Relative Importance of Gain in Fuzz Face

Frequently asked question on transistors: types, substitutions, how to test, use and misuse them.

Relative Importance of Gain in Fuzz Face

Postby modman » 26 Aug 2007, 22:54

No, sorry there a no myths to debunk, even after RG's Technology article. Because as with the wah technology article, in de midst of his mathetical deductions, there always a weak mojo or "sweet spot." Or it's something we heard from Fuller.

In the wah article, Mr Mojo appears in the guise of a faulty inductor which interacts postively with a fuzz face and can never be recaptured cause no modern inductor has those faulty characteristic anymore (except Teese?). In the Fuzz Face article, the ultimate dictum most people put over their bed is the magic hfe range for fuzz face. This is the crucial paragraph:

R.G. Technology of the Fuzz Face wrote:I've done a lot of circuit simulation on the FF, twiddling the values of the transistor gains, and looking at the clipping waveforms and resulting harmonic spectra. There is a definite sweet spot for musical sounding clipping at transistor gains of about 80-110. If you allow combinations of one high and one low gain device, the range widens out to 70 or so on the low end and perhaps 130 on the high end. Keeping in mind that preferences for distortion tone are definitely a matter of personal taste, the range of gains for unselected AC128's in this circuit would produce some really clunky-sounding devices.

This seems to be borne out in practice. Mike Fuller, maker of the Fulltone "69" pedal, posted his preferences for Fuzz Face transistor gain to the usenet news groups, and they fall right in this range. He noted that he feels that he can affect the relative amount of symetrical versus asymetrical distortion by selecting for non-identical gains in the two positions. (Mike also prefers transistors with only certain colored epoxy sealant, which I can't see making any difference except coincidentally, but then, who knows?)


I've really been looking for that sweet spot, but from the beginning builders were reporting great ff's with silicons and extremely high hFE's, initally even Jack Orman thought low gain to be essential to FF sound - though all the time people were building great si fuzzes with enormous betas:

11/18/1997 5:29 AM Jack Orman - Ampage
I've heard from people who made the FFace with 2N3906 trans. and were satisfied with the sound. However, its hfe of 200 sounds like a little much. I've heard that the best sound comes with the first transistor hfe= 85 to 95 and the second one at 100+. The MPSA18 has an hfe of 500! But then again, the BC108C trans. that are in my original 1972 FF also have an hfe of 500...  
 
The AC128 PNP Ge's that were in the original model rate out at 90.
(...)  
regards, Jack


11/19/1997 R.G.
Every MPSA18 I've ever measured has had an Hfe over 1000.  
 
The "two different gains" data was also posted by Mike Fuller about the FF some time back on the newsgroups. The first one can be as low as 70 and the second can be up to 140 (assuming that this is real gain, and not leakage as measured on a DMM gain test). In theory, the mismatch changes the amount of asymetry in the clipping.

the first transistor's properties dominate the "soft" distortion sounds, the second one's dominate the "harder" distortion settings. The soft distortion is the result of the mushy saturation of the voltage feedback first stage, and the harder distortion is the result of the cutoff clipping of both first and second transistors.  

(...)
 
Ge is indeed leaky. My last batch of AC128's was 75% unusable because of leakiness. Leakiness is exactly the same effect as putting a large value resistor from the collector supply to the transistor base, it simply adds extra current into the base - and hence the collector - of the transistor. As a result, the bias points shift. If this were the only problem, you could deal with it by tweaking the bias point. In fact, you see this in many "Golden Age" fuzzes, which have Ge transistors with NO bias into the base; it's not zero biased, the leakage is biasing it. The real problem is that the leakage shifts with temperature, and this fact makes speculation on the effect on sound unnecessary.


I conclude from this that the prime requirement for a transistor to function well in an original fuzz face circuit is not the correct gain, but rather the correct amount of leakage to bias the base.

