Russian GE transistors for dummies

[diy-tubes]

I'm trying to dig in Russian GE transistors theme, and think about collecting some info in one thread. Now I've got good collection of them and I have read many threads from runet (russian internet) about using nos ge transistors, measurements, successfull applications etc etc
Please let me know if it will be interesting. Being not a guru at all, I have some notes, I think our comrades from Russia (ex-Soviet Union) (gtlab.net, guitarplayer.ru, guitar.ru, vegalab.ru etc) will follow and help.
PS: Many thanks to Paul (chromesphere), sharing his experience, and great youtube channel!

[digi2t]

Excellent resource; http://transistor-spravochnik.ru/

Don't let all the Cyrillic scare you. Just use the search box in the upper left, and use Latin characters. Or use a translator.

[diy-tubes]

Here is my list (name, original name, hfe according to datasheet)
Russian letters (as it goes in Russian alphabet):
А=A
Б=B
B=V
Г=G
Д=D
Е=E
И=i (I)
Ж=J (ZH)
Л=L

Etymology of names: russian letter П (english P) meant transistor (полупроводник), it's early tradition from 60-s, next - МП (english: MP), modernized transistor, and ГТ (GT) and 1T (1T) germanium transistor (tradition of naming based on transistor material).
1T (not IT!) is a kind of grade (military grade, slected etc). Sometimes even usual ge transistors (П, МП, ГТ) have a rombus marking - OTK marking or military grade, selected version.
Next appears silicon transistors - their names starts with KT (english KT) and 2T (graded version, military etc, in russian tradition - "5" ("military") acceptance).
Example of KT transistor is famous sub for 2N5088: russian KT3102E with hfe 400-1000 used in russian BigMuff's.
Numbers after letter - number of design and specifications.

GT series (ГТ)
Many of them (not all) have 1T version, wich is the same but selected, graded version, sometimes even with mentioned in datasheet higher Hfe)

• GT108V (ГТ108В) - hfe 60-130
  GT108G (ГТ108Г) - hfe 110-250
  GT109E (ГТ109Е) - hfe 50-100
  GT109G (ГТ109Г) - hfe 50-100
  GT109V (ГТ109В) - hfe 50-100
  GT115V (ГТ115В) - hfe 60-150
  GT115G (ГТ115Г) - hfe 60-150
  GT124B (ГТ124Б) - hfe 71-162
  GT124V (ГТ124В) - hfe 120-200
  GT308B (ГТ308Б) - hfe 50-120
  GT308V (ГТ308В) - hfe 80-150
  GT308G (ГТ308Г) - hfe 90-200
  GT309B (ГТ309Б) - hfe 60-180
  GT310G (ГТ310Г) - hfe 60-120
  GT310E (ГТ310E) - hfe 60-120
  GT311D (ГТ311Д) - hfe 60-180
  GT311i (ГТ311И) - hfe 100-300
  GT311k (ГТ311K) - hfe 60-180
  GT311L (ГТ311л) - hfe 150-300
  GT402B (ГТ402Б) - hfe 60-150
  GT402G (ГТ402Г) - hfe 60-150
  GT402E (ГТ402E) - hfe 60-150
  GT402i (ГТ402И) - hfe 60-150
  GT403B (ГТ402Б) - hfe 50-150
  GT403G (ГТ402Г) - hfe 50-150
  GT403D (ГТ402Д) - hfe 50-150
  GT404B (ГТ404Б) - hfe 60-150
  GT404G (ГТ404Г) - hfe 60-150
  GT404E (ГТ404Е) - hfe 60-150

P series (П) seems to be old series.
Low noise GE transistors: П27, П27А, П27Б, П28 suitable for first input stage.

