Constructing a Coherent Amp Building Syllabus

[modestmelody]

So it turns out there is a pretty solid chance that I'll be able to create and design my own ISP (independent study course) with a professor on the Design and Construction of Tube Amplifiers (maybe effects as well).

I'm going to have to create a pretty serious academic syllabus, so lay it on me.

All the old tube manuals, the books on the various circuits, all the stuff that lays out the theory on how these things really work and what each component is doing, etc. I have to plan out a semester long syllabus (approximately 14 weeks) that requires approximately 10hrs of work per week that will bring me from knowing nothing to some goal-level of understanding on how to design and build amps.

I have until April 1st to create a syllabus, but I want to complete that over winter break so I can shop the syllabus with a few professors in the music department and the electric engineering department, making as little work as possible for them.

[Wakarusa]

Is your primary course of study music, electrical engineering, or something else? What's the goal of the course? I.e. how will the Prof(s) ultimately measure success or failure? I'd argue "working amplifier" is a poor metric, and "good sounding amplifier" even worse (who's to say what's good or whether you achieved your tonal objectives? If it does sound "good" how much of that is the amp and how much is your skill with the guitar?). Of course if your goal is really just to score semester hours for building an amp, then you might be able to cobble together a course that does that :D

If instead you go for success being measured by how well you understand design and construction theory (seemingly more appropriate for a college course), how broad do you want the scope to be? Take, for instance, construction: is the focus here on circuit layout for some desired performance metric (noise, stability, tone, dynamics), or should it include ease of construction? How about ease of automated construction? On the design side, is the focus on electrical characteristics, or on how to design to achieve a certain musical behavior? The latter metric being particularly difficult to measure (how do you measure whether you actually achieved "creamy distortion"?). If your course of study is music, rather than a course on amp build and design it might be interesting (and certainly more compelling from a research and learn perspective) to try to define tonal characteristics. E.g. on the music side do a study where folks describe amp behaviors and follow up on the electrical side by drawing conclusions on what specific design elements produce the described sounds/behaviors.

I'd think the problem isn't filling up 14 weeks, it's limiting the scope of the project to fit in 14 weeks.

[VacuumVoodoo]

I'd think the problem isn't filling up 14 weeks, it's limiting the scope of the project to fit in 14 weeks.

I agree, 140 hrs of study will get you through basics and general theory of tube amp circuits. If you could define the objective of the course it'll be easier to advise you.

As for literature on the subject a very extensive library is at http://www.pmillett.com/technical_books_online.htm

[modestmelody]

Is your primary course of study music, electrical engineering, or something else? What's the goal of the course? I.e. how will the Prof(s) ultimately measure success or failure? I'd argue "working amplifier" is a poor metric, and "good sounding amplifier" even worse (who's to say what's good or whether you achieved your tonal objectives? If it does sound "good" how much of that is the amp and how much is your skill with the guitar?). Of course if your goal is really just to score semester hours for building an amp, then you might be able to cobble together a course that does that :D

If instead you go for success being measured by how well you understand design and construction theory (seemingly more appropriate for a college course), how broad do you want the scope to be? Take, for instance, construction: is the focus here on circuit layout for some desired performance metric (noise, stability, tone, dynamics), or should it include ease of construction? How about ease of automated construction? On the design side, is the focus on electrical characteristics, or on how to design to achieve a certain musical behavior? The latter metric being particularly difficult to measure (how do you measure whether you actually achieved "creamy distortion"?). If your course of study is music, rather than a course on amp build and design it might be interesting (and certainly more compelling from a research and learn perspective) to try to define tonal characteristics. E.g. on the music side do a study where folks describe amp behaviors and follow up on the electrical side by drawing conclusions on what specific design elements produce the described sounds/behaviors.

I'd think the problem isn't filling up 14 weeks, it's limiting the scope of the project to fit in 14 weeks.

I agree, 140 hrs of study will get you through basics and general theory of tube amp circuits. If you could define the objective of the course it'll be easier to advise you.

As for literature on the subject a very extensive library is at http://www.pmillett.com/technical_books_online.htm

Definitely understood. Actually my course of study is chemistry, and I'll be working with my wave knowledge from Quantum Mechanics to apply here.

The goals and assessment are defined by me, so part of coming up with this syllabus is finding something that's reasonable and makes sense to fit into this time span. I'd like to say the goal is to understand how to create a functioning circuit electronically from scratch, and then how to adjust the way that circuit sounds as well. The course work itself, I envision, will be more about the design aspects of amp building, with an almost "lab" component simply learning how to safely construct electric circuits using voltages like these and simply having the experience creating the design. I'd almost look at like 10 hours of academic study on amplifier circuit design and 4 hours of lab learning practical building skills each week.

