Assistance Using Scope to Troubleshoot

[Chuck H]

I get sound at both the middle & one leg of jcm volume control. Is there a preference?

I don't understand this question. Are you saying that when you inject signal you get signal voltage at these locations? And what would be the criteria for preference? That information is a little arbitrary to the process just now. But that should be normal unless the volume knob is set to zero. In that case there should be no signal on the middle leg and the left leg (looking at the pot from the back).

[Perkinsman]

Let me start with the easiest stuff. I'm not injecting a signal yet, just plugging in a low wattage hi-fi speaker I use for testing small guitar amps for sound. I powered up & turned up the master & jcm volume to 2 so i could hear the amp thru the speaker. I tapped each preamp tube & each sounded like a dull thud except v2, which didn't make any shrieks but it just sounded different, maybe more "plinky" & more "live" if that makes sense.

As far as the shielding on the two locations Pedro noted, if you look at the white wires, they are different, theyre thicker with what looks like plastic shielding inside. Is that shielded wire that was used?

Also, i can't turn up too loud to simulate the feedback cuz I'll probably blow the speaker or disturb neighbors so i need to use the scope. I hooked up to a jack as suggested & have signal in now. Where do I hook up the scope? Black to chassis, probe set on x1, x10 or REF? Then what?

[The Dude]

The shield isn't plastic. It's actually a conductive shield around the center conductor. That said, I looked at your pic, and those wires are indeed shielded. You can see the shields soldered to ground.

If you can't turn the amp up to replicate the feedback, you're not going to see it on the scope. There's nothing yet to see except normal signal. Do you have a dummy load? If not, you're going to at least need to turn up the amp loud enough to get it to feedback.

[Chuck H]

This^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

You're going to need to replicate the problem if you're going to solve for it. If you need to talk with the amps owner to discover the exact circumstances then that's what you must do. This is step one.

[The Dude]

Normally, I wouldn't suggest this simply because the problem only appears to be on one channel of the amp. But, since it's a new build and gain structure is different on each channel (and since it's a very simple test), try temporarily unhooking the NFB wire from the speaker jack. It's the wire that goes to the 47K resistor. See if the feedback is still there. If that does take care of the problem, you'll need to reverse your OT primary wires.

[Perkinsman]

Normally, I wouldn't suggest this simply because the problem only appears to be on one channel of the amp. But, since it's a new build and gain structure is different on each channel (and since it's a very simple test), try temporarily unhooking the NFB wire from the speaker jack. It's the wire that goes to the 47K resistor. See if the feedback is still there. If that does take care of the problem, you'll need to reverse your OT primary wires.

Ok, I'll do that but first I need to be able to hookup the scope and duplicate the feedback. I have an 8 ohm 50W dummy load attached but not sure how to hookup the scope. Ground clip to chassis and x10 probe but what next?

[Enzo]

Don't be afraid of the scope. Really all it is is a voltmeter on a screen, a visual voltmeter. As the voltage you are probing changes, the line moves up and down. So you use it like a volt meter. Ground your probe and use the hot probe to touch points of interest in the circuit. Instead of reading the voltage as a number, you see it as a graph.

[Chuck H]

Don't be afraid of the scope. Really all it is is a voltmeter on a screen, a visual voltmeter. As the voltage you are probing changes, the line moves up and down. So you use it like a volt meter. Ground your probe and use the hot probe to touch points of interest in the circuit. Instead of reading the voltage as a number, you see it as a graph.

This is a good analogy. With the caveat that the scope could be damaged by high voltage at it's input. That's why the 10x probe is recommended. You'll want to keep your test points limited to the DC decoupled AC signal signal path for now I think.

And, as I mentioned, you're not even at a place in the effort where a scope is helpful yet. There are still possibilities to eliminate that don't require a scope. The first thing you need to do is recreate the problem the amp owner is experiencing, but at your bench instead of their playing space. This is the first step. Setting up the scope and THEN recreating the problem is sort of like getting your car repainted in case it get's scratched up in the future.?.

[Perkinsman]

Don't be afraid of the scope. Really all it is is a voltmeter on a screen, a visual voltmeter. As the voltage you are probing changes, the line moves up and down. So you use it like a volt meter. Ground your probe and use the hot probe to touch points of interest in the circuit. Instead of reading the voltage as a number, you see it as a graph.

To be honest, I’m not sure what oscillation even looks like on a scope, I’ve only used it to see AC sine waves & that was awhile ago. Do I set the master & channel volumes at the point where the feedback occurs? Will that damage anything if I leave it feedbacking for awhile as I probe? Do I just start at the input & follow the signal path to the speaker out? Are there any places I shouldn’t touch with the probe?

[g1]

You're up against a bit of a wall here. You don't know what oscillation looks like on a scope, but there is also the problem that the oscillation may be strictly acoustic feedback that will only occur with a speaker, not with a load. (as has been pointed out above)
So there is only a chance it will occur, and if it does, you may not recognize it.

A couple things may help you out. First, like nosaj said, scope the output of your generator to get a feel for what your basic sine wave looks and acts like.
Now at least you will know if the amp is producing something else.
Another thing would be to run the amp into a load, but also monitor it with your cheap speaker through a volume control in parallel with the load. Then you will at least be able to hear if it's feeding back or not.

[Chuck H]

Perkinsman,

You seem bent on using the scope for solving this problem no matter what. You're trying to solve the problem of "diagnosing this amps failure with a scope" rather than trying to diagnose the problem the amp has!!! What I'm trying to tell you is that the scope isn't part of the troubleshooting shooting process for this sort of amp problem yet. Why did you ask for advice about the amp problem if you're going to ignore it in favor of this strange path you seem to be on? (not rhetorical)

[Perkinsman]

[QUOTE Another thing would be to run the amp into a load, but also monitor it with your cheap speaker through a volume control in parallel with the load. Then you will at least be able to hear if it's feeding back or not.[/QUOTE]


I'd love to resolve this at the bench without using the scope...but I don't know how to replicate the feedback problem without playing thru a large enough speaker (which I don't have) to handle this amp & even if I could find one, I don't want to disturb my very pregnant neighbor....everyone is recommending that I use a scope.... so aside from using a scope, a dummy load, I'm not sure what else to do.
What are examples of the type of load you're suggesting?....and how would I monitor the feedback with the little speaker without blowing it?

[Helmholtz]

To me it is not clear up to now what kind of "feedback" you have. Could be acoustical (meaning feedback between speaker and PU or a microphonic tube) or purely electrical. How does it manifest/sound? Do you need to connect a guitar? How do controls' settings influence the oscillation? Have you changed input tubes to exclude microphonic tubes? Can you hear it with a dummy load and a speaker (with a 100 Ohm series resistor) in parallel?

An oscilloscope is a powerful tool to trace oscillation but requires experience. Amp oscillation may show in different disguises.

[Chuck H]

If the amp is unstable (parasitic oscillation) your little speaker should be sufficient to reproduce the problem and you usually don't need to plug in a guitar since these typically do not depend on volume or even input signal to actuate. Please get description of the problem and what typically causes it from the amp owner and report back.

[The Dude]

I'm starting to wonder if there is even a problem. Playing a guitar in close proximity to a cranked amp is often going to cause feedback. Is that what we have here? Is the feedback only when you stop playing the guitar, or is it constant once it starts? If you move further away from the amp, does the feedback stop?