• P27 (П27) - hfe 20-90
  P27A (П27А) - hfe 20-60
  P27B (П27Б) - hfe 42-126
  P28 (П28) - hfe 33-100
  P307B (П307Б) - hfe 50-150
  P307V (П307В) - hfe 50-150
  P414A (П414А) - hfe 60-120
  P414B (П414Б) - hfe 100-200
  P415A (П415А) - hfe 60-120
  P415B (П415Б) - hfe 100-200
  P416A (П416А) - hfe 60-125
  P416B (П416Б) - hfe 90-200

MP series (МП)
Low noise GE transistors: МП13Б, МП39Б  suitable for first input stage.
General usage, low frequency - МП21, МП40, МП41, МП42

• MP21A (МП21А) - hfe 50-150
  MP25B (МП21Б) - hfe 30-80
  MP39B (МП21Б) - hfe 20-60
  MP42B (МП42Б) - hfe 45-100
  MP15A (МП15А) - hfe 50-100
  MP16B (МП16Б) - hfe 45-100

The letters are given to distinguish them by Hfe, max collector-emitter voltage (like ГТ402Б (25V) and ГТ402И (40v)) etc
Many of them hard to find even in Russia (like popular ГТ402И (GT402i) which is good choice for Fuzz Factory clones).
Some of them have сomplementary types: ГТ402 (pnp) and ГТ404 (npn).

PS: Feel free to fix my mistakes, and add notes.

[DDD]

May I recommend you to put special attention to:
- GT308V (ГТ308В)
- MP39B (МП39Б)
As per my experience they are actually low-noise and stable ones.

[thermionix]

Cool thread, thanks for sharing this information.

[DrAlx]

May I recommend you to put special attention to:
- GT308V (ГТ308В)
- MP39B (МП39Б)
As per my experience they are actually low-noise and stable ones.

I had different experience with the GT308V.
The hFE mentioned in the list above seems to be off the datasheet.
But I think the datasheet also shows you only get those hFE when running at large collector currents (MUCH larger currents than a typical fuzz). See this post...
http://www.diystompboxes.com/smfforum/index.php?topic=108518.msg989605#msg989605

[digi2t]

I used the GT402/GT404 combo in my Schumann Two Face, and they sound great. It actually sounded better than jerm's original. Also used them in a Gemimi Dual Fuzz. Woolier than the AC128's, but not to the point of losing definition. Good all around.

Another great one, though tougher to find, is the MP20B. Those have some pretty good bark and bite in a fuzz.

EDIT:

Where the hFe numbers are concerned, for our purposes (fuzz pedals), always consider the lowest value of the stated range as what you would expect to get. Actually take another 10 - 20% off of that. I've tested a metric buttload of these Russian transistors, and while they always do quite well for leakage, they always come in low on my Peak meter. As such, I always order the higher gain versions to get transistors that come into the right zone, i.e. you want 100 to 120, then get the gain models spec'd at 90 to 200, not 60 to 180, etc.

[diy-tubes]

I had different experience with the GT308V.
.... See this post...
http://www.diystompboxes.com/smfforum/index.php?topic=108518.msg989605#msg989605

Yes, it's another story: Russian GE Сookbook I'll try to collect some usefull info about using russians in various schemas and make somethink like meta: will post links here.
Here is an example (Russian clone of fuzz face from famous russian radiohobby magazine "Radio"):
http://www.sugardas.lt/%7Eigoramps/article29/04.jpg
I think tricks with trimpots in collector chain and various nominals in base divider will do the job.
Some transistors are good for classic FuzzFace (like low gain MP42B) others for FuzzFactory  GT402 (B, i, G)

You're right, the hfe shown in my list is from datasheets, and more current gives more gain.
Here's another post with the diagram from PRR:
http://www.diystompboxes.com/smfforum/index.php?topic=99881.msg877117#msg877117
What about Fuzz Factory - I'm not sure about 1T308, but high hfe GT402 (B, i, G) many people liked in it.
I think something like P416 or GT311 may be useful here. Need some tests.

[duck_arse]

there is also this thread from a little while back:
http://www.diystompboxes.com/smfforum/index.php?topic=108054.0

diy-tubes, welcome to the forum (the greeting manners around here are slipping a little lately).

[digi2t]

there is also this thread from a little while back:
http://www.diystompboxes.com/smfforum/index.php?topic=108054.0

diy-tubes, welcome to the forum (the greeting manners around here are slipping a little lately).

OUCH!!! Papa duck has spoken!

Welcome to the forum diy-tubes!! 

[diy-tubes]

Thank you guys!!! 

So let it be not a flood I will add some links here. Nothing new, but a good place for noob to start.

Measuring GE transistors.

