An operational amplifier buffer (or voltage follower) is a unity-gain circuit where the output is tied directly to the inverting input, providing high input impedance and low output impedance to isolate a signal source from its load without altering the voltage. While it provides zero voltage gain (the gain is strictly 1), what it changes in a real circuit is the current delivery capability and the impedance profile. It takes a fragile, high-impedance signal that cannot supply meaningful current and clones it into a robust, low-impedance signal capable of driving heavy loads, long cables, or switched-capacitor ADC inputs.
Think of a high-impedance source like a weak water pump connected to a massive fire hose; the pressure (voltage) collapses the moment you open the valve (connect a load). The op-amp buffer acts as a secondary, high-pressure pump station that reads the weak pump's pressure and replicates it with massive flow capacity.
Selecting the Right Op-Amp for Buffering
Not every op-amp makes a good buffer. If you use a bipolar-input op-amp with high input bias current to buffer a 1MΩ piezo sensor, the bias current will create a massive DC offset voltage across the source impedance. You must match the op-amp's input stage technology to your source impedance and speed requirements.
| Op-Amp Model | Input Stage | Input Bias Current | Slew Rate | Best Buffer Application |
|---|---|---|---|---|
| LM358 | Bipolar | ~20 nA | 0.6 V/µs | Low-speed, low-Z general purpose (e.g., buffering a 10kΩ voltage divider) |
| TL072 | JFET | ~5 pA | 13 V/µs | High-Z audio, guitar pedals, and acoustic sensor buffering |
| OPA1612 | Bipolar | ~100 nA | 27 V/µs | Ultra-low noise audio line drivers and professional studio gear |
| MCP6001 | CMOS | ~1 pA | 0.6 V/µs | 3.3V MCU ADC driving, rail-to-rail battery-powered sensor nodes |
Notice the trade-off: JFET and CMOS inputs (TL072, MCP6001) offer picoamp-level bias currents, making them mandatory for sources above 100kΩ. However, bipolar parts like the OPA1612 offer vastly superior voltage noise performance, provided the source impedance is low enough (typically under 1kΩ) that the bias current doesn't generate excessive thermal noise.
The Loading Effect: A Worked Numeric Example
To see why a buffer is necessary, let's look at the math of the "loading effect" using a passive high-impedance source. Assume you have a sensor with a 100 kΩ output impedance generating a 1.00V DC signal. You want to read this with a microcontroller's ADC or a meter that presents a 10 kΩ input impedance.
The source impedance ($R_s$) and load impedance ($R_L$) form a voltage divider.
$V_{load} = V_{source} \times \frac{R_L}{R_s + R_L}$
$V_{load} = 1.00V \times \frac{10,000}{100,000 + 10,000} = \mathbf{0.090V}$
Result: You lose 91% of your signal before it even reaches the load.
We place a TL072 buffer between the sensor and the 10 kΩ load. The TL072 has a typical differential input impedance of $10^{12} \Omega$ (1 TΩ).
The voltage at the op-amp's non-inverting input is now:
$V_{in} = 1.00V \times \frac{1,000,000,000,000}{100,000 + 1,000,000,000,000} \approx \mathbf{1.00V}$
Because the op-amp operates in a closed-loop unity-gain configuration, its output drives the 10 kΩ load with an output impedance of roughly 0.1 Ω at DC.
Result: The full 1.00V is delivered to the load, and the op-amp sources the required 100 µA of current effortlessly.
Where You Meet Op-Amp Buffers in Practice
You will rarely see a buffer drawn as a standalone block on a high-level system diagram, but they are ubiquitous at the transistor level. Here is where they do the heavy lifting on the bench:
- SAR ADC Driving: Successive Approximation Register (SAR) ADCs use an internal switched-capacitor sampling network. When the sample switch closes, it demands a sudden burst of charge. If the source impedance is too high, the internal capacitor won't charge fully within the acquisition time, causing conversion errors. A CMOS buffer (like the MCP6001) acts as a "charge bucket," holding the voltage rock-solid during the sampling transient.
- Piezo and Acoustic Sensors: Piezoelectric transducers generate high voltages but have massive internal impedances (often >1MΩ). Without a JFET or CMOS buffer, the parasitic capacitance of a standard coaxial cable will form a low-pass filter with the sensor, rolling off all your high-frequency audio data.
- Audio Effects Loops: In guitar pedalboards, long cable runs and multiple true-bypass pedals create a massive capacitive load that drains high frequencies (the infamous "tone suck"). A JFET buffer at the input and output of the pedalboard isolates the guitar's pickups from the cable capacitance.
- Reference Voltage Distribution: When a single precision voltage reference (like a REF5025) needs to feed multiple ADCs or DACs, a buffer is used to prevent the varying dynamic currents of the converters from modulating the reference voltage and injecting noise back into the system.
Common Confusions and the Capacitive Load Pitfall
When discussing buffers, builders frequently confuse the analog op-amp buffer with a digital logic buffer (like a 74HC125). A digital buffer simply regenerates logic HIGH/LOW states using CMOS push-pull outputs; it cannot amplify or buffer a continuous linear analog waveform. Another common mix-up is confusing a buffer with a non-inverting amplifier. While a buffer is technically a non-inverting amplifier, its feedback resistor is 0Ω and its ground resistor is infinite (open), locking the gain strictly to 1.
The most common way to destroy an op-amp buffer's performance on the bench is driving a highly capacitive load (like a long shielded cable or a large filter capacitor) directly from the output pin. The capacitance interacts with the op-amp's internal open-loop output impedance, introducing a pole in the feedback loop that destroys phase margin. This causes the buffer to ring or break into high-frequency oscillation, turning your clean DC signal into a fuzzy, noisy mess on the oscilloscope.
The Fix: Always place a small "isolation resistor" (typically 10 Ω to 47 Ω) in series with the op-amp's output pin, before the feedback loop connection and the capacitive load. This resistor isolates the capacitive phase shift from the op-amp's output stage, restoring stability. For further reading on maintaining stability when driving capacitive loads, consult the Texas Instruments application note on op-amp stability.
Frequently Asked Questions
Can I use an LM358 to buffer a 3.3V microcontroller signal?
No. The LM358 is not a rail-to-rail output op-amp. On a single 3.3V supply, its output will top out around 1.8V to 2.0V, clipping your signal. Use a rail-to-rail CMOS part like the MCP6001 or TLV2371 instead.
Does a buffer consume power if the load is disconnected?
Yes. The op-amp will still draw its quiescent current ($I_q$) from the power supply to keep its internal transistor stages biased. For battery-operated IoT nodes, check the $I_q$ spec on the datasheet; a low-power part like the TSV6390 draws only 1.1 µA of quiescent current.
Where do I connect the feedback resistor in a buffer?
There is no feedback resistor. You run a direct wire (or a very short PCB trace) from the Output pin directly to the Inverting Input (-) pin. Keep this trace as short as possible to minimize parasitic inductance and stray capacitance, which can cause high-frequency ringing.
