Mastering the Inverting Amplifier Gain Equation: Derivation, Examples, and Bench Reality

The direct answer for the voltage gain of a standard inverting operational amplifier configuration is A_v = -(R_f / R_in). The negative sign indicates a 180-degree phase inversion between the input and output signals. While this formula is foundational in circuit theory, applying it on the bench requires navigating resistor tolerances, op-amp bandwidth limits, and unit conversion traps. Below is the complete derivation, symbol mapping, and practical design workflow.

The Core Inverting Amplifier Gain Equation

The inverting amplifier relies on negative feedback to force the inverting input (-) to the same potential as the non-inverting input (+), which is typically tied to ground. This creates a "virtual ground" at the inverting node. Because the ideal op-amp draws zero input current, all current flowing through the input resistor (I_in) must flow through the feedback resistor (I_f).

Applying Kirchhoff's Current Law (KCL) at the inverting node:

(Vin - 0) / Rin = (0 - Vout) / Rf

Rearranging for the voltage gain (A_v = V_out / V_in) yields the core equation:

Av = -(Rf / Rin)

Symbol Definition Table

Real-World E24 Resistor Pairings for Target Gains

In theory, you can pick any resistor values to achieve an exact gain. On the bench, you are limited to standard E24 (5% tolerance) or E96 (1% tolerance) series values. Here is a data-dense look at how standard E24 values map to common target gains, including the inevitable ratio errors.

Note: For audio or precision DC applications where a 1% error is unacceptable, use 1% E96 resistors or add a multi-turn trimmer potentiometer in series with R_f.

Assumptions, Limits, and Realistic Magnitudes

When the Formula Applies (and When it Breaks)

The equation Av = -(Rf / Rin) assumes an ideal operational amplifier. According to foundational texts like All About Circuits, this requires:

• Infinite open-loop gain (A_OL): Ensures the virtual ground at the inverting input is exact.
• Infinite input impedance: Ensures zero current flows into the op-amp's input pins.
• Zero output impedance: Ensures the output voltage doesn't sag under load.

In reality, a BJT-input op-amp like the classic LM741 has an input bias current of ~80 nA. If you use massive resistors (e.g., R_in = 1 MΩ, R_f = 10 MΩ) to save power, that 80 nA flowing through 10 MΩ creates an 800 mV DC offset error at the output. For high-impedance designs, switch to a JFET or CMOS op-amp like the TL072 (bias current ~5 pA) or OPA2134.

Realistic Answer Magnitudes

What does a "normal" gain look like?

• Attenuation (|A_v| < 1): Valid, but usually unnecessary unless you specifically need phase inversion alongside attenuation.
• Standard Range (|A_v| = 2 to 100): The sweet spot for audio preamps, sensor conditioning, and active filters.
• Extreme Gain (|A_v| > 1000): Highly problematic. Amplifying by 1000 also amplifies the op-amp's internal input offset voltage (V_os) by 1000. An LM741 with a 2 mV V_os will output a 2 V DC error before you even apply a signal. Furthermore, the Gain Bandwidth Product (GBWP) limits you. A 1 MHz GBWP op-amp running at a gain of 1000 only has 1 kHz of usable bandwidth. For gains >1000, cascade two stages (e.g., two stages of -31.6).

Unit Mistakes That Break the Math

Another frequent error is treating decibels (dB) as linear V/V. If a spec sheet calls for 40 dB of gain, plugging "40" into the equation as A_v is wrong. 40 dB equals a linear V/V gain of 100. Always convert dB to V/V first using 10^(dB/20).

Rearranged Forms and Design Workflows

On the bench, you rarely solve for A_v directly. Usually, you know the gain you need and the input impedance you want, and you must solve for a specific resistor. Here are the algebraically rearranged forms for design workflows:

For deeper insights into how feedback resistor values interact with op-amp stability and phase margin, refer to the Analog Devices MT-033 Tutorial. High-value feedback resistors interacting with parasitic input capacitance can create unwanted poles in the feedback loop, leading to high-frequency oscillation.

Worked Bench Problems with Unit Tracking

Problem 1: Audio Mixer Summing Node Design

Scenario: You are building an audio mixer. The source is a synthesizer with an output impedance of 600 Ω. You need a voltage gain of exactly -15. To prevent loading the synthesizer and causing high-frequency roll-off, the amplifier's input impedance (R_in) must be at least 10 times the source impedance.

Step 1: Determine minimum R_in.

• Z_source = 600 Ω
• R_in(min) = 10 × 600 Ω = 6,000 Ω (6 kΩ)
• Select the next standard E24 value up: R_in = 6.8 kΩ (6,800 Ω).

Step 2: Solve for R_f using the rearranged equation.

• Equation: Rf = -Av × Rin
• Substitute values: Rf = -(-15 V/V) × 6,800 Ω
• Calculate: Rf = 15 × 6,800 Ω = 102,000 Ω

Step 3: Select standard resistor and verify error.

• 102 kΩ is not a standard E24 value. The closest E24 values are 100 kΩ and 110 kΩ. Let's use 100 kΩ.
• Verify actual gain: Av = -(100,000 Ω / 6,800 Ω) = -14.70 V/V.
• Error: (14.70 - 15.00) / 15.00 = -2.0%. For an audio mixer, a 2% gain deviation is inaudible and perfectly acceptable.

Problem 2: Sensor Amplification and Rail Clipping Check

Scenario: A thermocouple amplifier outputs a DC signal of 45 mV. You are using a TL072 op-amp powered by ±12 V rails. Your breadboard is populated with R_in = 4.7 kΩ and R_f = 470 kΩ. Will the output clip?

Step 1: Calculate the linear voltage gain.

• Equation: Av = -(Rf / Rin)
• Convert to base units to be safe: R_f = 470,000 Ω, R_in = 4,700 Ω.
• Substitute: Av = -(470,000 Ω / 4,700 Ω) = -100 V/V.

Step 2: Calculate the ideal output voltage.

• Equation: Vout = Vin × Av
• Substitute: Vout = 45 mV × (-100 V/V)
• Unit tracking: 0.045 V × -100 = -4.5 V.

Step 3: Check against op-amp output swing limits.

• The TL072 is not a rail-to-rail output op-amp. On ±12 V rails, its guaranteed maximum output swing is typically ±10 V (leaving ~2V of headroom to the rails).
• Our calculated V_out is -4.5 V.
• Conclusion: -4.5 V is well within the ±10 V safe swing limit. The signal will not clip, and the inversion will be cleanly reproduced.