The Core Job: Amplifying the Voltage Difference

At its most fundamental level, what does an op amp do? It takes the voltage difference between its two input pins and multiplies it by a massive internal gain, outputting the result constrained only by its power supply rails. If the non-inverting input (+) is even a fraction of a millivolt higher than the inverting input (-), the output swings hard toward the positive supply rail. If the (-) pin is higher, it swings toward the negative rail.

Think of an op amp in open-loop (no feedback) like a highly sensitive, motorized water valve. A microscopic pressure difference between two pilot lines causes the valve to slam completely open or completely shut. Because a typical op amp has an open-loop gain of 100,000x or more, this "slamming" behavior is useless for linear amplification on its own. To make it useful, we use negative feedback—routing a portion of the output back to the inverting (-) input. This forces the op amp to continuously adjust its output to keep the voltage difference between the two inputs at virtually zero, creating a stable, predictable, and highly precise linear amplifier.

Bench Rule of Thumb: Never leave an unused op amp in a dual or quad package floating. The massive open-loop gain will cause it to amplify ambient RF noise, saturate the output stage, and waste power (or overheat). Always wire unused sections as voltage followers (output tied directly to the inverting input) with the non-inverting input tied to ground or mid-rail.

Pinout, Symbol, and Operation Regions

The standard schematic symbol for an op amp features five primary connections. While a single op amp requires five pins, you will almost always buy them in 8-pin DIP or SOIC packages containing two op amps (dual) or 14-pin packages containing four (quad). In a dual package, the two op amps share the V+ and V- power pins.

  • Non-Inverting Input (+): The signal reference. A positive voltage here relative to the (-) pin drives the output positive.
  • Inverting Input (-): The feedback node. A positive voltage here relative to the (+) pin drives the output negative.
  • Output (OUT): The low-impedance drive pin that sources or sinks current to maintain the input differential at zero.
  • V+ (Positive Supply): The upper voltage limit for the output swing.
  • V- (Negative Supply): The lower voltage limit (often ground in single-supply circuits, or a negative voltage in dual-supply circuits).
Op Amp Operation Regions (Assuming ±15V Dual Supply)
Region Input Condition (V+ vs V-) Output Voltage State Typical Output Current
Linear (Active) V(+) ≈ V(-) (within µV) Between -13V and +13V (dictated by feedback network) Sourcing/sinking up to 20mA-40mA
Positive Saturation V(+) > V(-) by > 1mV Clamped near V+ rail (e.g., +13.5V for standard, +14.9V for rail-to-rail) Max rated source current (often ~25mA)
Negative Saturation V(-) > V(+) by > 1mV Clamped near V- rail (e.g., -13.5V for standard, -14.9V for rail-to-rail) Max rated sink current (often ~25mA)

Selecting, Biasing, and Safe Default Part Numbers

Selecting an op amp requires matching its power supply topology to your signal. If you are measuring an AC audio signal or a sensor that swings above and below zero, you need a dual supply (e.g., +12V and -12V) so the output can swing negative. If you are reading a DC sensor into a microcontroller ADC, you will use a single supply (e.g., 0V and 5V). In single-supply AC applications, you must "bias" the input by creating a virtual ground at mid-rail (2.5V) using a resistor voltage divider, allowing the signal to swing up and down around that 2.5V center point.

Here are the safe default part numbers you should keep in your bench drawer, complete with their critical ratings:

  • LM358 (General Purpose): The undisputed workhorse. Works on a single supply down to 3V or dual up to ±16V. Inputs can go slightly below the negative rail. Limitation: Output cannot swing to the positive rail (tops out at V+ minus 1.5V). Cost: ~$0.15.
  • MCP6002 (Microcontroller Friendly): A modern 3.3V/5V rail-to-rail input/output (RRIO) CMOS op amp. Perfect for interfacing with Arduino, ESP32, or Raspberry Pi ADCs. Cost: ~$0.40.
  • TL072 (Audio / JFET): Excellent for audio mixing and active filters due to low noise and high slew rate. Requires a higher minimum supply voltage (±5V minimum) and inputs cannot reach the negative rail. Cost: ~$0.60.
  • OPA2134 (Precision / FET): High-end audio and precision DC measurement. Extremely low distortion and high output drive. Cost: ~$4.50.

Real-World Scenario: The Shunt Resistor Amplifier

Let’s walk through a classic bench scenario to see how theory meets reality when designing a complete application circuit.

