The Germanium Mystique

Germanium is actually a boring, silvery-gray color. Aren't you glad I added cheesy digital effects?

“Germanium” is a magical word among guitar gearheads, right up there with “pre-CBS,” “true bypass,” “matched tubes,” “point-to-point wiring,” and “scatter-wound.” And like those other phrases, it owes its cachet to a mix of fact and fancy.

Germanium was used extensively in electronic component production the 1950s and ’60s, but was abandoned in favor of silicon components, which boasted superior performance, consistency, and value. Today no one cares about rare and finicky germanium components except certain audio geeks — especially the ones who play guitars.

In the stompbox realm, you’re liable to encounter two types of germanium components: germanium diodes, which lend distortion pedals a softer, smoother tone relative to silicon diodes or LEDs, and germanium transistors, which are typically used in clones of ’60s drive and fuzz pedals. Germanium components are usually available only from specialized distributors, who inevitably sell them for many times the cost of their silicon equivalents.

Compared to silicon transistors, germaniums are . . . different. 

Ordinarily, I’m a debunker of magical “mojo” parts. Whenever a boutique pedal merchant hypes their product on the basis of some rare/antique part, my B.S. detector starts humming.

When they say, “You need to use ’70s op amps to get a good overdrive tone,” I reply:

“Bull!” :poop:

When they say, “My pedals sound fatter because I use only [tropical fish/oil-and-paper/Mesopotamian ceramic] capacitors, I cry:

“Double bull!” :poop: :poop:

When they say, “My pedal sounds warmer because I use carbon resistors.” I shriek:

“Triple crown of bull!” :poop: :poop: :poop: (P.S.: You are insane.)

But when someone says, “You can’t get the ’60s sound without old germanium transistors,” I mumble:

“Um, yeah.”

It’s difficult to describe the unique qualities of germanium transistors, though the words “smoother and spongier” seem relevant. They also have a distinctive harmonic profile. Their overtones shimmer sweetly. Feedback is rich and musical. They are unbelievably dynamic, yielding excellent and endlessly varied tones throughout the range of the guitar’s volume knob. Mmm.

Some of the classic effects that use germanium transistors.

Not all vintage fuzz pedals use germanium: Mosrite Fuzzrites, Jordan Bosstones, Foxx Tone Machines, and Electro-Harmonix Big Muffs are just a few examples of great vintage fuzzes that used silicon transistors. But the Maestro Fuzz-Tone, the Dallas Rangemaster, the original Tonebenders, and early Fuzz Faces all used germanium.

There are many good reasons why manufacturers ditched germanium. There’s much variation in quality from transistor to transistor. They’re noisy. Their tone changes with the air temperature. And the vast majority of surviving germanium transistors are of the PNP variety, which means they’re engineered for positive-ground circuits, which were common in the ’60s, but nearly non-existent today. You can’t use a pedal with PNP transistors with a conventional negative-ground power supply unless you jigger with the schematic (which is exactly what we do in the Tonefiend DIY Club project based on the Rangemaster).

(You can find NPN, i.e., negative-ground, germanium transistors, but they’re rarer and more expensive, and don’t sound any better than the PNP ones. If a) you dig the germanium sound, and b) might want to build more germanium-powered pedals in the future, I recommend getting into the PNP habit. Some online authorities claim it’s a bad idea to deploy PNP transistors in negative-ground circuits because they weren’t engineered to be used that way. I’ll spare you the sight of more “poop” icons, but suffice it to say that I’ve built many, many stompboxes of this type, and have never encountered problems. If you don’t trust me on this, maybe you can trust stompbox A-lister Robert Keeley, who uses PNP germanium transistors in a negative-ground circuit for his Java Boost pedal, one of the best Rangemaster-inspired overdrives.)