Sure the first link everybody knows from R.G. Keen with his method of measuring and why it is so important
In two words - germanium transistors has emitter-collector leackage and when we measure them with usual multimeter it's not taken in a result).
http://www.geofex.com/article_folders/fuzzface/fftech.htm

More info from Steve Daniels, Small Bear Electronics LLC
I'd say it a little bit more noob friendly (I mean his test rig).
http://diy.smallbearelec.com/HowTos/FuzzFaceFAQ/FFFAQ.htm

Simple method using usual multimeter like DT830
http://qrx.narod.ru/izm/dt830_i.htm
(It's in Russian so in a nutshell: insert collector and emitter (base is not connected!) in transistor socket and measure hfe (it's leackage). Then measure hfe with base inserted. Subtract "full" Hfe and leackage and get "real" hfe.

Sure, Peak DCA55 (and DCA75) measures germanium transistors and leackage
http://www.peakelec.co.uk/acatalog/jz_dca55.html

China clones (МК-168, MK-328, GM328A) of German Atmega based transistor tester. (Must measure ICE0 parameter! It's leackage). Make a search at aliexpress or ebay to find them.
http://www.diystompboxes.com/smfforum/index.php?topic=108076.0

Russian transistor testers: Л2 series. Everyone knows about famous Л3-3 (L3-3) tester, but in SU there were transistor testers like L2-23
http://www.rw6ase.narod.ru/00/prib/l2_23.html

Exellent youtube video from Paul (His Germanium Transistor Guide For Fuzz Pedals part2)
https://www.youtube.com/watch?v=kjG8eDVRieA

Folks, please let me know if I missed something!

[diy-tubes]

A pic showing meausrements of some russian germanium transistors with GM328

Morden versions of atmega328 transistor testers must meausre leackage Ice0 and give correct Hfe (with some notes) as discribed at docs
GM328A and other cheap versions with firmware higher than 1.10k must work out of box.
In theory, we can use any type of such china transistor tester with correct firmware and enough memory.
I've purchased DCA55 and GM328, will try them both.

[diy-tubes]

More about testers: always wanted to know how call theese probes and sockets from transistor testers.
So even if we build RG Keen or smallbearelec testers, it's comfortable to use probes similar to DCA55 probes and zif sockets used in chinese testers.
Google 'Clamp hook probe' and 'zif socket' you'l find everything we need. I think i will make zif socket addon for my DCA55 and probes to GM328 (just solder them to pins 1,2,3 under the socket).
Same things a re very useful when you build test rig for fuzz pedals projects. 
BTW very good transistor sockets suitable for germanium transistors with long legs are at russian Л2-23 tester front panel.
Do you still use sip or dip sockets? 

[diy-tubes]

More great info from Smallbearelec: how to coock germanium transistors for your fuzz pedal.
Breadboarding basics need for us to build testing rig:
http://diy.smallbearelec.com/HowTos/Breadboarding/BreadboardIntro.htm
Next simple and famous Rangemaster schematics:
http://diy.smallbearelec.com/HowTos/BreadboardRMs/BreadboardRMs.htm
And again FuzzFace (and idea of composing Darlingtone transistor allows to use nearly any of GE transstors we can get):
http://diy.smallbearelec.com/HowTos/BreadboardGeDarlingtonFFs/BreadboardGeDarlingtonFFs.htm
Speaking of Russians we can use GE transistors with low Hfe but low noise and designed for low freq amplification, widely used in Soviet audio gear: P27, P27B, P28, MP20, MP21A, MP42B, MP39B, MP13B (and many others with good tone, low leackage but not high Hfe).

[diy-tubes]

Hi folks, looked at 2N404 datasheet: hfe 24 min, what is common measurements for popular western transistors: 2n404, AC128, AC127 etc?

PS: Many ebay lots call MP16A a substitute for 2N404 (but I'd call MP16B with hfe 45-100 better choice), just want to compare russians with western analogues. Many popular things like 2N404 were developed as switching devices (sure, not only), but usable in sonic domain. Many russians like P27, MP11, MP39 are designed for low freq amplification, sometimes announced as low noise, but with lower hfe. Good choice for first stage or darlingtone?

[diy-tubes]

Finally I've collected all transistor testers I've looked for, and here is the short test. What we've got: Small Bear jig (as a reference), Chinese T3-T4 and GM328 transistor testers based on mega328 project and Peak Atlas DCA55, Peak Atlas DCA75.
I've measured the same transistor (Russian P29A germanium transistor), and what I've got:

Small Bear Jig

I've build it in 1590DD enclosure with two milliampermeters and external banana sockets for external multimeter connection (DC, mA). Switches: p-n-p/OFF/n-p-n, 100uA/500uA/external, leakage/Hfe.

