Big Muff versions - What really matters...

[Projectile]

I just spent the  whole weekend tinkering around with a Big Muff on a breadboard. My goal: to find out what the real difference is between all of this different versions floating around.

This was my first time with the Muff circuit. When I first decided to built it, I experience what most first time muff builders probably experience: The feeling of being completely overwhelmed with all of the different schematics! I found lots of information about mods and how the circuit works, but very little definitive info on what really makes a Green Russian a Green Russian, or a Triangle a Triangle, etc. This is not surprising, since there can be half a dozen part differences even within the same version. So, what does it all mean? That's what I was determined to find out.

I collected every single schematic of the transistor Big Muff Pi I could find and then set about swapping out every single component that differed between versions, putting them on switches and carefully flipping back and forth between them listening for subtle tonal changes. Where it seemed necessary (when components interacted), I sometimes put two of three components on one switch, so that I could hear the differences between swapping out all of them at once. After hours and hours of testing, this is what I found...

-------

Most of the parts changes between different versions make little to no difference in the sound of the pedal. The differences in most of these parts just lead to very small changes in gain. They did not really effect the overall tonal quality of the pedal. Any difference in sound was very subtle for most values, so unless you are trying to build a perfect replica of a 1974 Rams Head blah blah blah... it really doesn't matter. If you are more interested in just tuning in a great sounding Muff, then you can forget about 75% of the stuff that differs between schematics. It really does not make enough of a difference to worry about.

The transistors don't really matter whole lot either. I didn't have a lot to choose from, but the ones I did test sounded only very subtly different. And I really do mean "subtle." I can hear the difference between a JRC4558 and a TL072 op-amp like night and day. This was not nearly as big of a difference as that. Unless someone can recommend a transistor I should try that really makes a difference, I'd be just as happy with standard 2n5088s as anything else in a Muff. I wouldn't spend the cash on any exotic NOS transistors  in this case. It just doesn't seem to matter much.

So, what really does matter?

There are two main parts of the circuit that determine the sound of a muff: the gain section, and the tone stack. I'm going to talk about each individually.

For the gain section, the parts you should really focus on to sculpt the sound of the Muff are the capacitors in the feedback loop of the transistors. On the GGG schematic, these are C2, C5, C6, C8, and C9. Apart from the tone stack, that's really all that makes most of the difference between the different versions. Changing these 5 caps will get you 90% of the way from a triangle to a rams a head to a Russian to a modern reissue. If you want to know what values to experiment with, those are the key. After hours upon hours of experimenting, I'm convinced that variations these caps alone account for the vast majority of what people hear to be the difference between different muffs.

Of these caps, C2, C5, and C8 are the easiest to make a choice about. Their effect is fairly minimal, but noticeable. Higher values (560pf) roll off more highs in the clipping stages and therefore make the pedal sound warmer and more "vintage." These are the values of the caps for the Triangle, Russian, and some of the early Rams Heads. To get a more modern sound that "cuts" better, use lower values (470pf) for these caps. This is the value most of the Rams Heads and later versions use. They sound more "fizzy."

The caps that overall make the biggest (and most complex) difference are C6 and C9, the caps in series with the clipping diodes. Not only do these caps affect what frequencies get clipped, they also dramatically effect the bass response and overall gain of the pedal. These caps are .1uf on reissues and many rams heads, 1uf for the second cap on some rams heads, and .05uf on the Russians and the triangle versions (and possibly some rams heads). The biggest difference in changing these values is the bass response. Using .05uf caps here lets through quite a bit more bass and makes the pedal sound much more "woolly". I think this is why a lot of people swear by the Triangle versions and why the Russian versions are so highly regarded for bass players. It mostly just comes down to these two caps. The important thing about changing these caps is that they are highly interactive, so while changing one will produce a subtle difference in tonality, changing both makes a BIG difference. If you want a more bassy sound, use .05uf here. If you want a more controlled sound use .1uf.

The rest of the sound of the muff mostly comes down to differences in the tone stack. The major variations between versions mostly boils down to R18 and R19 (GGG schematic). All of the versions have the pronounced mid scoop characteristic of the big muff, but with some subtle differences. I recommend spending some time tweaking this section because what seems like small changes can become more noticeable after getting used to the sweep and playing with it for a while. Most Muffs use 22k for R18 and 39k for R19. This produces a the most dramatic scoop. My favorite is the Triangle tone stack which swaps R39k for R18. It has a little bit less of a dramatic "scoop", but still enough that it sounds like a Muff. The Green Russian version is a bit of a different variation. It uses 22k for R18 and 20k for R19, which boosts the bass up a little, flattens the scoop, and moves the scoop up higher in frequency. It makes sense that a lot of bass players would prefer this version. Another version which has a similar curve is the GGG tuned Big Muff (same as a late 70s version). This curve also boosts the bass a bit but flattens out the curve even more. I didn't really care for the GGG tuned version because it made the Muff start to sound more like a typical distortion and less like a Muff, but some people swear by it. It's a matter of taste. I recommend socketing these two resistors so that you can experiment with values.

