[tangent]

There have been a few people making "META42s" on protoboard now. The META42 is not a circuit design -- it's a PCB design only. We who make up the acronym don't deserve credit for the circuit design, so you shouldn't use our acronym to label amps not made on a META42 PCB. It detracts from those who created and helped popularize the circuit, and it confuses people who (correctly) equate "META42" with a particular circuit board design.

The generic term for the basic circuit design is "buffered multiloop amp". It's a bit ungainly, so if anyone's got a better name, I'm open to suggestions.

 

[carlo]

wow, didn't realize that. so who designed the amp? may be a good place to start when thinking of a name for the design. "buffered multiloop amp" is at least a concise and informative name though

EDIT - damn tangent, you have a nice site. just saw the schematic, gotcha. i have absolutely no suggestions for a clever name.

 

[tangent]

Who designed it? Well, lots of people.

Walt Jung came up with the op-amp-cum-buffer multiloop topology. I'm not sure who first introduced the Head* sites to it, but my guess would be ppl.

The idea of biasing an op-amp into class A is probably pretty old. The idea of using a pair of JFETs in a cascode configuration as a current source is also pretty old. So far as I know, though, ppl came up with the idea of combining these two ideas, instead of the simpler biasing methods used by others. There's a thread here at Head-Fi started by ppl with his initial experiments with this idea.

The virtual ground stuff is more or less ancient. It's all just straightforward refinements on the basic resistor-divider idea that you see in CMoy type amps.

We swiped the LED turn-off circuitry from one of ppl's amps.

Shamelessly.

See the META42 History page for more details of how all the pieces came together.

 

[Voodoochile]

Perhaps we should approach Walt himself???
Maybe he can contrive some sort of 'Jung-ism' for the design.

I guess my earlier suggestion for 'ppl's 42# MEAT' didn't fly so well.

Another thought. I just built what I would call "Apheared's #47" last night. Granted, there is no PCB for this, I used some Vectorboard. But the library at HW opens up with this notation: "[Editor: Apheared's 47 amplifier has a current-boosted output stage. It developed from his work with the pocket amp by Chu Moy.] " Now they don't list any specific PSU for this amp, so I did my own derivation, which I will describe later. But this begs the similar question, is it Apheared's #47 (based on his work with the C-Moy)... which appears to be based on circuits in TI datasheets, and so-on? At what point does a design cease to be one thing, and become another? The PSU I built for this amp last night, for example, is a cross between the C-Moy and BDX units, or the power section of the META42 PCB schematic. A railsplitter, driving a BUF634 for the vGnd, four 470uF Nichions, and a pair of polyester .47uF at the output.
So, is this 'Apheared's #47 following a CMoy/BDX/META42 PSU'?
Or... "Frankenstein's sweet-sounding person son"?

I'm honestly not trying to be a pain in the ass, but do wonder how these categorizations are arrived at, and/or how they can be somewhat consolidated. Last week I copied the schematic of the META42 on protoboard, with minor changes to accomodate BUF634s instead of EL200*, single-channel amps, and changed some of the R values in the multiloop config. I have no idea what to call this amp.

I love the PCB, BTW, Tangent. The second one is coming along nicely. It's good to have the first one config'd with resistor sockets to assist in picking values for this one.

 

[tangent]

Quote:

At what point does a design cease to be one thing, and become another?

That's why I focus on the PCB. It's the thing that we actually created, and it's the physical object being offered. The circuit we based the PCB on lives a life of its own -- as far as I'm concerned, it's public information.

Because the circuit design is independent of the PCB, you can see that when someone makes an amp based on this circuit it isn't "a META42". It's simply an amp based on a similar circuit design.

Several people have tried to get me to change the name of the PCB, but no one's trying to hang a name on the circuit design itself. I think that that would be a more appropriate division of credit than modifying the PCB's name to include all of the people who invented ideas in the circuit.

 

[Voodoochile]

Quote:

Several people have tried to get me to change the name of the PCB, but no one's trying to hang a name on the circuit design itself. I think that that would be a more appropriate division of credit than modifying the PCB's name to include all of the people who invented ideas in the circuit.

I couldn't agree more.

 

[andrzejpw]

So, what should people call their creations? I mean, like it or not, it seems that META42 has become synonymous with the amp. I mean, I remember reading ppl's post about his #42 a while back. . . so, should we call the amp the 42?

 

[geom_tol]

Maybe just 'multiloop'?
Buffering is unavoidable if doing a multiloop amp.

It's not very exciting though. Walt Jung calls it 'composite' amp in his book and the articles if I remember right.
'composite' is even less inspiring.

'Jung multiloop' sounds better maybe? Kind of philosophical.

Yea, 'Jung multiloop' get's my vote.

 

[andrzejpw]

eh, still not a fan of that. That's like calling a car a b16a, because that's what kind of engine it has.

 

[KurtW]

Well, there is something to be said about not having a name for an amp, regardless of what board is used. The sound quality of the amp will depend a lot on the components used and what kind of power supply it has, etc., not just whether it has a multi-loop topology or not. HeadRoom's Max and Little both have similar circuit topologies, neither one is multi-loop, but they certainly sound different from each other. Even when using the META42 board, there are lots of choices that are made that can change the character of the amp. I suggest people get out of the over simplistic mode and use a sentence or two (or a list of attributes) describing the amp that they are talking about, instead of trying to do the impossible by only using one word to describe it.

 

[CaptBubba]

Well because you can use just about any opamp/bufer for this amp it may be a good idea to add that into the title somehow.

Perhaps:

"Jung-type Amp: Multiloop" (yes, kinda repedative, but I don't know if Walt Jung made other designs so it is good to specify) with the OPA number and buffer # after it.

For example, one made with OPA2134s and 2 Elantec 2001s/chnl would be:
JAM 134/2*2001

If you just wanted to talk about the amp in general you could say JAM.

 

[Toe]

I kinda like andrzejpw's idea. "42" or "#42" could be used as a generic monkier for this entire family of amps. These days Apheared's original "#42" is more commonly refered to as the "A42", so I wouldn't worry THAT much about confusion.

People building amps based on any design more or less based on the whole buffered multiloop design shebang could just describe their creation as a "#42-style" or "#42-family" amp. This generic monkier could be used for the META42, the A42, ppl's 2001 space oddity / pocket amp, or whatever else someone comes up with.

 

[geom_tol]

I don't understand the choice of #42 for the generic name of these amps? I associate the #42 with an amp that Apheared made that is not the same as these Multiloop amps...

Why bring the #42 into it if it already is a name for a different amp? (For the naming of the meta42 pcb it makes perfect sense otoh)

 

[KurtW]

As I recall the A42 had paralleled output buffers but was not mulitloop. The META42 can have multiloop and paralleled output buffers or neither of these.

 

[Possum]

Actually, Apheared's 42 creation (ppl's 2001 space oddity?) didn't even have paralleled output buffers. In the first few threads discussing the 42, ppl had mentioned using multiloop. In threads indirectly related to the 42, discussing the EL2001 buffers, ppl mentioned parallelling the buffers as an improvement.