Maestro FZ1-B 4 trans restoration, factory schem and pcb photos inside

[joegagan]

a customer of mine got a freebie from a craiglist seller when he went to purchase a wah. the kid had this nice 69 ish maestro fuzz.
it was mostly complete, but all taken apart. luckily i found some gutshots online as i wanted to use the correct color wires for the restoration. luckily i found some correct knobs at my local amp guy's place. was a fun project!
there were numerous version of the maestro fuzz once it went silicon. i saw two, three, and four trans version all in this style case.

it gets those trashy buzzy 60s sounds really nice. the balance control mixes a bunch of clean in. when used this way, you get poofy cleanish notes with the buzzsaw sound hovering below the surface. it gates when you play lightly or turn the guitar volume down. the threshold between clean and total buzzsound is knife-edge, so it is really fun trying to ride that wave. i tried to show this in the video , see below.



this is the board before rewiring:




[/quote]

[joegagan]

[LucifersTrip]

very cool...it looks like a FuzzRite with impedance matcher and squelcher...what exactly does the squelcher do?

[joegagan]

brad and gus have both commented on this section.

brad/solidhex "Its a simple noise gate. Q3 and Q4 depend on each other to turn on. When no signal is present Q4 shuts off eliminating the Q3 emitter's connection to ground killing all signal. When a note is played current travels through the diode and raises the voltage on Q4's base turning it on. With Q3's emitter grounded through Q4 it then is on as well. The correct schematic is the first one with the 680K to ground and 3M3 to diode. I believe the .1 cap is just there to insure only direct current is traveling to the base. Its a cool idea but the circuit has a sort of anemic sound to it. Would like to hear an original unit. Its basically a Fuzzrite with some fun stuff added"

here was gus smalley's take:
"If you look close the circuit is a lot like the foxy lady, Shin EI, fuzrite  and other cascaded gain stages two transistor fuzzes.

  This one has an input emitter follower buffer and and the added squelcher section.  The mixing with this one is a dual pot the others have the outer legs of the mix pot to the collectors and the wiper to the volume pot

The circuit change I read in the ampage posts are

anode of diode stock connection

cathode of diode to 3.3meg to ground and .1uf to ground(3.3meg with a .1uf across it)

The cathode and 3.3meg and .1uf node to a 680K the other side of the 680K to the Q4 base

It looks like a balancing act.  It looks like you want enough current to barely turn on Q4 when not playing.
  When playing and the signal going more positive the increased current though the diode raises the voltage across the 3.3meg and the hold cap the .1uf(one of the RCs).  The node voltage increases and causes more current to Q4 base via the 680K turning it on.  The 3.3meg and .1uf and 680k control the turn off time and the turn on is faster via the diode.

Does that make sense how I worded it?"

[joegagan]

i found an old ampage thread that said the maestro part number for the diode cross refs to  1N457

[LucifersTrip]

brad/solidhex "Its a simple noise gate. Q3 and Q4 depend on each other to turn on. When no signal is present Q4 shuts off eliminating the Q3 emitter's connection to ground killing all signal. When a note is played current travels through the diode and raises the voltage on Q4's base turning it on. With Q3's emitter grounded through Q4 it then is on as well. The correct schematic is the first one with the 680K to ground and 3M3 to diode.

thanx for the solidhex quote...a very good explanation...a gate....should be able to play more start/stop choppy stuff.

I saved the other version of the schematic with the 3.3M and 680K reversed....
http://www.luciferstrip.com/fuzz/maestro-fz1b-schematic.jpg

...but good to know yours is correct. Here's the full thing without the tear over the squelch section:

[brett]

Hi
I'm no exert, but I wonder if you experimented with a 'reverse' squelcher you could get a simple compressor (by adding emitter resistance as the signal goes up). Maybe not appropriate for a fuzz, but useful in lots of other circuits.
cheers

[mac]

from facebook:

it's in my breadboard, 2n3904 and a ge diode. The diode does not alter the sound that much. I modified the squelcher, no 3m3 to gnd, 1uf cap from diode to gnd, and replaced the 680k with a pot, a lower value, say 220k, helps when I roll the guitar volume down.
...
With a 680k I was getting null sounds when I rolled the g-vol down. I think it is a bias thing. Also a big cap keeps more charge to feed the base... i guess

mac

[joegagan]

thanks , everyone.

mac, lucifer's t , i like the possibilities this circuit presents. seems like the public would enjoy having the 680K /gd resistor on a knob , labeled "gate" or something.

i would simplify by getting rid of the balance knob, i generally don't like  peanut butter in my chocolate ( clean dirty at the same time)

[joegagan]

brett, that is also an interesting idea. play around with it and let us know what you find!

[RedHouse]

Nice close-up (and clear) pic's Joe, thanks for posting them.

[joegagan]

thanks, brad. after i took these shots a bunch of other similar photos popped up that i hadn't seen before. but i think the trace side shot is fairly unusual.

[johnk]

i designed a pcb and built one of these (the 4 transistor version) using the factory schematic and while it works, IMO, it doesn't sound like the demos that i've seen on youtube. i used NOS 2N3381 transistors) it's quite a bit below unity volume even with the controls dimed and the fuzz is much thinner and brighter so i'm thinking that the factory schematic may have some errors?

[GGBB]

i designed a pcb and built one of these (the 4 transistor version) using the factory schematic and while it works, IMO, it doesn't sound like the demos that i've seen on youtube. i used NOS 2N3381 transistors) it's quite a bit below unity volume even with the controls dimed and the fuzz is much thinner and brighter so i'm thinking that the factory schematic may have some errors?

