Tripple fuzz build sounds wrong

[Jussi]

Hey,

At first I tried this circuit on breadboard and it sounded as described but now I have built it and it sounds more like harley davidson than what it should be   "Brrrrrrr". It sounds like transistor with too low bias, but it is still very controllable fuzz etc. Maybe less gain too.

When I started it up the first time I had minus and plus wrong way in power adapter jack. Maybe this broke some of the transistors? I've tried changing the pnp 2N3906 already but don't have other spare trannies atm.
I've also tried audio probe debugging (100 nF cap between testing point and amp input) and the sound was quieter than my input signal at 2N5089 collector, but was fuzzed well though. Broken transistor?

Here are voltages I've measured between the point and ground. I don't know if they are OK.


Sound sample of working build: http://www.jiggawoo.eclipse.co.uk/guitarhq/Circuitsnippets/tripple.mp3

[LucifersTrip]

Here are voltages I've measured between the point and ground. I don't know if they are OK.

are these before or after it was working well? did you test components other than the transistors to see if anything else was blown?

[Jussi]

The voltages are measured now when I've build it and its not working in the right way. I have only measured voltages, checked all resistances, checked that transistors are the right way and the whole circuit is like it should be.

But I don't know how the voltages should be. If the transistors were blown , woudln't it be seen from voltages..?

[LucifersTrip]

yes, you could prob tell the transistors were blown from voltages...those voltages don't look crazy, which is why I asked if you checked the other components.

having a set of voltages for when it was working would be a big help...then, you can see what changed and probably pinpoint the bad area.

so, the lesson is always take voltages when in sounds best, for future reference and to help others who might build it.

maybe the Signal Fuzz can give you hints
http://www.diystompboxes.com/smfforum/index.php?topic=99850.0

but since the parts count in small, I'd prob just breadboard again and compare voltages

...but do check to see if other components are blown.

[MaxPower]

Check the 1uF capacitor, especially if it's polarized.

Disconnect the first transistor from the others (maybe at the 1uF capacitor) and try the audio probe at the collector again. If you get buzzing/fuzz there, then you know the issue is in the first part of the circuit (since the fuzz is presumably supposed to come from the diodes).

[Thecomedian]

the voltages for the 906 seem wrong. the 904 is also going to have very "fuzz" attributes regardless of the diodes.

the 906 collector seems to be "hard" cut off to me. the voltage should be lower on collector compared to base, at least by 0.6-0.7v, and its actually higher by 0.3v. If its like you said, the circuit is already sounding like it sounds with audio probe at the output of the first stage, then the signal is just passing nearly unmodified through to the diode clipping stage, that is, the second transistor stage isn't doing anything. the 904 transistor has just barely enough voltage from E to B, but then it's voltage starved from B to C.

That's I think how it looks to me. I've been able to run miswired (maybe leakage activated) Ge fuzzes with the battery missing, but the transistors apparently in the path of the signal, and it sounded like I didn't even have the circuit in the chain at all, perhaps very slightly quieter.

I dont think it's the capacitor, if it was blown through to short circuit, the voltage on both sides of the cap would read the same, would they not?

Are the transistors clearly labeled? Did you perhaps swap places of one with another, such as 809 in 904 place and vice versa?

Accidental reverse voltage applied might not kill it. discrete transistors are hardier than ICs.

[digi2t]

Just a sidebar here;

This frequency trippler arrangement is also used in the Spaceman WOW Signal fuzz. The difference being that they used a germanium transistor post trippler, rather than the 5089 pre. I did a lot of transistor testing to get a nice fuzz out of the trippler arrangement when I cloned the WOW Signal, and in the end I found that a matched set of 2N5087 / BC546B, in the 300 to 325 gain area worked really well. Matching the two, to within two points (or less) gain-wise, yielded the best results.

I just found it to be a really finicky circuit transistor wise, but a great sounding fuzz though.

[MaxPower]

I just ran a dc simulation of it in Tina-ti. These are the results:

Q1 (2N5089):
Collector: 6.24 V
Base: 3.37 V
Emitter: 2.76 V

Q2 (2n3906):
C: 5.03 V
B: 4.48 V
E: 5.16 V

Q3 (2N3904):
C: 5.03 V
B: 4.48 V
E: 3.84 V

Seems pretty close to the voltages shown in the first post. Keep in mind Tina-ti isn't necessarily error free. Would the voltage drop across the 500k resistor be enough to cause the base to be lower than the collector of the 906? Or am I looking at it backwards?

[Jussi]

I just tried the pedal again and now it sounds quite similar than in the sound sample. I had resoldered everything once again. Thanks for the replies and simulation though! I learned a lot.

Only problem now is that it doesn't sound as good as in breadboard. It was much more full and smooth and juicy (compressed?) and I may have had the PNP transistor wrong way in breadboard. Is it possible it would have worked? Anyways I'm gonna order all parts again and try to achieve the sound again in breadboard. If I get it working, I'll post a schematic - it sounds much better and it's fun to play with.

[Jussi]

The problem was not enough gain. In the original circuit Q1 has 10k pot from emitter to ground and I thought max gain is with 10k resistance so I replaced it with 10k resistor. Now I shorted the emitter to ground and sounds good again.

Also if I have the 2N3906 PNP transistor backwards it sounds more smooth but different than by removing it from the circuit. But it may get damaged? Sounds like I'm making waves in ocean when I play Very hissy though.