olson fuzz master questions

[pinkjimiphoton]

man, i wish lucifer's trip was still around... dave! we miss ya! 

anyways, for no particular reason this caught my eye the other day scrolling thru the diystompbox forum on farcebook:




so i figured, since i finally have a wee bit of downtime between jobs, i'd slap one together.
so i vero'd it up.... this IS verified, with a caveat or two we'll get to later...





i changed the input and output caps from 3.3n to 39n for no good reason on the vero layout... i left the output cap that size, but made the input cap 3.9n ultimately cuz above that it was just too damned wooly sounding. also made the c resistors for the actual fuzz part a little bigger, i split the 10k resistors off to a 4.7k and a b10k trimmer "just in case" so i had a bit of wiggle room to mess with. this proved fortuitous with some ge's.

i experimented with a bunch of npn's... si doesn't work well in this circuit. ended up with 3 ac128's, all over the place in hfe... one's about 40, one's about 60, and one's about 100. i say about cuz they're ge. if ya BREATHE too close to them, the gain changes of course.... i socketed them, and found little difference swapping them around tonally, but ended up with the usual low gain in q1 and worked up from there. i did look up the 2sc175's, and their gain is in the ballpark with what i used.. of course i made the circuit npn so it would play nice with other stomps.

also went with an 18k rather than the 15k resistor from q2 b to ground, as it was handy and i was too lazy (some things never change) to dig thru my resistor drawer for the proper size.  close enough for rocknroll, right?

sound wasn't super impressive. not BAD, but not great. i found several bias points on each q that seemed to work out pretty well, still a work in process where that's concerned.

but the "fuzz" control just plain sucks. it doesn't really do much of anything, you have about 1/4 of its sweep that's actually useful, and below that, it just plain doesn't pass signal. looking at how its drawn, it looks weird to me...

so first question, can someone explain how the hell its supposed to work as drawn?

to ME, it looks like it should be varying the bias to q2, which it kinda does, if really @#$%in' poorly.
it seems to me a fixed resistor would work better.
further, the dc to the pot as its drawn damaged one pot, literally smoked the mother@#$%er.
so my spidey sense said, i bet this was drawn wrong <seen a bunch of those over the years, like the shatterbox... i know one of the guys here drew it, so i respect that and have no doubt they did it "right" to the original unit they traced, but still maintain it was a screw up by the designer in the first place that was reproduced for years by the fabricators... there's a thread here somewhere, i called mine the scatterbox cuz it sounded like shit.. but i'd found by moving ONE component, it went from useless 60's fizz to an actual decent sounding and useable fuzz>

so i looked around at the schematic, and to my understanding, that 1 meg fuzz pot should have a blocking cap at each end, really. yeah, without one, its a feedback loop for q2, but not a useful one. seems weird. i can't see any reason to add a pot that literally makes zero difference tonally...other than destroying it.

now, i am a fuzzaholic, but i'm no @#$%in purist. i don't gaf if something is "right", "original" or "correct to the period cuz i drew the mofo myself" or any of that crap.

half the damn drawings of schematics on the net are full of mistakes.... lets not go there now...lol

but anywho, looking at it, i pondered what would happen if i moved the fuzz pot second connection, on the schematic at the junction of the c resistor and c on q2 to the OTHER side of the 1u cap from q2 c to q3 b....

and bingo. suddenly, instead of a useless pot that barely does anything, it acts like a "contour" pot to the fuzztone itself. not a huge difference, but eminently useable by comparison to the schematic.

now, i'm too stupid to understand all involved in this minor circuit change... but it seems like it lets you blend some of the clean signal from the buffer stage (q1) around q2 to affect the overall tone... like a pan control like ya see in some of the super primitive fuzzes sometimes. does that seem right? it DOES make a huge diff in the overall sound of the pedal.

it went from super wooly and shitty sounding to a really nice overdrivey kind of fuzz, where adjusting the fuzz pot actually changes the tone somewhat (it didn't do anything but phart as drawn)... and the volume controls the overall level of fuzz as predicted. it also reacts properly to the guitar's knobs, which it did NOT do as drawn...

so i'm going out on a limb here, and postulating the schematic floating around the web is wrong, either because it was released wrong by olson to throw peeps off the tracks see RG's dirty tricks part of the geofex website <http://www.geofex.com/Article_Folders/dirtytrk.htm> or it was a mistake by a tracer, or it was built wrong by the original factory and nobody gave a shit cuz it was a @#$%in fuzz, and nobody cared or knew if it sounded right or not. its supPOSED to @#$% up the sound, right?

