Questions and rumbles about solid state amplifiers

[tca]

I've been playing again with small amplifiers. I've build a few chip amps and ALL of them sounded ok, battery powered and/or transformer powered. Depending on the pedal before it I can get some great sounds but nevertheless the feeling I get is that something is going to break inside that chip, I keep pushing it into distortion but the final result is that I'm never quite happy about the sound I can get out of it.

After searching I've came to the conclusion, you can correct me on this, that somewhere in the past that small transformer coupled guitar amps, the pignose amp for instance, disappeared into to oblivious. Nobody wants to build a discrete amplifier! What I mean is that between that 60's design and the chip amp, we still have all the textbooks class ab amplifiers, and these last amplifiers are not used anymore (I can't state any specific design made exclusive for guitar playing). Any examples?

Of course I could take a classical audio class ab amplifier (I've done that, BTW), from the Mullard book (transistor audio and radio circuits) and use it to play guitar, but that is to use a general audio amplifier and use it as a guitar amplifier. Taking in consideration that the majority of the pedals that we build and discuss in this forum are solid state, why aren't any such class-ab-design-for-a-guitar-amplifier? Of course chip amps are easy to build but the same happens with a Big Muff and it has the same number of BJTs of a typical class ab amp (4 if you are counting)!

Probably these are just thoughts of a deluded mind!

Comments and examples are welcome.

Cheers.

[R.G.]

After searching I've came to the conclusion, you can correct me on this, that somewhere in the past that small transformer coupled guitar amps, the pignose amp for instance, disappeared into to oblivious. Nobody wants to build a discrete amplifier!

The transformer-based transistor amps have faded out of DIY because industry doesn't use them any more, and it's really hard to get the transfomers, I suspect. DIY audio is very much a parasite on the larger electronics and audio industries.

What I mean is that between that 60's design and the chip amp, we still have all the textbooks class ab amplifiers, and these last amplifiers are not used anymore (I can't state any specific design made exclusive for guitar playing). Any examples?

I think I have several megabytes of schematics of solid state guitar amps. Maybe the problem is that these amplifiers were never designed explicitly for guitar playing, necessarily. They were adapted from the mainstream audio amplifier practice which was current when they were designed.

Of course I could take a classical audio class ab amplifier (I've done that, BTW), from the Mullard book (transistor audio and radio circuits) and use it to play guitar, but that is to use a general audio amplifier and use it as a guitar amplifier. Taking in consideration that the majority of the pedals that we build and discuss in this forum are solid state, why aren't any such class-ab-design-for-a-guitar-amplifier? Of course chip amps are easy to build but the same happens with a Big Muff and it has the same number of BJTs of a typical class ab amp (4 if you are counting)!

Specialization for use as a guitar amp is usually done by "dumbing down" a general hifi amplifier. Tricks like softening clipping, limiting or modifying frequency response, and increasing output impedance push the classical hifi designs toward the actions of a tube amp, which we all do think of as made for guitar.

... except they largely weren't. The tube-based guitar amps of the Golden Age were themselves taken from the hifi audio practice of the time and modified to be more cheaply made, and then "corrected" to the hard, industrial use conditions of guitar amps, I think.

[tca]

Hi R.G.,
 I knew you would answer.

I think I have several megabytes of schematics of solid state guitar amps. Maybe the problem is that these amplifiers were never designed explicitly for guitar playing, necessarily. They were adapted from the mainstream audio amplifier practice which was current when they were designed.

ah, so the my fundamental question remains. Is there a solid state amplifier specificity design for guitar playing? I mean, not adapted from the mainstream audio amplifier practice, made bottom up instead of top down (adapting a general audio amp)?

Thanks.

[Gus]

PRS had a solid state amp with transformer out.

If you look at solid state amp schematics you will see some use current feedback.

