International Journal of Advanced Engineering, Management and Science (IJAEMS) [Vol-2, Issue-6, June- 2016] Infogain Publication (Infogainpublication.com ) ISSN : 2454-1311 www.ijaems.com Page | 633 Distortion Analysis of CMOS Based Analog Circuits Suman Kumar Choudhary, Sathe R. B., Dongare M.P., Sonawane R. B. Department of Electronics Engineering, Amrutvahini College of Engineering, Maharashtra, India Abstract –The amplifiers are the vital part of the analog circuit designs. The linearity of the CMOS is of most important concern in the design of many analog circuits. There are several aspects regarding nonlinear distortion analysis in analog circuits implemented in CMOS technology. Basically, the investigations visualize the nature of the total harmonic distortion (THD) dependence on the amplitude and frequency of the input signals. In this paper, the basic building blocks of analog integrated circuits such as Common source amplifier with diode connected load and Differential amplifier with current mirror load have been presented for distortion analysis. The MOSFET model used for simulation is BSIM3 SPICE model from 0.13-μm and BSIM4 SPICE model from 22- μm CMOS process technology. HSPICE circuit simulator tool is used for distortion analysis of CMOS circuits. It is evident that the above function gives remarkable insight of the nonlinear behavior of the considered circuits and it is worth considering for further investigations. Keyword – CMOS circuits, frequency response, nonlinear distortion, THD, TSMC. I. INTRODUCTION Electronic circuits are inherently nonlinear. In the most general case such circuits, like any physical truth, can be described at least in principle, by nonlinear partial derivative differential equations (to say nothing about uncertainties, noise, and quantum effects etc). Fortunately, under certain theory regarding the amplitude and rate of deviation of the input signal associated to the physical dimensions of the devices, these equations can be reduced to nonlinear or linear usual differential equations (the lumped hypothesis). Generally, the approach of the designer about the nonlinearities of electronic devices ranges between avoiding them and using them. Distortion is one of the most essential undesired effects that appear in analog circuits due to non-linearity in the characteristics of the transistors or from the influence of the associated circuit (coupling component or load). As far as CMOS circuits are concerned, there a basically two type of nonlinear elements: resistive and capacitive ones [1, 2]. The major reason of nonlinear distortion comes from the nonlinearities of the resistive part of the model. However, the capacitive nonlinearity i.e., the nonlinearity of the q(u) characteristics (including here trans-capacitances as well) can influence the distortion in various applications. The main objective of this paper is to build a mechanism to investigate the distortion of CMOS circuits in deep submicron technology. In analog integrating circuits, replacing the bipolar transistors with MOS transistors, the problem of comparatively large values of the input bias, input offset currents and of the small value of the input impedance was solved, with the drawback of reducing the voltage gain caused by the quadratic characteristic of the MOS transistor working in saturation region. This information reveals the problem of applied input voltage levels. This effect will decrease if we scale down the size of MOS device. As the technology is now shrinking to nanometer range, there are some additional problems occur due to short channel effects such as body effects, channel-length modulations, signal-dependent capacitive effects and frequency-dependent distortions arising from the capacitive load in the CMOS circuits. In this paper we will present several aspects regarding distortion in a few simple CMOS circuits aiming at showing that the THD plot with respect to amplitude and frequency of the input signals is a remarkable apparatus to put into evidence interesting nonlinear behaviors [3]. II. CLASSICAL WAYS OF CHARACTERIZING DISTORTION Generally, the output waveform is not exact reproduction of the input signal waveform because of different types of distortion that may arise, either from the inherent nonlinearity in the characteristics of the transistors or FETs or from the influence of the associated circuit [3]. 2.1 Types of Distortion 2.1.1 Non-Linear Distortion This type of distortion results from the production of new frequencies in the outputs which are not present in the input signal. These new frequencies or harmonics results from the weak non-linear performance of circuit devices. This type of distortion is occasionally referred to as “Amplitude distortion”. When a sinusoidal input signal is applied to a circuit, the output signal will not only contain the ground harmonic