Non-zero Input Bias Current - Operational Amplifiers Types Tutorials Series

•In practice op-amps do not actually have zero input currents, but rather have very small input currents labeled I+ and I- in the figure at the left
–Modeled as internal current sources inside op-amp
–I+ and I- are both the same polarity
•e.g. if the input transistors are NPN bipolar devices, positive I+ and I- are required to provide base current
–In order to allow for slightly different values of I+ and I-, we define the term IBIAS as the average of I+ and I-
•Example: Given the op-amp shown in the bottom figure, derive an expression for vout that includes the effect of input bias currents
.jpg)
–Using the virtual short condition and KCL, we can write vIN/R1 = I- + (0-vOUT)/R2 or
vOUT = - (R2/R1)vIN + I-R2
–Plugging in values gives vOUT = - 20 vIN + 2 mV
Correcting for Non-zero Input Bias Current
.jpg)
•The effect of non-zero input bias current can be zero’ed out by inserting a resistor Rx in series with the V+ input terminal (as shown)
–This same correction works for both inverting and non-inverting op-amps
–We choose Rx such that the dc component on the output caused by I+ exactly cancels the dc component on vOUT caused by I-
–One can use either KCL (Kirchhoff’s Current Law) or superposition to show that choosing Rx = R1 || R2 completely cancels out the dc effect of non-zero input bias current
•KCL Method (inverting op-amp at left)
–vIN is applied to R1 and Rx is grounded
–v- = v+ = 0 – I+Rx due to virtual short
–Apply KCL to v+ input:
(vIN – v-)/R1 = I- + (v- - vOUT)/R2
–Solve for vOUT and substitute –I+Rx for v-
vOUT = - (R2/R1) vIN + I-R2 – I+Rx(R1 + R2)/R1
–Setting the dc bias terms equal yields
Rx = R1 || R2 = R1 R2/(R1 + R2)