Paul Decker wrote:
>
>
> I wonder if transistors don't care because generally they are connected to
> well regulated power supplies.
>
Power Tube filaments and particularly the thoriated Tungsten have a very
low resistance when cold. When power is applied abruptly as when the
switch is turned on, the current is many times the operating current.
This not only creates mechanical stress and physical movement from the
rapid heating, but magnetic currents that may cause the filament to move
physically as well.
A transistor has no such structure or even an equivalent structure and
thus no similar problems.
OTOH transistors are extremely sensitive to over voltage spikes. They
are also sensitive to power excursions exceeding their ratings as they
have little physical mass to absorb the extra heat compared to the
massive anode structure in a power tube. The die (Silicon chip) in a
power transistor is quite small and may be on the order of a 1/4" or 1
CM square. It depends on the heat transfer from the chip to the case and
from the case to a much larger heat sink. Even with exotic heat transfer
compounds the ability for rapid heat transfer is the main power limitation
in transistors. IF we could reduce the Delta T (Transfer the heat faster
from the chip through the case and into the heat sink) then we'd be able
to push the instantaneous power higher in solid state amps. One method
to increase the rating, or at least overhead is to refrigerate the heat
sink, but that creates its own set of problems. Even refrigerated, we
are still limited by the ability of the transistor to keep the Silicon
die cool. Often it's simpler and even cheaper to run a pair of those
expensive, high power transistors in parallel to safely run the power
level that is designed as the limit for one. That is why you find the
parallel, push pull, set ups in many of the high powered, solid state
amps. Voltage wise a single spike of very short duration will wipe out
a transistor where a tube would never have noticed it.
> My guess is if transisters were subjected to big spikes in current and
> voltage similar to tubes at power on state, there would be a higher number of
> transistor failures. But that is just a guess.
If they were subject to that kind of abuse we'd have no operational
transistor circuits running anything more than very low power. Remember,
just the static electricity generated from walking across a living room
carpet can wipe out a 200 watt RF power transistor. The two main
problems with high power transistors are keeping the voltage spikes out
and keeping the internal workings of the transistor cool.
>
>
73
Roger (K8RI)
> Paul (KG7HF)
>
>
>
>
>
>
> From: Charles Harpole <k4vud@hotmail.com>
> Subject: [Amps] amp advice from Alpha
> To: <amps@contesting.com>
> Message-ID: <COL107-W244DF73D48041F318C127AF7D70@phx.gbl>
> Content-Type: text/plain; charset="iso-8859-1"
>
> ... Transistors, apparently, don't care.
>
>
>
> 73
>
> Charles Harpole k4vud@hotmail.com
>
>
>
>
>
>
>
>
>
>
>
> ------------------------------
>
> _______________________________________________
> Amps mailing list
> Amps@contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>
>
> End of Amps Digest, Vol 81, Issue 47
> ************************************
> _______________________________________________
> Amps mailing list
> Amps@contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|