Other Parts Discussed in Thread: LPV821, LPV811, TLV2333, LM358, OPA627
I happen to have a design where the same input voltage is amplified with an inverting and a non-inverting amplifier (with similar gain actually).
The offset voltage of the LPV811/2 is naturally much bigger than from a zero-drift amplifier such as the LPV821.
In my case the parasitics would cancel out if there were a matching of the offset between the two channels.
The datasheet documents the offset distribution for channel A and channel B, pretty similar as needs to be.
But it does not state that the offsets (or bias currents) would be correlated, which I would expect they are due to symmetry (same die, same temperature, same supply, same common mode ...).
To which extent can it be expected that the non-idealities are "matched"?
Is an estimate of a factor 10 reasonable (i.e. 10% do not cancel out)?
Would a similar argument apply to say the TLV2333?
Are there OPAMPS for which such a correlation is specified (I didn't look everywhere)?