Search results for 2N3381 suggest that they are P-channel FETs. The schematic drawing indicates BJTs. I'm surprised it works at all, but I might be mistaken.

[zbt]

Mind for the board size Sir?
If I am not mistaken is around 51mm x 70mm, I guess

[duck_arse]

i designed a pcb and built one of these (the 4 transistor version) using the factory schematic and while it works, IMO, it doesn't sound like the demos that i've seen on youtube. i used NOS 2N3381 transistors) it's quite a bit below unity volume even with the controls dimed and the fuzz is much thinner and brighter so i'm thinking that the factory schematic may have some errors?

P L E A S E _ post your factory circuit diagram. the one shown in this thrad makes no mention of 2Nanything, so we have no idea what it is you have built.

I have two copies of that original circuit shown in the first post, one shows 680k//3M3 on the squelcher base, the other shows 3M3//680k. so, what does your circuit show?

[Rob Strand]

The four transistor version is called Fuzz Tone FZ-1B (version 2).

There's a few confusing points about this circuit:

Schematics:
1) There are a number of schematics with slight variations about the transistors and diode.
   I also saw one an input socket instead of a flying lead.
  The earlier schematic on stuck on the enclosure used xxx-00xxxx part number whereas
  the later schematics from the Maestro service manual use xxx-04xxxx part number.

2) The later schematics show the equivalent diode and transistor part numbers as
  1N457A and 2N3392.   The 2N3392 has a color dot for gain sorting which is not indicated
  on any schematics.

3) I don't have enough of my notes with me to resolve the parts on the earlier version, see 4.

Actual units:
4) Earlier units. The transistor packages are consistent with 2N3391/2N3392 parts parts.
   The color dots in this thread are yellow which I believe is a nominal hFE = 225.
   Do all units use yellow dot transistors?
   ==> EDIT:  it appear units were made with orange dots (hFE = 135) or  yellow dots (hFE = 225)
   The diodes have color bands which evaluate to 1N457.

5) A trace of the PCB in this thread agrees with the LucifersTrip's schematic:
   - the 3M3 going to gnd and the 680k between the Q4 base and the diode cathode.
   - the direction of the diode is as shown on the schematic(s):
     anode to Q3.c collector and diode cathode towards Q4.b.

There's not a lot of info on the bias voltages (I don't have my old notes and don't know if they showed voltages).

There doesn't seem to be any doubt that the 3M3 resistor should go to ground at least for older units, which goes against some of the schematics.

Some of the units which have 680k to gnd on the schematic attached to the unit actually have 3M3 to gnd on the PCB.

EDIT: I have seen a different PCB layout, probably later.  The 3M3 resistor on the schematic attached to the unit connects to ground.   The PCB wires the 3M3 resistor to ground.

I'm am yet to find a PCB from a real unit which has the 680k wired to ground.

As for getting the sound you want.  It's possible not all units used yellow dot transistor and the sound you like is actually a unit which has gains different/lower to the above unit.   Also perhaps you have high hFE transistors in your build which is pushing the bias points even further from what you like.   
EDIT: The lower gain orange dot transistors are going to give different bias voltages than the yellow dots.  I'd say enough to make the units sound a bit different.

Swapping the 680k and 3M3 to what is shown in the schematics might help - however that might just be fudging around the real problem!

Some edits were made to fill in some of the unknowns.

[Rob Strand]

Another point that occurred to me about that circuit:  The transistors Q3 and Q4 are operating at very low collector currents.  It is highly likely the in-circuit hFE is somewhat less than the nominal hFE at collector currents of 1mA to 2mA.  This is especially true for transistors back in the 60's.  The amount the hFE drop off at lower currents can have quite a bit of variability as well.

A typical value might be a little less than half the published hFE at 1mA to 2mA.  So orange dots could have low a current hFE of 54 and the yellow dots could have a low current hFE of 90.

Modern transistors often show less hFE drop off at low currents.  In some cases the hFE drop-off can be very little.  Different brands and batches show different amounts of hFE reduction.  The result is modern transistor are likely to show higher gains than the old transistors so it is likely the circuit won't bias properly with modern transistors.

Instead of trying to find magic low hFE transistors for the circuit it's probably easier to use low-ish gain transistor then decrease the collector resistor on Q3 to get the correct sound, or,  correct collector voltage on Q3.  Since we don't know what the collector voltage should be we would have to do it by ear after comparing it against a real unit.

The other transistor positions are affected as well but less so.  A lower collector resistor on Q2 is probably necessary to mimic the original units.

I mentioned earlier there are two PCB versions for this circuit.  The second/later  PCB uses different and more modern transistors.  I can't comment if these units sound better or worse than the earlier units.    We always assume the older units sound better but it this the case  here?


I dug up some old datasheets.  You can see the effect is significant.

[zbt]

I found this board on another forum, this picture is from Johann Burkard




FZ-1B but with only two transistors it seems like a different schematic

the other version from another forum, pcb also look different
https://www.ultimate-guitar.com/forum/showthread.php?t=2171191&per_page=20

[Rob Strand]

FZ-1B but with only two transistors it seems like a different schematic

the other version from another forum, pcb also look different
https://www.ultimate-guitar.com/forum/showthread.php?t=2171191&per_page=20

That's a "known" version.  Maestro call it the FZ-1B version 1 but some people refer to it as the two transistor FZ-1B.  It's actually quite different to the FZ-1B version 2 in this thread.

If you go to FSB, then find the section "Davidoff Library", you will find a thread called "Complete Maestro Service Manual, indexed PDF".  The PDF contains schematics for all the versions of Fuzz-Tones (perhaps not all correct despite the fact it's Maestro's documents).

The two versions I'm referring to with different PCBs are both FZ-1B version 2.