Amp Emulations

[R.G.]

It seems to be semi-irresistible for folks to take an amplifier schematic, delete the triodes and replace them with JFETs, diddle the biasing and then call it an XXXX emulator.

We know that JFETs and triodes do not distort alike, but the process of emulating an amp by rote seems to be satisfying to people.

I have this suspicion that the frequency determining elements filtering the distortion is what is so attractive, not the distortion. That being the case, I suspect that you could sub in for instance opamp distortion stages for the triodes/JFETs and have just as interesting SOUNDING "emulation", albeit not with as big an emotional hook.

Anyone ever tried this, other than Stephen Moeller?

[MartyMart]

Yes RG, the chap who put together the LXH2 JCM800 sim
Use's ( from memory ) 9 opamps for the preamp section and another 11 ! for the
speaker sim.
Sounds wonderful BTW !

http://www.diystompboxes.com/smfforum/index.php?topic=40339.0

Marty.

[Doug_H]

I had a happy accident with an op amp recently that sounds "marshally" in a scary good way. I didn't sit down and work out transfer functions or anything. Just simple bode and transient plots of another design of mine I was trying to improve and simplify. A few tweaks and it turned into this. Emulation? No. But I've campaigned against the use of the word "emulation" from day one, <sigh>...

The Moeller thing is interesting but way too complex for what it "emulates". For me it would be much simpler to just build an AC30 tube amp (or derivative with top boost pre and push/pull output stage) than to fool with a complex SS circuit that may or may not get you there. "Emulations"... <sigh...>

[Seljer]

The Moeller thing is interesting but way too complex for what it "emulates". For me it would be much simpler to just build an AC30 tube amp (or derivative with top boost pre and push/pull output stage) than to fool with a complex SS circuit that may or may not get you there. "Emulations"... <sigh...>

most of us are afraid of high voltage tube amps

[Rafa]

I built the Marshall (LXH2) sim two 9 opamps, great, amazing undescribable, wonderful, gorgeous...............
And right now Im gogin out of my house to buy the parts to build the stephen mioeller one
Cheeers
Rafa

[MartyMart]

But I've campaigned against the use of the word "emulation" from day one, <sigh>...
"Emulations"... <sigh...>

I'm with you on the valve amp track, having just built a small 5 watter that sounds "Superb"
However, the Bassman '59 and PV Classic 30 "sims" ! that I did , got very, very close to the tone
as compared to originals of both !  ( Bassman owner was gobsmacked ! )

Probably the "analog" sims are better than the "Digital" ones ...... POD anyone !!   

MM

[Xavier]

I guess I'm "one of those folks" that happen to like "emulators"

Really, I know what a tube amp is. I know these circuits (for the lack of a better term to describe them ) will never make you sound like a tube amp. I see them as nice sounding distortion pedals, which up to a certain extent have a similar kind of tone to some amps. The Dr. Boogey has a rectifier-ish quality to it that's for sure. I see them just as an extension of the other "normal" dirt boxes like a Rat or a DS1. I would never make a preamp out of a Thunderchief or a cauldron of Gain....

If I look back, 1 year ago I was holding a soldering iron in my hands for the first time, drilled a hammond box also for the first time. These "JFET circuits" are very helpful for some noobs like me because their archirtecture is modular, and it's easy to understand how they work, hence the interest. I'd love being able to design more complex circuits including hex inverters and opamps, however i'm not at that level yet.

We discover new ideas every day in this forum. You guys need to understand our excitement. I think this is the most informative forum I've ever seen. I've learned more about electronics in this forum in one year, than in all my life until now.

Thanks for all the help, in my book you guys rule.

[phaeton]

I've sort of noticed that circuits I build are more 'predictable' based on the filtering than one what actually 'powers' them, especially if you use diodes for clipping.  In fact, the more I learn about stuff, the more predictable things get, and the more that the magic and mystery goes away.

As far as "emulation", I'm useless.  I'm always trying to invent something new, to a fault. In vain.



Oh yeah.  I'm not ready to work on tube stuff either.  I'm not even sure if my DVOM can handle the voltages or if it will just go up in $300 smoke.

[WGTP]

What would be a better word to use???

I'm curious about the characteristics the stages with the negative feed-back and more complicated biasing add, that is sort of unique to these type circuits, but missing from other multi-stage distortions like the Obsidian or Shakka Express.   

