What Germanium PNP transistor do I use?

[ian field]

I thought people talked about power germaniums still being made and used.
The talk would have been in one of the sci.electronics.* newsgroups
sometime
in the past decade, but that's the best I can provide as a pointer.

Michael

Way back in the early days when silicon devices were new, I often wondered -
if silicon is so good, why is germanium still used in the power stages of
radios that have silicon RF/IF stages?

 

[Joerg]

Way back in the early days when silicon devices were new, I often wondered -
if silicon is so good, why is germanium still used in the power stages of
radios that have silicon RF/IF stages?

Old rule: As long as Ge was only one cent less than Si, they'd use Ge.
Maybe the fact that many cars were still equipped with 6V systems also
favored Ge.

 

[ian field]

Old rule: As long as Ge was only one cent less than Si, they'd use Ge.
Maybe the fact that many cars were still equipped with 6V systems also
favored Ge.

The 12V auto-electrics were well and truly established in the period I
remember, and don't forget the first ever transistor radio used a 22-1/2V
(tube) hearing aid battery - presumably the PP3 hadn't been invented yet.

 

[Joerg]

Joerg ([email protected]) writes:



But otherwise, it's a question like "where do I get a tunnel diode, I want
to build this circuit I found in a book or on the internet...". And the
poster doesn't realize how old the schematic is, or that tunnel diodes
at their best were mostly a novelty (with some real exceptions) and for
the hobbyist lasted about a decade (to stretch things a bit).

In order to stick with the circuit, they will have to work really hard to
find a tunnel diode, and may pay good money for it. And before that,
they really should evaluate why they want to make that circuit. IN most
cases, it's because they don't know better (because they want to build
something simple, and it certainly is a simple circuit).

As a kid I always laughed when I saw yet another circuit proposal with a
tunnel diode or a UJT. None of the stores had any, not even the big ones
in large cities. Looked like someone just had to publish it to show off.
I built my oscillators with scavenged AF139/239. Or sometimes tubes.

I thought people talked about power germaniums still being made and used.
The talk would have been in one of the sci.electronics.* newsgroups sometime
in the past decade, but that's the best I can provide as a pointer.

Most of what people thinks is fresh production is in reality just old
stock. When it's gone it's gone, plus they often want top Dollar.

 

[ian field]

As a kid I always laughed when I saw yet another circuit proposal with a
tunnel diode or a UJT. None of the stores had any, not even the big ones
in large cities. Looked like someone just had to publish it to show off. I
built my oscillators with scavenged AF139/239. Or sometimes tubes.

A few years ago I had a project published in Television Magazine which was
loosely based on a Programmable Unijunction Transistor (PUT).

The project met the need for testing opto-couplers that did not offer an
electrical connection to the photo-transistor's base terminal making the
usual "diode" tests impossible, this required a cheap simple and minimum
parts count pulse generator to drive the LED, granted regular UJTs are very
rare these days so initially I looked at the PUT which is basically a
thyristor with a negative (anode end) gate, my first thought is that an SCS
(silicon controlled switch) - a 4-layer thyristor device with both gates
available would be relatively easy for a TV engineer to obtain, and failing
that, one can be constructed easily with a complementary pair of small
signal transistors.

However as I looked at the circuit - and the complementary asymmetry of the
4-layer device, I realised that it was possible to turn the circuit upside
down and use a small TO-92 thyristor to the same effect. Thyristors of this
type are quite common in TV & monitor PSUs - not least as safety trip
devices, so almost by accident the project became a dual purpose component
tester.

Don't be too hasty to count out the tunnel diode - according the manual, the
Tek465 scope contains a couple of them.

 

[Rich Grise]

But otherwise, it's a question like "where do I get a tunnel diode, I want
to build this circuit I found in a book or on the internet...". And the
poster doesn't realize how old the schematic is, or that tunnel diodes
at their best were mostly a novelty (with some real exceptions) and for
the hobbyist lasted about a decade (to stretch things a bit).

In order to stick with the circuit, they will have to work really hard to
find a tunnel diode, and may pay good money for it. And before that,
they really should evaluate why they want to make that circuit. IN most
cases, it's because they don't know better (because they want to build
something simple, and it certainly is a simple circuit).

I thought people talked about power germaniums still being made and used.
The talk would have been in one of the sci.electronics.* newsgroups sometime
in the past decade, but that's the best I can provide as a pointer.

Apparently, they're available _SOME_where - at least the audiophools seem
to think so:
http://www.chandlerlimited.com/products/germanium_tceq.php

(although, a glance at the price tag kinda makes me think they've set
up their own custom fab line or something ;-) )

Are schottkys not good enough?

