For tube sound, which is more important: preamp or power amp?

If the Tube is driving a signal then the Tube will wear more compared to it glowing with the just the heater filament being lit up while waiting for the signal.
.The powering on and off does cause some stress to the tube from the Hot/Cold part of that process.I know people with Guitar amps that have NEVER changed a Tube yet..after years of use.

Regarding the statement by Mr. Carver about the 50+ year longevity of the power tubes in his amplifier, I would point out that in general there are statistical approaches that are commonly used in deriving published MTBF (mean time between failure) specifications which produce misleading results.

For example, many of the mechanical (non-SSD) hard drives used in computers have published MTBFs of 1,000,000 hours.  Or in the case of some Western Digital Gold Drives I recently purchased, 2,000,000 hours.  2,000,000 hours corresponds to 228 years of continuous operation.  Does that mean these drives can be expected to last anywhere close to that amount of time?  Of course not.  What it means (and I'm choosing the following figures for purposes of illustration) is that if say 100 drives are operated simultaneously one of them can be expected to fail after slightly more than 2 years.

Similarly, **if** Mr. Carver's 50 year figure is based on similar statistical methodology, and given that there are 12 power tubes in a stereo pair of these amplifiers, it would mean that one tube can expected to fail after a bit more than 4 years.

On the other hand, though, if we assume an average of say 1 hour of use of Mr. Carver's amplifier per day his 50 year figure would correspond to 18,250 hours.  I suppose that if the tubes are driven very lightly, and his design includes good soft start provisions, numbers approaching that figure might not be totally unreasonable.  For example, I recall reading that the Western Electric version of the 300B power tube, when used under recommended conditions, can last for 40,000 hours.  (I'm not sure, though, if that applies to the original 1930s version, or to the relatively recent reissue, or to both).

In any event, personally I would consider Mr. Carver's claim as simply indicating that the KT120 and KT150 tubes that are offered with the amplifier will last considerably longer than in most other applications.

Regards,
-- Al
 

I think if a tube is on, it's wearing out…I'd like to see what at atmasphere says about this.

If it has B+ on it, then that is the case, slower if there is no B+ (filaments degrade over time as well).

Al's comments reflect the actual reliability, but in this case a KT120 is in use and in practice they don't seem to hold up as well as the KT88. I suspect that the tubes in this amp will hold up only slightly longer than in an amplifier where the tubes are biased a bit hotter, assuming the tubes don't have a defect which can shorten their life, which many modern tubes do.

I own a Single Ended amp with 2 power tubes and a single input tube running in Class A (also with a tube rectifier) putting out around 12 watts per side, and have no idea what sort of tube life I should expect. No specs are available from the manufacturer (Dennis Had working away in his lair apparently hand wiring these things himself…maybe with a helper) which is fine with me as the amp is so good, but I still wonder. I run the thing at least 8 hours or so a day…every day…even holidays, unless I'm not at home.

As the old adage goes: Garbage in, garbage out. I have found the preamplifier to have more influence on the sound than the power amp. A warm, dimensional, tube preamp, feeding a good solid state amp, works for me.