[Mark Hammer]

The thing to remember about "emulators" is that:

• a great many desirable amps are essentially textbook, appnotes-based applications of a couple of fairly standard and similar tube circuits; differences between their front end are often (though certainly not always) simply a matter the various permutations and combinations of gain-stage, tonestack, volume control.
• a huge part of the "magic" of any tube amp lies in the power supply, and how tubes respond to variations in that; highly doubtful that 9v would reliably mimic any and every amp's response just by swapping FETs for triodes.
• another huge part of tube sound is the output transformer and any negative feedback applied from the speaker side of the transformer; I have yet to see any amp emulators using transformers in creative ways
• a third huge part of any amp is the speakers; many amps from the "golden era" were electronically identical to other big name amps based on the same tube manual circuits, except they were stuck with crappy speakers in crappy cabs, often with crappy output transformers
• everyone here tends to use them as either a pedal going into a front end, or if fed directly to a power amp, that "power amp" is generally an LM386 rather than a long-tailed phase-splitter and a pair of 6550's or EL84's; it is rare that anyone uses them "as if" it were the front end of an entire gigging amp

My intent is not to diss emulators or op-amp vs FET emulators, merely to note that there is a whole lot missing from the equation.  Next time you want a romantic evening, throw a pair of high heels, a bottle of perfume, some naughty undergarments, and two small silicon-filled pouches in a bag and try to persuade yourself you're with the woman of your dreams.  Doesn't quite feel complete, eh?

I have no doubts that op-amp circuits could accomplish all one wishes to accomplish with FET-for-triode substitutions.  Certainly hasn't stopped Sans Amp, has it?  On the other hand, those simple FET-for-triode (let's just call 'em "F4T") circuits make it easy to start thinking about how the sequencing of the basic building blocks can result in different sounds.  In other words, they are great to think with, and the F4T approach has lent clarity to that thinking.  In some ways, they are also pleasingly uncomplicated, or at least we haven't been tempted to introduce complications in the same way we might with op-amps.  I'm actually quite glad that slew-rate, voltage-swing, input impedance, and all the other op-amp properties have NOT entered into the fray.  To the extent that this group tends to think about FETs in a rather monolithic way (i.e., a J201 is a J201 is a J201, the way a 12AX7 is a 12AX7 is a 12AX7), that has been very helpful at this stage in the game.

Happily, as Marty and Xavier have pointed out, you get outcomes that are either remarkably similar in many sonic properties, or else if not same-sounding then good-sounding. 

[MartyMart]

Jfet emu's have a major "flaw" which is being almost zero percent repeatable with even exact
components , due to the differences in the fets.
OK for a "DIY" one off, but if you were to go commercial !!  
But then you can have someone check all the fets and match them ... by the thousand !
I love the OD/Dist from Jfet's , they dont seem to "Fart" like so many diode clipping circuits do, on the
decay stage of chords etc ....

MM.

[Jay Doyle]

What would be a better word to use???

Distortion circuit? 

I think the major flaw is that you don't have anything in there replicating the pentode power amp section or the output transformer.

Plus, when put in front of a tube amp, these circuits essentially become pre-pre-amps and are still contoured by the used amps' tone stack as well.

And I REALLY do not like the trim-on-the-drain biasing approach.

But that is just me. Caveat Emptor.

[Doug_H]

I see them as nice sounding distortion pedals, which up to a certain extent have a similar kind of tone to some amps.

I agree 100%.

My problem with the word "emulation" is that it always invites comparison to "the real thing". As good as some of these pedals sound, they will always fail an A/B comparison to "the real thing" IMO. I know we know what we mean by "emulation" in here. Although I don't want to wordsmith this to death, it would be nice to have a more accurate word to describe the original intent- that is to use some ideas found in amps in pedals.

Personally I've gotten away from the jfet stuff because I think they are more complex than need be to build a "good sounding" pedal. Some of my old favorites are still the one or two stage boosters & fuzzes. Some of the simple stuff I've built I like as well or better than the stuff that requires stringing 4 or 5 jfet stgs together. The jfet circuits are a fun learning experience though, I'll grant you that.

[mojotron]

It seems to be semi-irresistible for folks to take an amplifier schematic, delete the triodes and replace them with JFETs, diddle the biasing and then call it an XXXX emulator.

