Why passive component has better drift characteristics than active components had?

Question

I came across this paragraph in the classic reference Op Amps For Everyone by Ron Mancini where it says on the first page:

Many attempts has been made to make a stable amplifier, but temperature changes and power supply voltage extremes experienced on phone lines caused uncontrollable gain drift. Passive components had much better drift characteristics than active components had, thus if an amplifier's gain could be made dependent on passive components, the program would be solved.

I don't quite get what that means. I'm going to take a few assumptions here and hope someone can confirm or correct my understanding.

1. Say when there's a voltage applied across a passive resistor, its resistance changes within some specified tolerance as it heats up overtime by the current. Is that why we see or what we call the drift in the input voltage vs. output current?

2. Let's take a BJT as the active component. Be it configured in common collector, emitter or base amplifier, the only variable in any of the gain equations that is not passive is the \$ \beta \$ or the Hfe. (e.g. \$ A_v = {R_E \over {\alpha r_m + R_E + {R_B \over {\beta + 1}} }} \$ for C.C.). So the volatility or stability of \$ \beta \$ is essentially the main indicator for the drift?

3. Because the \$ \beta \$ is more unpredictable than heating does to a resistor's resistance, hence the drift characteristic in a passive component is better than an active component?

Answer 1

I think that the unseen context of the quote needs to be revealed to be able to adequately refer to it in your question. The quote appears to be talking about telephone lines and this suggests to me that the quote is meaningless without more context. Moving away from the quote....

In short, passive components don't suffer from anything like the problems that transistors and diodes suffer from. Diodes and transistors have very non-linear characteristics (e.g. forward diode conduction V/I ratio) and these "characteristics" are very much temperature affected at the atomic level and are uncompensatable other than by using high gain and feedback to linearize things (as a control system would). The diode equation: -

At a very fundamental level, the temperature (in degrees above absolute zero temperature) are involved at the very heart of things. Whereas, for a resistor at the fundamental level: - $$I_R = \dfrac{V_R}{R}$$

So, resistors very are linear and, if there is any temperature related resistance change phenomena that is purely down to how the resistor material was mixed or produced (in simple terms). In other words they are basically linear right from the start whereas diodes are just simply highly non-linear right from the start and always.

Answer 2

Its just a guess, but maybe the author means that passive components are mostly linear (i.e. parameters change linearly with voltage or current), whereas with active components you face nonlinearities because of the supply voltages that limit the linear operation of active devices. Drift characteristics are better controllable in a linear system.