In article . com,
w_tom
wrote:

[snip]

Semiconductors are manufactured in repeated cycles from room
temperature to upwards of 900 degrees. Now that is real thermal cycling
- and it is not destructive. Do same items get thermal cycled when
temperature varies less than 100 degree?

Although I agree in general terms with your thesis, I would apply the above
with caution. The manufacturing process is quite different to the use.

Most domestic audio SS devices will be made of silicon-based
semiconductors. These will normally be designed to work with device
temperatures up to the order of 200 C - so quoting '900' (with no units)
would be misleading. The reality is that either the device, or its
electrical connections, or its bonding to the pack, may fail if you go too
much above the order of 200 C.

In practice, SS devices tend to be designed to be cycled many many times,
and, provided the designer chose and used them with care, there is a good
chance they will work for a very long time. Devices in circuit locations
where the power dissipation does not vary much tend to operate with low
device temperatures. Devices in, say, B or AB output stages will have their
temperatures vary far more with signal than between being powered up or
not.

I have two 30+ year old receivers. In daily use. No SS device failures so
far. I have two 25+ year old 200W power amps. One device failure last year
(intermittent) - which may have been the soldering onto the PCB, but I
suspect was the bonding inside the pack. The device in question was a class
A, voltage stage device, not an output device, used well within spec, so I
put this down to the normal laws of statistics for real-world devices...

So 1 transistor in about a hundred, in 25-30 years of daily use, often
being switched on/off more than once per day.

Hence if someone is worried by thermal cycling of the SS devices in their
audio amplifiers, then the best advice is that they should either never use
the units, or only play low-power 1kHz sinewaves into high impedance,
resistive, speakers. :-)

OTOH If you want to play music, then I regret to say that 'thermal cycling'
of the output devices generally goes with the territory.

Of course, you could go to class A output. But then the operating
temperatures are elevated almost all the time the unit is used. The
lifetime of devices is a strong function of device temperature, so the
chances are that the MTTF will be much lower for such 'thermal cycle
avoiding' designs than if you'd used a well designed B or AB unit. If you
used one that was on continuously, I have my doubts that it would run
without problems for 25-30 years - but might if designed with this in mind.
Would probably cost you far more overall, though.

In my experience, the bias levels in the decent amps I know of all settle
in a matter of a few seconds to minutes after they turn on. After that, it
is down to the music, the weather, and your central heating. :-)

So I'd agree. Just choose the music you want to hear, and enjoy it. :-)

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc. http://www.st-and.demon.co.uk/JBSoc/JBSoc.html