i don't have the capability of doing video these days unfortunately, but if ya mess with this circuit, its way better with the mod. in this case, on the vero, move the fuzz 1&2 connection from c1 to j17 instead, and you'll immediately see which variant works and doesn't.

also, as shown on the schematic, it SMOKED q3 at one point. i never before encountered a ge getting hot enough to burn my finger!!!!! surprisingly, tho, once it cooled down again after i swapped that one wire,  the q still works. weird. fuzzes are weird. add germanium, and shit gets REALLY weird.

but anyways... this is what i done did, and the results...

so... is the fuzz control supposed to be a bias control? or do ya think the connection was inadvertently (or nefariously) moved to the wrong side of the 1u blocking cap before q3?

am i crazy? <duh> or on point? what do ya think?

look forwards to your observations, fam.

LLAP

[ElectricDruid]

Pink Jimi wants *us* to tell *him* about fuzz?!? I carnt 'ardly believe wot I'm seein'! No wonder it's so quiet around here!!

Sorry mate, no clue. Not really my department. I leave the ge stuff to folks like you and stick to silicon!

[pinkjimiphoton]

Pink Jimi wants *us* to tell *him* about fuzz?!? I carnt 'ardly believe wot I'm seein'! No wonder it's so quiet around here!!

Sorry mate, no clue. Not really my department. I leave the ge stuff to folks like you and stick to silicon!

lmao... its been a while since i messed with any ge fuzzes, i always end up using si these days on the commercial stuff! i JUST found my box of ge transistors, lost on my bench for a year or so


Hahahah my stoned ass forgot that i AM running this at 9v. i didn't even notice the 1.5v batt originally.

when all was said and done, i tried a bunch of fuzz pots there, most worked about the same, from 10k on up.

when all was said and done, i moved lugs 1&2 to the node after the blocking cap before q3. its not adjusting the bias as it was,
but it makes a nice softness adjustment to the fuzztone itself there. its boxed. @#$% it.

the vero is right, but it really does come down to needing fairly leaky germanium to work. at one point i DID have a si in q2 messing around and got it to pass signal and a decent fuzz, but the "fuzz" control worked very poorly. maybe the last 20% of the pot, below that, it was off, at least at audio frequencies.

coulda went with a smaller pot, and a big ass resistor, but wiring it the other way gave me a much more useable fuzztone.

[antonis]

Pink Jimi wants *us* to tell *him* about fuzz?!?

He's got a bottomless audacity,...!!~!

P.S.
With all the respect Jim, don't do it..
You call us for theoretical analysis on your billionth build.. - it is unfair, if not dismissive..

[pinkjimiphoton]

.............and don't forget, no f'n couth!

i been thinkin' <always dangerous> about upping all my old layouts n stuff, warts n all for yet another generation of fuzzaholics, warts n mistakes and all.

but this frustrating legacy may be too much for us mere mortals to survive the onslaught of. many have dumb mistakes on them that i'd find while building them. BUT this led to me being able to find out how to fix issues, and work around 'em....

but should i do that to yet another generation of fuzzmongers? i still await approval to release some of the newer stuff, if they're not gonna use it where i work, i don't see any reason not to give it to the world, but then i get yelled at by the corporate part of the whole mess. yikes.

i miss being a simple fuzzmonger. my mortgage prefers getting paid tho.

ANY ways... hope yous all is well, and be's staying dat way! <3

[iainpunk]

Hahahah my stoned ass forgot that i AM running this at 9v. i didn't even notice the 1.5v batt originally.

i wish i got 10ct every time i used the wrong voltage on old circuits, i'd be rich (probably not), but i mostly get confused between 1.5 3 and 6 volts. i generally know its an odd voltage, but mix up the options.

cheers

[pinkjimiphoton]

Hahahah my stoned ass forgot that i AM running this at 9v. i didn't even notice the 1.5v batt originally.

i wish i got 10ct every time i used the wrong voltage on old circuits, i'd be rich (probably not), but i mostly get confused between 1.5 3 and 6 volts. i generally know its an odd voltage, but mix up the options.

cheers

hahah right? i probably should try it right with a single 1.5v batt, but... i already boxed it lol