High power Germanium power amplifier (any thoughts??)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
C

CaterpilarSK

Member
Joined 2018
I dont think anyone has been in my situation more of because these very high power germanium transistors are kind of a rarity and there is not many of them at all..

I got a hold of a factory matched pair TESLA 7NU74. TO3 style 50W PNP transistors. Uce 70V with a Rbe of 30 ohms at least thats what the datasheet states (yep thats right none of that low residual current silicon rubbish).

I couldnt find ANY high power germanium amplifier schematics. What I plan to do is salvage a well known silicon based amplifier schematic and adapt the power output section for the 7NU74s (PNPs) unlike for the bootstrapped NPNs.
More likely I will avoid any feedback.

I will use silicon semiconductors for doing the preamplification and the drive and the bias (transistor diode bias). Ill be using BD139,BD140 to do the driving of the finals (7NU74) and probably do some easy preamplifying stage with a BC547 and a current source so I have stron signal going to the drivers aswell. At least thats what is my plan

If anyone has a actual schematic that would be a high power germanium amplifier (30W and so on) I would appreciate if you would show it to me and let me know your thoughts and ideas on this. I have not come up with a schematic yet, but its on the list of things to do.

Specs of the granny tranny:
-Ucb: 90V
-Uce: 70V (with the 30ohm Rbe)
-Ueb: 15V
-Ic: 15A
-Icbo max: 1000uA
-Ucb: 6V
Hfe: 50-130
Ie: 10A
-Uces max: 1V
Ft: 150kHz

Got a little tired of looking at the transistors in the box so I want to do some adventure stuff with them :).

Edit: I have no intentions in using any kind of iron in my build. No transformers beside the power supply.
 
Last edited:
indianajo

indianajo

Member
Joined 2010
Look for a geloso G-253-hf amplificatore stereofonico. Schematic has been on here, no use posting it again. Uses ac128 ac129 as drivers and ac139 quasicomp pnp outputs.
Better use for these is as repair parts for some old person's blown up hifi. They didn't last very long.
 
C

CaterpilarSK

Member
Joined 2018
Well wow. Even symmetrically powered ones! Not bad. I'll look into it tomorrow in my crude simulator. Unfortunately I don't have many GE transistors besides this pair of pretty big ones.

I can't really do the preamp and driver out of germaniums and I'll try my best to avoid that. Thank you kind sir.

A note: I'll be likely to use BD139 and BD140 as the drivers. Majority of amplifiers use the "variable diode" technique of biasing the output stage. Germanium transistors as far I know have rather low junction voltage drops unlike the semicoductor counterparts. I wonder if I can still stick to the variable diode or I if I should go with the potentiometer that would be connecting the two bases together of the drivers (BD139,140). I still have the junction drops of the drivers there to deal with so I think it should be okay to bias the thing with the variable diode method (or amplified diode... The transistor with the pot).

Thank you in advance
 
indianajo

indianajo

Member
Joined 2010
CaterpilarSK said:
A note: I'll be likely to use BD139 and BD140 as the drivers. Majority of amplifiers use the "variable diode" technique of biasing the output stage. Germanium transistors as far I know have rather low junction voltage drops unlike the semicoductor counterparts. I wonder if I can still stick to the variable diode or I if I should go with the potentiometer that would be connecting the two bases together of the drivers (BD139,140). I still have the junction drops of the drivers there to deal with so I think it should be okay to bias the thing with the variable diode method (or amplified diode... The transistor with the pot).
Click to expand...
The geloso G-243 used a 500 ohm resistor between the bases of the germanium drivers. The pull up driver had 68 ohms to centerline, the pull down driver had 10 ohms. The pull down driver had 68 ohms to negative rail.
Class B was popular in germanium power push pull amps. People were still buying radios with 10% HD in those days.
 
Last edited:
JMFahey

JMFahey

Member
Joined 2009
Sadly G244 HF is an EL84 based tube amp.

So far your BEST bet is Hohner´s Premier 30 amplifier suggested by OldDY

Your current germanium transistors are about 50% stronger than original 2N4241 so you have an adequate safety margin, look no further.

Do not modify biasing, get the proper NTC resistor, it´sout of range of any possible "amplified diode" you can kludge today.

Oh, and the 30 ohm BE suggsted is not loss, by far, but an external BE connected resistor to absorb high voltageb leaks (which are way smaller) and allow it to reach 70Vce.

FWIW a 2N3055 needs 100 ohm BE for the same reason, again to reach 70Vce.
 
M

Mark Tillotson

Member
Joined 2018
Just be sure to have good thermal management. If the die limit is 70C, I suggest aiming for a case max temperature of 40C, and implement thermal protection. TO3 sockets are a good idea to prevent any over-temperature risk from soldering - the bond wires in a 15A device can transfer a lot of heat.
 
JMFahey

JMFahey

Member
Joined 2009
OldDIY said:
...bias the thing with the variable diode method - 2 diode.
Click to expand...
Sorry but no, not here.
First you would need **germanium** diodes or at least a Germanium transistor as thermal sensor but in any case, sensing a junction is not enough, Germanium power transistors NEED "overcompensation" because they are extremely easy to go into thermal runaway, not only Vbe changes with temperature but also leak increases and that is not even detected, so classic compensation relies on Thermistors which allow to "build" any compenastion curve you need.

I have actually designed and built Germanium output amplifiers in the late 60´s and remember there was an equation to calculate the proper network and optimum Thermistor.

No, I don´t remember it by heart (50 years ago :eek:) although I might rewrite it myself from scratch if needed.

But you already have a working solution if you build the Hohner amplifier!!!!
Don´t reinvent the wheel, specially to build "just one".

FWIW the Leak amplifier suggested above ALSO uses Thermistor compensation :D , it was the accepted solution way back then.
Even early Silicon amplifiers used Thermistor compensation (think Vox, Acoustic, Standel or Philips-Fapesa-Miniwatt amplifiers.

Usable with transformer driven amps where diodes were not. (thermistors are bidirectional while diodes are not).
 
PRR

PRR

Member
Joined 2003
Paid Member

Attachments

  • GoogleGePowAmp-42.gif
    GoogleGePowAmp-42.gif
    142.8 KB · Views: 584
U

U130421

Disabled Account
Joined 2015
attachment.php

attachment.php


This is my first amplifier built. It figured in a booklet by Valvo transistors . The outputs AD149, were the most powerful ones available then in Beirut. It delivers 20w with 4 ohms.
 

Attachments

  • ad149_umgeb.png
    ad149_umgeb.png
    22.1 KB · Views: 3,039
  • ad149.JPG
    ad149.JPG
    61.3 KB · Views: 4,842
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom