Design Of Low Noise Integrated Amplifier Used For Photoelectric Detection

With the development of detection technology, more and more weak signals are processed in basic research, industrial application and so on. Noise becomes the main limitation in amplifier, which could affects the accuracy of usable signal of circuit easily. On the basis of noise theory and analysis of noise in CMOS circuits, this paper presents several methods that are used to reduce low-frequency noise and offset of amplifier. These methods include optimized structure of op-amplifier, autozeroing (AZ) technique, correlated double sampling (CDS) technique and chopper stabilization (CHS) technique. After analysis and comparison, this thesis presents a low noise circuit used for weak signal amplification in photoelectric detection. The amplification circuit uses the chopper stabilization technique to reduce the offset and low-frequency noise. It applies modulation to transpose the signal to a higher frequency where is no1/f noise, and then demodulates the signal back to the baseband after amplification. Since the noise and offset are modulated only once, they are transposed to the Since the noise and offset are modulated only once, they are transposed to the high-frequency, which can be easily removed by a low-pass filter, leaving the amplifier ideally without any offset and low-frequency noise. The circuit includes four parts:Preamplifier, High-pass filter, Demodulator and Switch capacitor low-pass filter. The input modulator chooses optical modulator instead of general MOS switch modulator which could introduce the residual offset to the output due to its nonideal effects. After design, the chip was taped out using the CSMC0.6μmDPDM mix signal process. The simulation and test results show the low noise amplifier based on chopper stabilization could reduce the noise and offset, and the equivalent input noise is66.1nV/(?) for a chopper frequency of50kHz.