I have experienced some annoying features of this INA.
When there are no protecting diodes, it is enough to kill the amplifier by...
- have an input signal connected and no supply voltage or in brown out
- a rectangle shaped common mode step of several volts (totally within allowed CMR range)
- Even the glitches of a function generators output, when changing output range, was enough to blow the INA! Unbelievable!
When you use series resistors and protection diodes, it degrades performance significantly.
With low valued series input resistors (because of noise & offset), it is impossible to design a effective ESD protection,
as differential input voltages of more than 1V(!!!) will destroy the input stage immediately (in nanoseconds?)
(when employing an INA it is mostly because of significant amplification)
So this INA has superior performance - as long as you don't put it in a real circuit!