Part of the problem with measuring or "charting" component settling is that audio is a very dynamic signal. As such, there is little consistency in amplitude, frequency or duration of the signal being passed through these devices. On top of this, one might listen for a few hours on Monday and then not have the chance to put the system to use until the week-end. Once the week-end comes around, the system gets a good work-out. As such, the duty cycle and consistency of use also come into play.
The reason that i bring this up is that i've had people come in with units that were 10 years old but were never used much at all. When i put them on the bench, they operate like brand new units. Within a matter of minutes or multiple hours ( depending on the specific unit ), i can sometimes see HUGE differences in how the circuits perform. In effect, those 10 year old but barely used units are going through break-in just like a brand new unit would.
Along those same lines, i work on many units right out of the box. After a reasonable warm-up period, i calibrate and align the circuitry and send them out the door. It is VERY common for a customer to come back in with a unit a month or two later with that same unit and it measures and performs very differently. Since i told them that this would happen and i cover this under warranty, i re-align the unit again from top to bottom. Once i do this, it will typically hold these settings much better / more consistently than if it were brand new. It may require further adjustments later on down the road, but not with the drastic shifting that it underwent during initial use. That is because most of the parts had fully "settled in" but a few others were still shifting at a reduced rate.
For the record, i get to watch digital "phase locked loop" and "quartz locked" circuitry drift all over the place every day of the week on multiple units a day. While these types of circuits are supposed to be "rock solid", and that's why they use them for generating reference frequencies and speed control stability, some designs are MUCH better than others. Some circuits are right on the money in a matter of minutes whereas some take several hours of actual use to stabilize. The funny thing is, some of those that are the least stable can be found in some of the more expensive gear. Like anything else, you don't always get what you pay for. Unfortunately, not every audiophile or consumer electronics enthusiast know and realize this.
The only thing that you can be assured of when you buy electronic componentry is that it WILL "settle" at least a small amount. That is why each individual component that went into making the assembled component and the assembled component itself is alloted a specific "tolerance" for normal operation. If the unit does not meet spec and is out of tolerance, that is because parts have "settled" or "shifted" further out of tolerance than they should have. As mentioned above, this typically happens due to thermal or amplitude stress in most cases. Sean
>
PS... I'm done on this subject pretty much forever. If someone can't follow along with the information presented in this thread or in the archives, so be it. They are either too dumb to comprehend the subject at hand or are too stubborn to admit the truth. I don't think that you'll find ANY Electrical Engineer or service technician with any amount of experience that will argue with any of the points that i've brought up. That should tell those that challenge the "theories" that were presented here where they really stand.
