Re: Howlers

On Thu, 12 Apr 2007 01:37:45 GMT, James Eades wrote:

On Thu, 12 Apr 2007 00:36:20 +0100, spam@xxxxxxxxxxxxxxxxxxxx
(Jonathan L Cunningham) wrote:

Andrew Stephenson <ames@xxxxxxxxxxxxxxxxxxx> wrote:

In article <1hwf1y6.115rd6vrd2h7wN%spam@xxxxxxxxxxxxxxxxxxxx>
spam@xxxxxxxxxxxxxxxxxxxx "Jonathan L Cunningham" writes:

I'd be interested to know if anyone has a transistor radio made
in, say, the sixties that still works. [...]

Would the mid-70s do? A standard AM/FM portable jobbie, made by
some (West) German outfit -- it lives in the kitchen thus is out
of eyeball-reach -- and run for most of its life on mains power,
if that matters. It has lived a sedentary life, with vibrations
being limited mainly to self-made noises and temperatures around
what we humans would consider comfortable, ditto humidities.

Thanks.

About 30 years then. I predict it will fail sometime in the next 300
years. And would almost bet a pint that it will fail in less than 1000
years. Let me know if it doesn't. :-)

I'm wondering if my radio had some germanium transistors, and whether
that means it should fail sooner. By the 70's, I think the vast majority
of transistors were silicon. IIRC, the switch from germanium to silicon
coincided with a switch from mostly pnp to mostly npn transistors in
circuits. Not sure why.

Because npn transistors were supposed to be faster?
__
JamesE

No. It's practical rather than theoretical physics and/or electrochemistry.

Transistors are made by starting with a collector that's bulk silicon,
putting a base on it, then putting an emitter on that. (That's why a TO3
package has the collector connected to the case; the bulk material sits on
the case structure.) Bulk silicon tends to be NPN, because the dopants to
make it N are cheaper and dissolve more evenly than the ones to make it P.
PNP silicon transistors have to be made by either starting with P bulk
material, or by starting with weakly-doped N and overdoping a P structure
on top -- either more expensive material or an extra step, either way
costing more. PNP silicon transistors are still more expensive than NPN
ones, although the price of either is now remarkably low.

P material also has less electron mobility than N, so, yes, silicon PNP
transistors are slower than NPNs. They also have higher bandgap voltages
and therefore get hotter for the same current through them. The difference
is fairly small in modern stuff.

For years microprocessors and similar complex circuits were made with
N-channel field effect transistors, because the channel is effectively bulk
silicon. CMOS (Complimentary Metal Oxide Semiconductor) technology was more
expensive than either NMOS (using N-channel only) or TTL (using almost
entirely NPN transistors) for a long time. Nowadays they have the
processing down to the point where the speed and power saving of
complementary circuits overrides the extra cost.

Germanium has the opposite qualities as far as the dopants go. In the early
days one found quite a number of circuits with germanium PNPs and silicon
NPNs. Forgetting which bias was which could get you in deep trouble.

Regards,
Ric

--
Posted via a free Usenet account from http://www.teranews.com

.

Relevant Pages

  • Re: So what exactly do scientists know about global warming?
    ... >transformer coupled germanium designs! ... Silicon came in when I was at school. ... Hundreds of transistors could be formed side-by-side on a wafer of silicon, ... Silicon power transistors manufactured by the homotaxial process were very ...
    (uk.business.agriculture)
  • Re: What Germanium PNP transistor do I use?
    ... Way back in the early days when silicon devices were new, ... radios that have silicon RF/IF stages? ... out as being good for not much more than audio, ... could actually build audio output amplifiers out of the transistors. ...
    (sci.electronics.basics)
  • Re: Wilson Current Mirror - some confusion
    ... doesn't depend on its own VBE. ... similar to the other transistors to do its job. ... LTSpice will plop down an "NPN" that doesn't show the Early effect, ... Take what you've got and use a real BJT or else add this: ...
    (sci.electronics.basics)
  • Re: How would one use NPN transistors only?
    ... >I am familiar with the typical amplifier using NPN, ... It just makes good common sense, ... >What about amplifiers that use ONLY NPN transistors? ... >There are a few people now building amplifiers with only NPN ...
    (sci.electronics.design)
  • Re: Has physics a future?
    ... "technology" had nothing to do with it, this was initially pure theoretical solid-state physics. ... Having characterised it Physics played catch-up trying to explain how it worked. ... It was intended as a means of making cheaper transistors but made far better more consistent transistors. ... The larger the area of silicon used the more likely it was to include an imperfection in the silicon and be a dud. ...
    (sci.physics.relativity)
Loading