yet another OS MUTANTES thread

[pinkjimiphoton]

i had to get that out of the way...
always loved the os mutantes fuzz sound, so i found this:



which of course, meant i had to build one. so i did yesterday.



other than a couple things in the BOM beneath being mislabeled, the layout itself is built and verified.

used next closest values in a couple spots, 3.9n for 3n, 2.2k for 2.5k , had to make a 3.3m resistor with a 2.2 and a 1m in series, used a 10k trimmer. built initially with bc547's, tried a gamut of silicon, and the fuzz part was great, but the treble boost part shrill and unmusical. i tried it as a hybrid, sounds great with bc109/108 pair, but in that case needs a cap to bleed off the shrill high end. there's quite a bit of tone for such a simple circuit... but i didn't want to deal with that unmusical treble, a shrieky kinda "presence" like on a marshall jcm800.. so i went germanium, with some vintage npn's a freind salvaged from an old thomas organ. that did it. brought the whole thing to life. so... sockets are your friends. i made this so you can use just one piece of SIP socket for both transistors. i used my iron to poke out the one between them just to make it idiot proof.
gotta wait to check the hfe's, my meter is down in the basement with another project.
also added simple standard power supply filtering, a couple diodes, a current limiting resistor and a bleeder and filter cap. since the unit already has a 47u filter cap on board, it's very quiet. i will add a 100u cap right across the power rails at the power jack too, probably overkill but i like fuzz pedals to be quiet as possible when not making some raucous @#$%in' noise.
how does it sound? really good. different from almost every other fuzz i've built, with two distinct personalities between the filtered and fuzzed setting. one knob. volume. that's it. perfect for kamikazee studio here.
the treble boost side is vaguely klon centaurish tonally, with a hint of nasal mid sounding very much like an old marshall right on the breaking point, the fuzz side (bypassing the small "filter" cap) is big and wooly like a big muff and has a real nice sustain with a pronounced buzz to it. at least into an overdriven amp.
as with virtually all fuzz, it sounds like poo without hitting a somewhat distorted amp. but add it to a cranked up tube amp... and .... yum.
stupid pedal trick coming soon, for now tho, here's some os mutantes:

if you want the exact kinda sound in this video, use bc108 and 109. socket. experiment. you'll find it.

[Gus]

Jimi

I am not a fan of how the transistors are biased.  TAOE has a good page about why not to bias this way.

If you want to do a simple test note the temperature in the room and the collector voltages when you had it sounding good, do this with a fresh battery
Then heat the circuit and measure the collector voltages
Cool the circuit and measure the collector voltages
try a range of 60 F to 80 F to start then what you might measure ona sunny summer day on stage oputdoors
Do you have one of the infrared temperature meters you can sometime find for under $20.00 from HF?  You can use them to measure the transistors without contacting them.

You might want to use low gain Si transistor or even med power Si transistors

[LucifersTrip]

but the treble boost part shrill and unmusical

but i didn't want to deal with that unmusical treble, a shrieky kinda "presence"

sounds cool that you got this one how you like it...

most likely, the reason you're getting the harsh highs is that your schematic is missing C5 (which shouldn't be .1, but much smaller)

[pinkjimiphoton]

Jimi

I am not a fan of how the transistors are biased.  TAOE has a good page about why not to bias this way.

the grounded emitter bias you mean gus? or am i missing something... what is TAOE?
i like grounded emitter circuits in some cases... really depends on the transistor, i've found.

If you want to do a simple test note the temperature in the room and the collector voltages when you had it sounding good, do this with a fresh battery
Then heat the circuit and measure the collector voltages
Cool the circuit and measure the collector voltages
try a range of 60 F to 80 F to start then what you might measure ona sunny summer day on stage oputdoors
Do you have one of the infrared temperature meters you can sometime find for under $20.00 from HF?  You can use them to measure the transistors without contacting them.

my meter has a thermometer function with a little sensor to it, i'll give it a shot.
so far, i've found the 5 k (10 k on mine) bias trimmer doesn't seem to matter a whole lot tonally... the lower you set it, the louder it gets. i may just swap it to a 10k pot so i can tweak it externally if it acts up. thanks bro!

