Bill, and all,
So, absent migration, do transistors "wear out"?
That's a rather philosophical question.
One might argue that over the years there will be corrosion, radiation damage,
who knows what else, and after thousands of years the transistors will no
longer be usable. Also, in typical ham use, there is frequent thermal cycling,
with all the expansion and contraction that means, and depending on construction
and mounting technique this might eventually lead to fatigue cracks, the silicon
chip detaching from the carrier, and so on.
There are relatively well established curves relating thermal stress and current
stress to lifespan. What's clear is that transistor's lives are drastically
shortened by high temperature. Used in a comfortable temperature range, to
nearly stop migration effects, they last very long. One could say that an
amplifier using transistors can be designed for pretty much any desired service
life, by making the transistors run cool enough. On the other hand, equipment
that will see relatively few hours of actual use - like most ham equipment - can
be down-designed to a price target by accepting a relatively short life span,
running the transistors rather hot.
I can provide some first-hand anecdotical information: I bought a new Kenwood
TS-450SAT in 1991. Very soon later I set up a packet radio BBS, and from then on
this radio was used for over 12 years for the BBS, running 24/7, and adding up
easily 8 to 10 hours of daily TX time, at full 100W output. Almost all of my
message and bulletin forwarding, with typically 6 to 12 forwarding partners, was
on HF. Mostly 40 and 20 meters, also some on 15 and 80. Bandswitching was done
automatically, by CAT. The antenna was a non-resonant delta loop, tuned by the
radio's atomatic antenna tuner. My BBS was almost permanently linked to one of
my partners, 24/7. I left the BBS running continuously, even when travelling
abroad for several weeks. So my radio must have added up around 40,000 hours of
full power transmitting, and well over 100,000 hours of total operating time.
That's far more than most ham transceivers are ever required to do. And since
packet radio involves huge numbers of brief full power transmissions,
interrupted to get the acknowledge frames, my radio's final transistors were
subjected to an enormous amount of thermal cycling.
After the packet radio era pretty much ended, and I too shut down my BBS, this
radio reverted to the typical occasional ham operation, in SSB voice and some
digimodes, slowly adding some more hours.
And after all this, the final transistors are working exactly like they did when
new. No degradation can be noticed at all. Instead several other components of
the radio have needed replacement, among them several electrolytic capacitors,
switches, controls, the DDS ICs, and a ceramic filter. So I'm a witness for the
fact that RF power transistors, correctly applied, have a long life span, longer
than that of several other component types. Which doesn't mean that it's
infinite, of course.
Now go and ask broadcasters. I would bet that there are some who have clocked
200,000 hours or so on their solid state transmitters, at full power, without
observing any transistor wear effects. Anyway broadcasters must have the most
experience with RF power transistor life expectancy.
Manfred
========================
Visit my hobby homepage!
http://ludens.cl
========================
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|