A few more highly subjective observations:

    • It’s almost always a bad idea to substitute a silicon transistor for a germanium in a schematic. You can often jigger the circuit to sound good with silicon, but silicon will almost certainly sound awful unless you adjust other values. In fact, much of the bad rep of silicon fuzzes has to do with several decades of lousy-sounding Fuzz Faces that committed precisely this sin. 

    • One of the best things about the Rangemaster (and our Fiendmaster project) is that it requires only one germanium transistor to sound awesome. Two-transistor Fuzz Faces and three-transistor Tone Benders are trickier, because you not only need to round up more fussy parts, but voice them to play well with each other. However, there are a number of updated schematics that attain a vintage germanium sound with a single germanium transistor alongside one or two silicon ones. Sometimes these sound even better than the all-germanium originals.

    • Every germanium transistor sounds a little bit different — or a lot different. And yes, there is a depressingly high percentage of crappy ones. But if you buy from reputable dealers, they’ll generally replace a dud. Some dealers even pre-test every single transistor.

    • One the other hand, different models of germanium transistor may not sound as different as you might suppose. In my experience, almost any properly functioning germanium transistor can sound great in a distortion pedal. To cite one particular example: a certain British manufacturer sells Fuzz Face clones for the equivalent of US$600, based on his use of two rare old germanium transistors. I had an opportunity to listen to one of these side-by-side with a pedal made from BYOC’s $95 ESV Fuzz Kit, with its relatively common AC127s. The pedals sounded frickin’ identical. I mean, really, really identical. The point is, you can make great pedals using the (relatively) readily available germanium transistor supply. If anyone tells you need $200 transistors for best results, well, I’m going to have to start wielding that :poop: icon again!

25 comments to The Germanium Mystique

  • Tonechaser

    I already plugged my iron in. This DIY club is the best thing since sliced bread.

  • Tom

    Without naming names or getting into who makes what, what do you think of today’s “modern” Fuzz Face clones, such as those with the BC-108 nerve centers? Are players who use those missing the boat completely? Are there ‘face clones out there today who have that “magic?”

    • joe

      Well, I have to admit I’m more partial to the germanium sound in Fuzz Faces, but a lot of folks love the BC-108/109 ones, which were widely used by the early ’70s, and which Jimi himself may have sometimes used. I don’t believe there’s anything special about that transistor — any modern transistor of similar gain yields similar results. (While BC-108s aren’t as expensive as germanium, they too are long discontinued and sell for a premium.)

      My very personal and subjective take: The magic of a good Fuzz Face has to do with its phenomenal dynamics, which are less evident with a silicon transistor. But if you’re less concerned with the ability of the circuit to clean up a LOT when you back of the guitar volume, you might be real happy with the silicon variant. Though you might be even happier with a good Tone Bender…