You might want to use low gain Si transistor or even med power Si transistors

i tried pretty much every si i had in the house... some sounded great, some were ridiculously shrill, or had a "farty" kinda undertone on the bass boosted setting.
these cheezy ge's are supremely low gain... q1 is hfe 16, q2 hfe 36 both with about 9ma of leakage. these crappy low gain ge's really brought the circuit to life!!
i tried with some medium gain bc547's, 5088-89, 3904, bc 108-109, even some fets and mostfets. the best sound was the ge's, and the lower the gain, the better it sounded!! weird little circuit.
i wanted to build it pretty much as it was in the schematic... i see dave posted a somewhat different one.
one thing i can say tho, is it absolutely nails the os mutantes tone with the si in there... it's a bit too brite for my taste, but i love how it cleans up.

[pinkjimiphoton]

thanks for the headsup dave.... i will try seeing if maybe tacking a cap in there will help. i'll play around some and report back.
this is actually a very groovy sounding little box!

[pinkjimiphoton]

tried a bunch of caps, to my ear, 2.7n did the trick. box time. thanks dave!!!! 

[Kesh]

most likely, the reason you're getting the harsh highs is that your schematic is missing C5 (which shouldn't be .1, but much smaller)

that mod probably came about because CC Dias Baptista suggested removing it in his design notes, but in that case replacing it with a voltage divider.

http://bestnetworx.com/uploader/files/48/mutantesfuzz.pdf

[pinkjimiphoton]

thanks kesh.

i ended up adding the cap, used 2.7n as it seemed to work best, tried everywhere from 100p to .1u. took out the stridency without killing that razor edged treble.

tried a mess more transistors, still, lower gain leaky ge seems to rule in this circuit.

as it stands, i'd like it just a LITTLE louder than it is... do you guys think if i ditched the 1.5k to ground on the output it would jack the output slightly?

i mean, right now, it's about unity gain with the volume maxed. i'd like to be able to push it just a little bit harder, but higher gain transistors don't make it any louder really.

[tubegeek]

... what is TAOE?

TAOE = The Art Of Electronics, a/k/a Horowitz and Hill. A Very Good Book.

Long story short, their advice is, if your biasing scheme depends very much on a particular value of leakage or hfe it's not the best way to do it. I think it's in the very first "bad circuit examples" section.

PS: you had me at "Os Mutantes"!

[pinkjimiphoton]

... what is TAOE?

TAOE = The Art Of Electronics, a/k/a Horowitz and Hill. A Very Good Book.

Long story short, their advice is, if your biasing scheme depends very much on a particular value of leakage or hfe it's not the best way to do it. I think it's in the very first "bad circuit examples" section.

PS: you had me at "Os Mutantes"!

i think i downloaded that book back when doing so was still possible, i have to look! 

this thing very much depends on leakage i think. i tried a bunch of transistors, and the leaky low gain ones reigned supreme.

i can't believe i did this, but i made a total rookie mistake... the 25k resistor on the right of my layout is shown going from collector to output. i screwed up, it needs to be moved to the other side of that 10u cap, or the diode clipper is effectively out of the circuit... and it leaks dc thru when ya turn the knob... 

total facepalm.

so... i figured it out and fixed it. when all was said and done, i changed out the 22k resistor i'd had there for 10k, and ditched the 1.5k to ground. that did the trick, and brought it to life. now it even sounds good into a clean amp, and has TONS of balls. very psyched!!!

here's a CORRECTED and VERIFIED layout. i call my variant the "bat macumba" cuz it SOUNDS like bat macumba.

[Kesh]

Long story short, their advice is, if your biasing scheme depends very much on a particular value of leakage or hfe it's not the best way to do it. I think it's in the very first "bad circuit examples" section.

I think fuzz tends to break the rules of good design. Lots of fuzz circuits are particular about hfe and leakage, and awkward to bias.