A story, a friend bought a 2x12 fender like cab.  They bought some nice 12" speakers.  I installed the speakers and to test we used a small Peavey amp IIRC about 10 watts with a 8 inch.  I disconnected the 8" and connected up the speaker cab.  It sounded good.
IMO you need a good speaker(s) for a guitar or bass amp.

I think good speakers matter more than the amp type.  I tend to like 12" speakers for guitar and 15" speakers and bigger for bass.

[tca]

Let me be more specific. I have this on my breadboard



If my calculations are correct it can deliver up to 5W with less than 100mA quiescent current, it sounds terrible for simple audio but sounds great for guitar. You may recognize the circuit driver part. Rolling down the guitar volume it sounds with much more character than any chip amp that I've build, and if you crank the volume up you'll get a good not to fuzzy overdrive sound. And that was the sound that was trying to get. I'm using a BOSS OD-3 to overdrive it and I'm still thinking what to put before it, a booster and an active tone control?!? That is what I mean my a bottom up guitar amplifier: you fix the output power, chose the power voltage and what to drive the power amp (typical tone), etc. One could the same with a chip amp, but that simple class ab topology with all its defects adds something that is hard to get with a chip amp.

Of course one could argue that it sounds better because  it has 3 stages that tunes the input signal, typical my chip amps only have a buffer and the chip itself, but you can get much more control for tone shaping in this case.

I think I have several megabytes of schematics of solid state guitar amps.

Still looking for more examples. Could you share or point to some refs?

Cheers.

[R.G.]

Try googling "peavey schematic". That should bring up a number of them.

I think I'm still a little unclear about what you mean by "designed for guitar". Many solid state amps have been designed for guitar use. However, the basis was general audio power amp practice. Do you mean preamps, the whole input-to-output, or just the power amp?

[tca]

I think I'm still a little unclear about what you mean by "designed for guitar". Many solid state amps have been designed for guitar use. However, the basis was general audio power amp practice. Do you mean preamps, the whole input-to-output, or just the power amp?

I mean the whole input-to-output, I'm thinking globally, not just taking a audio amp and making it suitable for guitar. Probably my thinking is wrong... there is no such thing as a specific guitar amp topology.

P.S. (edit)

Just got the Peavey Standard schematic. Let me look inside.

[R.G.]

You're right - there is no such thing as a specific guitar amp topology, unless you take the very few kinds of tube amps as an example, and those are heavily hifi (such as it was at the time) derived.

On the other hand, let me pose another question. Is a great guitar sound still great went recorded and played back through hifi gear? Or is a miked guitar amp on stage still a killer sound when it's played through the main PA system?

These would seem to say that once a good guitar amp sound is created, it's preserved by high quality, high fidelity amplification and wideband, level response speakers, right?

One question that's worth thinking about is this: where is the magic of great guitar sound hidden? It's certainly not all in one place. Some of it is in the preamp circuits, some in the oddities of the power amp and ideosyncratic made-for-guitar speakers, and probably more in the player's mind and fingers.

I think speakers are underappreciated as the source of great guitar tone. They are really musical instruments in their own right, and they flavor and color everything that goes through them.

[PRR]

> I have this on my breadboard

If it sounds good, do it.

It lacks any real audio NFB, which is probably a good tone.

I doubt it comes close to 5 watts clean, but what's in a watt?

(It has 5-7 ohms internal resistance both ways: the 4.7 at T4, and the 680 divided by T3 hFE. So if it could really pump 18V/2.828= 6.36Vrms toward a load, this voltage is split between internal resistance and load impedance. Figure nominal 8/(6+8) or about 0.57 of 6.36. 3.63Vrms in 8 ohms is 1.6 Watts RMS at clipping.)

It is maybe barely short-proof (mostly because of those large internal losses). Short-proof is a major advantage in hi-fi and *essential* on stage.

C3 does not bypass all the way to bass. The main effect may be that input impedance falls at low frequency. (R3 is "Miller effect", divided by T1 voltage gain.) That's not a terrible thing. If you want to try full bypassing, you need to shunt the 2.5 ohms of T2 emitter. In fact to flatten the input impedance you must get much less than 2 ohms at the lowest frequency of interest (82Hz). That may be impractical. (I've had that cap physically larger than the whole amp.)

