Influence of DC offset on bias

Kirkus, I'm not experiencing a problem, just searching for an explanation on why after setting bias identically, I have to then pinpoint the adjustment while listening to program material. I believe Marakanetz answered my question. All of my amps are gone over and have been recapped and upgraded throughout. However resisters are usually left alone unless they're blown so I'm assuming it stands to reason the adjustment is much more critical; or is it? You tell me. I thought a stilted offset might require a stilted bias but apparently that's not the case. The bottom line is, and whatever the reason, on every one of my amps, perfect channel balance results in a stilted condition. Very close, but stilted. The channel with more offset always has a higher bias also.

Well, bias of course affects distortion performance, and I have seen many, many amplifiers where the final quescient bias level when warmed up varies with how much load there is on the amplifier when its warming up, and ends up being very different from the value and conditions specified by the manufacturer in the biasing procedure. These thermal subtleties seem to be lost on a great many engineers, and a manufacturer's audiophile "brand reputation" seems to have very little correlation with competency in this area.

As far as offset goes, most modern amps *shouldn't* exhibit a change in distortion performance as a result of an offset adjustment. But many early solid-state designs, including the GAS amps, don't have a current-mirror on the input diff-amp, which means that the balance of quescient current between the diff-amp pair is highly dependent on component tolerances . . . and offset adjustment. And even-order distortion products rise pretty quickly as the diff-amp becomes unbalanced.

With a distortion analyzer, oscilloscope, and a bit of experience, one can easily see the different effects of different distortion mechanisms on the amplifier as a whole. But without this, you're pretty much just "stirring the soup" of whatever flaws the amplifier has, and getting whatever floats to the top.

But if my procedure addresses or "corrects" the situation, is that not a best case scenario? Are you speaking of inherent characteristics of a given topology not equipped to deal with certain potential drawbacks, or are you referring to an unseen but needed repair? What's the option to stirring the soup?

Excellent read. I have an 80's integrated here that is very clean and shows no issues, but the left channel dc offset at turn-on goes to over 200mv and then eases it's way back down to around 10 like the other channel. Thing is, it takes about a half hour to do so. Any ideas where to start?
Thanks in advance.