Does anyone know if the input offset current of an op-amp drifts with
temperature? Or will it stay within the +/- rating of the datasheet
spec while the input bias current drifts vs. temperature? I think
JEFT and MOSFET bias currents from gate leakage double every 10=B0C?

Kingcosmos wrote:
> Does anyone know if the input offset current of an op-amp drifts with
> temperature? Or will it stay within the +/- rating of the datasheet
> spec while the input bias current drifts vs. temperature? I think
> JEFT and MOSFET bias currents from gate leakage double every 10=B0C?

Most opamps have an initial offset spec (like millivolts=20
offset either way at 25C), and a temperature drift spec=20
(like microvolts per degree C, but it can be in either=20
direction, so it might improve on the initial offset), and=20
some have a time drift spec (like microvolts per hour/day at=20
constant temperature, also in either direction). This last=20
spec is usually only for the ones that have offset nulling=20
terminals).

Bias current is something else, entirely, than offset=20
voltage. Bias current also has an offset spec as well as a=20
maximum at various temperatures.

On Aug 6, 10:11 pm, Tim Wescott <t...@seemywebsite.com> wrote:
> Kingcosmos wrote:
> > Does anyone know if the input offset current of an op-amp drifts with
> > temperature?
>
> Every parameter of an op-amp, including size, drifts with temperature.
>
> > Or will it stay within the +/- rating of the datasheet
> > spec while the input bias current drifts vs. temperature?
>
> If the data sheet guarantees the input offset current over temperature
> then it should stay in bounds, yes. Read your data sheet carefully.
>
> > I think
> > JEFT and MOSFET bias currents from gate leakage double every 10=B0C?
>
> Possibly, and it may even matter with a JFET. With a MOSFET it'll go
> from negligible at 100C to several orders of magnitude below negligible
> at lower temperatures. Meanwhile, your offset voltage will be atrocious.
>
> --
>
> Tim Wescott
> Wescott Design Serviceshttp://www.wescottdesign.com
>
> Do you need to implement control loops in software?
> "Applied Control Theory for Embedded Systems" gives you just what it says.
> See details athttp://www.wescottdesign.com/actfes/actfes.html

Most datasheets I see will state a +/- rating for both input bias and
input offset currents, and a graph later in the datasheet showing
input bias vs. temperature. I would simply guess that the input
offset currents would drift in the same manner. Rule of thumb maybe.