Mojo Ge trannies

[makaze808]

Hello. I've just had a long breadboarding session and was shocked how poor some Ge's sounded in some circuits, similar gains and leaks but a horrible tone from them. I really thought that maybe all the mojoness of certain Ge's was a little hype but now I suspect not.

So a simple question, can a NPN Rangemaster sound as good as a PNP OC44 build, are the OC44 builds that much better?

Why does a trannie sound so different from other trannies.

I suspect I'm in for an extended shopping spree.

Cheers, Woz.

[GREEN FUZ]

Isn`t it all down to biasing?

[frank_p]

You have to measure their characteristics.  Google technology of the Fuzz-Face at GEO and go see how to use and build the recommended tester.  Then if you do a search on this forum, you'll have tons of other info (bias, etc.).  Measure your transistors first before going crazy on shopping.

[darron]

i think the noise figures can be upsetting sometimes comparing Ge to Si. There's some truth to the mojo in that Ge's saturate in a totally different way. They do perform differently when you stat pushing them to their limits. I could be wrong, but they seem amplifier upper harmonics a little bit better.

[frank_p]

Ah! Sorry I misinterpreted the post, but:  Thanks for asking that (not so) noob question.  GE transistors varies a lot from one and an other: that can cause a lot of frustration.  But the (real) question is (I suppose)  what characteristics are we are searching for in a Ge transistor  that is (so) different from a Si one (without the inconveniences inherent from them)...

The same question I should ask myself...

[Solidhex]

Hello. I've just had a long breadboarding session and was shocked how poor some Ge's sounded in some circuits, similar gains and leaks but a horrible tone from them. I really thought that maybe all the mojoness of certain Ge's was a little hype but now I suspect not.

So a simple question, can a NPN Rangemaster sound as good as a PNP OC44 build, are the OC44 builds that much better?

Why does a trannie sound so different from other trannies.

I suspect I'm in for an extended shopping spree.

Cheers, Woz.

  This is a good question. We're not talking about silicon vs germanium or external biasing here. He wants to know why germaniums of different types sound better in some circuits while having near the same gains/leakage. The question is what other characteristics about germanium transistors specifically contribute to their tone? Unfortunately I don't have an answer haha but hopefully someone can shed some light on it 

--Brad

[darron]

like i said, i think the main differences are that they saturate totally differently, they may have different effects on harmonics. can anyone advise on research for linearity?

[makaze808]

Thanks for the reply's. I love to be able to see in my mind what a mojo tranny is doing that is different for another. I hope someone can shed some light.

I've shopped for a NOS 2n404a as I'm desperate to build a brick (doug hammonds') and I have no doubt that without it it doesn't do something and that something is something special.

[brett]

Hi
there are several things that a vintage Ge BJT does differently to a modern Si.
(i) the hFE is much lower (a few times lower for SS devices).  For some circuits (e.g. FF), this makes the input impedance low, loading th pickups
(ii) leakage (current from the collector to base) is much higher (100 to 1,000,000 times higher).  This allows leakage-current biasing (e.g. FZ-1)
(iii) the base-emitter voltage drop is lower (0.3 V vs 0.7 V).  This makes distortion asymetrical in some circuits (e.g. FF)
(iv)  the transition frequency, Ft, is much lower.  Often only 0.5 to 1 MHz (compared with 100MHz).  This may reduce gain at high frequencies.
...some other things I can't remember just now...?? 
cheers
PS Noise is higher, but that isn't usually of any benefit.

[Solidhex]

Yo

  Just to keep this thread on track. We're not talking about silicon vs germanium or comparing the two. Its strictly a question about germanium transistors. The question posed in beginning of the thread which I'd be interested in hearing a discussion about is what characteristics other than leakage and gain contribute to the tone. Maybe its something that can't be answered for sure but not all germaniums sound the same in circuits. Even when their gains and leakage are similar. What other info on datasheets will give us insight into that?

--Brad

[brett]

Hi
sorry I didn't read the topic thoroughly.  The answers have some similarities, though.

RE: gain. 
There is hFE and there is gain.  e.g. Q2 in a fuzzface produces less gain than its hFE.  Also, every device has a hFE-Ic (maximumSS gain-collector current) relationship.  Some are "bendy", with a big rise in hFE as Ic goes up, and others are flatter.  I suspect that bendy devices sound much better, and that the circuits that employ a range of dynamic Ics (e.g. the FF) are using this characteristic to advantage.

RE: leakage
Many people get very worried about leakage.  I don't (too much).  You can't build an original FZ-1 without leakage.

The transition frequency (and related Miller capacitance) probably affects highs in some circuits.  How important is it?  Important enough that it is emulated in the Axis Face and other circuits with a cap bypassing the collector resistor.

cheers

[makaze808]

Hi, for me hearing how poor sounding some devices were, no matter what I did really made me wonder how authentic substituted trannies are.
I have built some Rangemasters for freinds with generic devices and they sound similar and good, well no one is complaining. But I haven't heard a OC44 Ge rangemaster.

But now I want to buiild a tone bender, wanting to avoid bipolar power issues I looked straight for the NPN version, at fuzz central. while the "2n404 trannies I now know are used in some good pedals i just wonder how close it would sound to one built to the original; specs. I know the answer is breadboard and see, but if the answer is mainly it sounds ok, then I don'tt see the point.

In my quest for a BRICK, I have been assured from voices who have the respect on here from those who are experienced that without the specified trannies and diodes any result is not capable of prducing the effect that a spec'd  the circuit can produce.

[brett]

Hi
the human mind is an amazing thing.  If you expect differences, they will be there.  Always. 
That's mojo.

PS some people have also commented that Japanese devices (2SBXXX) and Russion devices (GTXXXA) are much more consistent and higher quality (narrow hFE, low leakage).  I few have ventured to use them and declare that they sound good, too.  Maybe, just maybe, the emperor has no clothes !

[bigchasbroon]

i have 4 ac128s I am build a rangemaster and I reckon ill build it then try all four and see is theres a difference.

[jrod]

Hey makaze808!
What kind of poor qualities are you hearing in different devices? Are you sure you have them biased correctly?
I ask because I have spent hours  upon hours swapping different devices in a Fuzz Face circuit. I have tried different gain, leakage, different manufacturers of the same part number, 2n's, GT's 2SB's, OC's etc etc etc and can hear only very very subtile differences, except when they are biased differently. I have found that ANY substitute works in a Fuzz Face (at least) as long as the hfe and leakage are ok.

Thanks,
jrod

[brett]

Hi

ac128s

These are some of the most variable devices ever made.  About 1/4 of NOS either don't work at all or are virtually useless (hFE<20 or leakage >1000uA).  In a rangemaster, working AC128s will sound different because the hFE is all over the place (30 to 130).  If you have devices with a narrow range of hFE (say, 80 to 100) that are not too leaky, I believe that they are likely to sound very similar.  But why not check it out and let us know?

Don't forget - Without a double blind test, you'll get the result you expected (ie meaningless).
cheers