The Setup: You need to measure up to 10A of DC current using a 50mV / 10A shunt resistor. You are reading this with an Arduino Uno, which has a 5V, 10-bit ADC. At 10A, the shunt outputs 50mV. You need to amplify that 50mV to 5.0V to use the full resolution of the ADC. This requires a non-inverting amplifier with a gain of 100.

The Circuit: We use the standard non-inverting formula: Gain = 1 + (Rf / Ri). To get a gain of 101 (close enough to 100), we select standard 1% resistor values: Rf = 100kΩ and Ri = 1kΩ. We wire the shunt to the non-inverting (+) pin, place the 1kΩ resistor from the inverting (-) pin to ground, and the 100kΩ feedback resistor from the output to the inverting (-) pin.

The Numbers: 50mV input × 101 gain = 5.05V output. On paper, this is perfect.

The Outcome (What Went Wrong): You grab an LM358 from your drawer, power it from the Arduino’s 5V pin (single supply), and apply 10A to the shunt. The Arduino serial monitor reads an ADC value of roughly 716 (which corresponds to 3.5V), not the expected 1023 (5.0V). You check your wiring; it's flawless. You swap the LM358 for a fresh one; same result.

The Diagnosis: You fell victim to the LM358’s output swing limitation. The LM358 is not a rail-to-rail output op amp. On a 5V single supply, its internal output transistors drop about 1.5V, meaning the maximum output it can physically achieve is 3.5V. Your circuit is saturating against its own internal ceiling.

The Fix: Swap the LM358 for an MCP6002. Because the MCP6002 features a true rail-to-rail output stage, it will swing within millivolts of the 5V positive rail, giving your Arduino the full 5.0V signal you designed for.

Bench Testing: How Op Amps Fail and Multimeter Diagnostics

Op amps are generally robust, but they fail predictably when abused. The most common failure modes include input overvoltage (exceeding the supply rails on the input pins, which destroys the input differential pair), output short circuits (driving the output directly to ground or a low-impedance load beyond its 25mA limit, melting the internal output transistors), and ESD damage (particularly on CMOS and JFET input stages like the MCP6002 or TL072).

While a multimeter cannot test an op amp's dynamic AC performance or open-loop gain, it is highly effective at identifying dead shorts and blown junctions. Here is how to test a suspected bad op amp on the bench:

  1. De-energize and Discharge: Remove all power from the circuit. If large capacitors are present in the power supply filtering, discharge them with a bleeder resistor. Testing a powered IC in resistance/diode mode will yield false readings and can damage your multimeter.
  2. Test the Input Stage (Diode Mode): Set your DMM to diode test mode. Place the black probe on the V- pin. Touch the red probe to the IN+ pin, then the IN- pin.
    • For BJT-input op amps (like the LM358), you should read a forward voltage drop of roughly 0.6V to 0.7V. This is the base-emitter junction of the input transistors.
    • For JFET or CMOS-input op amps (like the TL072 or MCP6002), you should read OL (Open Loop) because the gates are insulated.
    • Failure indicator: If you read 0.000V (a dead short) between the inputs, or between an input and the supply rail, the input stage is blown.
  3. Check for Output Shorts: Still in diode or continuity mode, check the resistance between the OUT pin and V+, and the OUT pin and V-. You should not see a dead short. A low resistance here indicates the internal output push-pull transistors have melted together, usually from driving a heavy capacitive load or a direct short to ground.
  4. The Ultimate Functional Test (Voltage Follower): If the IC passes the DMM tests but you still suspect it is faulty (e.g., it has high offset voltage or internal leakage), breadboard it as a voltage follower. Tie the OUT pin directly to the IN- pin. Apply a known DC voltage (e.g., 2.5V from a potentiometer) to the IN+ pin. Power the IC. Measure the OUT pin with your DMM. It should match the IN+ voltage within a few millivolts. If it deviates significantly, or if the output is stuck at a rail, the IC is internally damaged and must be replaced.
Safety Warning for Mains Sensing: If you are using an op amp to amplify a shunt resistor on a mains-voltage AC line (e.g., measuring current on a 120V/240V water heater), standard op amps will expose your low-voltage microcontroller to lethal mains potential if the shunt ground reference shifts. For any circuit connected to non-isolated mains, you must use a dedicated isolation amplifier (like the TI AMC1301) or a Hall-effect current sensor (like the ACS712) to maintain galvanic isolation.

Understanding what an op amp do on the bench goes far beyond memorizing the golden rules of ideal op amp theory. By respecting the physical limitations of output swing, input common-mode ranges, and supply topologies, you can move past frustrating "why isn't this working" moments and design robust, precise analog front-ends for any digital system.