Metropoulos Forum

Haha, don't know about addict just yet..

I'm not sure of its applicability in the quest for jimi-tone put it sure has a nice flavour to it anyway. Put my JTM45 back to KT66-spec yesterday (been running el34's since the 100 went to 6550's) and with that always on fuzz in front of it.. soo nice!

In terms of Jimi's marvelous tone, which is what we are discussing here right? Especially in regard to the later years (69-70) this fuzz-solution (originally it was roughly the same circuit as fuzz-face, but made by JEN) is definitely missing something. That 'something' is evident in the Octavio with the octave disabled, but then there is 'too much' of what is missing from the other one if that makes sense.. My next approach will be building something with inspiration from the Octavio and the helios (thus having lots of Axis nuance) but with the ability to easily switch transistors. So yeah, for the ease of things I will probably build two, one for experimenting with Si and one with Ge. As for the switching: I will implement the always ON but in a different sense.. will get back with info on that.

Uni-Vibe: yes, I got a bit gutted last week when a deal on a Castledine Supra-vibe went south and in (a probably highly emotional) response I decided I was gonna build my own clone. Jokes aside, it will have the whole power-supply on board, will follow the original layout quite closely but will be built on PPH (since I don't etch..). Some substitutions for parts (especially transistors) have been made due to availability and I have values to try both the 'early' and the 'later' circuit, however my transformer has 18V secondaries more in line with the later circuit. Have 2 sets of photocells on the way. First batch of parts got here yesterday so I'll start tinkering with it next week. If it sounds good I might go through the effort of etching a board and source correct parts down the line. In all honesty Im doing it mostly for the sake of the project, its fun! And it will keep me busy enough to not do something stupid like ordering parts for ANOTHER way too powerful amp.. I am actually pretty happy with my Sweetsound Mojo-vibe as is, Bob Sweet sure made a fine product. Maybe we should open up a Uni-Vibe thread once we get our projects rolling, so we don't go off-topic in the fuzz and modulator-threads?

I like the idea of a separate Uni-Vibe thread.

Was fortunate to pick up an original Shin Ei Uni-Vibe unit a few weeks back, incredible tone! I would highly suggest going with an original unit even though they are pricey. Look at how much you are spending on multiple amps and other gear, I think it's worth it.

Hello !

i'm thinking that it's maybe time to relaunch the fuzz project
Dave, if you read me, i Wonder if by chance you'll have a bit of time for it, and if you still have my fuzz

Xplorer, I was thinking about that myself, and yes I have all the fuzz parts you sent. It's about time I got em up and running.

daveweyer wrote:Xplorer, I was thinking about that myself, and yes I have all the fuzz parts you sent. It's about time I got em up and running.

Sweet !!!!

I missed this "announcement", so sorry for the apparent lack of interest. I will be very excited to hear what comes out of this!

To answer the question from the other topic.

If you want move volume from the fuzz just change the output voltage divider (8k2/470 Ohm). raise the 330/470 to 1k its gonna have more volume output. You can also use a pot on Q1C a-la RangeMaster,(like explorer posted) but biasing will get a bit harder.

For more fuzz well you want to lower the voltage on Q1, from my experience changing the 33k/Q1C resistor value has really little effect on gain(mostly bias and impedance). you can raise its value that will lower the Q1 bias for more fuzz and the input impedance will raise a little bit. you could also change the 100k to a lower value, that will lower the bias(increase fuzz) and raise the gain a tiny bit and it will lower the input impedance a bit.
You can also try to have transistor with lower VBE that will offset the bias a bit and increase fuzz.
And then there's the higher HFe too.

Of course the 3 transistor Fuzz from Dave is THE circuit for huge gain/fuzz. Can't wait to see it in action.


I got myself some AC128 and AC127 transistor the other day and i made a new fuzz (and it do have more gain with the first transistor engagted, and a nice headroom) and mysteriously it do BOG quite well(wich was not my goal with this one lol)
It has an Ge input buffer(wich has a bypass switch NOT shown in the schematic, wich really do make for interesting always on fuzz). Well just try it if you want, the input bass control can be omitted.(and the polarity of the capacitor is wrong, i draw it quickly for a friend)

http://www.mediafire.com/view/dgla3e3cs0r6cn1/Fuzz.png

Xplorer
I just measured the current gain of the 2N525s you got to use in the West Coast Fuzz.
It turns out they are in the 50 to 55 range.
This may explain some of the issues you were having. A NOS 2N525 I have here measures 65 to 70.
To use the lower beta 525s, you might want to use a 2N591 in the second slot. The ones I have here measure over 100hfe.
Just for fun I measured an old NKT275 I got out of one of Jimi's FF units, 77 hfe.
From what I read, the NKT275 is pretty scarce and valuable.
I also had some OC71s from the 60s, and they measure about 50.
FYI

great to see such valuable and interesting info tek !

also, definitely give it a try to a Dunlop blue type of classic fuzz, maybe with your ac127 and 128, with dave's mod ... i can't go backward now, ever ...