Further evidence that a certain amount of gain is not imperative to create a fuzz effect, can be found on the other end of the spectrum. It's the famous Piggybacking thread

[quote="Brett @ diystompboxes:"Piggybacking" trannies for lower gain
« on: 06-01-2004"]Simple. Or so I thought.... Well, my Hornet had Q1 and Q2 hFEs of 130 and 340, so I piggybacked then with 2 extra tyrannies of hFE = 200 and 340, respectively. The Hornet was now more of a Wasp, and sounded great. I bundled it up and sent it off to the guy I was building it for. Then I started to wonder what I had done to the hFE of those transistors. A few checks revealed some things that STUNNED the *^%@ out of me.

I found that piggybacking two similar transistors (PN100s, hFE=350 and 430) to those I'd used for Q2 in the Hornet gave a hFE of either 6 or 12, depending on which transistor piggybacked on which. So piggybacking DIDN'T HALVE THE GAIN. Also, the Hornet worked great with a hFE at Q2 of about 10. (What!!??)[/quote]

Adding a resistor between the emitters, allows you to increase the gain. I did some experiments with some various silicon transitors

    Take two identical transistors, tie the BASES together, cut off one COLLECTOR, connect a 3k to 6k resistor to the EMITTER of the collectorless devices, and connect that to the EMITTER of the other transistor. By varying the resistance, you can dial in the gain you want... It is my experience that identical transistor work better in getting a good fuzz sound.


Without the resistors, you would get hfe's in the range of Brett's (5-15). Below graphs shows the effect of the resistor on the gain.

Image

So, if we can mimic both leakage (big resistor between B and C) and low gain with silicon, where's that germanium mojo? This is a question I've been asking myself some time now. The piggybacking experimenters at diy were happy to find low gain silicons and settled for that. Which is fine, but...

Is there reason to assume germanium still has something to offer? Is this a path worth persuing?

just throwing a bone, hope nobody gets hit :D
Last edited by modman on 26 Aug 2007, 23:21, edited 1 time in total.
"The whole point of diy is diy. It's not dsoiyathodtr - do some of it yourself and then have others do the rest" (paulc)

For this message the author modman has received thanks:
sinner (10 Jul 2012, 09:44)
User avatar
modman
a d m i n
 
Posts: 4301
Joined: 19 Jun 2007, 17:57
Has thanked: 3108 times
Have thanks: 549 times

Postby soulsonic » 26 Aug 2007, 23:03

I have been pondering these very same questions lately. Let the experiments begin!
User avatar
soulsonic
Old Solderhand
 
Posts: 3530
Joined: 27 Jun 2007, 04:38
Location: Buffalo, NY
Has thanked: 56 times
Have thanks: 269 times

Postby MoreCowbell » 26 Aug 2007, 23:25

Its funny you bring this up, as yesterday I was reading through some of those same exact threads and coming to similar conclusions / hypothesis. I've been able to replicate some of germanium "sag" and "thickness" with silicon devices, but it's something I'm still not 100% satisfied with yet. As soon as some space is freed up on the breadboard I'm gonna start back up with the experiments.
MoreCowbell
Transistor Tuner
 
Posts: 1107
Joined: 06 Aug 2007, 15:39
Has thanked: 0 time
Have thanks: 11 times

Postby vanessa » 27 Aug 2007, 00:03

I was always fascinated by this but never got a chance to experiment with. The end of that thread sort of died out with little success if I remember. But it seemed there was something there that could be the building block of something.

The only thing I've really found that germaniums have over si's besides lower hfe are they tend to have a brighter dare say wider less compressed tone over si's, but not by much. Maybe a cap adjustment could open things up a bit?

Modman are you saying that tying the two transistors together in series with a variable resistor yielded you higher gains than bret's?
vanessa
Cap Cooler
 
Posts: 468
Joined: 08 Jul 2007, 18:51
Has thanked: 0 time
Have thanks: 10 times

Postby modman » 27 Aug 2007, 00:18

Yep, click the graph for bigger view.

it shows the effect of raising the value of the resistor between the emitters - higher resistance, higher gain. The graphs shows that if you 2-3k for Q1 and 5-7k for Q2, you'll have "perfect" GE gain value on your SI Fuzz Face.