C3 = 10uFd appears to cause a bass-cut below 160Hz, which IS very customary in guitar amps.

When you add up production cost and full complete abuse-proofing, the chips win the day for any mass-market product. Current-limit, SOA limit, thermal tracking and shut-down are included for 2 cents, much less than discrete protection or warranty repair costs.

In DIY we don't care so much about assembly cost or trips back to the bench.

[tca]

> If it sounds good, do it.

I will.

> It lacks any real audio NFB, which is probably a good tone.

Yes.

> I doubt it comes close to 5 watts clean, but what's in a watt?
I didn't say it was 5W clean! But you are right it gives an almost a 1W clean.

> It is maybe barely short-proof (mostly because of those large internal losses). Short-proof is a major advantage in hi-fi and *essential* on stage.
I get that.

> When you add up production cost and full complete abuse-proofing, the chips win the day for any mass-market product. Current-limit, > SOA limit, thermal tracking and shut-down are included for 2 cents, much less than discrete protection or warranty repair costs.

ah, but those chips do not sound that good, or do they?

> In DIY we don't care so much about assembly cost or trips back to the bench.

[tca]

These would seem to say that once a good guitar amp sound is created, it's preserved by high quality, high fidelity amplification and wideband, level response speakers, right?

Of course but the creation of the guitar sound is the problem we have in hands.

I think speakers are underappreciated as the source of great guitar tone. They are really musical instruments in their own right, and they flavor and color everything that goes through them.

The way I see it, for those DIYs that can make a 5eurs amp that sounds good, buying a 75 times priced speaker (or more)  is out of the budget, specially in Europe where guitar speakers are very expensive, but we can still do a lot with regular "hi-fi" speakers.

[R.G.]

Of course but the creation of the guitar sound is the problem we have in hands.

Yes. My point was that this can be done without a power amplifier, only the preamp sections, and then this signal can be reproduced at larger volume by the much more common high fidelity audio amplifiers around us.

The way I see it, for those DIYs that can make a 5eurs amp that sounds good, buying a 75 times priced speaker (or more)  is out of the budget, specially in Europe where guitar speakers are very expensive, but we can still do a lot with regular "hi-fi" speakers.

I agree. My points in saying that were that if all you have or can afford is a hifi speaker, you will need to do more work in the low power stuff before the power amplifier/speaker setup to get the sound you are looking for, and more over, that it is *possible* to do so. There are many who would dispute that second point loudly and long.

[bhill]

Even with the based on hi-fi designs you find in SS amps, the pre amp is usually filtered to eliminate higher frequencies. Guitar frequencies pretty much are non-existant above about 5khz, so why bother amplifying them. The speakers chosen for most amps either 10" or 12" (25cm or 30cm) are bass drivers, and most of them bandpass themselves to below 5khz. The amps designers should take this into consideration in the design phase, but most seem to want to tack on useless bells and whistles, like clipping or like 10 band graphic equalizers that only emphasize hiss for the top 5 bands (can you say Peavey? Or Mesa?) As an aside, you might want to think about that a bit when you are trying to build a graphic eq pedal. But getting back to the subject, I have found I can get an identical (to my ears) sound out of either my SS amps or my old BF Fenders. The secret is to use the SS amp in it's clean range, and use pedals to modify the sound. That way you can avoid the ice picky harshness of SS amp distortion.

Of course, when you throw DSP's and amp modeling into the mix it's a whole different game...

[Paul Marossy]

Have you ever looked at the Lab Series L5 amp? It has an awesome crunch tone - think early King's X guitar sound. BB King and Allan Holdsworth were also notable users of that amp. Bob Moog/Norlin design. No output transformer. Granted it's complex, but it was designed specifically for guitar and to sound great, which I think it does.

http://www.diyguitarist.com/Schematics/L5-Schematic.jpg

[Seljer]

ah, but those chips do not sound that good, or do they?

