Tube Version of OP Amp Circuit

[silverfox]

So I got the schematic from the Pedal Section of Forums here: i'm building a distortion pedal but i've hit a design snag...

What would the tube version of this schematic look like, generally speaking? I don't need to see a specific power supply for the circuit but wonder what it would most resemble? A high gain **** what amplifier preamp stage. And I do understand the highly speculative nature of the question but believe some good approximations will be proffered.

[mhuss]

You'd need a lot of tubes. A "tube equivalent of an op amp" depends on how faithfully you want to duplicate the typical 10,000++ Av (voltage gain) of an IC op amp. To minimally duplicate the op-amp functions, you'd need a differential amp (e.g., a long-tailed pair often used as a PI), plus an output buffer/cathode follower to provide a low output impedance.

[Jazz P Bass]

Some further reading on tube opamps:
http://www.analog.com/library/analog..._chh_final.pdf
http://www.forsselltech.com/media/at...s/VT_Opamp.PDF

[NateS]

i came close to going down this path w discrete power mosfets before I realized i was on the path to reinventing the wheel a few month ago. (albeit a much higher voltage wheel than you can buy for the price of half a dozen mosfets). If you want active summing or differential amps with high voltage swings and without huge input resistors it might be necessary. I chose trying to avoid that requirement.

[J M Fahey]

The pedal you posted has 4 Op Amps.
There used to be tube Op Amps, here´s one:

The schematic is:

So now you know: you need 4 of these , plus a couple extra tube diodes (let´s go all the way) plus enough floating power supplies to feed all of them.
Still interested?

[The Dude]

You could use an old floor monitor for a chassis.

[Enzo]

OK, if we start with the premise that you don't know the underlying electronics, it is a fair question. Op amp ICs have taken the place of tubes in our circuits. But the question assumes the technologies are parallel, and they are not. I like to eat a good meal. To me, meatloaf and mashed potatoes (with of course lots of gravy) is a great meal. I also like fettucine alfredo with lobster. They are equivalent in that both are satisfyingly hearty and taste good. one is based upon meat and potatoes, while the other is pasta and seafood. Both are starch and protein. Now imagine I ask "how can I make lobster fettucine with alfredo from meatloaf and potatoes?" We are kinda at that point here. There is a functional equivalence between the technologies, but not one to one.

Please take this post at face value, I am not trying to be snarky, I am trying to be real. The tube circuit Juan posted used to be what you got when you wanted op amps. But even using them as is, the circuits still are not straight swaps.

Your circuit uses the common 4558 IC, and here is a data sheet.
http://www.ti.com/lit/ds/slos073f/slos073f.pdf

The inner circuits schematic is on the sheet. And if you are interested, you can compare the data sheets of other brands of the same IC. Note all the transistors that go into making one of them.

[silverfox]

Survey Says!

Okay. This is great as I didn't know why it would not be possible to simply come up with the proper biasing for the same number of tube count vs. OP Amps in the schematic.

Thanks all for the in depth but understandable explanations.

Silverfox.

[teemuk]

There used to be tube Op Amps, here´s one:

Indeed. The performance of those simple opamps would also be rather pathetic compared to modern opamps with plenty of linearizing tricks in the sleeve, albeit they add a significant amount of complexity. But they had to make those tube thingies somewhat simple and compact so that compromise was evident. If a modern opamp has dozens of integrated transistors then an equally performing tube opamp would likely have to have as many tubes. It would be the size of a refridgerator, weight a ton and require its own power plant.

e.g. modern opamp:


vs. simple opamp (current sources would be plain resistors):

What would the tube version of this schematic look like, generally speaking?

Generally speaking, the main thing is negative feedback. The feedback path from output to inverting input would be analogous to feedback path from plate (output) to grid (inverting input). But tubes and modern opamps aren't exactly similar (e.g. tubes have much less open loop gain and linearity) so the outcome wouldn't be exactly the same. But the main motivating idea behind those circuits is the utilization of negative feedback and that can be done with plain vacuum tubes just as well.