Cheers!
Rich

 

[Rich Grise]

Way back in the early days when silicon devices were new, I often wondered -
if silicon is so good, why is germanium still used in the power stages of
radios that have silicon RF/IF stages?

At the time, Ge was the only way that they could make cost-effective
power PNPs. They still had some bugs to work out with making high-power
Si PNPs. (they've apparently figured it out sometime in the last 30
years or so, however. ;-) )

Hope This Helps!
Rich

 

[Michael Black]

Way back in the early days when silicon devices were new, I often wondered -
if silicon is so good, why is germanium still used in the power stages of
radios that have silicon RF/IF stages?

I thought that was due to the curve of a new device. Transistors started
out as being good for not much more than audio, and low power at that.
They worked at it, until there were better frequency capability and they
could actually build audio output amplifiers out of the transistors. Witness
all those hybrid radios (not just car, but there were early lunchbox style
walkie talkies that were hybrid), where the audio stages were transistorized
and maybe some of the lower frequency stages, while tubes remained in
most of the radio stages.

Silicon came along, and they had to start all over, working at improving
their frequency range and power capability (though likely the curve wasn't
as slow since they'd practiced on germanium). By then, they could make
fully transistorized radios, but not always with silicon.

Look at the early transistor audio amplifiers, and they were basically
copies of tube amplifiers, ie transformer or RC coupled, with an output
transformer. A few years later, they were DC coupled, and depending
on the design may not have even had a coupling capacitor at the output.

Michael

 

[ian field]

At the time, Ge was the only way that they could make cost-effective
power PNPs. They still had some bugs to work out with making high-power
Si PNPs. (they've apparently figured it out sometime in the last 30
years or so, however. ;-) )

Hope This Helps!
Rich

In the period I'm thinking of transformer coupled O/P stages still hadn't
been completely outdated by complementary pairs so they could have chosen
whichever polarity suited best - it just seemed to take a while to develop
silicon transistors that could supersede germanium in power stages.

 

[Michael Black]

A few years ago I had a project published in Television Magazine which was
loosely based on a Programmable Unijunction Transistor (PUT).

The project met the need for testing opto-couplers that did not offer an
electrical connection to the photo-transistor's base terminal making the
usual "diode" tests impossible, this required a cheap simple and minimum
parts count pulse generator to drive the LED, granted regular UJTs are very
rare these days so initially I looked at the PUT which is basically a
thyristor with a negative (anode end) gate, my first thought is that an SCS
(silicon controlled switch) - a 4-layer thyristor device with both gates
available would be relatively easy for a TV engineer to obtain, and failing
that, one can be constructed easily with a complementary pair of small
signal transistors.

However as I looked at the circuit - and the complementary asymmetry of the
4-layer device, I realised that it was possible to turn the circuit upside
down and use a small TO-92 thyristor to the same effect. Thyristors of this
type are quite common in TV & monitor PSUs - not least as safety trip
devices, so almost by accident the project became a dual purpose component
tester.

Don't be too hasty to count out the tunnel diode - according the manual, the
Tek465 scope contains a couple of them.

But that was one of the exceptions. IN the hobby realm, they were a novelty
and nothing that was published could not be done with some other device.

In that scope, it's either in the trigger system, or it's generating a pulse
(I forget, though I know at some point I did read what the tunnel diode
was doing).

ANd yes, if you needed to replace that tunnel diode, you'd need a tunnel
diode. But, even for that, there seems to be a fair consensus that getting
a tunnel diode at this point would require effort at the very least.

Michael

 

[Michael Black]

In the period I'm thinking of transformer coupled O/P stages still hadn't
been completely outdated by complementary pairs so they could have chosen
whichever polarity suited best - it just seemed to take a while to develop
silicon transistors that could supersede germanium in power stages.

There may have even been some chasm to leap to NPN. So many of the early
circuits did use PNP transistors that one might suspect there was some
problem in making NPN. But perhaps it was just lethargy, they started out
with PNP and kept with them up to a certain point. Or those positive
ground cars made PNP a better choice at the time.

Michael

 

[Joerg]

The 12V auto-electrics were well and truly established in the period I
remember, and don't forget the first ever transistor radio used a 22-1/2V
(tube) hearing aid battery - presumably the PP3 hadn't been invented yet.

Then you must be young. My first car was built in 1969 (I got it around
1980). A Citroen 2CV, 6V electrics all through out. Number of electronic
components in that car before installing the radio with its Ge
transistors: Zilch. Not even a diode.

