, to the reverb send signal.

good luck

[pdf64]

The ‘splatter at high signal level’ symptom may indicate parasitic oscillation, ie an instability that requires a regular signal for oscillation conditions to be met.
Need to scope the output.

[SoulFetish]

The ‘splatter at high signal level’ symptom may indicate parasitic oscillation, ie an instability that requires a regular signal for oscillation conditions to be met.
Need to scope the output.

Good point! I may have misread his description. If the noise is only audible, riding on signal output, then I think you're probably right.

[Helmholtz]

A lot of manufacturers use a low value capacitor in series with a selected value resistor and place the combination in parallel with the anode load resistor as Helmholtz correctly states.
Using the formulae F (3db cut off point) = R (resistor value) x C (capacitor value) a rough HF cut can be introduced to help stop parasitic take off in the stage it is fitted to. (This obviously ignores anode impedance etc).
It can simply be used as a treble cut as well.

Something went wrong with the formula above. The product R*C has the dimension of time (seconds) and is called the time constant of the filter. The corner frequency is given by
f = 1/(2*pi*R*C).

[Jon Snell]

"Something went wrong with the formula above. The product R*C has the dimension of time (seconds) and is called the time constant of the filter. The corner frequency is given by
f = 1/(2*pi*R*C)".
You are correct but the time constant was meant as an example that can easily be related to a point at which the frequency response is effectively slowed down or reduced to help avoid parasitic oscillation. The full formulae that you rightly applied can be too complicated for some that may be learning as to why these little circuits are used.

[Helmholtz]

I think it's essential to understand that the corner frequency varies with the inverse values of C and R. More C,R -> lower frequency.
The effective R is typically dominated by the output impedance of the stage (often around 40k).