Jim, Don,
It seems to me that this is a function of the design and construction of
the regulator, including choice of components, not the method of AC/DC
conversion.
The quality of design and construction of course is the most important
factor in determining the likelihood of failure of a power supply, much
more important than whether it's switching or linear. And I can't really
tell which type fails more often, because in 37 years designing and
building both types of power supplies, and repairing lots of them, I can
only say that both types fail a lot. Even nicely built professional
linear power supplies (such as US-made Kepcos) have landed on my repair
bench.
But I would like to insist on my point, because it's factual, supported
both by practical experience and by logic: In the event of a failure,
switching power supplies almost always go DOWN in voltage (often to
zero, sometimes to a lower-than-normal voltage, and often they start
pulsing at low voltage), while linearly regulated power supplies very
often fail with the voltage going UP. Of all failures I have seen in
linear supplies, a good 30% has been with the voltage going up.
The technical fundamentation is simple: A linear power supply has a
transformer, rectifier and filter, producing a voltage that's much
higher than the nominal output voltage, followed by a regulator circuit
with a series pass element. If this pass element fails shorted, the
output voltage goes up to the full unregulated voltage. And the pass
element works under high stress, because it has to dissipate a lot of
power. So it fails easily.
Instead a switching power supply has a small, high frequency
transformer, fed by a high frequency (25kHz to 1MHz) generated by power
switching elements. The secondary voltage on that transformer is
rectified and filtered, then delivered to the output without further
regulation. There is no pass element, eliminating the main failure mode
of linear power supplies that results in high output voltage. Regulation
is done by adjusting the duty cycle of the switching elements. If the
power devices of a switching power supply fail, the output will go to
zero, no matter in what way they fail (shorted or open). That's an
obvious safety advantage over linear supplies.
In both types a failure of the control circuitry can cause a rise in
output voltage. But since the control circuitry works at very low power,
it's not nearly as failure-prone as the power devices.
The bottom line is that switching power supplies are far less likely to
expose a radio to overvoltage, although it's likely that their overall
failure rate is higher than that of linear power supplies, owing to
their higher complexity. Where there are more parts, there is more that
can go wrong.
A high power linear power supply should always be fitted with a
crowbar-type overvoltage protection. A switching supply can also be
fitted with one, but it's less necessary.
Manfred
========================
Visit my hobby homepage!
http://ludens.cl
========================
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|