      • I agree on the Fuzz Face. I have an original that I have had for many years. I also received a new Dunlop Fuzz Face that I got from someone for Christmas this year (Santa must really know what I like). The new Fuzz Face is actually surprisingly good  it  is the red one that uses germanium transistors not the blue one with silicon transistors. It is built on a circuit board not a point to point like my old one but it has the trimmer potentiometers and is all regular components (no surface mount devices). It also like my old one is battery power only. I have played it side by side with my old Fuzz Face and while they are different sounding the new Fuzz Face still sounds quite good in it’s own right. I would characterize it as being a bit more “immediate” in it’s response (I suppose that could be due to the aged caps and carbon comp resistors in the old Fuzz Face). I used carbon/zinc batteries in both units and even swapped the batteries half way thru testing to eliminate that variable. I was amazed there was not a bigger difference between the two Fuzz Faces. They differences are more like the differences in two fresh ground coffees of similar origins rather than coffee and tea. My benefactor was actually planning on getting me the Joe Bonamassa  Fuzz Face but could not find one available at the time which is fine with me I am not a big fan of Joe’s anyway (he’s OK but not a god IMHO). The JB Fuzz Face is supposedly optimized for humbuckers and I mostly play P90s so I think the regular Fuzz Face is better for me anyway. I have been using it inline with my regular guitar signal before it goes into the GK-3 for my Roland GR-55 when I am using it with a full range amp/spkr setup. It gives the internal FX something to chew on (I need to figure out a sling or small bag to hold it when I play standing up). This new Fuzz Face satisfies my Jones and lets the vintage unit reside safely away in my studio for recording. I would very much like to find some old school carbon/zinc batteries rather than the “heavy duty” ones you  get today as I would like to A-B them. I am of the opinion that the carbon/zinc batteries maximize the interaction with the guitars volume control which I believe is an issue of “loading the pickups” and changing the frequency response and ringing associated with the pickup’s inductance and impedance it seems to me to be more than just “cleaning up” when the volume control is turned down). I think the impedance or internal resistance  of those old style batteries was low enough to affect the interaction between the pickup, volume control and fuzz circuit. I am speaking here of the resistance across the battery’s poles rather than the series resistance that limits current flow or causes a voltage drop in circuit analysis it’s also what slowly drains the battery when not in use and reduces shelf life (old style batteries had a shelf life that was much shorter than even today’s carbon type batteries and that was possibly due to this internal resistance being lower IMO). I can definitely tell a difference  between modern alkaline and carbon/zinc batteries in this new Fuzz Face. Seems like it might be possible to emulate this with power supply mods but I have never heard of anyone doing it this would be something different than the “sag control” that is used in boutique fuzz boxes. That is a project for another day in the future.

        • joe

          Hey Matthew — thanks for you observations. The Fuzz Face is SUCH a fascinating circuit, because it’s so very interactive with the guitar, and with itself. Don’t be surprised if it shows up as a DIY Club. If what you say is true, that’s awesome that you can get decent Fuzz Faces again at a sensible price. I love, love, LOVE the Fuzz Face I made from Build Your Own Clone’s ESV Fuzz Face kit.

          I have to admit I’m real skeptical of the sonic difference between battery types — or more specifically, that any differences stem from anything other than the typically higher voltage of a brand-new alkaline battery relative to a non-alkaline. Yes, you can absolutely hear the difference between a brand new alkaline with a charge of 9.5v and anything with a lower charge. But please, prove me wrong with some audio clips! 🙂

          BTW, that sag control is real easy to add to any fuzz, and it’s especially nice with Fuzz Faces. The brilliant Dano from Beavis Audio Research wrote a great article on this technique.

