Models are very often used in engineering. A model is a way of describing the behavior of a system, often by using mathematical formulae. A well-known model in physics is F = m x a, which describes the relation between the force acting on a mass, and its acceleration. Ohm’s Law can also be considered as a model. What a model does not do is explaining what is happening inside a system. Another limitation of a model is that its accuracy might not always be sufficient. We know e.g. that Newton’ law does not not work well for very tiny masses at atomic scale.
Models are used to make predictions and calculations. In fact, the exponential law that relates the diode current to the diode voltage is a model. However, making calculations with that formula is rather cumbersome. It might give an accurate result, but in electronic engineering, high precision is not always needed. When designing a circuit, an approximate calculation suffices. In a later stage, the circuit can be simulated or even built, to get an idea of the ‘digits behind the comma’. Therefore, models will be used that approximate the behavior of the diode. The purpose of these models is not to explain the physics of the working device. Their purpose is to simplify calculations. The simpler the model, the simpler the calculation. The more complex the model, the higher the precision of the calculation.
The first diode model: The ideal diode
The first diode model or the model of the ideal diode is the simplest. It only contains an ideal switch. This switch is closed at forward bias: The current is positive, and the voltage across the diode is 0 V. At reverse bias, the switch is open. There is no current flowing through the diode, and the voltage is negative. This characteristic resembles that of a practical diode, but the approximation is rather coarse. Yet, in practice it is the most frequently used model, certainly in power applications where voltages vary from tens to hundreds of a Volt. Notice that this model corresponds to the simplified property of the diode: It allows current only in one direction.
The second diode model
If more accuracy is desired, the second diode model is used. The difference with the first diode model is that the knee voltage is no longer neglected. In the second model, the ideal switch is put in series with a voltage source, of which the value corresponds to the knee voltage. For Silicon diodes, this value is 0.7V. So, when the diode conducts a current, it has to be positive (ID > 0 A) and the voltage equals the knee voltage (VD = 0.7 V for Silicon diodes). When the diode is off, there is no current (ID = 0 A) and the voltage is below the knee voltage (VD < 0.7 V for Silicon diodes). The second diode model is used in applications where the voltages are relatively low, say a few V.