PHYSICS · Multi-part · Fermi University Physics · Electronics, Instrumentation & Communication · Feedback, operational amplifiers and analogue signal processing · Sensors, detectors and experimental instrumentation
Electronics for Measurement, Ch Noise and the signal-to-noise ratio
Problem
An electret microphone capsule behaves, for noise purposes, as a capacitance Ccap=50pF at the node where the transducer's signal appears. It is biased through a resistor Rbias=10MΩ at 290K into a JFET-input amplifier with en=15nV/Hz and in=5.0fA/Hz, both referred to the amplifier's input. Downstream of this node the amplifier's own noise rolls off as a single first-order pole of corner frequency fc=20kHz; that roll-off does not act on anything at the bias node itself. The microphone's open-circuit sensitivity is 10mVPa−1, referred to the same node. Treat all three noise sources as independent.
((a)) Find the bias resistor's Johnson noise density eR and the corner frequency fRC of the low-pass that Rbias forms with Ccap at the node. Compare fRC with fc.
((b)) The resistor's noise reaching the node is shaped by that Rbias–Ccap low-pass, not by fc: use the kBT/C result to find its rms contribution at the node. The current noise in is injected at the same node and sees the same low-pass, but with a DC gain of Rbias instead of one; extend the noise-equivalent-bandwidth argument to find its rms contribution.
((c)) The amplifier's own en is applied directly at its input, undiminished by the node's impedance, and is limited only by fc: find its rms contribution over the amplifier's noise-equivalent bandwidth. Combine all three rms contributions in quadrature and identify the dominant one with its share of the noise power.
((d)) Find the smallest sound pressure detectable at a signal-to-noise ratio of one.