          • For me the differences in batteries are more feel than sound. Like I said the FuzzFace seems even more responsive to guitar volume and tone controls like the interaction has more to do with reactance effects between the capacitance and inductance of the guitar electronics. You would never get  this interaction with active pickups or a guitar with a preamp for instance as the reactance of the guitar electronics is eliminated by those buffer circuits. I will try to record it for you but it will have to be done with a guitar as re-amping will not duplicate the changes in the guitar volume and tone controls.
            This is something I have struggled with as an electronicist working with guitar amps and FX, there is much more going on with a guitar than a simple signal source, the pickup is a coil of wire that exhibits both inductive and capacitive properties as well as the impulses they try to reproduce. Picking a guitar string on an electric guitar gives a constantly changing signal from the initial attack of the pick (or finger in my case) which gives a big spike in the signal to the sine wave like signal as the string vibrates and eventually diminishes naturally or is damped. All those portions of the string movement cause the magnetic field around the pickup to change creating the signal voltage we amplify or effect. All this is happening in a coil of wire that is an inductor with properties beyond simply reading the string. Any inductor has a certain amount of ring to it where the signal current  is creating it’s own magnetic field that in turn modifies the original signal. If you pulse an inductor you see echoes of that pulse gradually reducing in strength until it is unmeasurable there are some electronic theories that imply that they never totally disappear unless you damp them out. There is also a certain of amount of capacitance in a pickup which while measured as an overall amount for the pickup is actually a whole series of small capacitors formed between the various portions of the windings, their magnitudes are a function of the uniformity or non-uniformity of the windings and the insulating material between the metal of the windings (air,the lacquer coating insulating the wire and the wax or even epoxy used to stop vibrations resulting in “pickup squeal”). As evidence to this there are the many pickups today that are hand wound to capture the vintage sound of pickups when the winders were not as precise as they are today the “scatter winding” technique used by boutique builders and even the mismatched windings in humbuckers. Tube guitar amps are similar affected as the power and output transformers also exhibit these inductive properties, especially the older vintage designs. When manufacturers started applying modern technology to remove these and other artifacts the amps were decried as sterile and lifeless but were in fact just actually reproducing the signal more accurately. Many boutique amps today have switches or controls to reduce or eliminate the negative feedback in the power amp which is seen as good engineering by the engineers. A good example is blackfacing a silverface CBS Fender amp one of the biggest changes was removing caps on the plates of the output tubes that CBS Fender engineers put in to make the amps look better on their oscilloscopes when testing them on non-inductive dummy loads rather than speakers (to keep from going deaf). The caps stopped a parasitic oscillation that never occurred when speakers were connected because the inductive and capacitive properties (reactance) of the speakers eliminated them. But the players could hear the change right away and sold the silverfaces and went back to the old blackface pre-CBS amps (it killed the Fender sparkle and shimmer in the high end). This gave CBS Fender a big black eye in the industry. Actually all they needed to do was put a .1 uf cap in series with a 2 watt 8 ohm resistor and hang that in parallel with the non-reactive dummy load when testing to simulate the reactance of the speakers (something you always did when testing audiophile tube amps for the stereo trade).
            Most of these differences are difficult to quantify but players and listeners are able to discern them. Add to that the interaction between the player and what he/she hears when they play as a form of feedback and you have a very complex system to describe. Beyond that most players have a recognizable sound that fans can pick out regardless of the amplifiers, guitar and FX in use. Even my daughter when she was a five year old (20+ years ago) could pick out Jimi Hendrix as soon as he started playing on a tune in the first 10-15 notes and she was not fooled by imitators or songs she’d never heard previously. I was one proud Papa when she awed my friends shouting Jimi Hendrix in her little kid voice (she’s still a big Hendrix fan). When I listen to recordings of myself where I have a lot of FX on my guitar I am always a bit amazed at how little difference they can make. Of course you can “play to the FX” which accentuates the FX sound but when I relax into a jam it usually minimizes the FX sound.
            It seemed to me at one time that perhaps a mechanical guitar picking machine might help solve some of these issues but in fact will solve nowhere near all of them. That is really what makes an electric guitar such a wonderful instrument, there are so many little things going on that in a way it is like playing a new device every time you pick it up. There is so much subtlety and nuance involved to allow your artistic expression and individuality be maximized and that is what makes electric guitars the greatest instrument ever devised IMHO. Just my 2 cents worth.

  • Tom

    I recently got a ‘Face clone for the first time since playing guitar since the early ’90s, and it was a revelation, doing things that I’ve never heard a distortion or overdrive pedal do. It was like the amp tripled in size, and the pedal was sitting back, smiling and saying, ‘Now you just play guitar, son, and we’ll take it from here.’ Big fun.

    • joe

      I know exactly what you mean, Tom! I talk a lot here about the value of dynamic pedals that respond differently according to how you play. Another way of saying that is, a lot of overdrives and fuzzes have so much damn gain, your signal is always compressed. Your tone implodes when it should explode. Mine isn’t a “retro for retro’s sake” attitude — the simplest, most dynamic circuits just sound livelier, more complex, and less fatiguing to the ear.

