Other Parts Discussed in Thread: OPA145
Application is measuring unknown impedances using an variable feedback resistance (any of 8 resistors switched in with reed relays). The parasitics cause errors in the measurement. I have modeled parasitic series combination of C and R in parallel with the device under test (Rdut or Cdut), and in parallel with the feedback resistor (Rf). Also, have modeled the parasitics with gain rolloff. Labview fitting program is used to get the C and R parasitic values. However, the model must not be correct, since the fit values for the parasitics are not constant with different measurements of Rdut and Rf. Also using an input and output voltage follower. Measurements are from 20Hz to 250kHz. I expect some influence, at least from the phase due to rolloff of the 6.5MHz BW.
Is there a better model out there? I'm using one from "Design with operational amplifiers and integrated circuits" by Franco, p266, eqn 6.15a. It has A(jf)=A0x1/[1+jf/fB]. I'm using a form of A0 similar to eqn 1.62, accounting for Rf and Rdut, as well as the input and output impedance (rd=>Zd, and r0=>z0), and substituting Zf and Zdut including parallel parasitics for Rf and Rdut.