[pinkjimiphoton]

this one seems pretty forgiving, but yah, it definitely is a bitch to bias with some transistors!!

can't wait to crank this puppy up today.

[smurfedelic smurfberry]

this one seems pretty forgiving, but yah, it definitely is a bitch to bias with some transistors!!

can't wait to crank this puppy up today.

Can't wait to hear it too, play some Os Mutatesey stuff too ..  That english version you posted is so funny when you've listened to the one in portugese for years.

[pinkjimiphoton]

i dunno if i can play any of it, lol.... that chap is quite fluid and speedy, and i am old and arthritic!
but i'll try and do a stupid pedal trick tonite..

and as soon as i figure out where to stash it, i got a whole bunch of os mutantes effects i found earlier i will post... octave box, phaser, phaser/flanger etc..

[Gus]

did you try med power transistors?
maybe  something like TIP29s

[pinkjimiphoton]

yah, i tried everything from 16hfe ge's to darlingtons my meter can't even read cuz they go too high.
tried some lower gain bc108's (9's? shoot, not i forget) and it was ok, but kinda strident.
tried also 3904's around 120 or so, bc 547's around 300, 2222's, all kindsa stuff.
i sat there and plugged in random npn's for about 3 hours.
to MY ear, the best sound was still the ge ones, leaky and low gain, so that's what i boxed earlier.
SCREAMIN' fuzztone. different from the usual fair, more overdrive-y kinda, cleans up real nice, too.
the phat setting is kinda wooly, but great for some things. i put the filter on a spst footswitch i'd salvaged somewhere.
i don't know if these transistors are magical or something, but heating them with the tip of my iron almost against them doesn't seem to make any
difference whatsoever, normally getting within a foot of 'em makes 'em act up.

this thing is also wicked sensitive.. if i play another guitar within a couple feet of a guitar plugged into it and turned on, it actually picks up the unplugged guitar
from across the room. i love it.

i found some other CDDB effects, will upload soon. working on a one-chip octave up that was apparently in a brazillian version of elector.
just now working up a vero for it, adding a charge pump power supply so it's self contained... the build plans want you to use two batteries tied together for +/- 9v and a 0v ground, so i used RG's method with a charge pump and a 3906 as a switch connected to the input jack ring.

gus, this is a really cool sounding box, have you tried one out yet? you might like it.
i am SURE mark hammer would LOVE it.

[tubegeek]

Long story short, their advice is, if your biasing scheme depends very much on a particular value of leakage or hfe it's not the best way to do it. I think it's in the very first "bad circuit examples" section.

I think fuzz tends to break the rules of good design. Lots of fuzz circuits are particular about hfe and leakage, and awkward to bias.

Absolutely true! I was only just answering the question, not making a recommendation, but I can see how that might not have been clear. H&H is a good source for the rules we love to break.

[aishabag23]

Hey Jimi! I've been away for quite a while. I came back to this build with all the mods you've got posted here. It rips! Like you, I spent hours and hours auditioning transistors and settled on some 2N1986's (hfe 85 for both). I had to reverse the pot wires to get it to go louder CW, quieter CCW. I changed out the 47uf cap for a 100uf cap. I was literally shocked at how quiet it is once it's in an aluminum enclosure. I thought I might be asking for trouble putting it into a 1590a, but it's working great. I can't thank you enough for all the helpful info you're always posting, and for helping me out when I just can't figure stuff out on my own.

I etched a lil' tiny faceplate for it with some copper clad board fragments I had laying around. My wife painted it. I love this thing! Thanks again, friend.

[pinkjimiphoton]

awwww, thanks bro!!

glad ya posted yer build!! now i may make me one too!! i am on a 1590a kick hard lately. its most definitely outta control, but something about being able to triple my pedal intake at 1/2 the weight just kinda makes me smile in my sickest places

hope the FDE is still working for ya! i never found that other chip for it.

rock on aisha!

[duck_arse]

don't let it fall in the coffee!

it's a nice looking box, there. nice dots.