  • dave

    Note: Early Mosrite Fuzzrites used germanium….and they sound way better. Similarly, one of the greatest Fuzzrite variants, The Orpheum, was available as ge and si.
    Not all vintage fuzz pedals use germanium: Mosrite Fuzzrites, Jordan Bosstones, Foxx Tone Machines, and Electro-Harmonix Big Muffs are just a few examples of great vintage fuzzes that used silicon transistors. But the Maestro Fuzz-Tone, the Dallas Rangemaster, the original Tonebenders, and early Fuzz Faces all used germanium.

  • erik

    Joe, can you devise a project with a cheap piezo sensor and simplest hi-Z amplifier. “Impossible Guitar: Concrete Block & Piezo” has dispelled the myth of the guitar wood role. The next legend is the role of a special piezo sensor. FIIO makes cheap and good headphone amps, they possibly can be made hi impedance.

  • Elliot

    stumbled in here researching germanium transistors, needless to say I will be frequenting this site regularly in the future.

    but I was hoping you could elaborate on your comments about using PNP germaniums in a negative ground circuit. Im in love with germanium and want to build as many fuzzes drives and boosters as I can with them, but I greatly dislike using batteries and have enough power supplies to hang myself. I’d love to just daisy-chain in with the rest of my board if at all possible.

    I used to have a Dunlop germanium Fuzz Face and also agree that for the price it sounds fantastic. what I did was file a slot in the bottom and stuck a Visual Sound 1-Spot battery clip adapter in it. this worked great, not even any extra noise that I could detect, but its a bit cumbersome. it would be nice to use a standard 9volt boss-style plug in future projects.

    • joe

      Hi Elliot — I’m glad you find some of this stuff useful!

      This is kind of confusing, but…all the CURRENT projects on this site work with regular 9v power supplies, even the PNP-transistor Fiendmaster. It’s schematic has been jiggered to compensate. If you compare its schematic to that of the original Rangemaster, you’ll see similar components and values, but they’ve been sort of “flipped upside down.” Some folks with far more tech knowledge than I warn against deploying PNP transistors this way, but I call BS. Many good boutique pedals use this technique, and I’ve built dozens and dozens of pedals this way, all of which sound great and perform consistently.

      But why bother doing it that way? Why not just use NPN in the first place? First, it’s way easier to find vintage PNP transistors than NPN ones, and the Fiendmaster was specifically designed for the easiest-to-find germanium transistors, the PNP AC-128. But now companies like Mammoth have come up with new production NPN transistors that sound GREAT. (At least, the dozens I’ve used have been great, though some users have reported encountering bad ones, which, of course, any reputable vendor will promptly replace.) So if I were to revisit the Fiendmaster circuit, I’d probably switch to an NPN schematic.

      The other reason: Many pedal geeks really want an exact-as-possible replica of the ’60s originals, and all those original Fuzz Faces, Tone Benders were PNP/positive ground. Manufacturers didn’t really commit to negative-ground circuits until around the time silicon replaces germanium in the late ’60s. I find this motivation sort of silly, but hey, whatever.

      But having said all that, the reason I emphasized the word “CURRENT” above is because the next tonefiend project (created by my pal Mitchell Hudson) will be super-vintage, dual-PNP Fuzz Face with an old-style positive ground circuit and no AC power jack. Like many modern players, I hate burning through eco-hostile batteries, but this circuit has a very low current draw, and a single battery lasts for months and months. (Plus, the pedal may sound even cooler when the battery starts to weaken.)

      Hope that helps!

      • Elliot

        Everything I’ve read so far here has been very useful! Excellent stuff here.

        I looked over the feindmaster build and breadboarded it out (fun circuit to tinker with), but I don’t quite understand the changes that you made. I would have expected to just flip the electro cap, but it seems you’ve inverted nearly the whole circuit. I guess my question is, if I wanted to “jigger” with a fuzz face circuit to make it negative ground what do I need to change? Or if I wanted to use a PNP germ in a standard negative ground circuit what kinds of alterations do I need to consider or are you saying that a PNP can be used in a circuit designed for NPN without any changes?

