Issues with the Maestro FZ-1S (low fuzz volume) - DIY version of GGG

[CosmicEffect]

Hi everyone !

I'm actually having some issue with a DIY of the Maestro FZ-1S reproduction, based on the schematics of General Guitar Gadget (true bypass version) http://www.generalguitargadgets.com/pdf/ggg_fz1s_sc_bp.pdf.
I decided to do the PCB myself, but i duplicate the schematics.

Here's mine, done in Easy EDA:


I have done some modification :
- I wire the "+" pin of C12 with R19, because they weren't connected on EasyEDA. So the PCB was made wrong.


The sound is normal when the footswitch is off. And when i turn it on, i have an amplified clean sound, without any fuzz. And when i turn the balance knob, i have a barely audible fuzz tone.
I verify an re-verify that every transistor was on the right way. And i tested them with an multimeter. Everything seems ok.
I made an Audio Prob, and saw that the volume of the down channel (the one after R3) was very quiet after the collector of Q3.

Any idea ?

(Also, i'm new here so if i forgot any information, i'll give you without problem !)
(Also n°2, English is not my native language. So I apologize in advance for the mistakes )

Thank you for your help !!!

[FSFX]

You say you did your own PCB. Did you have the balance pot (which is a dual one) connected correctly.
For balance to work, one half of the pot must be wired opposite to the other half.
By that I mean:
- Pin 1 on one half should go to ground and input signal should be on pin 3.
- On the other half of the pot, pin 3 should go to ground and input signal should be on pin 1.

[CosmicEffect]

Hi FSFX,

Yes, i did this modification, cause i saw that on the original schematics of the FZ1-s (I should mention that on my modification list)

[duck_arse]

welcome to the forum, CosmicEffect.

we can't see your thing unless you post up some photos of your build and the offboard connections. and, as ever, we want your voltage measures around the circuit, all transistors, supply, etc. and a board layout if you have an image of that would be helpful.

[CosmicEffect]

Hi,

So i made the different measurement of the transistors.
I made it without sound singal. I only connected the 9v power supply.
I also put on below some pics of my PCB, and the layout.

Tell me if you need anything else.

xxx(black wire) - xxx(red wire)
H3 - H1 = 8.7V

H3 - Q1[base] = 2.7v
H3 - Q1[emitter] = 2.21v
H3 - Q1[collector] = 8.7v

H3 - Q2[base] = 1.2v
H3 - Q2[emitter] = 0.61v
H3 - Q2[collector] = 4.5v

H3 - Q3[base] = 5.9v
H3 - Q3[emitter] = 6.1v
H3 - Q3[collector] = 6.6v

H3 - Q4[base] = 0.39v
H3 - Q4[emitter] = 0.4mv
H3 - Q4[collector] = 8.6v

H3 - Q5[base] ≃  -100mV
H3 - Q5[emitter] ≃ 1.0mv
H3 - Q5[collector] ≃ 0.1mv

H3 - Q6[base] ≃ 0.5mv
H3 - Q6[emitter] ≃ 0.1mv
H3 - Q6[collector] ≃ -100mv

[FSFX]

The voltage on Q4 collector seems to indicate that Q4 is not biased into conduction.
In fact all of the voltages around Q3 and Q4 look strange.
Still, the voltage you read on the base of Q3 could be loaded down by your DMM and shows lower than what it actually is because of the 1M base bias resistor.

[CosmicEffect]

I agree. I spotted this spot with my audioprob and saw that there was a big drop in volume.
Just checked the 2N5089 datasheet again though, and it seems to be in the right direction.

Could R16 be responsible for this?

[FSFX]

Could R16 be responsible for this?

R16 is fine. It is actually good there. Q4 would benefit from a similar resistor in its emitter to stabilise the bias.

The circuit around Q3 and Q4 should work OK as it is. It quite severely squares off the waveform and has quite a lot of gain so it doesn't need much input to give a good distorted output.

[duck_arse]

your circuit dia shows R16 as 470R. please advise the actual value you have fitted ..... it might be the source of your bias woes.

Q4 would benefit from a similar resistor in its emitter to stabilise the bias.

only if the correct value was used, tho.

[antonis]

With all respect to GGG and JD Sleep, I don't like Q4 biasing configuration...
(unless it aims for fully asymmetrical clipping..)

[FSFX]

The whole area of the circuit around Q3 and Q4 is very weird from a design perspective but it does hold true to the original FZ1-S circuit so I suppose that is what was intended.
No experienced electronics designer would do it that way but most of these pedals were not designed by electronics engineers anyway.

The biasing of both transistors is really poor unless you intentionally want to create severe distortion.

The biasing of Q4 has no stabilisation but as it appears to be such as to bias off Q4 in its quiescent state anyway, then stabilisation of the bias is irrelevant here.

