Fundamentals of Feedback Amplifiers

Feedback amplifiers form the backbone of precision electronic measurement systems. By returning a portion of the output signal to the input, these circuits create a closed-loop configuration that dramatically improves performance over open-loop designs. The core concept is straightforward: a feedback network samples the output, compares it to the input, and adjusts the amplifier's behavior to minimize error. This mechanism underpins the high accuracy required for voltage and current measurements in scientific instrumentation, industrial control, and test equipment.

Negative vs. Positive Feedback

The two primary types of feedback serve different purposes. Negative feedback subtracts a fraction of the output from the input, reducing gain but providing stability, linearity, and bandwidth extension. It is the dominant choice for measurement circuits because it suppresses distortion, reduces sensitivity to component variations, and lowers output impedance. Positive feedback adds the feedback signal to the input, increasing gain and creating oscillation or hysteresis. While useful in comparators and oscillators, positive feedback is generally avoided in precision measurement due to instability and nonlinearity. Measurement systems rely almost exclusively on negative feedback to ensure predictable, repeatable results.

Loop Gain and Closed-Loop Transfer Function

The performance of a feedback amplifier is governed by its loop gain, defined as the product of the amplifier's open-loop gain (A) and the feedback factor (β). The closed-loop gain is given by A_cl = A / (1 + Aβ). For large loop gain (Aβ >> 1), the closed-loop gain approaches 1/β, which is determined solely by passive components in the feedback network. This insensitivity to variations in the amplifier's open-loop gain is the key to measurement accuracy. For example, an op-amp with open-loop gain of 100 dB (100,000) and β = 0.1 yields a closed-loop gain of approximately 10, with the actual value deviating by less than 0.001% for a 10% change in open-loop gain. Such robustness makes feedback amplifiers indispensable when repeatability and low drift are required.

How Feedback Improves Measurement Accuracy

Measurement accuracy depends on an amplifier's ability to deliver a faithful, scaled replica of the input signal without adding error. Feedback amplifiers address four critical factors: gain stability, linearity, distortion, and bandwidth.

Gain Stabilization and Linearity

Open-loop gain varies with temperature, supply voltage, and frequency. A feedback loop reduces this variation by a factor equal to the loop gain (1 + Aβ). For instance, if the open-loop gain changes by 10%, a loop gain of 1000 reduces the closed-loop gain change to 0.01%. This stabilization directly translates to more accurate voltage and current measurements over temperature and time. Similarly, nonlinearities in the amplifier's transfer function are attenuated by the same factor. The feedback network itself can be made from precision resistors with low temperature coefficients, ensuring that the closed-loop gain remains linear across the signal range. This linearity is essential for applications like data acquisition and sensor conditioning, where the relationship between input and output must be strictly proportional.

Distortion Reduction and Noise Performance

Harmonic distortion, caused by nonlinearities in the amplifier's output stage, is reduced by the feedback mechanism. Negative feedback causes the amplifier to generate a correction signal that cancels the distortion components. The resulting total harmonic distortion (THD) is typically reduced by a factor of (1 + Aβ). In high-precision measurements, such as audio testing or strain gauge signal conditioning, feedback amplifiers can achieve THD levels below 0.001%. Feedback also affects noise, though the effect depends on where the noise originates. Noise injected inside the feedback loop is suppressed, while noise from the input source or the feedback network is amplified. Proper design places low-noise components at the input and uses feedback to minimize internal noise contributions.

Bandwidth Extension

The closed-loop bandwidth of a negative feedback amplifier is approximately the open-loop gain–bandwidth product (GBW) divided by the closed-loop gain. This means that feedback extends the useful frequency range beyond the open-loop bandwidth. For example, an op-amp with GBW = 10 MHz configured for a gain of 10 yields a closed-loop bandwidth of 1 MHz. This wider bandwidth allows measurement circuits to accurately capture fast transients and high-frequency signals, such as those from switching power supplies or high-speed current sensors. The trade-off is that excessive feedback can lead to instability, which must be managed through compensation techniques.

Feedback in Voltage Measurement Circuits

Accurate voltage measurement relies on amplifiers that provide precise gain, high input impedance, and low offset. Feedback amplifiers are used in several fundamental topologies.

Non-Inverting and Inverting Amplifiers

The non-inverting amplifier uses a resistive divider from output to inverting input to set the gain: V_out = V_in × (1 + R_f / R_g). Its high input impedance (often megohms) minimizes loading on the voltage source, making it ideal for measuring the output of sensors or voltage references. The inverting amplifier, with gain = –R_f / R_g, offers lower input impedance but simpler biasing and common-mode rejection. Both configurations rely on negative feedback to achieve the exact gain dictated by the resistor ratio. For the most demanding applications, discrete precision resistors with low drift are used to ensure gain accuracy better than 0.01%.

Instrumentation Amplifiers

When measuring differential voltages in the presence of large common-mode signals (e.g., bridge circuits, shunt-based current sensing), an instrumentation amplifier is required. This circuit uses three op-amps with feedback networks to provide high common-mode rejection ratio (CMRR), typically exceeding 100 dB. The first stage provides differential gain with high input impedance, while the second stage removes common-mode voltage. Feedback ensures that the CMRR remains stable across frequency and temperature. Instrumentation amplifiers are widely used in data acquisition systems, medical devices, and industrial process control where small voltage differences must be resolved accurately.

