How useful is an Oscilloscope for working on Guitar Amps?

[Chuck H]

Chuck H say's "thank you" for this useful post.

[Gingertube]

I would'nt be without my CRO - Before I paid out heaps for a tube tester, the CRO was the ONLY way I had to check for low gain in a stage due to a weak tube and weak tubes anywhere can dramatically affect your sound.

I did a fix up on a Twin reverb recently - the guy was technically savy and had done a full recap on his old amp but it just did'nt pick up sound wise as he had hoped. A weak phase splitter tube and one weak preamp tube were his problem. I found that with the CRO fairly quickly. Now I have a tube tester I'd probably find it faster but the CRO did the job.

Be aware of the limitations of the fancy new digital CROs (like I have). A fellow tech having spotted my CRO went into raptures about the CRO's ablility to do an FFT on the waveform data and give me a spectrum analyser type dosplay and distortion values etc. - until I pointed out that the 8 bit digitiser meant that I would'nt see any distortion product below -40dB. Thats OK for guitar amp stuff but virtually useless for HiFi amp work. The HiFi guys are using 22 bit sound cards with purpose built CRO front ends and fancy audio analysis software for this sort of work.

They are a very handy tool. For the day job I need to service equipment on aircraft. A battery powered Hantek CRO is what I use for that.

We also recently bought one of those little hand held things, a "Sainsmart", cheaper and better than the unit referred to above but really these things are just toys.

Cheers,
Ian

[Steve Conner]

the 8 bit digitiser meant that I would'nt see any distortion product below -40dB. Thats OK for guitar amp stuff but virtually useless for HiFi amp work. The HiFi guys are using 22 bit sound cards with purpose built CRO front ends and fancy audio analysis software for this sort of work.

I spent a lot of time messing with soundcards, but eventually I had to get a real distortion analyser. scopeblog

The THD+N figures published for soundcards are from tests carefully done to show the sigma-delta converters at their best. Testing hi-fi amps is a quite different story, it brings out the worst in them. I found that the high frequency noise and distortion from my M-Audio Audiophile Firewire was worse than my prototype amps.

[loudthud]

Here's an old story from my days at Tektronix. A 465 was brought to Customer Service that had been in a fire. The knobs were melted and the line cord was damaged. After the knobs and line cord were replaced, the scope powered up and met a majority of it's specifications. An ipod has no knobs.

[Chuck H]

An ipod has no knobs.

And is much more likely to be rescued from a fire.

[knutknut]

I have read and followed this thread with interest. As an ex computerised-machine-tool engineer, turned amp repair man, I would not want to be without a least one scope (I have 4). However, a word of caution. Any tool is only as good as the person using it, and oscilloscope can show things you may not understand. It's easy to say hook up the scope to show what's happening, but interpreting the results can often take a lot of experience. Anyone dabbling would probably benefit from a night school course in basic electronics before seroiusly using a scope IMHO.

[JCHaywire]

Troubleshooting 5F4 build with a Hitachi V-212 analog O-scope

It's easy to say hook up the scope to show what's happening, but interpreting the results can often take a lot of experience. Anyone dabbling would probably benefit from a night school course in basic electronics before seroiusly using a scope IMHO.

That would be a fairly good description of me. This is the second iteration of a build of the 5F4 Fender amplifier circuit. My first attempt made sound, didn't smoke (finally did drop wax from the choke), but was quite unsatisfactory. As is often the case, the second attempt went much better. I built a proper (somewhat proper) turret board and laid out the components somewhat similar to a layout diagram posted by Weber Amps online for the 5F4. I learned quickly that working from the schematic and only referring to the layout diagram is a no-brainer. For the second build I used a highlighter and colored every portion of the schematic as it was finished. This one turned out better. Not a whole lot better, but I am learning a lot as I go, letting my curiosity lead the way.

My question is a basic one. This thread mentions clipping on positive and negative peaks as well as asymmetrical clipping. With one probe on pin 7 of V1B monitoring the sine wave output from an android running Audio Tool I see, a nice round input. With the other probe on pin 2 of V2A I note the positive peaks only are clipping as I bring up R16 to about half-way. I'm just getting started in this endeavor, so I want to know what I SHOULD be seeing, where I should be looking, and in which order--to rein in this build. I'm a veteran of dozens of tiny radio transceiver, antenna tuner, and various amateur radio kits, but this is my first build where I am flying solo a bit more. Please point me to any tutorials I may be overlooking to bring me up to speed here. Thank you. I chose this forum because the conversation, although a couple years old, appears exactly on track with what I am doing.