The collector load resistors are a magnitude higher than would normally be used in audio common emitter amplifier stages.

Running simulation models of these two stages with a variety of silicon transistors shows that it has sufficient gain to create a fully clipped asymmetric output from an input signal of just a few tens of millivolts.

I did originally post these observations but deleted them after looking at the original FZ1-S circuit and deciding that was they way they intended it to be and any changes to the circuit design or component values would be an unwelcome deviation from what the FZ1-S was and how its fuzz sounded.

   

[duck_arse]

With all respect to GGG and JD Sleep ....

none due to Maestro?

[antonis]

I made an Audio Prob, and saw that the volume of the down channel (the one after R3) was very quiet after the collector of Q3

Without particular measurements, I'd start experiment by lowering R14 value..

But here, Q3 looks like been bottomed (only 500mV V_CE availiable) so I'd try raising R14 value..

[FSFX]

I made an Audio Prob, and saw that the volume of the down channel (the one after R3) was very quiet after the collector of Q3

Without particular measurements, I'd start experiment by lowering R14 value..

But here, Q3 looks like been bottomed (only 500mV V_CE availiable) so I'd try raising R14 value..

The low reading on the base of Q3 is due to the DMM loading it down as it is biased via a 1.5M resistor.

In reality it actually works OK with 2N2222, 2N3904 and 2N5089 transistors and produces a lot of gain but the output waveform is severely distorted as would be expected with all of those weird values.

[antonis]

The low reading on the base of Q3 is due to the DMM loading it down as it is biased via a 1.5M resistor.

I was talking about Collector - Emitter voltages..
6.1V on Emitter calls for an about 13mA Collector current hence Q3 forced into saturation mode..

[FSFX]

We are not meant to be redesigning a classic pedal like the FZ-1S.

All of the component values shown are as they were on the original circuit.
The circuit works as it should with those values.

It is a fuzz, not a linear amplifier so it may look a bit weird but it will work.

If it does not work for the OP then there is a wiring issue or faulty transistors.

With 2N5089 transistors the voltages on Q3 should be approximately as follows.

Collector  760mV
Base        615mV
Emitter      25mV

Anything vastly different from this indicates faulty components or wiring errors.

[CosmicEffect]

your circuit dia shows R16 as 470R. please advise the actual value you have fitted ..... it might be the source of your bias woes.

Q4 would benefit from a similar resistor in its emitter to stabilise the bias.

only if the correct value was used, tho.

thanks for pointing it out. I noticed that my resistor package was mislabeled (I put 470k instead of 470 ohm).
Rookie mistake, but now the fuzz is audible!

The second problem is that the sustain is very low. Barely after playing a power chord, the sound disappears...

I may be trained in electricity, electronics is really not my field...

(aside from that, thanks for all your replies!)

[bmsiddall]

I had similar happen to my build.  Try increasing the 47k resistor on q4's base to give it a little nudge (i used a 100k trimpot).

[duck_arse]

your circuit dia shows R16 as 470R. please advise the actual value you have fitted ..... it might be the source of your bias woes.

Q4 would benefit from a similar resistor in its emitter to stabilise the bias.

only if the correct value was used, tho.

thanks for pointing it out. I noticed that my resistor package was mislabeled (I put 470k instead of 470 ohm).
Rookie mistake, but now the fuzz is audible!

The second problem is that the sustain is very low. Barely after playing a power chord, the sound disappears...

I may be trained in electricity, electronics is really not my field...

(aside from that, thanks for all your replies!)

well, bad transistors aside, I'd be checked all the other resistor values for correct colour bands. I'd also be keeping an eye peeled for misplaced caps/values. and another round of voltage measures, please. meter loading effects notwithstanding.

also, I find your board layout method a little odd. personally, and I have no formal qualifications to fall back on, I'd have put the Q3 resistors near Q3, instead of half the board away. I had trouble tracking down your resistors while checking.

[duck_arse]

more to the previous - a few things.

if you have a three pole switch, you should probably wire the pedal as true bypass.
wiring the whole board with black is cool, as they say, but does no favours to those debugging/inspecting.
very much check your solders on R14, the holes look dry of solder.
what are the markings on C6, please? I always suspish caps when I can't see the writing, esp when there is only one of that kind on-board.
also suspish blue resistors when they only have 4 bands. please measure the resistance of R2.

but nevermind all that. I'd bet on the fact [!!] that the two greencaps at C10 and C11 are not the correct values. they should be 680pF, but I'd bet on at least 68nF being fitted. can you please confirm the markings on them?

and having gone all over your board this evening, I again say - you need a rethink on your layouting. my opinion only, but none of your far flung placements make sense. group a transistor and it's biasing parts in a lump, then link the lump to the next lump. you will end up with a less pretty pattern of parts, but also a much much less complicated mess of traces and vias.