How to interpret scope waveforms for guitar amps

[overtone]

Is there a source for understanding how to interpret scope waveforms specifically for guitar amps?

I have just got my first scope, a Tek 465 that arrived today, and have searched the subject here and through google but with no luck. Maybe I am using the wrong search words. I am new to this.

Here are my first test waveforms taken on a junk SE Champ circuit. I could hear the test signal coming out of the little pea size OT, even though I had a dummy load :-)

Image 1. Sine wave measured at input
Measured at the Output:
Image 2. A wobbly form which is, I think, due to the strange oscillation-like noise this amp sometimes makes at zero volume. It settles down by turning it up.
Image 3. Volume at 9 o'clock: a drop down peak emerges at the bottom & the upper waveform seems to swing more than the lower half.
Image 4. The glitch gets larger as the gain is rolled to maximum, now the lower half of the waveform seems to have more swing than the upper one.

Any pointers to this could all mean, or should I just scope every amp that I can get my hands on to get a grip on it?

Best, tony

[nashvillebill]

3 and 4 are way into the clipping territory.

Try turning your sig gen down, does the output clean up?

[olddawg]

Do you have the scope's operator's manual? You should probably be using a 10:1 probe. Then you need to calibrate the symetry by hooking the probe on the calibration stub. There should be a little screw on the probe so you can get a perfect symmetry out of the sine wave generator in the scope. Once that is done put channel 2 on the amp output, trigger to channel 2 and slowly bring up the input signal until it clips on the output then back it off. I would normally do it with the tone controls in neutral position an the volumes all the way up. Start with the signal generator at zero output. Generally you would use a 1khz signal. Depending on what you are trouble shooting for you will set up the amp differently, but if you just want to see the max clean output under load that is a start.

[teemuk]

Looks like pretty typical output of a single-ended amp driven to clipping.

[overtone]

Thanks for the quick replies so far and the advice which I will follow.
Yesterday and today I read the book from R.A. Penfold. I have downloaded the Tektronix manual and have a pair of 1:10 probes and was using those, but did not calibrate them so thank you for pointing me in the right direction.

The Signal was at 4V with 0,2 Div V. I think it was a 1k signal, but as I said this is my first time with this kit and the guitar tuner was not interested. It sounded like a little 1k whine coming out of the puny OT.

[nashvillebill]

Try setting your sig gen signal to around 1 volt or even 1/2 a volt. And if I'm interpreting the dial right, fully up to 200 would be 2 kHz so yep you are at 1kHz.

[loudthud]

1) Normal Sine wave
2) Can't explain
3) Non-symetrical square wave (good). A symetrically clipped wave sounds bad. Note that the average of the signal above ground must equal the average of the signal below ground. So it may look like the wave is moving up and down as the wave changes shape. This is normal for any amp that is not DC coupled from input to output.
4) The little glitches can come from the output transformer, an inductive load resistor or even the cable used to connect the load to the amp depending on where you connect the probe. Since the glitch is only on the falling edge, I would suspect the output transformer forms the glitch when the tube cuts off to zero current because of the overdrive. Usually, you ignor such glitches when they are only several microseconds long but if they get too big (hundreds of volts) they can kill your output transformer.

[overtone]

second scope readings

That was naughty of me to hit the input with 4V, a little embarrassing too.
Below are new images, two channels, with 0,5V signal and the spike only starts to appear at about 3 o'clock on the volume.
I am not over concerned about Channel 2 on the first image, I can hear the test-amp oscillating internally when the volume is at zero, I have not tried to remedy that - I just turn it up. This amp is a mess of brass nails and the wiring looks like my tagliatelle from last night.
Below are overviews of the scope, in case anyone wants to squint at the settings or the probe compensation.

The probe compensations looks to be OK, but I had a little difficulty making out the square wave form. At least the lines appear to be parallel, as in the manual, so I think it is good for now.

Thank you Loudthud, that kind of interpretation is what I am looking for. I have Merlin's books, which contain some scope images, so I think I am set. I still wonder if there is a collection of images of guitar amp waveforms - the good, the bad and the ugly.

Best, tony

[teemuk]

I still wonder if there is a collection of images of guitar amp waveforms - the good, the bad and the ugly.

