Do you leave your components on 24/7?


Lately I've been leaving my components on all the time, on the assumption that a) they'll be ready when I want to listen, and b) the on/off cycle ages the equipment (tubes, anyway) faster than leaving everything on. Is the latter a reasonable assumption?
128x128cmjones

03-05-11: Tmsorosk
Paper ... You may claim thermal switching is B.S. but several manufacturers of equipment I own as well as my long time dealer tell me that leaving things on reduces brake down rate , Iv'e followed there advice for many years and have found this to be true. What do they no eh . Do you no a manufacturer of S.S. equipment that states otherwise . Class A amps that run very hot may be the exception.
if you just took the word of the manufacturer yet don't know what failure is that is being prevented by 24/7 operation, then you don't know whether you are just accepting a bill of goods. if you uniformly believe what you read, then you sound like a prime candidate for $15,000 interconnect cable.

the reality is that product failures are often characterized by a "bathtub" curve where there is a period in which you have a higher failure rate early (this is referred to as "infant mortality"), a period where the predominant failures are random failures, and a period where failures result from the product just wearing out. a company that has reasonable quality control processes should screen for infant mortality. this screening is typically referred to as product burn in. this kind of burn in is different from the meaning of "burn in" as used by audiophiles. burn-in testing often tends to weed out infant mortality by testing at elevated temperatures and/or voltages. i can believe that there are manufacturers who don't have such good quality control processes and they ship products out without burn in testing. my amplifier and pre-amplifier are by bryston. one good thing about bryston is that they openly disclose their processes and they describe the burn in process that they follow before shipping products. the company that supplies your products may not do any such burn in.

proper burn in testing will have a lot more to do with the likelihood of reducing product failure than this 24/7 stuff that you are repeating.

here are the comments from pass labs on product life from one of their power amplifier owners manuals:

So how long will this hardware last? It is our experience that, barring abuse or the odd failure of a component, the first things to go will be the power supply capacitors, and from experience, they will last 15 to 30 years. Fortunately they die gracefully and are easily replaced by a good technician.

After that, the longevity will depend upon the number of operating thermal cycles, but we can say that we have had amplifiers operating in the field in excess of 20 years with no particular mortality, and we don’t have good information beyond that.

More to the point, you should not worry about it. This is a conservatively built industrial design, not a frail tube circuit run on the brink of catastrophic failure. If it breaks, we will simply get it fixed, so sleep well.

what pass describes is a wear out failure mode (capacitors) and the possibilty of random failure (which they say they will fix).


03-05-11: Tmsorosk
And as far as a sonic difference goes , my own ears as well as other well respected audiophile friends have tested the affects of turning off large power amps and the results are consistent , etched sound for hours to days , one of the few things we agree on . Have you done such tests for your self , Or are you a theoretician.
the reason why i mentioned theory was to suggest that the stuff about how equipment sounds better after warm up may not be complete bs - that there is some theoretical basis for the claims. for my own part, my view is more like that of cmjones: after 15 minutes i'm not going to remember how the system sounded then with enough detail to make a meaningful comparison with how the system sounds later.
Mcintech, I recently talked to a tech who told me that long storage can be far worse on the caps then leaving a piece of gear on 24/7.

03-05-11: Hifihvn
The link to a post Almarg pointed out covers the way it is IMO. Some electronic components (semiconductors) do not like the big thermal swings, and may fail sooner. Others may settle, and break in to the on always operation since they stay hot all the time. With this case, those semiconductors may break when they shrink during cool down, since they have never been allowed to contract during a cool down, and expand during warm up, since this is what they have become accustomed to. Think of it as a super miniature bridge without expansion joints, for thermal change, while others may handle the temp swing like a bridge. That one poster is an engineer(Almarg link) that specializes in this. And like he says, it varies from component, to component. I'd have to agree with that. It makes sense.
i have not read the almarg posting to which you are referring but the failure mode to which you are referring is highly unlikely in practice because semiconductor manufacturers typically manufacture devices to iso 9001 standards. so they do burn in testing to screen out the kinds of failures that you describe (which are typically "infant mortality" type failures) [i briefly described burn in testing in another posting but burn in testing by electronic manufacturers is different from "burn in" testing as the term is used by many audiophiles]. furthermore, semiconductors are tested over a temperature range that is wider than they are likely to encounter in actual use.

but failures can occur at various stages. the semiconductors are assembled onto boards and the boards are assembled into systems. burn in testing would need to be done to screen for infant mortality at each stage. a given audio equipment maker may not use such quality control processes (it does after all increase costs because some products will fail) but bryston does do burn in testing before shipping products (according to the information on their website).
I should have emphasized that I was referring to vintage gear (my specialty, and what I listen to at home), and since I typically like to buy once (I drive a 16 year old Corolla) long term reliability is paramount 'cause this boy don't like to repair his own gear.

The Nagra pyramids run so cool I doubt you'd see heat damage for over 10 years or more.

My shop is also McIntosh Labs and several other brands, and the same applies. After SEVERAL years, you can tell which units never get turned off, even CD players in the power supply areas.

If you only keep new units for a few years, leave it on.
Paperw8, This thread and post was the one I believe. [http://forum.audiogon.com/cgi-bin/fr.pl?aamps&1296956886&openfrom&13&4#13]