A Complete Noob's Attempt at a Fetzer-ized Boogie Preamp

[karis12]

I need you guys to tell me how badly I'm doing at this DIY thing.

First off, here it is:



The back story:

I'm really, really desperate to either buy or create a pedal/circuit design that's aimed at recreating the Boogie Mark series distortion sound, which is a combination of many cascaded gain stages (as many as 6), the traditional F-style TMB tone stack acting as a guitar "shape" control (in other words, what frequencies from the guitar end up getting boosted to hell by the succeeding gain stages), and finally the 5-band graphic EQ that does the actual EQ-ing.

I've learned enough commands on LTSpice to come up with virtual emulations of the Mark V, based on schematics floating around the interwebz, and while I know that Mesa intentionally includes errors in them to prevent counterfeiting, I did come across something that's supposedly verified. I put it together in Spice, and entered values corresponding to my favourite guitarist's settings based on a product demo video (hint: his initials are J.P.), and ran some virtual tests on it. I found out he sets the TMB in such a way that it acts as a treble booster going into the succeeding gain stages, which contributes to the characteristic sound of said guitarist.

I was curious enough that I also pulled up a schematic of my personal amp (a British design with the TMB stack near the end) and simulated it in LTSpice with settings designed to get it as close as possible to the virtual Mark preamp. I also added a treble booster simulation in front of it to mimic the EQ curve of the Mark's TMB "shape" controls at the aforementioned settings. It lined up pretty close, which encouraged me to build a physical op amp treble boost based on the simulation. This treble-boost-into-driven-amp method was the closest I got to the sound so far.

Going further, I took the revised Fetzer valve design (more or less), plugged 6 of them into the basic schematic in place of the 12AX7 simulations, and tweaked some values to get a similar EQ and phase response from the circuit. The photo above is what I came up with.

Notes: the drain and source resistors were calculated using the RunOffGroove method, based on the values that show up when plugging the J201 models into a virtual testing circuit. Also, it would probably be immediately obvious just how little I know about circuit designs, so I'll have to apologise in advance for the mistakes and redundancies.

I'm pretty sure I'll need to find a way to reduce the stages to 4, since I'll be plugging this into standard guitar amp inputs. The typical guitar amp's TMB controls with the natural mid scoop shall serve as this pedal's "Graphic EQ," with the pedal TMB controls again serving the purpose of "shape" control.

TL; DR please roast me 

[duck_arse]

are you at all familiar with the Dr Boogie circuit?

you probably want to drive the tonestack from the output of a source follower, or similar. also, your J3 has no gate bias. R24 seems to be shorted. R33//R34//R44//R35 could probably reduce to two resistors.

[karis12]

are you at all familiar with the Dr Boogie circuit?

I am, and as far as I know, it's meant to emulate the sound of a Dual Rectifier, which is a completely different topology than the Mark series.

The tone stack is placed early in the circuit to serve as a signal "shaper" (at least in the real version), dictating which frequencies from the first gain stage gets to the succeeding ones. I'm sure it'll bring the entire signal down a few dB's, but I think, basing from the design I have with me, it'll more or less work the way I want it to.

R24, as well as the values of C1, is supposed to be switchable to allow a MkIV and a IIC+ mode, but since I haven't even gotten to breadboarding this yet, I wouldn't know if it's worth it.

You seem to be right about the J3 gate bias, I'll see what adding a 100k resistor to ground does to it.

[287m]

noob ask, not to hijack thread

just out of curiosity
to build something like this badass boogie, must be jfet based circuit?
can't be opamp based? why?

[robthequiet]

My humble suggestion would be to build the circuit without the MK2C+/MKIV switch, as the difference may not be as dramatic as simply tweaking your tone shaping for optimal sound and use the tonestack to tweak. A "bright" switch and a "boost" switch could be useful. I have a MKIIB that has so many switches for this and that but ultimately I run it with the graphic eq disabled and clean channel only with most switches "off". The fundamental sound is so good you get most of your control from the guitar. In a pedal chain, if necessary, add a boost/drive pedal (TS or KLON types) at the input for more dramatic shift into the overdrive. The ROG Thunderbird was crafted after a specific Marshall amp and they have a good writeup on the project worth reading.

[BubbaFet]

To bring a design from the metaphysical, into the physical, involves a border crossing, and as we all know, there is always trouble at the border  .

I applaud your intent, and encourage you to build and test, one stage at a time. Ask questions as you go. You will learn more than you possibly can imagine.

Fets are fabulous, IMO. You might want to change from Fetzer Valves to Tillman style. Then you can Boogie Till' the cows come home  .

[ashcat_lt]

noob ask, not to hijack thread

just out of curiosity
to build something like this badass boogie, must be jfet based circuit?
can't be opamp based? why?

You could get close to the same sound.  The SansAmp is opamp based, and it has a Mesa mode.  I think that's also more like the Dual Rectifier.

You cannot, though, just drop an opamp into a schematic in place of a tube the way you can with a transistor.  I think that's part of the point here.  To do it with opamps you'd have to completely redesign the circuit.  It wouldn't actually work anywhere near the same, even if it did sound really close.

[PRR]

J3 sure needs a gate-to-ground resistor.

[karis12]

Added a 100k resistor to J3 gate, so that should sort it out.