How that for ease of boutique production? :wink:

Whether that's enough for "GE sound" is a further question.

zj
"The whole point of diy is diy. It's not dsoiyathodtr - do some of it yourself and then have others do the rest" (paulc)
User avatar
modman
a d m i n
 
Posts: 4301
Joined: 19 Jun 2007, 17:57
Has thanked: 3108 times
Have thanks: 549 times

Postby JHS » 29 Aug 2007, 18:54

The RM classic Fuzz has trannies with rel. low hFE, put in some BC549C and you have the Stonefuzz.

You can put in every trannie in a FF, Ge, Si, FET, Mosfet, Darligton or mixed and it will sound, as long if the BIAS for the trannies is set to the correct point. Maybe some parts must be slightly altered to match those high hFE trannies.

It's such a simple circuit, so breadboard it, play with the values and the trannies and let your ears decide what sounds best...

JHS
JHS
Cap Cooler
 
Posts: 484
Joined: 14 Jul 2007, 12:20
Has thanked: 0 time
Have thanks: 3 times

Postby snail » 30 Aug 2007, 15:00

Don't Germs have a differente frequency responce to them compared to SI?? I guess this is also part of the equation to the FF sound. Some Germs will reject bass while others will reject treble and so on, so I feel that somehow they also act as a filter, voicing the fuzz according to their own (faulty) characteristics.

snail
snail
Solder Soldier
 
Posts: 202
Joined: 30 Aug 2007, 01:41
Has thanked: 0 time
Have thanks: 1 time

Postby soulsonic » 30 Aug 2007, 22:03

I agree. Some germaniums had really significant capacitances that has a definite effect on frequency response. For example, an AC128 has a maximum Collector capacitance of 200pf and a bandwidth of 1Mhz. An OC44 has a maximum Collector capacitance of only 14pF and a bandwidth of 8Mhz. Compare that to a 2N3906 which has a Collector capacitance of 5pF and a bandwidth of 250Mhz!
User avatar
soulsonic
Old Solderhand
 
Posts: 3530
Joined: 27 Jun 2007, 04:38
Location: Buffalo, NY
Has thanked: 56 times
Have thanks: 269 times

Postby analogguru » 30 Aug 2007, 22:18

An OC44 has a maximum Collector capacitance of only 14pF and a bandwidth of 8Mhz.


From where to you take this data ?
And...: As far as I know the HF-datas are only valid in a base configuration...

analogguru
There´s a sucker born every minute - and too many of them end up in the bootweak pedal biz.
analogguru
Old Solderhand
 
Posts: 3285
Joined: 26 Jun 2007, 14:58
Has thanked: 0 time
Have thanks: 116 times

Postby soulsonic » 30 Aug 2007, 22:39

That data comes from "Towers' International Transistor Selector" published by TAB Books in 1975.
The Collector Capacitance (COB) is listed as being measured "Normally Emitter Open Circuit"
The High Frequency bandwidth (FT) is defined as "Frequency at which Common-Emitter current gain drops to unity".

I just wrote what the book says, sorry if it doesn't match up with reality. It still proves the point as far as the discussion is concerned.
User avatar
soulsonic
Old Solderhand
 
Posts: 3530
Joined: 27 Jun 2007, 04:38
Location: Buffalo, NY
Has thanked: 56 times
Have thanks: 269 times

Re: Relative Importance of Gain in Fuzz Face

Postby HydrozeenElectronics » 11 Feb 2009, 03:45

Great article. I piggybacked a couple of 2n5088's in A FF Configuration and it sounded great!

thank you so much.
User avatar
HydrozeenElectronics
Resistor Ronker
 
Posts: 269
Joined: 11 Feb 2009, 01:56
Location: Vancouver Washington
Has thanked: 1 time
Have thanks: 0 time

Re: Relative Importance of Gain in Fuzz Face

Postby Mugshot » 06 Jun 2009, 10:02

raising this thread! any development over this? :whappen:
signatures are for the weak.