Dave, glad that you could have a look at my fuzz and find what was wrong ! unfortunately i don't have any other of these transistors and i doubt i'll find them and in the correct gain.
was the rest of the circuit correct ?

well, you may have the most rare and valuable NKT275 ever, in the whole world .... it was in one of jimi's fuzz ? ha ha

Here are some pictures from Dave, about the west coast fuzz Hendrix prototype, refreshed, and apparently it sounds great !

http://imgur.com/a/gykZ2

Well I finally got the West Coast 3 transistor FF up and running, Xplorer has posted a couple pics.
As you can see, the first transistor is a 2N525 with a beta of 71, the second transistor is a 2N591 with a beta of 108, and the third is a 2N525 with a beta of 51.
-1.2V on collector of Q1, -5V on collector of Q2.
You can completely overdrive the input of your amp with the fuzz control turned all the way down to zero, forcing all the distortion to occur in your preamp stages--or, you can go the opposite direction and let the transistors do all the work. Any option in between is available.
I can re-post the schematic if you like.
I'll work on getting some licks played through the box so you can hear it too.

schematic and clips would be great. I'd buy a kit if it were available

here is the schematic, slightly refreshed :

http://i.imgur.com/jcA6mEf.jpg

Well for all you cats who can dig the 3 transistor fuzz schematic I asked Xplorer to post it, with the few modifications I did since the last time.
As you can probably see, it is really just an extension of the original FF unit that Jimi used. The first transistor is still running at maximum gain with its emitter grounded, and its collector resistor is approximately the same as the original, adjusted for the right collector voltage, which always seems to come out around -1.2 volts for the best sounding distortion.
Because I wanted to take advantage of the gain of the second transistor independently of the amount of feedback used to control the fuzz level, I needed to add a third transistor as an emitter follower. This adds another inversion in the collector circuit, which is then in the right phase for feeding back to the base of Q1 to adjust the fuzz level. You can see that the feedback loop has been divided, the top loop only feeds back DC to the base of Q1, just to establish its bias and provide sufficient current to turn it on, and the lower loop provides AC feedback to determine the gain/distortion of Q1. This whole business allows you to have a great range between clean and dirty, and keeps the transistor biasing stable, even with devices of varying gain.
The second transistor is now free to be run just as the first, with an AC grounded emitter, but without shorting out all the feedback to Q1 and causing it to run at maximum gain, which is what happened in the original FF circuit.
This means you can run the second transistor at full gain, and the first transistor at its lowest gain, for a nice taste of crunch, sort of like you get with a regular amp stage just on the edge of clipping. You have total control independently of the gain of the first and second transistors.
Then, you can adjust the 10K pot in the third stage to get as much signal as you want to drive your Marshall input stage, from just a touch, to complete saturation, tube saturation in this case.
The Fuzz control simply applies less AC feedback to the base of Q1 as you raise the fuzz level, and Q1 goes into asymmetrical clipping. A 1 meg pot provides enough resistance in the Q1 base circuit i.e. the summing node, that for all intents and purposes there is no negative feedback at full resistance, and the gain of Q1 goes to maximum.
As many pedal builders have discovered, the gain of Q1 has a lot to do with the sound of the pedal. If you get too much gain there, you get "metal" distortion, especially if the transistor has a really good frequency response, like a silicon NPN device.
Hendrixy tones happen in the range of 80 gain for the first stage.
The second stage can be higher, I used 108 gain for that stage. The third stage in this circuit has a gain of 51, just enough for the proper amount of negative feedback to send to Q1.
By the way, the circuit is always on, that 33K resistor just bypasses the 1 meg pot for "clean", and leaves it in the circuit for "fuzz".
By having the circuit always on, you get just that little boost in the clean mode that you can't get with a hardwire bypass, as in the original FF.
In the pedal I made for Jimi, I had to guess at the gain value for Q2 since there was no knob for adjusting that. There was also no input pot on Q1, I just used 10 or 20K resistors there.
There was also a fixed gain adjuster when you switched from clean to dirty, so that the levels would come out the same, but there is no need of that on this circuit because you will surely want to fiddle with all the controls for every mood.
Jimi wanted a unit with all the adjusters on it, but I never got one finished for him, and for the time being had to make do with the existing shells with the two knobs.

So there you have it in a nutshell. If folks want to build these things, they'll need to find a source of decent germanium devices which have known hfe characteristics from device to device. The right hfe is pretty important to get these circuits to sound the same from unit to unit.
That was the problem with the original FF pedals, you had to try a whole bunch to find ones with the right gain structure. Jimi had a whole stack of them but he would only use ones that were especially marked, unless they were all broken and he was in a fix.
One of my jobs for him was to get the gain structure right in the huge box of FF units. I think they eventually all got done, but by then he had some other favorites, including the 3 transistor West Coast Fuzz. Roger Mayer also made a 3 transistor fuzz which showed up at West Coast Organ and Amp for repair. If I remember correctly it had a little add-on circuit board, and the extra transistor was in front of Q1 instead of after Q2.
Little trivia for you.

great tech infos !! well explained, thanks

so Q3 is just here for the right negative feedback phase, and allow Q2 to work a special way ?

also, it would be interesting to see what roger's fuzz was on a schematic, the principles, to compare with his current axis fuzz and such, but i think that they're different from that principle. not sure.

i suppose that applying your other fuzz mod ( the one made in the blue Dunlop recently ), on this three transistors fuzz : that could be really great, what do you think ?