Underperforming Maestro FZ-1B

[Mark Hammer]

I perfed up a little 5-way fuzz with a quintet of 2 and 3-transistor classics that could be selected with a rotary switch.  It has:
- Shin-Ei FY-2
- Silicon Tonebender Mk II
- Bosstone
- Green Ringer
- Maestro FZ-1B

All are working and performing just fine, with the exception of the FZ-1B, which barely produces a grunt, and nothing one could ever mistake for fuzz.  Max output level is barely equal to bypass level.  I'm using the schematic shown here, which has several corrections and changes from the original Maestro factory schem, but corresponds to the circuit photos here: http://www.diystompboxes.com/smfforum/index.php?topic=61138.msg479711#msg479711

The only real change I made was using 2N3391 transistors, with an hfe in the mid 200's, instead of the recommended 2N3860.  The 3391s have a BCE pinout, which I was attentive to, so I know the low output is not a result of pinout issues...or at least I'm reasonably confident.  So, I'm wondering if the problem lies in the biasing of this particular transistor.

Any ideas?

[PRR]

http://www.diystompboxes.com/wiki/index.php?title=Debugging

[Aph]

I tried it. Circuit as shown.
Balance all the way to fuzz:
2N3393's (175 to 190 hfe) = Somewhat fuzzy... volume can give more than unity gain.
2N5088's (500 hfe) = more fuzzy... volume can give more than unity gain.

Balance all the way to clean (either transistor): volume all the way up gives way less than unity gain.

Even with the 5088's, it wasn't super fuzzy. But, I've played through other "vintage" type fuzzes that sounded similar.

[PRR]

> Balance all the way to clean (either transistor): volume all the way up gives way less than unity gain.

In that setting, it is a unity-gain follower with about 3:2 loss after that (10K into 33K||50K). So significantly shy of input.

[Mark Hammer]

In which case, should something be changed to improve upon that output?

[Aph]

Are you getting unity or less than unity with the Balance control all the way to fuzz, Mark?
If you lift the line that goes from the .15 cap to the top of the volume pot, you'll get a lot more gain (even with the balance control all the way to "fuzz")... but then the balance control becomes superfluous. Maybe move that to the input like some fuzzes have it.

[thermionix]

Maybe dumb questions here.  Could the Balance be rewired to a blend, instead of one end going to ground?  Are the clean and dirty parts out of phase?

[PRR]

> improve upon that output?

What output?

The clean output at gain <2/3.... you may just be asking too much of two transistors.

However-- is that Volume pot of any use? If not, then this mod gives unity (0.95) gain in DRY.


{EDIT -- this may have too much dirt leakage.}

Gain in DIRT should be ample. If that is what you are asking, specify, and share your voltages.

[Aph]

Maybe dumb questions here.  Could the Balance be rewired to a blend, instead of one end going to ground?  Are the clean and dirty parts out of phase?

I tried that. Unfortunately, the clean sound was still dirty with not much gain.
This mod will give plenty of make-up gain even with the "Dirt" turned down a bit. Still not the greatest fuzz sound though, IMHO.

[Mark Hammer]

Following up on the suggestion to consult the repair FAQ, I find that Q2 in my FZ-1B meets voltage expectations, but Q1 has only a 400mv difference (in the right direction) between base and emitter voltage.  Is that way off the expected 500-700mv range? or normal?  And if off, what manipulation brings the difference up in the desired direction?

[JustinFun]

The FZ1B runs on 3v not 9v, doesn't it? Could the increased headroom at 9v be why it's underwhelming as a fuzz?

[JustinFun]

The FZ1B runs on 3v not 9v, doesn't it? Could the increased headroom at 9v be why it's underwhelming as a fuzz?

Scratch that, I was thinking of the FZ1

[Aph]

Following up on the suggestion to consult the repair FAQ, I find that Q2 in my FZ-1B meets voltage expectations, but Q1 has only a 400mv difference (in the right direction) between base and emitter voltage.  Is that way off the expected 500-700mv range? or normal?  And if off, what manipulation brings the difference up in the desired direction?

Just as a reference, here are the voltage readings I got after breadboarding the circuit you posted (the circuit does work):

Using 2N3393's
Q1
C: 9.5
B: 7.25
E: 6.88

Q2
C: 3.77
B: 1.5
E: 0.9

[Mark Hammer]

I'm getting 7.54 for Q1-B and 7.14 for Q1-E.  So, not that much different from you, and maybe even a little better, in terms of B-E difference.

[Aph]

I'm getting 7.54 for Q1-B and 7.14 for Q1-E.  So, not that much different from you, and maybe even a little better, in terms of B-E difference.

Yes, and the gain of your transistors are higher than mine.
Here's what I get with the real high gain 2N5088's:

Q1
C: 9.5
B: 8.5
E: 8.25

Q2
C: 2.24
B: 1.7
E: 1.1

And the circuit works with these, also.

[Mark Hammer]

Looks like I'm going to have to re-check my wiring.