[Mick Bailey]

After reading the posts it struck me that if the intention is to produce a tube equivalent of the functional aspect of the circuit, then the challenge isn't to replicate the function of an op-amp, but to replicate the function of each section of the circuit in a form suited to tube technology. The fact that an op-amp is used for a certain function doesn't necessarily mean that it can't be effectively replaced by a much simpler circuit.

To clarify, lets say we had an op-amp unity-gain audio buffer. You could say; "well this op-amp has 30 transistors, so we need 30 triodes and as many discreet components". Or you could take a look at the function the op-amp performs in this role and determine that only a single triode is needed and a handful of passive components.

[J M Fahey]

FWIW I've used plenty of the simple Op Amps posted by Teemu
When I started doing this in the early 70s, the only Op Amp which was readily available, I mean massive and at low cost was today's much despised 741 .
Way back then, *everything* was made with them, including expensive Recording Studio stuff.
And it worked, within its limits.
Not in the spec wars game, but one limit I met often was that they were not low noise enough for transformerless balanced microphone inputs, so I just made a 3 transistor front end, and as Teemu said, with a resistor replacing every current supply.
Don't remember the explanation, but PNP input transistors were better.
It also helped lowering idle current through them.
Quoting from memory, I1 would be 47K, R2 and I2 would be 4K7 , Q1/Q2 would be BC157 , Q3 BC147 and of course, "+Vcc" would be -15V and "-Vcc" +15V because we just reversed polarity .
Don't ask way too much from it, but with a closed loop gain of about 10X it lowered the noise floor a lot, compared to going straight into the 741 .
Difference was simply from "hissy" to "nobody complains"
Around 1975 there was an Audio-dedicated Op Amp, much wider response, lower noise and distortion, the uA739.
I started seeing them in Kustom amps, some PA mixers, they were also in Craig Anderton's project book .... but not available (read expensive/exotic) in Argentina, so the discrete simple Op Amp put the bacon on the table.

And FWIW my standard Power Amp is *still* a "Power Op Amp" , a couple BC556 driving a TIP31C to make the basic gain block, and TIP142/147 Darlingtons to provide the muscle, go figure

Those Studio guys who wanted better, had to build *their own* discrete Op Amps and use them everywhere.

One example, the Jensen Transformers "990":

[nashvillebill]

The ultimate irony would be somebody building a tube circuit to try to emulate the circuits on runoffgroove....

(for those who may not know runoffgroove, many of their circuits are attempts at making SS pedal versions of tube amp preamps; I have not tried to make any of the runoffgroove pedals so I cannot comment on the technical validity of their designs)

[nashvillebill]

oops, somehow it double-posted.

[tedmich]

of course their are current craz...audiophiles who insist on op amps without horrid integrated circuits, thus the fully discrete component op amp by Burson

described in self righteous detail here:
Burson Discrete Opamps

other things being made more complicated than they seem to need to be may include Phil Marchand's tube crossover

which probably sounds great as most of his stuff does

[Leo_Gnardo]

One example, the Jensen Transformers "990"

Thanks for mentioning that one Juan - I made a little custom mixer for a music producer using 990's and wow - the 990 is a "sound-good box". What comes out sounds better than what goes in. Yeah, I know that's not supposed to happen. And it can't change garbage to magnificence, hardly. But our ears were very pleased with the results. I got mine from Hardy - licensed by Jensen - Hardy builds simple mic preamps with them. Worthy of experimentation for those inclined to do so. Caution, they are BIG compared to the usual op amp, because they're built of discrete components in a one inch cube. Also expensive, not the buck-apiece op amp you're used to. Worth it for the sound quality.

Thanks also for mentioning the Philbrick tube op amp. There was an article in one of the Audio Amateur magazines IIRC, maybe Glass Audio, that covered these @ 20 years ago. Or it might have been in Vacuum Tube Valley. You saved me a couple hours of digging thru my old mags.