It has been a while since we've had an interesting circuit analysis problem on the group, so I thought I'd see if I could stir the pot a bit.

Back in 2004, Bob Pease had this column:

Then, just last December, he got this response:

Let me post a circuit so we can all discuss the same thing:

R1 R2 1 ___ 3 ___ Vin- o----o-|___|-o---o------o-|___|---. | | | | | |\\ | '---|-\\ | |A >---------o--- Vout .---|+/ | |/ | | ___ | ___ Vin+ o----o-|___|-o---o------o-|___|---o 2 4 | R3 R4 o GND

Bob Pease had R3 labelled R1 and R4 was labelled R2, but I've made them distinct for greater generality. I've also given the opamp a gain of A, which in the ideal case would be infinite, but for real opamps, it could be desired to let it be some function of frequency. In Pease's columns, he seems to assume it's infinite.

In last December's column, Eero-Pekka Mand says a number of things about the input impedance at Vin-.

Assume A is infinite, and R1=R3=1000, R2=R4=100000.

Eero-Pekka says that if Vin- and Vin+ are driven with "equal and opposite in phase" signals, the input impedance seen at Vin- is 500 ohms. Is he correct?

He also says "When inputs are driven to 10 V dc each...if gain = 1 (all four resistors equal), then impedance is 2/3 R."

I believe the impedance he refers to is still the impedance at Vin-, and R is the value to which the 4 resistors are now set. Is he correct?