Construction Practices

[Reeltarded]

What kind of music do you like?

[martin manning]

(1) when building a guitar amp one should not attempt to amplify what the pickup (acoustic, coil or microphone) is producing with any degree of accuracy.

An amp for an acoustic instrument or vocals (a "PA") is a different bag. These are like Hi-Fi- the pickups are designed to capture the natural sound of the instrument, and the amp to reproduce it cleanly. One exception I can think of is amplified harmonica, where it is typical for the both the mic and the amp to be overdriven. For electric guitar, the pickup is very much part of the instrument, and the pickup and the input stage of the amp collaborate to some degree.

(2) If any stage in the amplifier does provide distortion free amplification it should immediately be hunted down and killed like the misfit it is.

This is not always true. In early guitar amp circuits (1950's to early 1970's) typically there is one attenuator located between the first and second preamp stages. In the late 1950's and early 1960's, guitarists found they liked the sound they got with the volume turned up, where significant distortion was produced in the following preamp stages as well as in the phase inverter, the output stage, and the speaker. In many cases the tone was good but the volume became ridiculously high and out of balance with the other players. "Master volume" topology solved this issue, wherein a preamp volume control allows independently setting the degree of preamp distortion, and a second control sets the overall volume by controlling the signal level going into the phase inverter or power stage. This results in the distortion being primarily preamp-derived, with the output section being relatively clean.

[teemuk]

(1) when building a guitar amp one should not attempt to amplify what the pickup (acoustic, coil or microphone) is producing with any degree of accuracy. (2) If any stage in the amplifier does provide distortion free amplification it should immediately be hunted down and killed like the misfit it is..

It's all a matter of design goals in the end. Electric guitars may sound quite boring and "non-traditional" without considerable signal mangling by various distortion methods and heavy-handed EQ'ing.

The same design principles are a complete disaster when you try to design something which's application is signal reproduction without any coloration. PA and monitor amps come to mind at first but the same "sterility" in reproducing the instrument's own sound is often also preferred by those playing acoustic instruments, keyboards, lap steels, maybe bass guitars... and of course it's always dependent on musical context in the first place.

So after reading up on grid clamping..... .... I am asking why drive a control grid so hard it draws appreciable current.

Grid signal swing is evidently in that range once the power amp is driven into clipping.

You will learn that a great deal of guitar amp design comes from "abusing" circuits so that they distort in some preferable way.

Designs prone to grid conduction are often the same ones everyone's been copying practically since early 1960's (if not longer) so I guess it's just a proven concept to start working on when design has certain types of goals. There's obviously very little point in it if you wish to design differently.

Wouldn't a simple diode like a 6AL5 or silicon javascript:emoticon(':x') (wash my mouth out) provide the same distortion?

It's a whole another can of worms to open if we start to discuss would modeling a certain effect provide the same performance as the original effect would itself. In the end it falls down to realization of the concept. And with this topic you will likely face an attitude where certain group of people are never satisfied until they get the "real thing".

But yes, personally I think there can be and are many different methods one can use to reach a certain goal. If you want to emulate grid conduction or perhaps even operation of the entire power amp as system it's entirely possible but naturally the details and design of your emulating concept determine how closely it performs in comparison to "real thing".

So yes, nothing stops you from using diodes. There are really not too many rights or wrongs in this. It's wrong if it's unreliable. And that's pretty much it. How something sounds is entirely a matter of taste.

[EtherealWidow]

Let's talk about the TW Express.

[ToneMerc]

Why even go down this rabbit hole; there a solid 60+ year history of guitar amplifiers with perceived or real questionable engineering or reliability practices that has given us some of the greatest music ever heard. ......end of story. Thus, pick one and start soldering

TM

[cbass]

What kind of music do you like?

Both kinds country and westeren swing
Don't matter who's in Nashville cause Bob Wills is still the king

[dave g]

So after reading up on grid clamping.....a term unknown in engineering of my day since it would operate a tube outside of its specifications.... I am asking why drive a control grid so hard it draws appreciable current. It has a hard enough life being so close to the cathode and its wires warp so easily. Wouldn't a simple diode like a 6AL5 or silicon javascript:emoticon(':x') (wash my mouth out) provide the same distortion? I'll wait for the answer for that one.

It doesn't draw appreciable current if there is no driver. The output impedance of either side of a long-tailed pair phase splitter is high enough that the grid will not be able to draw a significant amount of current - this is what causes the grid clamping distortion in the first place. Yes, you can probably get some cool sounds out of a tube diode since it follows a 3/2 power law. A silicon diode has an exponential I-V characteristic and will clamp much more abruptly, which is why we like to use tubes in the first place. That said, a clipping diode alone isn't going to do it for most guitarists. A lot of guitarists actually do want an amplifier that has a lot of low-distortion clean headroom, but gradually and gracefully distorts when the input signal level is increased. Hence, guitar amps usually employ amplification stages which are pretty linear for a large region of their operation, and gradually enter into distortion for large signal levels.

You might want to watch this video to get an idea of what a lot of guitarists want an amplifier to do and sound like:

https://www.youtube.com/watch?v=Y2_12Ler9B8

It's only YouTube quality audio, but hi-fi sound is not necessary to get an idea of how he is using the amplifier.

It might be most useful as a hi-fi guy to not think of a guitar amplifier as a "true" amplifier. Think of it as one half of a musical instrument, the other half being the guitar itself.

[surfsup]

no pedals in that video either...

[Meat&Beer]

no pedals in that video either...

Wah! At 1:07 he doesn't kick in anything aside from his bridge pup? The more I look & listen to these things, the more I conjure and conspire the next build...

Upon further watch before posting, dude is quick with the guitar vol pot...

[Phil_S]

He's a player's player, IMO. Every time I listen to that video, I think that my feeble attempt at a TWX came out really well. I can get tone like that, but I can't seem to play it like he does.

I built mine a few years ago. It's got more hiss than I'd like and that I didn't know how to resolve. After much thinking and searching, I finally concluded it needs 1W or 2W plate load resistors which I have but am too lazy to install.

['67_Plexi]

When asking about construction techniques in guitar amps and apparent non-optimality, one must remember the extremely fine lines between laziness, idiocy, and genius.

Ha ha !!

Right on.