Abstract - A revised model for the current-voltage (I-V) characteristics based on the current EKV model is presented. A EKV model uses a linear combination of two logarithmic functions to interpolate between two regions of operation to generate a single-piece model which allows continuous derivatives with respect to the external bias voltages. This model is rather convenient for circuit simulation for its simplicity and continuous nature. However, the original EKV model does not have the required precision compared to the simplified, symmetrical two-piece model, which is commonly used in SPICE. Thus, a modified EKV model has been developed to obtain the precision required as well as preserving the original advantages of such a model.
INTRODUCTION
To simulate MOS transistors in a circuit, we need a model which can generate results in a reasonable amount of time while the accuracy is maximized. Although the complete charge-sheet model can produce the best accuracy, it is rather computational intensive attribute to the fact that it contains a high-order of polynomial as well as implicit expressions for the surface potentials. And, to be sure, the latter requires iterative techniques to be evaluated, which is completely unsuitable for circuit simulation.
Till now, one of the most popular SPICE MOSFET Level-3 model is the Simplified Symmetric Model [1] – [4]. This model as named suggested is simple and reasonably accurate within a certain range of operation, namely, the weak and strong inversion. Yet, this model completely ignores the moderate inversion have we not stretched the boundary between strong and moderate inversions. Because of the nature of this model, a discontinuity occurs when we go from the weak inversion into the strong inversion. In addition, even inside the strong inversion, there are two equations required to describe saturation and non-saturation regions. Hence, in this regard, this multi-piece model does not really simplify the matter for it complicates the derivatives of the current with respect to the external biases.
A EKV model solves the problem. The initiative of an EKV model is not to improve the accuracy but to be able to collapse the simple model discussed above into a single-piece which has a continuous derivative with each of the externally applied bias voltage, and most important of all, it describes the moderate inversion. To do so, it interpolates between the strong and weak inversion to approximate what is happening in the complete charge-sheet model. Hence, it is rather a mathematical effort as opposed to physics to arrive such a model. Nevertheless, the present EKV model, though simple, is biased either toward the strong inversion, depending on some parameters. But if we can fix this problem by mathematical means, an EKV model is reasonably suitable for circuit simulation if the precision expectation does not exceed that from the symmetrical model.
In the following section, the revised EKV model is presented, and compared to the complete charge-sheet model as well as the simplified symmetrical model. In section III, the results and precision issues from the modified EKV model will be discussed.
To see more, please click HERE to download the entire Microsoft Word 97 document.