| eCircuit Center |
|
|
About SPICE |
SPICE Basics | Running SPICE
| CIRCUIT COLLECTION |
SPICE Commands | SPICE Demos and Downloads |
Op Amp Input Offset VoltageCIRCUIT
OP_VOFF.CIR Download the SPICE file Build any opamp circuit, apply 0V to its input, and what do you expect at the output? Although you'd be tempted to say 0 V, there's actually an error voltage present at its output. What causes this error? You can trace the error back to a number of unbalances in the op amp's internal transistors and resistors. To account for this in a circuit design, the net error is modeled as an offset voltage, Voff, in series with op amp's input terminals. How will it affect your circuit? That depends on the op amp itself and your circuit design.
INPUT OFFSET VOLTAGE The input offset voltage can range from microvolts to millivolts and can be either polarity. Generally speaking, bipolar op amps have lower offset voltages than JFET or CMOS types. The offset voltage is modeled in series with one of the op amp input terminals. Which one? Although the net effect is the same at either input, it's much easier to analyze Voff in series with the positive (V+) input. Why? The resulting circuit with Voff at V+ looks just like the non-inverting amplifier configuration. The analysis for this circuit is a simple one.
AMPLIFIER WITH OFFSET VOLTAGE Ignoring Voff for a moment, is the circuit (shown above) an inverting or non-inverting amplifier? The answer is yes, both! With the input signal source set to 0 V ( shorted ), the inverting and non-inverting amplifiers look the same. The analysis for offset voltage is independent of the amplifier configuration. You can predict the error at the output Vo by the equation for the non-inverting amplifier Vo_error = Voff ( R2 / R1 + 1 ) What danger is this equation warning you about? If you have a large signal gain in your circuit, the amplifier will increase the error Voff along with the signal. CIRCUIT INSIGHT Run a simulation of OP_VOFF.CIR. Voltage source VOFF models the offset voltage, initially set to +1 mV. With R1 = 10k and R2 = 100 k, what is the error at the output V(4)? The output error gets bumped to a whopping 11 mV as predicted by the equation above. Choose a different gain and/or offset voltage. Run a simulation. Is the output error what you expected?
In some applications, the output offset may be unacceptable. So how do you get rid of this error? One solution involves canceling it using another voltage of the same magnitude and opposite polarity. But the exact magnitude and polarity of the offset is unknown! You'll need a range of values (using a potentiometer, for example) to cover your basis. The question is how to get the canceling voltage into the circuit? One simple way is via a resistor R3 to the op amp's negative input.
For simplicity, VPOT represents a voltage generated by a potentiometer divider circuit. The output due to VPOT is simply
which is the same equation as for the inverting amplifier. HANDS-ON DESIGN Insert VPOT and R3 into the circuit by removing the * at the beginning of the statements. Run a simulation of OP_VOFF.CIR. Voltage VPOT generates a ramp that goes from -0.2V to +0.2 V. This simulates you turning the potentiometer over the full range. Check out the output voltage V(4). At some point, does the output get trimmed to 0 V? Suppose you have a poorer grade op amp with a maximum offset of 5 mV. Change the VOFF value to 5MV. Run a new simulation. Does the output get trimmed to 0V? If not, increase VPOTs range beyond -0.2V to 0.2V. How big of a range to you need to cancel the 5mV offset. Another way you can adjust the cancellation range is to adjust R3. The equation above tells you that decreasing R3 increases the gain from VPOT to Vo.
HOW DOES R3 EFFECT GAIN? What effect does R3 have on the gain of your circuit? That depends on which gain we're talking about. There are three gain of interest here. ( Notice that looking left from the op amps' negative input, R3 is effectively in parallel with R1.)
CIRCUIT INSIGHT What are the above equations telling us? If you don't want your gain severely affected, keep R3 as large as possible. You can check the influence of R3 by setting VPOTs's ramp limits to 0V and 0V. Then, with R1 = 10k, R2 = 100k and VOFF = 1mV, the expected output error is 11 mV. Run a simulation to see how R3 = 1MEG or 100k changes the output voltage error?
OFFSET VOLTAGE DRIFT Just because you trimmed out the offset voltage, doesn't mean all is tranquil on paradise island. The input offset voltage will drift with temperature; you have no control over this. But, knowing your overall error budget, you can select an op amp with a low enough offset drift for the intended temperature range.
INPUT BIAS CURRENT Input offset voltage is not the only error of the op amp's input. The input stage is made of transistors, requiring a finite amount of bias current for operation. The circuit above assumes the bias is negligible. However, real op amps have bias currents to be reckoned with. The good news is there are clever techniques you can use to minimize and cancel out these errors too. (See Op Amp Input Bias Currents )
SPICE FILE Download the file or copy this netlist into a text file with the *.cir extention. OP_VOFF.CIR - OPAMP OFFSET VOLTAGE * * AMPLIFIER CIRCUIT * R1 0 2 10K R2 2 4 100K XOP1 3 2 4 OPAMP1 ;V+ V- VOUT * * OPAMP INPUT OFFSET VOLTAGE VOFF 3 0 DC 1MV * * OFFSET COMPENSATION * POTENTIOMETER DIVIDER - VPOT *VPOT 10 0 PWL(0MS -0.2V 10MS 0.2V) *R3 10 2 1MEG * * * OPAMP MACRO MODEL, SINGLE-POLE * connections: non-inverting input * | inverting input * | | output * | | | .SUBCKT OPAMP1 1 2 6 * INPUT IMPEDANCE RIN 1 2 10MEG * GAIN BW PRODUCT = 10MHZ * DC GAIN (100K) AND POLE 1 (100HZ) EGAIN 3 0 1 2 100K RP1 3 4 1K CP1 4 0 1.5915UF * OUTPUT BUFFER AND RESISTANCE EBUFFER 5 0 4 0 1 ROUT 5 6 10 .ENDS * * ANALYSIS .TRAN 0.1MS 10MS * VIEW RESULTS .PRINT TRAN V(4) .PROBE .END
© 2002 eCircuit Center |
|||||||||||||