Precision Voltage References

Feedback amplifiers are also central to the design of voltage references, such as bandgap references or buried zener references. An op-amp with negative feedback regulates the output voltage by comparing it to a stable internal reference (e.g., a bandgap or zener diode). The feedback loop corrects for temperature and load changes, achieving temperature coefficients as low as 1 ppm/°C. These references are used as the baseline for analog-to-digital converters, digital-to-analog converters, and calibrators. The feedback amplifier's role is to buffer the reference element and provide the required current without disturbing its accuracy.

Feedback in Current Measurement Circuits

Current measurement poses different challenges, including the need for low input impedance, high dynamic range, and isolation. Feedback amplifiers address these through several specialized configurations.

Transimpedance Amplifiers

For measuring current from photodiodes, ionization chambers, or other current-source sensors, the transimpedance amplifier (TIA) is the standard. The feedback resistor (R_f) converts input current to output voltage: V_out = -I_in × R_f. Negative feedback holds the input (usually the inverting terminal) at virtual ground, providing low input impedance and minimal voltage burden on the current source. This is critical for accurate low-current measurements, where even millivolt drops would introduce significant errors. The feedback capacitor (C_f) in parallel with R_f provides stability by limiting bandwidth and attenuating high-frequency noise. Careful selection of these components yields femtoampere-level sensitivity.

Shunt-Based Current Sensing with Feedback

In power systems, current is often measured indirectly via the voltage drop across a precision shunt resistor. A feedback amplifier amplifies this small voltage (typically 10–100 mV) to a usable level. To maintain accuracy, the amplifier must have low input offset voltage and drift, as the shunt voltage is small. A common topology is the differential amplifier with a high-gain feedback loop that rejects common-mode voltage (often present in high-side sensing). The gain is set by precision resistors, and feedback ensures that the amplifier's common-mode rejection remains high across the full supply range. Modern current-sense amplifiers integrate the feedback network on-chip, achieving typical offset voltages below 50 µV and gain errors under 0.1%.

Feedback in Current Transformers

For AC current measurement, current transformers (CTs) provide galvanic isolation. However, the burden resistor across the CT secondary must be precisely defined to maintain the turns-ratio accuracy. A feedback amplifier can be configured as an active burden, using a transimpedance stage to set the effective burden to a low, stable value. This approach reduces phase errors and extends the measurable frequency range. The feedback loop also compensates for the CT's magnetizing inductance, improving linearity at low frequencies. Such circuits are used in power monitoring equipment and protective relays.

Practical Considerations and Limitations

While feedback amplifiers dramatically improve measurement accuracy, they are not without constraints. Understanding these limitations is essential for reliable circuit design.

Stability and Compensation

Excessive negative feedback can cause instability, leading to oscillations or peaking in the frequency response. The phase shift around the loop must be less than 180° at the frequency where the loop gain drops to unity (the gain crossover). To ensure stability, feedback amplifiers often include compensation networks, such as a capacitor in the feedback path or a dominant pole in the amplifier's internal circuit. These compensation elements reduce the loop gain at high frequencies, ensuring a phase margin of at least 45° to 60°. In measurement circuits, where wide bandwidth is needed, designers may use decompensated op-amps with smaller internal compensation, but then must ensure the closed-loop gain is high enough to maintain stability. Phase margin can be measured and simulated using Bode plots or Nyquist criteria.

Slew Rate and Bandwidth Trade-offs

The maximum rate of change of the output voltage (slew rate) limits the ability to track fast signals. For a given closed-loop gain, the full-power bandwidth is determined by the slew rate: f_max = SR / (2π × V_peak). In current measurement circuits, especially when sensing fast transients (e.g., from a switching converter), the amplifier must have sufficient slew rate to avoid distortion. However, high slew rate often comes at the expense of higher power consumption and noise. The gain–bandwidth product also imposes a limit: the closed-loop bandwidth cannot exceed GBW divided by the noise gain. This trade-off forces a choice between high gain at low frequencies or lower gain with wider bandwidth.

Noise and Offset

Every feedback amplifier contributes noise from its internal transistors and resistors. The input-referred noise (voltage and current noise) is amplified by the noise gain (1 + R_f / R_g). In low-signal applications, such as measuring microvolt-level voltages or picoampere currents, noise can dominate. Designers must select amplifiers with low voltage noise (e.g., < 1 nV/√Hz) or low current noise (e.g., < 1 fA/√Hz). Offset voltage and drift due to temperature are also critical. Chopper-stabilized or auto-zero amplifiers use feedback to continuously null the offset, achieving typical drift below 0.05 µV/°C. These techniques extend the usable dynamic range of measurement systems.

Conclusion

Feedback amplifiers are not merely optional components in accurate measurement systems—they are the enabling technology. By stabilizing gain, linearizing the transfer function, reducing distortion, and extending bandwidth, negative feedback transforms a basic amplifier into a precision tool. Whether measuring DC voltages with instrumentation amplifiers, photocurrents with transimpedance stages, or high currents with shunt sensors, the core principle remains the same: closed-loop control suppresses errors and delivers reliable, reproducible data. Understanding the fundamentals of loop gain, stability, and practical limitations allows engineers to design measurement circuits that achieve the accuracy demanded by modern scientific and industrial applications.