Using the Junction Cap Model

This section describes how to use the junction cap model statement.

General Syntax

The general syntax for including a diode element in a Star-Hspice netlist is:

Dxxx nodeplus nodeminus modelname <<area=>val>

+ <<peri=>val> <<pgate>=val> <<dtemp>=val>

+ <<off>=val> <<IC=>val> <<m=>val>

Juncap Model Syntax

The juncap model statement syntax is:

.MODEL modelname D level=4 <keyword=val>

Examples

.model MD D level=4

+AB=2E-12 LS=2E-6 LG=1.3E-6 DTA=0 TR=30 VR=0.3

+JSGBR=1.2e-3 JSDBR=1.3e-3 JSGSR=1.1e-3

+JSDSR=1.3e-3 JSGGR=1.4e-3 JSDGR=1.4e-3 NB=1.6

+NS=1.3

+NG=1.3 VB=0.9 CJBR=1.2e-12 CJSR=1.2e-12

+CJGR=1.3e-12 VDBR=1.6 VDSR=1.3 VGDR=1.2 PB=0.5

+PS=0.6 PG=0.4

Setting Juncap Model Parameters

Theory

~

This section summarizes the elementary physics of a junction diode. Refer to semiconductor textbooks for additional information.

Generally, the current voltage characteristics can be represented as follows:

For V<V D , the charge of the junction capacitance is described by:

JUNCAP Model Equations

JUNCAP Model

The JUNCAP model is intended to describe formed by the source, drain or well-to-bulk junction devices, limited to the case of reverse biasing of these junctions. Similar to the MOS model, the current equations are formulated and AC effects are modeled via charge equations using the quasi-static approximation.

In order to include the effects from differences in the sidewall, bottom, and gate-edge junction profiles, these three contributions are calculated separately in the JUNCAP model.

Both the diffusion and the generation currents are treated in the model, each with individual temperature and voltage dependence.

In the JUNCAP model, a part of the total charge comes from the gate-edge junction very close to the surface. This charge is also included in the MOS model charge equations and is counted twice. However, this results in only a very minor error.

In the next section, the model equations are presented. Correct operation of the model in a circuit simulator environment requires some numerical additions, which are described in the section on implementation. Any fixed capacitance that is present on a node (e.g., metal-1-to-substrate capacitance) must appear in a fixed capacitor statement or must be included in INTCAP. They no longer form the JUNCAP model in contrast to the old NODCAP model.

Nomenclature

The following table lists the electrical variable parameters:

The following table lists internal variables and parameters:

ON/OFF Condition

Circuit solution involves a process of successive calculations. The calculations are started from a set of "initial guesses" for the electrical quantities of the non-linear elements. The devices start in the default state.

Example

DC Operating Point Output

The DC operating point output facility gives information on the state of a device at its operation point.

Temperature, Geometry and Voltage Dependence

The general scaling rules, which apply to all three components of the JUNCAP model, are:

Internal Reference

The internal reference parameters for the bottom component are specified by:

Similar formulations hold for the locos-edge and the gate-edge components. Replace the index B by S and G , and the area AB by LS and LG .

For the locos-edge :

For the gate-edge :

JUNCAP Capacitor and Leakage Current Model

In the charge description, the following internal parameter is defined:

In order to prevent an unlimited increase of the voltage derivative of the charge, the charge description is in two parts: the original power function and a supplemented quadratic function. At the cross-over point between these regions, indicated by Vl, the following parameters are defined:

Similar expressions exist for the locos-edge and gate-edge charges, Qjsv and Qjgv .

The total charge characteristic can be described by:

Using elementary mathematics, we can derive from Equation 12.63 (above) simple equations for the capacitance of the bottom area:

Similar expressions exist for Cjsv and Cjgv .

Total Capacitance

The total capacitance can be described by:

*Bulk to source or bulk to drain diode current.

Diffusion and Generation Currents

With the scaled parameters of the preceding section, the diffusion and generation current components can be expressed as:

The first relation concerning the diffusion component is valid over the whole operating range. The second relation, describing the generation current, shows an unlimited increase in the derivative of this function at V=V DB . Therefore, the power function is merged at V=0.0 with a hyperbolic function in the forward bias range. The exponential part is divided by . This enables a gradual decrease in the generation current component.

The hyperbolic function is used. The parameter B controls the decrease of the current for voltages V>0.0 for all generation components. The value of B is fixed and set to 2 in the model. The continuity constraints of function and derivative in the merge point lead to the following relations for F sb and V ab :

The generation current voltage characteristic in the forward region becomes:

Final Model Equations

The final model equations for the currents of the bottom area are:

Similar expressions exist for the locos-edge and gate-edge components.

The total junction current can be expressed as: