Physical explanation of amp's break in?

Paulwp, I hope you accept the fact that components do initially shift their values, in predefined tolerance range? You don't have to call it break-in and it doesn't automatically assume any sound improvement. But if the same kind of components will tend to settle in the same direction, and to achieve their natural point of settling after about the same hours of use, where do you see a contradiction that during design period an "already settled" prototype is evaluated and being worked on, so the aim of the designers is not how it sounds just assembled, but after known and measured period, specifically for the parts used?

No, I don't accept it as a fact, not do I deny it. Theories to justify the belief in burn-in are of no interest to me. I can't think of a less interesting subject actually, but I suppose I would pay attention to some test measurements. An amp had better perform to specs when first plugged in by the buyer. Then, so long as it continues to do so, it shouldn't sound different from day one. That's been my personal experience of using amps, though of course, I havent done a controlled analysis.

Take two identical amps. Do a DBT to verify that you can't tell them apart. Break one of them in. Then do a DBT to see if you can tell them apart. Do this with a significant number of listeners or a significant number of times. Come back and tell me the results.

Or, show me some bench test measurements in the range of audibility.

Oh, tube amps may change. I guess it takes tubes a few hours to settle in, and then they start downhill.

But, again, who cares? For me, it's like shoes. Do shoes break in? Of course. Should you go ahead and buy a pair that hurts hoping they will break in the right way? I never do. And I never keep a component that sounds wrong hoping it will break in.

Sim Audio amps all have a notoriously long break-in period, 600 hrs minimum. Also after you power them on from a cold state they take a couple of days to settle in. Look at it from the manufacturer's perspective. If it wasn't true and borne out by experience then why would they state it in their documentation again and again. There is no upside to fabricating something like this.

SimAudio's components seem to take a long time to burn in. This was detailed in a review of the Magnum Dynalab 208 Receiver in Stereophile (Jan. 2001). SimAudio designed both the pre and power amp section for the MD208. The reviewer felt the sound was changing and not that great. Here's a snippet of the review. SimAudio's designer himself, Vince Stables, speaks to it himself. If you are going to believe someone, believe the guy who designed it and is most familiar with it.

from Stereophile's Review.....
"I'm glad I didn't, but for a while I experienced the symptoms of aural indigestion: amorphous bass focus, limited bass extension, incoherent soundstaging, a lack of high-end extension, poor low-level resolution, and, most notably, a limited volume range. There seemed to be an optimal volume setting at which the MD 208 evinced a realistic, coherent tonal balance and dynamic range; below that, the presentation lacked presence, body, and tonal coherence; above it, the tonal balance seemed to go out of kilter, as everything seemed a touch too loud and glary.

Vince Stables of Simaudio concurred, and seemed to be making a mental check mark next to each symptom as I detailed them over the phone. "Yeah, it takes a long time because there's a lot of Teflon used on the internal wiring in the preamp stage, those 4oz copper tracings on the PC board take forever to burn in, and you hear it all because there's no filter caps acting as sonic Band-Aids, and no corrective feedback save at the output stage. So it's much more revealing of nuances in the burn-in process as things charge up; eg, transformers, the power-supply section. And it doesn't harmonically sound right—it sounds sort of outside the music. The soundstage starts off very small and it doesn't permit a lot of microdynamic detail. It's almost like it's one step behind the speaker. If you turn the volume up, you won't be able to play as loud because it sounds out of sync from the music. And then, after about a month or a month and a half, as it warms up, it starts to jive and everything becomes more cohesive."

For those users who have never heard an amp change sound over time from new to burned in......some can hear it and some don't. But please don't tell us who have heard it though, that it doesn't happen.

Well, I guess sean has checked out, but just as I do not discount knowledge derived from practical experience, he should not discount what's learned in college. As he says,there are no degrees granted in "common sense, problem solving, application of knowledge" but there are most certainly courses in such subjects. Indeed, many specific facts learned in school become obsolete by the time one gets a job, and the real benefit of school is that it teaches you how to learn, which you should continue to do all your life. Also, having a degree does not prevent you from doing all the things that sean does to gain knowledge.

I mentioned Taguchi design philosophy previously: very quickly here is the jist of it.

You build a circuit (or anything else, like a camera lens system) and test it.
You find that some component, say R235, critically affects performance.
It's been suggested that the circuit designer should now begin to tweek R235, and other parts of the circuit to optimize performance.
WRONG! says Taguchi. Send the circuit designer back to the drawing board to modify the design so that R235 does not need to be tweeked.

The purpose of inspection and test of samples of a product as it comes off the production line is it make sure that the design and the production process are producing satisfactory results. If a problem is found, you don't fix the unit before shipping. (That was the old idea of quality control). You junk it, and fix the design or production process.