        • joe

          Yes — you flip the entire circuit AND reverse any polarized caps. It’ll be the same when switching between NPN and PNP Fuzz Faces. Just Google “fuzz face schematic npn” and “fuzz face schematic pnp.”

          • Elliot

            Ok, ater studying yur scedmatics and te fuzz face schematics fr a couple hours, it looks tome like everythig going to ground now goes to 9v and everything going to 9v now goes to ground (so just flip the + and – wires), then flip any polarized caps (and diodes i’m guessing. so if i wanted to use a PNP i a circuit like an Electra with the diode pair connected to ground, would those now connect to 9v or still to ground?

          • Elliot

            Sorry about that, not sure what happened. What I was trying to say was after studying your schematics and the fuzz face schematics for a couple hours, it looks to me like everything going to ground now goes to 9v and everything going to 9v now goes to ground (so just flip the + and – wires) then flip any polarized caps. (and diodes I’m guessing…) so if I wanted to use a PNP in a circuit like an Electra with the diode pair connected to ground, would those now connect to 9v or still to ground?

  • Digital Larry

    Hi Elliot, I found an “Electra” fuzz schematic that has a couple back to back diodes to ground on the other side of a capacitor from the transistor. I would keep the grounded end of that pair connected to ground. It is a signal path rather than a power/biasing path.

    This circuit depends mostly on the diodes for its distortion. I would expect less “Germanium” character from replacing the transistor than if it were a pure transistor circuit.

    That said, don’t let me stop you from trying. You may just hit upon the next big thing!

  • Digital Larry

    Transistor fuzz circuits where the emitter (the lead with the arrow) is connected directly to power or ground are the most sensitive to the transistor type and characteristics.

    The Electra circuit posted here:

    has a 680 ohm resistor between the emitter lead and ground. This provides “negative feedback” which reduces the stage gain and makes the overall circuit less dependent on the actual transistor used. I didn’t say “not at all dependent”, just “less”.

  • Elliot

    Thanks Digital Larry, that’s the same circuit I was working off of. So I swapped all the +/-connections, and based on your info about the transistor emitter I removed the 680 resistor to get more gain from the transistor (AC128) which sounded rather similar to the fiendmaster build (no diodes at this point so I could hear the just the transistors gain). Then I added a 100K pot between the collector and the 3.3K resistor for gain control. Finally adding two 1N34 germanium diodes to ground and… I’m in love! Though it could use a tone control imo. Worth a try if you like sweet germy goodness, it’s like the ultimate germanium fuzz.

  • I’m not so convinced that you couldn’t engineer a silicone based fuzz to sound identical to a germanium. It would be really hard to compare the 2 otherwise since they would have a different range of operation and require different biasing. How did you compare the two in your experiments.
    Not trying to argue for the sake of it. I’ve actually wondered about this so I’m curious.

    • joe

      My testing has consisted simply of building them and listening to them. I think I’ll whip up a non-Ge one and add it to this mix for reference.

  • Re: Graveyard of dead batteries.

    A sag circuit really does something. It makes tone more mellow, “spongier”. I called it “syrupy” for the fuzz face.

    If this adds additional compression as sag does in a tube circuit, I don´t know.

    A circuit that does exactly that has been developed by Tone God. He calls it “punisher” a dynamic power supply sag simulator.

    On fuzz face tone in general, I believe (after studying R.G. Keen´s “reference” article) that the circuit topology of the fuzz face itself coins an iconic tone, except they fouled it up in using monstrous gain transistors. Some even darlingtons :shake:

    But I have noticed several times that people who re-issue effects based upon proven designs sometimes create the impression that not too much brainwork has gone into the redesign despite a hefty price tag, so you end up with effects that are far below what they could be.

Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>




Click to upload a JPG

This site uses Akismet to reduce spam. Learn how your comment data is processed.