Are input offset voltage and offset voltage temprature drift correlated of an opamp?
Let's say I know the polarity of the offset voltage from that information can I infer the polarity of the offset voltage temprature drift? For example, let's say non-inverting input is more positive than the inverting input. In that case is it safe to assume that the offset is going to increase towards the non-inverting input as the temprature increases or are they completely unrelated?
Re: Are input offset voltage and offset voltage temprature drift correlated of an opa
The implementation affects the offset and offset drift dependency. For example, for an emitter-coupled differential pair, the drift and offset are proportional; however, for a mosfet source-coupled pair several design parameters including the overdrive voltage, transistor size mismatch, and the threshold mismatch, determine the drift and offset dependency.
Re: Are input offset voltage and offset voltage temprature drift correlated of an opa
I guess there is no straighforward answer to that question. My impression is that the dependecy of drift and offset heavily depends on the topology of the opamp. Is that kind of correct?
My comment is about a simplified model for the transistor but as far as I know it includes several offset sources. (Reference: Analysis-Design-Analog-Integrated-Circuits by P. R. Gray)
I conclude from your comment that op-amps with factory offset trim may not act according to derivations similar to those given by the above reference, am I right?
Re: Are input offset voltage and offset voltage temprature drift correlated of an opa
Thanks for mentioning "Analysis and Design". I see that Vos is assumed to be proportional to Vt for BJT input stages, what I considered as special case in post #2. My general comment is mainly substantiated by the specification of modern OPs which don't seem to show a systematic dependency of Vos and dVos/dT. I guess it's due to the main offset voltage being trimmed off.
I think that's the case and the trimming approach may affect the offset and its drift relationship. According to page 235 of the book, when using an external potentiometer to null the offset, since the temperature coefficient of the nulling resistor is different from that of the diffused resistors, a resistor-mismatch temperature coefficient is introduced which can make the drift worse than it was without nulling.
Re: Are input offset voltage and offset voltage temprature drift correlated of an opa
The quality of the op amp and its design and fabrication
all come into play.
An op amp with a natural nil offset, and one that has a
natural front end skew that's trimmed in, will have
different drifts. Even on the same design and same
wafer. Vbe1=Vbe2 is a better bet to hang tough
than Vbe1=Vbe2+f(Rmumblemumble).
Op amps with wider Vio will tend also to spec and
test sloppier drift (if they test drift at all - not a
good bet on commercial grade). Every piece is a
sum of nonidealities, with the ones failing to spec
being thrown out (or sold on the gray market by
the cousin of the third shift test operator).