So that's basically it. If you want to experiment with different versions of the Muff sound, all you really have to swap are [C2,C5,C6] (fizz) and [C8,C9] (bass response), and [R18, R19] (scoop). That covers most of the tonal range from a Triangle to a Rams Head, to Russian, to modern reissue.

Other mods:

I tried a bunch of mods, but the only one that really stood out was removing the diodes (and associated cap) from the feedback loop of Q2 or Q3. This is an awesome mod that really changes the tonal character of the pedal. Removing the first set of diodes basically turns the Muff into a Colorsound Supa Fuzz, which is awesome! It completely changes the tonal character of the pedal in such a cool way. It's so easy to just put it on a switch that I think it would be silly not to include it if you were building of modding the pedal. Highly recommended!

Removing the second set of diodes produces a less usable sound, but it's still pretty cool. It's kind of a low-fi, blown speaker kind of sound, which I personally like, but this kind of thing is largely a matter of taste. I definitely recommend trying it out, at least.

Also, note that when removing the first set of diodes, the cap in series with diodes in the second loop (C9) now becomes critically important for shaping the bass response of the pedal. I found that the more vintage value of .05uf here let through too much bass. This mod sounds much better with the standard .1uf or even .22uf for C9 in my opinion.

Hope this information helps people out there. This is the kind guide I wanted when I started looking into Muffs, so I thought I'd share my findings with everyone. Happy building.

[guitarify]

Thanks for posting this.
I've yet to build a Big Muff exactly for the reasons you mentioned. It just seemed like a daunting task to look through all the versions and figure out what to try.

[Earthscum]

+1 Good post... many of the things I've found, as well.

I do have to argue about the transistors. I did notice a difference in the clipping sections by trying low and high gain transistors. Other than gain relation, these don't seem to make much difference in sound to me, either.

However, I did notice improved bass 'clarity' by using MPSA13 in the first stage with stock values. This  is the same difference I noticed in the Bazz Fuss as well. All the other transistors I tried, again, too subtle to matter much.

And, coming in last... err, in the last stage, I've found this transistor to be the one that makes the most difference in tranny types. This stage doesn't get masked by clipping down the line. 2N3904's and 2N5088's sounded the same. BC109, on the other hand, spiced the tone with, well, BC sound... to me, they have a rattle-can kind of sound to them. The MPSA13 let a bit more bass clarity through (not  more bass, just more apparent and clear). 2SC1815Y's sound a bit softer than the 2N's, yet still fairly clear. The more I use them, the more I like the Japanese transistors for softening things up, but not to the extreme that Germanium does... nice and subtle, just enough to ward off ear fatigue during long blasts.

I love the BMP circuit... lots of different flavors just by swapping out a cap or two. I tried all sorts of things with the clipping stages...one I remember liking was heavy filtering (1n) in the first stage feedback, and a larger cap with the diodes, then squeezing the highs back into action in the second stage by using 150p cap and  a .047u cap with the diodes (I think I got that right. Diode caps may be backwards, though).

[radio]

Hello

Thanks for your insight

On my BM,I do have 2 switch for the 2 BTB diodes.

With only the last diode pair , you can get pretty close to older

Steve Hackett sounds.  I quit like the "diode-less" muff too.

Still have to try to make the feedbackcaps switchable.

I however made the 2 caps between stages switchable

100n to 10n ,the difference is a slightly clearer tone.

Regards JM

[Projectile]

Thanks for the insight Earthscum. I'll admit, I didn't mess around with transistors as much as other parts, and everything I tried was high gain  (that's what others said sounded good in Muffs). I do have a bc109 lying around, so I'll try that one in the last stage like you suggested. I'm not sure I could afford to do the whole pedal in bc109s though. Those are getting a bit pricey. Do you know of any cheaper alternatives?

I'll see if I can dig up an mpsa13 as well, and I'll also try that cap combination while I still have it on the board. The muff is a really great circuit to mess with.  There are so many great tones that can be gotten out of it. I could see the need for building more than one Muff.

