AD822 Input Impedance - variation from typical spec and how to measure it?

Document created by analog-archivist Employee on Feb 23, 2016Last modified by analog-archivist Employee on Feb 23, 2016
Version 2Show Document
  • View in full screen mode

would like to know more about the AD822’s input impedance. In the datasheet, 
the AD822’s input impedance is a typical spec, do you have any more information
regarding an estimated variation of this impedance? What's the best way to
measure this parameter?


We do not measure the input impedance directly. An indirect way of inferring
the common mode input impedance is as follows:
. Input Impedance = Change in applied voltage / change in input current.

So you can estimate the input impedance from looking at Ibias versus VCM curve.
Of course, this simplistic model may not apply in all cases. Sometimes the
typical datasheet spec is also based on detailed knowledge of the process
corners (simulation results).

The input bias current and input resistance are DC specs. A simple model of the
op amp input can be something like attached document.

Some of the bias current you see is fixed, and some of it will vary with the op
amp’s input impedance. Needless to say there is no physical resistance and what
you see is mainly due to leakage currents through the oxide/base current of
transistors, ESD leakage current etc.

From our understanding, many applications do not need to know the exact input
impedance. For example, in a transimpedance amplifier circuit, the designer
would be more concerned on keeping the input bias current negligible vis-a-vis
the photodiode current rather than knowing the absolute value of input
impedance. Furthermore, if the photodiode is operated with zero bias or a fixed
bias then the input common mode also does not vary so a fixed base current
model works perfectly okay. Likewise in a non-inverting amplifier the gain
resistors would be designed to carry say at least 100X times the worst case
input bias current to guarantee an accurate amplification.