Fuzz box guitar pedal project

Thread Starter

Roob

Joined Mar 21, 2015
38
Hi All,

I built the fuzzbox pictured below. I used ZTX751 pnp transistors in place of the AC128 in the schematic (waiting for delivery of some NOS AC128). It sounds pretty nasty, very temperamental and liable to oscillation, it drops in and out of fuzz and some of the pots have a weak effect. My first pedal build so a massive success and it feels like it go well with other transistors or tweeking.

A question:

Should I have attached the emitter of the pnp transistors to the positive battery line and the collector pin to ground?
 

Attachments

ian field

Joined Oct 27, 2012
6,536
There is an offer of 25 matched pnp/npn pairs of these untested on ebay, am I likely to get 50 duff ones? I've heard they are unreliable...

They are new old stock from the 70s, the rectangular heat sink k variety...
In what alternate dimension do untested matched pairs exist?
 

Thread Starter

Roob

Joined Mar 21, 2015
38
That is a very good point. The ad says they don't know the matching parameters. I feel that these untested germanium resistors, which are readily available may be tested and found to have the low hfe values that none of the pedal heads want and then sold as the untested so that poor mugs then buy in the hope of finding a good one. I've spent £10 or so on some untested ac128 and I'll test them and report back.
 

Thread Starter

Roob

Joined Mar 21, 2015
38
This is the ad:

10 vintage germanium transistors AC128
,
metal case.

PNP, Manufacturer: TUNGSRAM

New old stock (NOS), not tested, not used, sold as it is.

For vintage radio purpose or guitar effects
(fuzz-face, treble booster)

Will ship worldwide!

We'll see....

Question- how do I measure the hfe?
 

ian field

Joined Oct 27, 2012
6,536
This is the ad:

10 vintage germanium transistors AC128
,
metal case.

PNP, Manufacturer: TUNGSRAM

New old stock (NOS), not tested, not used, sold as it is.

For vintage radio purpose or guitar effects
(fuzz-face, treble booster)

Will ship worldwide!

We'll see....

Question- how do I measure the hfe?
There was a tester published many years ago in Radio Constructor that nulled out the collector leakage before measuring the gain.

If I could remember the author's name, a search would probably be easy.

An archive of RC is on the web - but not in PDF form.
 

Dodgydave

Joined Jun 22, 2012
11,285
You can use 2n3906 as a substitute for the Ac128, they are silicon pnp, with a hfe 140+.


to measure the hfe you need a meter, some DVMs have them built in.
 

ian field

Joined Oct 27, 2012
6,536
You can use 2n3906 as a substitute for the Ac128, they are silicon pnp, with a hfe 140+.


to measure the hfe you need a meter, some DVMs have them built in.
Many guitarists absolutely insist on genuine germanium transistors - and yes, they can tell the difference!

A faint possibility of getting surplus germaniums is NOS AF11x family types such as the AF116 etc.

These parts are particularly prone to tin whiskers growing inside and shorting to the can. They're typically RF/IF transistors and have a 4th lead to earth the can. When the short occurs and stops it working, some bodgers simply snipped the screen lead, others strap the EBC leads together and zap the screen lead with 47uF charged to 50V or so, both cures have issues so most repairers took to substituting BF451.

Chances are any NOS of the AF11x parts will have whisker shorts as well - but for audio frequency use, you can just ignore the screen lead anyway.
 

Thecomedian

Joined Oct 12, 2013
22
http://www.geofex.com/article_folders/fuzzface/fffram.htm

We can understand this today. 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.
Does this still matter in this circuit? I think it might, since the maker decided on using AC128's for the leakage and gain in them, as well as the voltage drop of Ge compared to Si playing a role. The trick afaik is to try out varying gains of transistor in those two key spots, where the fuzz action is supposed to happen, and decide which sounds best.

http://www.zvex.com/fuzz.html

The geofex link has a simple two transistor tester for leaky germanium.

You can use 2n3906 as a substitute for the Ac128, they are silicon pnp, with a hfe 140+.


to measure the hfe you need a meter, some DVMs have them built in.
fuzzcentral.ssguitar.com/axisface.php

Transistor Selection
For Q1 I followed Jack Orman's "YAFF" lead and used a PN2369A NPN Silicon transistor. The gain of this transistor ranges from 40 up to 120...sounds a lot like the good ol' Germanium transistors, huh? This transistor will work great in the Q1 position. For Q2 I'm using the BD139 NPN Silicon transistor, as suggested by Brett Robinson. The gains of these are usually around 130, which will work perfectly for Q2. These low-gain transistors will help to keep the fuzz from being too harsh, which is a common problem with Fuzz Face-type circuits that use modern high-gain transistors like the MPSA18.
Having too high a gain can actually be problematic in the original circuit, of which the direct-coupled stage is a relative, with the AC128s and everything. I have no personal time with this specific circuit, so couldn't say whether replacing them with higher gain transistors sounds better or worse, but AC128s are not cheap, so they probably are in that circuit for reasons.
 
Last edited:

Dodgydave

Joined Jun 22, 2012
11,285
They were use a lot in the 70'S, cos they made the best sound due to the fact there made from Germanium doped with Arsenic, so dont put them in your mouth...
 

ian field

Joined Oct 27, 2012
6,536
http://www.geofex.com/article_folders/fuzzface/fffram.htm



Does this still matter in this circuit? I think it might, since the maker decided on using AC128's for the leakage and gain in them, as well as the voltage drop of Ge compared to Si playing a role..
Its not so much the voltage drop on a germanium junction as the softer knee on the germanium curve.

An article I read recently showed the scope traces from diode clippers made with germanium, silicon, and Shottky devices.

The germanium clipper sort of pinched the top of the AC waveform narrower with a pointy top - silicon and Schottky pretty much produced squarewaves.
 
Top