Electronic – Error in this explanation of instrumentation amplifier

A- this topology is typical of the input stage of a three-op-amp instrumentation amp implementation. The final stage, which is a differential amplifier, is missing, and is presumably replaced by the differential input of the ADC. There is a clear error on OA1, as others have described.

The addition of R3 serves to provide some gain. Without R3, you simply have two input amps configured as voltage followers.

There could be two reasons why this is used. One is that the input impedance of this stage is very high, especially if FET op amps are used. If there is concern that the ADC doesn't have high enough input impedance, this is a nifty solution. The second is that some gain may be necessary to provide enough resolution at the ADC.

The disadvantage is that R1 and R2 should be precision matched, or you may screw up your CMRR. Typically, people looks at this particular issue in the context of the difference amp stage on the output of the IA. Without doing the math, I suspect it holds here as well.

A instrumentation amplifier is more than just a opamp hooked up as a differential amplifier. The inamp puts buffers in front of each of the diff amp inputs. This presents a high impedance to the outside, and also eliminates the cross impedance between the inputs of a bare diff amp.

In theory you could make a inamp with three opamps, but in practise that still won't be as good. A real inamp benefits from all the components being on a single die, which help in matching thermal characteristics and ballancing the gain of each input to get better common mode rejection. Precision inamps go thru a laser trimming process in production to get the absolute and relative gains just right.