TSV912: Offset spec in datasheet and product folder not consistent

Kevin,

I see the disagreement between the front page and electrical characteristics table. Front page will be 25C value (and typical) unless specifically stated otherwise. I will check to see which value is intended and more importantly which one is implemented in test. This could take the weekend plus a couple days.

Kevin,

This correction task has been assigned and should be in work now. I'll keep you advised of the progress as I receive it.

If there is a histogram plot, their limits come from ATE and should be the right test limits - here the +/-1.5mV max match the spec table, so the front page +/-1.9mV is probably needing to change to +/-1.5mV (maximum)

HEre is that kind of odd unipolar drift histogram again, I would think this is either  magnitude plot (and should be labelled as such) if the actual drift is +/- limited. Being unipolar is actually kind of odd for physical op amps - but maybe so? 

Michael,

Figure 2 is the absolute value of offset drift. ABS( (Vos[-40C] - Vos[+125C]) ) / 165C

Vos drift is not unipolar and may not be linear. Should it be labeled as a magnitude? ; probably although it is inferable. 

But still the peak bar is not at 0uV/C centered, so if you flip this around to be bipolar it would show bimodal peaks at +/-0.4uV/C. probably not correct, here is an example showing more easily interpeted Vos drift, So the actual vio over temp shows you do have some outliers over the full range and many are not straight lines - many are over a smaller range, but the full temp range line fit shows a non-Guassian distribution. This was a tough one as we tried to not have those outliers - especially on dIos/dT.