 

[Joerg]

But that was one of the exceptions. IN the hobby realm, they were a novelty
and nothing that was published could not be done with some other device.

In that scope, it's either in the trigger system, or it's generating a pulse
(I forget, though I know at some point I did read what the tunnel diode
was doing).

ANd yes, if you needed to replace that tunnel diode, you'd need a tunnel
diode. But, even for that, there seems to be a fair consensus that getting
a tunnel diode at this point would require effort at the very least.

Effort plus a well padded bank account. Unless you can find a similar
scope with a shot CRT on EBay for parts. But then shipping will eat your
money.

 

[Joerg]

A few years ago I had a project published in Television Magazine which was
loosely based on a Programmable Unijunction Transistor (PUT).

The project met the need for testing opto-couplers that did not offer an
electrical connection to the photo-transistor's base terminal making the
usual "diode" tests impossible, this required a cheap simple and minimum
parts count pulse generator to drive the LED, granted regular UJTs are very
rare these days so initially I looked at the PUT which is basically a
thyristor with a negative (anode end) gate, my first thought is that an SCS
(silicon controlled switch) - a 4-layer thyristor device with both gates
available would be relatively easy for a TV engineer to obtain, and failing
that, one can be constructed easily with a complementary pair of small
signal transistors.

However as I looked at the circuit - and the complementary asymmetry of the
4-layer device, I realised that it was possible to turn the circuit upside
down and use a small TO-92 thyristor to the same effect. Thyristors of this
type are quite common in TV & monitor PSUs - not least as safety trip
devices, so almost by accident the project became a dual purpose component
tester.

I needed something similar at a client. At my lab I have one, a plain
old pencil shaped signal injector. But new TSA rules consider that a
dangerous items. So.... I made sure the client had some CMOS logic at
hand. Took a single Schmitt inverter (but a 74HC14 would also have
sufficed), one resistor, one cap and bingo, I had a new pulser.

Don't be too hasty to count out the tunnel diode - according the manual, the
Tek465 scope contains a couple of them.

In their haydays those big scopes fell into the category "money is not
an objective"

 

[ian field]

Then you must be young. My first car was built in 1969 (I got it around
1980). A Citroen 2CV, 6V electrics all through out. Number of electronic
components in that car before installing the radio with its Ge
transistors: Zilch. Not even a diode.

Call that a car - its French FFS!

 

[ian field]

I needed something similar at a client. At my lab I have one, a plain old
pencil shaped signal injector. But new TSA rules consider that a dangerous
items. So.... I made sure the client had some CMOS logic at hand. Took a
single Schmitt inverter (but a 74HC14 would also have sufficed), one
resistor, one cap and bingo, I had a new pulser.


In their haydays those big scopes fell into the category "money is not an
objective"

Maybe I got the number wrong - its not one of those big "double decker bus"
scopes you wheel about on a trolley.

 

[Joerg]

Call that a car - its French FFS!

Don't know what FFS means but that was a great car. 16 horses, about
50mpg on regular unleaded and transported a huge fridge that a guy
couldn't get into his VW bus. He was pretty embarrassed.

Heck, I could even adjust the shock absorber strengths on it with
getting dirty.

 

[Michael A. Terrell]

Don't know what FFS means but that was a great car. 16 horses, about
50mpg on regular unleaded and transported a huge fridge that a guy
couldn't get into his VW bus. He was pretty embarrassed.

16 HP? My dad's lawn mower is 18 HP.


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[Michael A. Terrell]

Then you must be young. My first car was built in 1969 (I got it around
1980). A Citroen 2CV, 6V electrics all through out. Number of electronic
components in that car before installing the radio with its Ge
transistors: Zilch. Not even a diode.

US cars were 12 volts a lot earlier than that. My '61 Ford Galaxy was
12 VDC. I could dig through my collection of Sams Photofact car radio
manuals to find the exact dates, but 12 VDC was in wide use when car
radios were still using vacuum tubes in the US.

--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[Michael A. Terrell]

In the period I'm thinking of transformer coupled O/P stages still hadn't
been completely outdated by complementary pairs so they could have chosen
whichever polarity suited best - it just seemed to take a while to develop
silicon transistors that could supersede germanium in power stages.

US car radio audio output transistors went from PNP Germanium to NPN
Silicon in one step. The first I saw was the Bendix with the early
TO-220 where they bent the leads right at the body to fit the spacing on
a TO-3 transistor and had an almost 100% field failure rate within a
year.


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida