Death by Audio Fuzz War schematic

[gigimarga]

No problem, these misunderstandings happen often when people from different backgrounds get together, whose first language is not always English. It's very cool to have people from all over the world on here!

"Laid-back" means calm and relaxed - I just meant that people will help you if you just relax and wait a little while.

Thx a lot Taylor for your kindness
After I wrote the answer I realized that my logic is out again (here it's 3:00 AM): thinking to Smokie's song "Laid back in the arms of someone" it's hard to believe that "laid back" can to mean what I supposed...

[gigimarga]

I gave up...I tried to bias every transistor...it sounds better with 430K/100K and 820K/180K pairs...and without 27K resistor and the pair of the Ge diodes it sounds like crap.

[analogguru]

Why don´t you connect the transistors as shown in the schematic ?
Did you see with your own eyes that the transistors are 2N 5089 ?  I didn´t.
What if they are BC 549´s ?
And if they would be marked as 2N 5089 I would measure them for pinout before I believe that they are reversed.

analogguru

[gigimarga]

Why don´t you connect the transistors as shown in the schematic ?
Did you see with your own eyes that the transistors are 2N 5089 ?  I didn´t.
What if they are BC 549´s ?
And if they would be marked as 2N 5089 I would measure them for pinout before I believe that they are reversed.

analogguru

Thx a lot for your advice!
I will make this tryout tonight...

[gigimarga]

I reversed all 7 transistors (emitters to ground and collectors to V+) and all is dead

The voltages, using a 9.00V battery, are now the following:

1. all emitters at 0.03V
2. all bases at 0.56V
3. Q1's collector varies from 0.05V (gain pot to minimum) to 4.82V (gain pot to maximum)
4. Q2, Q4, Q5 collectors' at 0.03V
5. Q3's collector at 0.05V
6. Q6's collector at 0.3V
7. Q7's collector at 0.02V

Maybe I am stupid, but it seems to me that this Fuzz War is made from 7 cascaded LPBs with the collectors resistors increased 10 or 18 times...do you think that's possible that they marked in a liar manner? 

Thx again all!

[Paul Marossy]

do you think that's possible that they marked in a liar manner?

Possible. Might be one of those "dirty little secrets" that RG mentions at GEO. 

[Gus]

The bias used is not what I would "design" a circuit with.
  You need to understand the pros and cons of the different bias methods used with transistors

[analogguru]

The bias used is not what I would "design" a circuit with.
  You need to understand the pros and cons of the different bias methods used with transistors

analogguru wrote at the other forum on Sun Jun 22, 2008 9:14 pm

That´s the big problem with Fuzzes:
Nearly with every sh*t you can achieve a (distorted) fuzz sound.

And this design is sh*t... The DBA-guy should visit an electronics school before he starts producing and selling.

Since I don´t want to waste my time by analyzing, here is the:
Death By Audio - Fuzz War schematic

As you can see, it is very temperature sensitive and if you would assemble R 19, the complete bias of this stage will be changed at higher levels.

Sorry that I didn´t draw jacks and switching for this junk.

and later:

In short:
Every transistor stage is temperature sensitive.
As a countermeasure you have the possibility of current feedback or voltage feedback.

Current feedback is achieved by a resistor between emitter and ground. The resistor must be significant compared to the collector resistor - 180k to 390 ohm is a relation of 1 : 40000 and therefore not significant.

Voltage feedback is achieved with a resistor between collector and base - like it can be seen in an EH Big Muff Pi stage. (If the collector voltage increases, the base voltage gets higher too by this reducing the collector voltage).

[gigimarga]

Thx a lot both of you, Gus and analogguru!
The biasing techique is still a great unknown for me, but I'm working on it.

Anyway, my great problem on Fuzz War remains: "Why in the soundclips it sounds great and my clone, made after a good traced schematic sounds like crap?"
Maybe sometimes what you see isn't the truth 

[Gus]

What temperature was the original built at and what hfe transistors were used? 

Look at page 71 and 72 in "The Art of Electronics".   Also read pages 83 and 84.

I just went to google books and typed in "The Art of Electronics"  I could read the pages I posted.

[gigimarga]

What temperature was the original built at and what hfe transistors were used? 

Look at page 71 and 72 in "The Art of Electronics".   Also read pages 83 and 84.

I just went to google books and typed in "The Art of Electronics"  I could read the pages I posted.

Thx Gus...I will read the pages soon!

[teemuk]

The way I see it, the transistors are intentionally reversed, which still makes them work like ordinary NPN transistors (the junctions still have the polarity of a NPN transistor), except they now have the junctions reversed and will not work as ideally. Generally, the reversing causes the device to have a way lower speed and beta (gain) - the latter property, I think, is kind of what you need when you cascade that many gain stages together.

[gigimarga]

Thx teemuk...I am almost sure that they are reversed (from their site, regarding SOUNDWAVE BREAKDOWN): "...It uses specially maatched transistors turned backwards to get it's bombastic sound...", but I think that's a clue how to bias them because, as analogguru and Gus said yet, this biasing is very instable.

Thx a lot all!

[gigimarga]

Playing around only with one stage with reversed transistor of the first five I think that it sounds much better with 56K-82K from emitter to Vcc instead of 100K/180K...

[gigimarga]

Because I have seriously problems with my head, I replaced all 180K with 200K trimpots and, using a 1P12T rotary switch to run around all 7 stages of Fuzz War, I biased by ear every stage in the next mode: first stage, first two stages, first three stages,...

I didn't use the Ge diodes from the sixth stage because it sounded too smooth with it and I have enough smooth fuzzes

After about 2 hours of tweaking it sounded OK, but i still have seriuosly problems with it:

1. the "Drive" pot must to be a trimpot because it acts very harshly (abruptly)
2. as analogguru and Gus said before it's very instable biased (I touch a trimpot a little by mistake and it sounded like crap)
3. the decay of notes, after a long sustain that it has, it's strange sometimes (like in a Dynacomp)

By the way, anyone knows another raw and compressed fuzz like that?

The trick seems to be to use transistors with some gain because it matters very much, as aron said in his article about reversed biasing.

Thx a lot all for your help!

[analogguru]

Maybe the Seymour Duncan - Power Grid could do the trick ?

analogguru

[gigimarga]

Maybe the Seymour Duncan - Power Grid could do the trick ?

analogguru

Thx a lot! On youtube it seems to sound OK, but I will try to test a real one...

[DougH]

Maybe you could try cascading  a number of  properly (i.e. stable) biased stages together on a breadboard to see what it sounds like. You could reverse C-E as on the FW to slow it down etc and see what it sounds like. Someone mentioned verbiage of "specially matched transistors". That implies that he's looking for specific hfe values in the transistors he uses. Without that info, this 'design' is useless as it won't bias properly unless hfe (and temp) is correct. Which is why it is such a p.o.s. to begin with.

[analogguru]

Maybe you could try cascading  a number of  properly (i.e. stable) biased stages together....

Maybe you won´t believe it, but  exactly this is done inside the power-grid.  5 or 6 stages with a gain of 10 each, low pass filtering between the stages.

analogguru

[gigimarga]

Maybe you could try cascading  a number of  properly (i.e. stable) biased stages together....

Maybe you won´t believe it, but  exactly this is done inside the power-grid.  5 or 6 stages with a gain of 10 each, low pass filtering between the stages.

analogguru

Do you have a schematic of it?