We know that JFETs and triodes do not distort alike, but the process of emulating an amp by rote seems to be satisfying to people.

I have this suspicion that the frequency determining elements filtering the distortion is what is so attractive, not the distortion. That being the case, I suspect that you could sub in for instance opamp distortion stages for the triodes/JFETs and have just as interesting SOUNDING "emulation", albeit not with as big an emotional hook.

Anyone ever tried this, other than Stephen Moeller?

The word 'emulator' or 'sim' should not be applied to these circuits... I like the term 'adaptation...'

So R.G. - have you built one?

They are great pedals - compared to the everyday stuff like TS, DS1..... diode clippers... Sure, in a 'purest' sense - these calling these emulators is like calling my kids' finger painting an emulation of the impressionists, but the better ones, of these pedals, put a crappy amp well into the ball park of a given sound in a lot of cases.

[ubersam]

... I like the term 'adaptation...'

I was thinking along the same lines, a solid state adaptation of tube circuitry.

[petemoore]

  All Tube amps are experiments. Many designs, some of them good, were abandon for one reason or another. Others have stood the test of time and may have benefitted from field application feedback guiding alterations seen to be improvements to an original experiment.
 Control field tests are implicitly inconclusive because one to the next off the line [for the most part] the amps will exhibit differences, to one extent or the other, strict control environments for these tests are rare compared to the wildly diverse [most often not subcontexed because of the impossible complexity/non-congruity of subtexting wildly varying control elements] control elements in real world, like speaker response, pickup height, string thickness...TONS of etcetera that has been rehashed odd infinitum...
 I'd get the Fet's on the board and basically biasing, with 'larger' value source resistors/'smaller value caps' [or the inverse...you can decide whether to multiply or divide to tune values]...on all stages.
 Stick Q's in sockets and bias, adjust Q choice or source R or Drain R to get a bias on it.
 Then start at the front and put higher gain Q or lower R source [mess with bypass cap if any], rebias drain...see if you like the gain structure to be high from Q's 1-? [how many?].
 Reset that to 'normal', then try increasing the gain of later stages.
 At this point you probably have voicing wants, a good time to try some filtering alterations on the SP...TC yet?
 Just start with something close to something that works, then diddle with that experiment 'till you don't like these suggestions anymore or you see my point!
 I had a MkII Master Lead, I thought that's what it was and how it should work...because ____ ...until...
 I took that spike in the head treble content and tamed it while making ChI volume usable...
 I had a Plexi didn't work at all like Bills Plexi.
 Jfet's are different than tubes, I'd be moslty worried about finding 'it' in a Jfet circuit than...
 My 2c+...messin' about with those source R/C values and voicing/gain structuring might be more fun/easier to bias/get to breaking over to distortion more to 'your' liking that sticking to the part of the schematic between tubes and ground...Drain bias V and R sure seems to want to and get to go all over the place...lol...
 
 

[petemoore]

They are great pedals - compared to the everyday stuff like TS, DS1..... diode clippers.
  They can be !! They do well to be all figured or fiddled with IME.
  What's THiS Jfet Going to do? What does that do to the other JfetS? What's response is my amp capable of with all this going on?

[aron]

Jfet emu's have a major "flaw" which is being almost zero percent repeatable with even exact
components , due to the differences in the fets.

Has anyone done a test on these? From my limited building, I have found the J201 batches to bias pretty darn close lately.

[mojotron]

...
Next time you want a romantic evening, throw a pair of high heels, a bottle of perfume, some naughty undergarments, and two small silicon-filled pouches in a bag and try to persuade yourself you're with the woman of your dreams.  Doesn't quite feel complete, eh?
...

Well Mark, I'll just have to try that out this weekend

[mojotron]

Jfet emu's have a major "flaw" which is being almost zero percent repeatable with even exact
components , due to the differences in the fets.

Has anyone done a test on these? From my limited building, I have found the J201 batches to bias pretty darn close lately.

Yep, not that I build a lot these days, but I have found that the Fairchild J201 parts do have some level of consistency with the last few batches I got from them.

Don't worry folks, I'm sure next year we'll be building BSIABIIs with 12AU7s and cd4007s in place of JFETs. 

Seriously though, at some point a few of us will have a good idea of what to do with stuff like this, do some actual design with that idea and come up with something really great. I'm pretty jazzed about that.