In what context? You need to think about the function of the circuit: if the task is to amplify stuff with zero distortion then any skewing from the sine wave format is bad. If task is to produce distortion then all bets are off. A typical tube amp can easily produce hard clipping in almost square wave form, crossover distortion, and similar other waveshapes deemed "bad" by people who never took a time to look at a scope and just repeated the generic tube amp myths ad nauseam. Similarly, asymmetric or symmetric clipping or soft or hard clipping can not be condemned good or bad as things are not that black and white and a lot of the tone also depends on many other issues than how the signal distorts.

If you want to use the scope to follow the mere existence of the signal through the signal path, search for hum, noise, oscillations, weird attenuation points, pulse data driving some circuits, etc. it can be used for that and again the context defines what you should be seeing in the screen.

What's really good or bad depends on the context. Perhaps you just need to scope through various sections of all kinds of amps to see how they typically function and what types of output you would expect from certain types of circuits. Then you start to have a better overall idea what you should be seeing and what you should not be seeing.

[Enzo]

Kep in mind that a scope display is just a graphic representation of voltage - a way to watch voltage changes at audio frequencies. You can find collections of waveforms for amp testing, but those are mainly aimed at high fidelity amps. For example, run a square wave in, and if the output has rounded off corners, it means the high freq response of the amp is reduced. From flat. And you will find collections of sine waves, with various harmonics added in. All well and good, but of limited use to us. Guitar amps are neither hifi, nor flat, and those would not be desirable anyway.

COnsider your volt meter. If I said I had 390 volts in my amp, that would mean nothing. It only takes on meaning when I compare it to a standard. I have 390 volts where 485 ought to be - that's a problem. I have 390v where 395 ought to be - perfectly fine. I have to know what I am looking for. You will learn what the traces mean, it is not all that complex, and learning to use the scope will take a little while. But if you find for example that you have an excess of second harmonics in your output waveform, OK, now what? If your square wave test shows a sloping top, indicating a weak bottom end say, OK, now what? What the scope won't tell you is this waveform means an off value screen resistor, or a weak cathode bypass cap. Your experience and knowledge of circuits tell you that. If you have a weak bottom end, then you have to troubleshoot to find out why.

COnsider also, that the waveform might show a symptom - clipping perhaps - but won;t tell you where. You have to hunt that down. And some things can be caused by multiple causes. More than one thing could cause the parasitic oscillation you might see. The scope is a very powerful tool, but it is an extension of the mind.

I'll try to find that electronics data book and post it, I think I linked it here a while back. Full of formulae, the waveforms, terminology, etc.

[overtone]

Thanks teemuk and Enzo, I am taking this all on board.
The point about context for good / bad and generic myths is understood. I will scope through all the stages of all the amps I have, in order to get this under my skin and get myself some first-hand experience.

At my present stage of scope experience, which is about 24 hours, the comments from Loudthud are like goldust. I had no idea what a spike may, or may not, signify and whether to get worked up about it. Enzo, it finally sunk in to my mind that the scope display shows voltage changes at audio frequencies! I guess one needs a mentor. I will consume the Tek manual today, it is very well written with many examples, very tasty.

This new bit of kit is already widening my understanding and approach to guitar amps,

best, tony

[J M Fahey]

You *already* did the best: getting a scope and hooking it to your amp.
All else comes from practice and a little educated guessing.
Use the real speaker that belongs there, play your guitar and your signal generator, vary volume, tone controls, everything, and soon you will start to match what you see to what you ear.
And start challenging a lot of myths.
Now you can easily check that an overdriven tube amp does produce what's basically a "square wave" , none of those nicely rounded waveforms so many *draw* , that it's spikey, unsymmetrical, changes pulse width or duty cycle, in PP amps introduces crossover, you name it.
You will also clearly see the waveform correcting effect of negative feedback, the typical diode clipping waveform, the SS power amp one, and many more.
Plus ripple, oscillation, bad ground, interference, etc.
You will be your own best teacher.
Congratulations again.

[Ronsonic]

Connect your amp to its speaker. Connect the scope to the speaker leads. Plug in your guitar and start learning stuff. You'll have to fuss with the scope to get a good picture. Hit single notes, harmonics, chimes, play clean, crunchy, distorted. Start getting a feel for what those all look like on the scope. This will also help show you how to set your sig gen so that you can get useful info from the scope. For years I've gone to other tech's benches and readjusted their sig gens when I catch them trying to test a guitar amp with 1K Hz. There's no guitar at 1K, that's most of the way up the neck.

Keep experimenting and having fun with this.

[Jazz P Bass]

Good point.
Musical Note Frequencies - Guitar and Piano