The way I may envision would be like...

2 weeks of electronic circuits in general getting me up to speed and reviewing all of that

2-3 weeks on circuits specifically for audio purposes, maybe start looking at effects and discussing how the simple effect circuits change the wave form of what's coming through (maybe build some simple effects here for experience)

2-3 weeks of understanding amplifier circuits in general and how they work (maybe build some very simple kits/circuits here)

2-3 weeks on guitar specific amplifier circuits, start getting into more complex circuits


Remainder -- look at popular modifications to amps and discuss (or attempt to pin point) how these modifications are adjusting the sound, more complex circuits and modification, begin designing my own amp, etc.

Final would be to design an amp on paper, go over the circuit and approve the design, build the amp. Once it's working, demonstrate it, discuss what I'd like to change about the sound and propose ways to adjust the circuit to achieve that, and finally modify the design to, in some way, get closer to the goal.


Is that reasonable?

[GenoBluzGtr]

google and find (then download) the US Navy NEETS modules. They cover lots more than what you need, but there are a few on circuits, soldering, troubleshooting, and maintenance. If you can find old copies the predecessor to that is a two volume training course on Navy Electronics. I have old copies from when I taught the Electronics course in my Navy days, but I don't know if any are available still.

I think the place to start would be basic electronics theory. Why electrons do what they do, and then transition into individual component theory. THe uses of resistors, capacitors, inductors, transformers, diodes, etc.... different basic circuits for power supply, filtering, rectification, amplification, etc... You can then move into techiques for construction... soldering, IC Chips, PCB vs Tag boards vs PTP, etc...

Lots to do in the area of conversion of signals from electronic to physical (transformers to speakers relationships and the impact of impedance, loads, etc.) Even different construction techniques for components, especially transformers and how that impacts the output signal. Different speakers to best 'display' your design's output.

You can spend alot of time on the various different types of circuits used in the most famous amplifier designs... their differences and similarities.

I think a great project would be to construct two identical circuits, using tubes for one and transistors for the other (with all other components being identical) and try to document the physical differences in the output waveforms and signal characteristics to illustrate the possible reasons for preference for tubes by most guitarists.

[Wakarusa]

The next question is how many credits is the course good for? (these things were worth 3 when I did 'em back in the day). Running with that assumption, let's say that for 3 credits your level of effort should be the equivalent of 3 hours of "classroom" per week and, say, double or triple that in study time with perhaps 1 hour/week of "lab". So total investment of around 170 hours, or about 21 man-days of work.

Compared to the Navy courses mentioned by GenoBluzGtr, that's about the same amount of time Uncle Sam gives your average squid to learn the fundamentals.

So I'd think the problem is again one of scope and your plan overly ambitious (or, at the very least, more work than you'll receive credit for). One way we might want to think about this is from the perspective of the role and expectations for your average student -- if we agree that an undergraduate's job is to learn an existing body of knowledge and a graduate student's job to extend that body of knowledge through research (the dreaded thesis!), then I'll suggest choosing a small sample of existing amplifier circuits and developing a full understanding of them.

So, drawing from the course of study you outlined maybe something like this:
- basic electricity/electronics, electrical safety
- operation of vacuum tubes (and maybe transistors. Bonus points if you do both BJT and FET)
- how to perform AC and DC analysis of a circuit
- operation and application of test equipment
- A full analysis of a few existing designs. One interesting line would be to start with a 5F6-A Bassman, then a Marshall JTM45, then a JMP. Because each is derived from its predecessor you can show this progression in your work. When I say "full analysis", that means to sit down with paper and pencil and work out the expected DC operating points, stage gains, effect of NFB, input and output impedances, phase shift and frequency response of each stage, etc. The analysis of the tone stack alone should prove quite a challenge (how're your linear algebra skills?).
- In parallel, learn basic construction techniques. Now build either one of each design, or build one circuit that can be all three with the flip of some switches and perhaps a tube change or two.
- Now go to the lab and measure the circuit you built. This assumes you've got access to a lab with the right test gear. For the full work up you'll need a lot of stuff. DMM, oscilloscope, spectrum analyzer and signal generator at a bare minimum. Even better if you have access to a distortion analyzer, LCR meter, and other goodies.
- For your write up, explain the differences between your expected results (the full analysis) and what you measured on the bench. See if more expensive coupling caps make a difference in measured results :messedup.

For bonus points, choose one of the circuits and then find a few common mods for it. The JMP would be easiest since there are zillions of mods for 'em. Same deal as before with a full analysis of the modded circuit followed by measuring the results on the bench.

An alternative course that covers a lot of the same ground but is cheaper/easier to implement would be a comparative analysis of a variety of tone stacks. You could easily chew through 14 weeks of study on this subject alone.