JCM800 2203 late 80s - background crackle noise

[Helmholtz]

I thought same, but was told the divider's path to ground is in the scope, not the probe. And on AC coupling, that path is after the cap. Without the schematic it's harder to describe.

That makes sense for the internal attenuator - but not for the voltage divider in the probe.
Just connect a 10:1 probe to a 9V battery and measure the voltage at the BNC contacts.

BTW, scope input and probe voltage limits typically refer to DC + peak ACV.

[g1]

That makes sense for the internal attenuator - but not for the voltage divider in the probe.
Just connect a 10:1 probe to a 9V battery and measure the voltage at the BNC contacts.

As far as I know the probe has only a series resistor.
In your 9V battery test, if I put a cap in series with the meter probe, what DC voltage will appear at the cap?

For the record, this is one of those issues I hope to be proven wrong about.

[nosaj]

More than you ever wanted to know about probes
https://download.tek.com/document/02...bes-Primer.pdf
nosaj

[g1]

It's not really the probe we are worried about here, but the Scope's coupling cap and it's voltage rating when set to AC coupling.

Here is a better explanation from an actual scope manufacturer:

"In DC mode the 9M resistor in the probe forms a 10:1 attenutor with the 1M input impedance of the scope. In DC mode both AC and DC signals will be attenuated by a factor of 10.

In AC mode the blocking capacitor is switched in before the 1M input impedance of the scope. AC signals will be attenuated by a factor of 10, but from a DC point of view the input impedance of the scope is (ideally) infinite due to the capacitor. In other words the input to the scope will see 400V not 40V. I would not worry about a overload at 40V, but 400V could cause problems.

The reason we (and all other scope manufacturers) have the capacitor at before, not after the attenuator is to present a high impedance (ie not load) to DC signals when in AC mode."

https://www.picotech.com/support/vie...22c9ec47#p9915

[Helmholtz]

Ok, so I did some measurements.

My Tek 100:1 probe has an internal 9.9M/111k voltage divider (so I was right here).

BUT the 10:1 probe only has a series resistor.

Furthermore the scope input measures open with AC coupling, meaning that the DC blocking cap is wired before the 1M input resistance and DC voltage will not be attenuated.

So except for the 100:1 probe, the scope input sees the full DC voltage and you are mainly right.

Thanks for the useful info!

[Greg_L]

Well I didn't measure any DC that exceeds my scope and I got new diodes coming tomorrow, so we'll see how that goes. Fingers crossed.

[g1]

To be honest, I've never personally observed that scope precaution, am unaware of the voltage rating of my scope input cap, and have never heard of anyone blowing up a scope coupling cap for that particular reason.

As you say, 100:1 probes usually have an internal resistor to ground, as do 'high voltage' probes, so no problem there. Also differential probes get around the issue.

[Helmholtz]

To be honest, I've never personally observed that scope precaution, am unaware of the voltage rating of my scope input cap, and have never heard of anyone blowing up a scope coupling cap for that particular reason.

I typically rely on the probe's voltage limit assuming that takes care of the scope's limit as well.

[Greg_L]

I had a problem with my last scope, and it was around the same time I last probed some high DC in an amp. Now to be fair, it was one of those Asian Hanteks. I don't know for certain that I fried it with DC, but it could be. It was a little buggy before that. Maybe the DC pushed it over the edge. But it wasn't a ton of DC. IIRC it was the B+ in a Deluxe Reverb, so not super high power stuff. It could have just been a coincidence. I bought another one, the one I have now, and it's been great.

[Helmholtz]

But I decided to go all in and set it to AC coupling and I could zoom in on the DC trace. AC coupled, 10x probe, down to 200mv/div and there it is....lots of hairs and crap squirting off the DC voltage.
I also tried the back end of the last B+ dropping resistor - same result. DC signal with lots of scratchy hairs coming off of it.

It's the same hairy crap that shows up at the output.

And when I scope the output at the speaker jack and listen at the same time, the spikes in the waveform correspond with the sound it makes.

That result seems to confirm the arcing theory.

Invisible arcing can even happen within caps.

[Greg_L]

That result seems to confirm the arcing theory.

Invisible arcing can even happen within caps.

I've tested and/or replaced EVERY SINGLE cap in this amp. Filter and bias caps got changed immediately. The big can filters were bulging and the bias caps were original. I'm wondering if the filter caps going near death hurt the PT or a diode.

[Helmholtz]

Unfortunately not every new part is a good part and arcing problems don't show at low voltage.

Do you see the spikes on B+ with power tubes pulled?

[Greg_L]

Unfortunately not every new part is a good part and arcing problems don't show at low voltage.

Do you see the spikes on B+ with power tubes pulled?

Yes. The spikes are in the amp everywhere.

[Helmholtz]

Yes. The spikes are in the amp everywhere.

That's good, because without power tubes the OT is essentially out of circuit.
To further exclude OT primary to secondary arcing, disconnect the primary CT.

[nosaj]

I've tested and/or replaced EVERY SINGLE cap in this amp. Filter and bias caps got changed immediately. The big can filters were bulging and the bias caps were original. I'm wondering if the filter caps going near death hurt the PT or a diode.

that being the case definately replace the diodes after Helmhotz's request on the OT. Diodes are cheap.
nosaj