[Projectile]

double post

[Hides-His-Eyes]

Great thread, thanks. I dived in blind once with a tonepad pcb and no knowledge of what I wanted and wasn't very impressed with the result; perhaps with this and another PCB at some point I'll end up with the muff I'm looking for (even if I have to socket a couple of caps!)

[WhenBoredomPeaks]

Once i started to make a "custom" Muff, i shooted for triangle spec but as i found an article about mods i changed some components while soldering and substituted some values where i didn't had the original values.

Long story short, i went with 270pF for C6 and C9 and it sounded kinda horrible but when i changed them back to 560pF it sounded like a Muff should.

Also i used sockets for the diodes and tried some stuff in there, the Germanium sound is definitely present with Ge transistors but overall i liked the stock diodes the best.

[dthurstan]

Great post. I've amm & ahh about building a muff. This post gives you some straight answers to whats what.

About transistors, I have read that the circuit is bias so that the gain of the transistor isn't important. I've been reading the Art of Electronics by Paul Horowitz and Winfield Hill, which mentions that in a "good (well designed) circuit" doesn't rely on a specific Hfe. This immediately made me think of a Fuzzface & Big Muff. ff's need a certain Hfe to sound right where Muffs don't because of biasing. However, Earthscum does notice a difference mainly with the last stage (which looks like a LPB to me).

I was wondering if you changed C1, C4, C7, C12 & C13 (GGG schm). I would have thought changing these coupling caps would also change the bass responce?

[alex frias]

I've changed the first 3 coupling caps, to half the value, and my Muff sounded more articulated with less "mushness". Using BC550 left me with less white noise.

[tuckster]

Hi Projectile,

thank you for this nice read!!!!!!
I built the different muffs last week on a breadboard too.
The pictures and pdfs on http://www.kitrae.net/music/music_big_muff.html are very nice for this.

I stick with a ram's head with three switchable clipping stages and different diods, transistors and caps.
1st stage = nice fuzz
2nd stage = regular muff....
3rd stage = heads down!

Add the tiny tremolo with a momentary switch and you have a very cool box

[Gila_Crisis]

really an interesting post about the BM!
what you share about the circuit it's just the same feelings i have since a lot! those caps in the gain stages and the tone control are the effective things that affect the BM-Sound!
so that i built "My" BM with 2 switch: one changes the resistors on the tone control between russian/triangle values (R18,R19), and the other the caps in series with the diodes and in/out caps (C1,C6, C9,C13).

it looks like this:


C2, C5 and C8 are 560pF, which i like a lot, and the rest its a mix between triangle (collectors side on the gain stages), ram's head (emitter sides on the gain stages) and russian (Q4 stage)

and just in case this is my BM-lounge:
http://www.aronnelson.com/gallery/main.php/v/Gila_Crisis/Big_Muff/

[caspercody]

Thanks for doing this, and posting your findings!!

[maarten]

congrats on your report! It is compact, yet seems to be accurate and relevant. Indeed good enough for me to store. Wish we had more reports
like these.

Maarten

[tuckster]

This is very usefull for tinkering around with caps, diods and so on:



Right now I only have one for the film caps (12pin rotrary switch) but I'm going to build one for diods too.
For resistors it's easier to buy a decade resitorbox because it's kind of annoying to diy except you only use a few values.

Also give this doc a try:
http://www.beavisaudio.com/bboard/projects/bbp_DiodeMods.pdf
Especially Jack Orman's Warp Control did a nice job in a muff with one clipping stage.

[Hides-His-Eyes]

For resistors it's easier to use a pot. 

[tuckster]

For resistors it's easier to use a pot. 

yes but what if you don't want to check the pot with a multimeter everytime you found a desired value. these things were "invented" for a reason.

[Electron Tornado]

This thread might help, "Technology of the Big Muff - What Does What?":

http://www.diystompboxes.com/smfforum/index.php?topic=78111.0

[jrod]

Thanks, projectile!!!

This is some great information. I am in the process of building several Muffs, but had to move. Once I get my work space back up I will certainly use this thread as a guide!

[edvard]

Awesome post!!

I too, have been wondering what exactly makes for the differences when experience tells me there just isn't that much to it.

I agree with Earthscum that the BC- transistors DO have a different sound, however subtle.
My favorites are the BC239/309's and someone else pointed out that bandwidth on those drops dramatically at very low I_c.

As I get older, I've noticed I've gotten MORE sensitive to high frequencies, rather than less so the "trashier" something sounds, the less I like it.
Thanks for the tips, I'll be bookmarking this thread...