I will revisit this message in a short moment when I'm at my PC with photos and show what is requested--anything to make my process more visible to those who may see the message here.

A couple of the photos are the FIRST build--using the first circuit board I found that was big--an old Hammond organ board. That build was gutted, the chassis reworked and made neater. The other photos are my currrent build. The turret board was made from a PCB that I sanded the copper from, cut some nickel/brass studs I had to a workable length and superglued into 7/64" holes in three rows. Not the best layout I've seen by a long shot--but in the end it made a pretty solid board. The bias pot is a 10K 10-turn pot I had laying around. Novel, but certainly not necessary. The schematic I used is also attached.

[Alan0354]

To me, scope is useful for troubleshoot and if something goes wrong. If the amp is already working ( which is the easy part), you need to tweak the sound ( this is the hard part), I found scope is almost useless. When I bring the amp up, I use scope to work on noise, stability, basic clipping etc. to make sure they all are good. Then I put the scope away.....more because I have no room on the table and it's almost useless to tweak sound.

That's just me. Forget scope, I don't even use LTSpice to tweak sound anymore. I used LTSpice to design filters and look for certain frequency profiles before, it didn't mean anything to the sound!!!!

[JCHaywire]

Schematic for my Conn 5F4 Build

Schematic.

[Chuck H]

The bad news is that there is no definition for what you SHOULD be seeing on the scope WRT guitar amps. They're not the same thing as other electronic devices where finite parameters dictate spec'd performance. Guitar amps clip and distort in all manor of ways. What your scope image looks like with one good sounding setting may be entirely different from another good sounding setting. Asymmetry is good, or bad. Symmetry is also good, or bad. Seeing either at different stages is good, or bad. Some great sounding guitar tones have scope images that no sane interpreter of electronic information would deem proper function of anything. In this light you can pretty much ignore knutknut's commentary. An understanding of electronics won't really help you to determine if the scope image you're seeing is a good tone or a bad one. Too many variables. It's also unreasonable to site a particular amp and ask what the scope image should look like. Guitar amps have a VERY wide range of operator interaction compared to other amplifiers. Just changing the tone stack settings into range extremes can can alter the scope image dramatically. To that end, even asking how a particular amp should scope with specific settings is also not likely to generate answers. We're pretty good with guitar amps here, but we haven't mapped things out in such finite detail.

Scopes used for tone tuning in guitar amps are pretty much limited to identifying peaks, overshoots and symmetry of clipped wave forms, phase alignments, phase margins at frequency and the simple stuff like signal amplitude. WRT repair it's the same as would be for other amps. Things like oscillations and odd distortions can be "seen" on a scope and this allows for isolation in the trouble shooting process. There's a ton of unmentioned goodness about scopes I haven't mentioned here, but my main point is that there is no magic wave form your supposed to see. Even ugly and typically undesirable things a scope can show, like crossover distortion or overshoot spikes, can be part of a "good tone".

There's still plenty that I didn't cover here. It's often helpful to post scope images and get the opinion of experts with more experience. More important would be to use a scope for whatever you can and have it on hand for testing something specific when a situation (or a post responder) calls for it.

[Alan0354]

Yep, as I said before, scope is for trouble shooting, looking for oscillation, find out where you loss the signal, noise. Mainly to bring up the amp to make sure it is working properly. Then it's time to put the scope away. Getting the amp to work is just a very small part, tuning the amp for sound is where the magic happens and that's the hard part. I won't say scope is useless, if you loss the signal somewhere, a scope is "the" instrument you need to trace out the signal.

I can even say that after the amp function electrically, most of the electronic theories are out the window also. I tried using LTSpice to design a circuit to match the frequency response some known network. I can match it on the computer, but when I put it in the circuit, it's not the same!!! I don't even use LTSpice anymore.

That's the discouraging part of guitar amp. I don't have golden ears, that's the weakest part of me. My ears are very relative. It's like my ears can adapt to the sound, so I cannot even trust myself to judge. I resorted to build 3 preamp into my Bassman, one 2204 front end, one Dumble front end so I can use them as the standard so I can A/B compare with my design. But still, it's like different days, I hear it differently.