Measure leakage and get about 20uA

Now collector current (480uA)

And the same measured via external multimeter:

Let's calculate: Collector Current - Leakage current and divide to 9mA: 480-20/9=51.1 So we've got Hfe about 51.

Chinese GM328 and T3 (T4).
GM328 (1.12k firmware). Hfe 53 (it considers leakage (Ice0) 24uA)

T3 (T4) (stock firmware) Same Hfe 53


Peak Atlas DCA55/DCA75

Peak Atlas DCA55 gives us same Hfe=53, leakage=0.023mA (23uA)



Peak Atlas DCA75 gives us next measurement:
PNP Germanium BJT
Red-B Green-E Blue-C
hFE=64 at Ic=5,01mA
Vbe=0,344V at Ib=5,00mA
VceSat=0,021V at Ic=5,0mA and Ib=1,00mA
IcLeak=0,022mA   

But DCA75 is pro gear with curve tracing features, so we can simulate SmallBear jig (as it's shown in Paul's video:
https://www.youtube.com/watch?v=qIrHZm6067o

Here is the plot:

Vce=9,00V curve is the SmallBear test (Ib, two points: 4uA and 9uA; Vce=9v). More accurate Hfe=51,51 at 0,4887mA

What we've got: more or less all the testers can do the job. I like DCA55 and DCA75 more, to my taste they look more professional. DCA75 can simulate conditions of real schematics, and it's really very cool thing.

Speaking of FuzzFace (and other Ge fuzzes) - what are transistors operation mode (Ib and Vce)?

[ilcaccillo]

Hi, thank you so much for your post, it's a great test.

I have both the DC75 and aan ATMega based Chinese meter.

I'm a newbie could you please explain why did the DC75 measured higher HFE than any of the other meters?

It seems that the Small bear jig, Chinese GM328 T3 and DC55 all measure the same value, why is the DC75 value higher and how can we interpret that when measuring with that unit?

Or what meter should we follow for simple Germanium transistor selection?

Thank you so much

[diy-tubes]

DCA75 measures transistors with higher Ic (DCA75 Ic=5mA and DCA55 Ic=2.5mA) it gives a bit higher measurements. Sure DCA can make measurements with Ic like DCA55 when using curve tracing.
Chinese do the job similar to DCA55. I think we can use any fo rough measurements.DCA75 is good for simulating "real" conditions (within it's limits of 12v etc).
Please take a look at groupdiy discussion, i've asked for background of the question, it may be useful.
https://groupdiy.com/index.php?topic=65803.msg834158#msg834158
Thank you for your feedback! I hope someone will correct me if i'm mistaken.

[ilcaccillo]

DCA75 measures transistors with higher Ic (DCA75 Ic=5mA and DCA55 Ic=2.5mA) it gives a bit higher measurements. .DCA75 is good for simulating "real" conditions (within it's limits of 12v etc).

In terms of the Ic used in the HFE and leakage measurement what are the meters that are closer to the "real" conditions in Fuzz pedals, the DCA55 and Chinese at Ic=2.5mA or the DCA75(without using curve tracing) at Ic=5mA ?

thanks

[ilcaccillo]

I measured different transistors today using 3 different testers the Peak DCA75, the Chinese MK-328 and the RG Keen Geofx circuit and method.

With the Peak DCA75 I always have higher HFE than any of the other 2, Leakage measurments are always roughly the same between the DCA75 and MK-328

The strange thing is the Geofx circuit, it always, and always measures a much lower HFE value than any other tester, and always gives a much higher leakage value, sometimes 2x more.
Circuit can be found here:
http://www.geofex.com/article_folders/ffselect.htm
Taking into account this circuit alone you would be trashing lots and lots of transistors that have perfectly fine values in the other testers. This circuit really makes the transistors look bad, probably make them look worst than what they really are.
The circuit measures HFE at a base current of 4.046uA   

So what measurements should I trust for Fuzz pedals?
whats going on with the Geofx circuit why does it make transistors look so bad? Is there any chance any assumption might be wrong in the testing circuit or that it maybe is outdated?

thank you