My greatest challenge is figuring out how much saturation I want, and how much I ought to expect per JFET stage. I know I don't want to plug this many stages into an amp input because it's pretty much a downscaled version of plugging a distorted amp's effects loop send jack into a clean amp's guitar input jack. In other words, tons of unusable gain, and mucho feedback. Is there any way to reliably predict this through math, or should I just breadboard/prototype it and hope for the best?

I just remembered I wanted to add to it an AMZ tone control based on the Muff seesaw, but with adjustable mids. That'll require a source follower JFET to use properly, correct?

[anotherjim]

As I understand it...
A source follower only gives the tone network a reasonably low source impedance. The network can still be made to work with a high source but it screws any calculated values you may have for the R's & C's.
If you directly copy the network values from somewhere, you should also copy the source & load resistances.

I'm curious how resistor values like 11929 came about. You know it's 12k right?

[karis12]

As I understand it...
A source follower only gives the tone network a reasonably low source impedance. The network can still be made to work with a high source but it screws any calculated values you may have for the R's & C's.
If you directly copy the network values from somewhere, you should also copy the source & load resistances.

I'm curious how resistor values like 11929 came about. You know it's 12k right?

It's set to be as close to 4.5v to the simulated JFET, and it's the same as the source resistor not being a preferred value.

[jubal81]

As I understand it...
A source follower only gives the tone network a reasonably low source impedance. The network can still be made to work with a high source but it screws any calculated values you may have for the R's & C's.
If you directly copy the network values from somewhere, you should also copy the source & load resistances.

I'm curious how resistor values like 11929 came about. You know it's 12k right?

It's set to be as close to 4.5v to the simulated JFET, and it's the same as the source resistor not being a preferred value.

I suggest multiplying your supply voltage by .6 and using that as the drain bias voltage target.

[BubbaFet]

Real world JFETs usually have quite different parameters than the Spice model. So do not get too invested in getting everything perfect in Spice. Start building your first stage. You might discover that you want to use a different JFET, use different R-C combinations, etc...
JFETs are like people.... similar,  but different from one another.

[KarenColumbo]

I wonder if you'd like to have a look at http://www.runoffgroove.com. Those guys have a thing for emulating classic amps with 20 parts and a budget. Sounds good, too. One of my favorite sites out there

[karis12]

I wonder if you'd like to have a look at http://www.runoffgroove.com. Those guys have a thing for emulating classic amps with 20 parts and a budget. Sounds good, too. One of my favorite sites out there

I actually took my Fetzer stages from here. The only difference is that now, I'm attempting to use the revised version instead of the old one. The closest to my current attempt is the Uno circuit. I may have to check it out and see if there's anything in there of use to me. Thanks for chiming in.

[Agung Kurniawan]

noob ask, not to hijack thread

just out of curiosity
to build something like this badass boogie, must be jfet based circuit?
can't be opamp based? why?

You know Dream Theater?
Their guitarist John Petrucci use Mesa Boogie Mark:V. And he love the MkIIC+ Mode. I love that mode too
I have once design some distortion pedal using opamp that sound the same as MkIIC+ mode do. But I didnt get the high end punchy sound that MkIIC+ have. If you want you can experimenting with the gain stage, clipping diode and some band-pass filter.

[karis12]

noob ask, not to hijack thread

just out of curiosity
to build something like this badass boogie, must be jfet based circuit?
can't be opamp based? why?

You know Dream Theater?
Their guitarist John Petrucci use Mesa Boogie Mark:V. And he love the MkIIC+ Mode. I love that mode too
I have once design some distortion pedal using opamp that sound the same as MkIIC+ mode do. But I didnt get the high end punchy sound that MkIIC+ have. If you want you can experimenting with the gain stage, clipping diode and some band-pass filter.

Yes, John Petrucci's sound is what I'm chasing after with this little project.

[Agung Kurniawan]

I put it together in Spice, and entered values corresponding to my favourite guitarist's settings based on a product demo video (hint: his initials are J.P.)

Ah, I didnt read that.
Well, I'll be waiting for this project. Since your project would be sound very close to the original Mark:V

[karis12]

UPDATE:

I found the values that sort of work for what I want, and here is my updated circuit:



At least according to my personal tests, and some simulating on the SPICE program, the bypass capacitors on each JFET's source pin will work for the general resistor values typically used on J201/J202 transistors for a FETzer valve-type configuration (820-1.5k). It introduces some gain boost and a slight distortion that helps achieve the amount of saturation I want.

I built the first two stages + tone stack into a separate pedal, because I want to try it out on my tube amp to see if it gets me any closer to my desired boosted sound. The latter stages I decided to breadboard first.

As I mentioned in my original post, the EQ will be controlled by the amp it plugs into (for now). I'll try a Big Muff style control in the future and see how far that can take me.

With this setup, I can achieve a sound that is reasonably close to the saturation of a Mesa Mark-type amp. The sound of palm-muting, as well as full chords, is excellent, even. The problem is the sound of the pick attack when it comes to single-notes.

I'm not sure if I'm pushing the FETs too hard, or if it's an inherent characteristic of the FETzer valve, but I do get a similar response with my favourite runoffgroove.com build, the Thor, every time I try to boost it with an overdrive. Again, excellent saturation, palm-mutes, and full chords, but the pick attack sounds strange. The resonant frequency is too high to sound convincingly like the amp it's attempting to mimic, and ends up sounding "fake," for lack of a better term.

Hopefully, members of the forum who are also involved with ROG can help me with this. Alternatively, can anyone suggest alternative circuit designs that can also help me achieve a similar amount of saturation and sustain without this pick attack issue?

Thanks in advance.