FiveseveN wrote:
considering the low quality of the transistors and caps

Are Asian electrons inferior to American electrons?
User avatar
Mugshot
Breadboard Brother
 
Posts: 97
Joined: 18 Mar 2009, 08:17
Has thanked: 1 time
Have thanks: 1 time

Re: Relative Importance of Gain in Fuzz Face

Postby DrNomis » 04 Mar 2011, 13:07

For what it's worth, I did some experiments with transistors with different gains in my Germanium Fuzz face build, I had one transistor with a gain of 108 as Q1, and another one with a gain of 119 as Q2, I got a low gain bluesy sound with not much sustain to it, next I tried using a Ge transistor with a gain of 139 as Q1 and another one with a gain of 148 as Q2, this gave me more sustain.... :)

Also, from my experiments here's what I found, not that in all cases Q2 was biased to have -4.5V on it's collector:

Q1's gain affects clipping symmetry, the more gain, the more asymmetrical the clipping becomes.

Q2's gain affects the onset/hardness of clipping, and the amount of sustain, the more gain, the harder the clipping and the longer the note played sustains.... :)


I tend to like high gains in my Fuzz Face builds..... :)

Another couple of things that I have noticed is:

The Q1 Collector resistor tends to increase the voltage on Q2, if it is made larger in value, it also has an effect on clipping symmetry.

Q2's collector resistor decreases the voltage on Q2's collector if it is made larger in value, but, it also has an effect on the output level.

The 100k feedback resistor may also have an effect on Q1's collector voltage.
Genius is not all about 99% perspiration, and 1% inspiration - sometimes the solution is staring you right in the face.-Frequencycentral.
User avatar
DrNomis
Old Solderhand
 
Posts: 6886
Joined: 16 Jul 2009, 05:56
Location: Darwin,Northern Territory Australia
Has thanked: 123 times
Have thanks: 397 times

Re: Relative Importance of Gain in Fuzz Face

Postby Hides-His-Eyes » 04 Mar 2011, 14:17

Nowhere in this thread is the lower forward voltage of Ges mentionned; I was under the impression that could affect the asymmetry?
Testing, testing, won too fwee
User avatar
Hides-His-Eyes
Tube Twister
 
Posts: 2015
Joined: 02 Feb 2010, 13:34
Has thanked: 53 times
Have thanks: 84 times

Re: Relative Importance of Gain in Fuzz Face

Postby DrNomis » 04 Mar 2011, 16:35

Hides-His-Eyes wrote:Nowhere in this thread is the lower forward voltage of Ges mentionned; I was under the impression that could affect the asymmetry?




You have a point there, Hides-His-Eyes...... :)


To be honest, I didn't see any effects on asymmetry caused by the lower forward voltages of Ge transistors in my experiments, I did a quick A/B of the clipping generated by two Fuzz Faces, one had Ge transistors in it, the other had Si transistors in it, and guess what?, the sharpness of the clipping was practically the same..... :)


The funny thing is, although the clipping looks identical in both cases, the Si transistors definitely sounded harsher than the Ge transistors to my ears, although I made sure that both Fuzz Faces had the same control settings... :)
Genius is not all about 99% perspiration, and 1% inspiration - sometimes the solution is staring you right in the face.-Frequencycentral.
User avatar
DrNomis
Old Solderhand
 
Posts: 6886
Joined: 16 Jul 2009, 05:56
Location: Darwin,Northern Territory Australia
Has thanked: 123 times
Have thanks: 397 times


Return to Transistors

Who is online

Users browsing this forum: No registered users and 0 guests