Those chips sound perfectly fine if you keep them clean and let the preamp circuiry do the clipping to make it guitar-amp-like. Letting them clip usually doesn't lead to optimal results.

[Kesh]

You may recognize the circuit driver part.

fuzz face?

As Seljer says. I built a chip amp for the power part of a guitar amp, and concentrated on pre amp and tone stack for "tone" (as well as speakers). In my case this was tubes though. Sounds awesome.

[tca]

Thank you for all your replies.

@Paul Marossy
I'll dive in that schematic.

@Seljer
> Those chips sound perfectly fine if you keep them clean...
That is also my feeling. That's why I thought building a class ab amplifier, then I could use the power amp to also tone the signal, besides all the things PRR said about non chip amps.

@Kesh
Actually is muff fuzz variation, if you take out the 2.7k from the second transistor emitter and the 10k collector resistor. I've breaboard that circuit just to see how it sounded and really liked the fuzz-overdrive sound that you can get out of it. I've tried to do same with the fuzz face circuit using low hfe transistors but the result was not so convincing (I've used hfe=40 BJTs). There is something special the way those transistors are coupled, the feedback loop (base 1st BJT to the emitter of 2nd BJT) makes this type of topology very responsive. BTW, if you look at the schematic of the Peavey Standard, a similar topology appears (fuzz bias, lack of better name). In the case of the Peavey AP there is boostrap-for-gain configuration on the beginning of the schematic and the vibrato has also a fuzz bias configuration. I've also played with the boostrap-for-gain (eg. Escobedo's circuit) and that configuration has some killer OD.

@bhill
About the frequency response... I'm not convinced that the upper bound for the frequency range of the guitar signal is 5KHz. Take the Fender Twin amp,  I'm pretty convinced that there is not such a cut off in that amp. Please feel free to comment on that.

[R.G.]

@Seljer
> Those chips sound perfectly fine if you keep them clean...
That is also my feeling. That's why I thought building a class ab amplifier, then I could use the power amp to also tone the signal, besides all the things PRR said about non chip amps.

The first amplifier I'm aware of to employ this technique was the Thomas Organ Vox amps from the late 1960s. As long as a high feedback amplifier (which performs like the "chip amps" if protections are not activated by some signal) never clips, it puts out what it is told to. Contrary to what is politically correct in the music world, this is not necessarily bad.

@bhill
About the frequency response... I'm not convinced that the upper bound for the frequency range of the guitar signal is 5KHz.

That's an approximately correct figure, give or take a bit, from the inductance and self-capacitance of the pickups. It's demonstrable. I've played my strat into a software oscilloscope and then did an FFT/spectrum on the result, and found that with volume up and tone up, it had a half-power point about 6.3kHz, depending on the note and playing. That's consistent with humbuckers having larger self inductance and self capacitance, and a lower cutoff point.

However, with all kinds of modifiers after the guitar, especially distortion pedals creating higher harmonics, what we hear today is indeed higher, with added content that was not in the original guitar signal.

Take the Fender Twin amp,  I'm pretty convinced that there is not such a cut off in that amp. Please feel free to comment on that.

It probably does have a higher cut off, as Fenders were adapted from existing hifi amps of the time, and they didn't try to cut them back.

[Keppy]

I've noticed that the discussion of frequency response often confuses the reduction of certain frequencies with the elimination of those frequencies. The treble frequencies of a guitar pickup (and most guitar amps & speakers) are definitely reduced. That doesn't mean they're not there, or not important. Half power is a lot different than no power.

[R.G.]

I've noticed that the discussion of frequency response often confuses the reduction of certain frequencies with the elimination of those frequencies. The treble frequencies of a guitar pickup (and most guitar amps & speakers) are definitely reduced. That doesn't mean they're not there, or not important. Half power is a lot different than no power.

Quite true.