This task can be automized by PSpice. That way an information about the yield can be made, respectively the circuit can be optimized in view of the yield already during the development phase.
The Sensitivity function locates the sensitive componants of a circuit. Only those componants, whose tolerances have a big influence on the function of a circuit, are then fitted with expensive componants and a small tolerance. For componants with less of influence on the total function can be budget cost-saving options.
Without complex test series it is possible to determine with the simulations of the PSpice Advanced Analysis long-term effects of the circuits and so to reach a high quality for an optimal price, within the specification. In the Optimizer we differentiate according to two methodes. The circuit can be optimized, as an objective function or a reference curve is given. The Optimizer function then determines the componant parameters that lead to the best compliances with the specifications.
With the Parametric Plotter multidimensional sweeps are shown grafically and so the changings of the circuit as a function of several variables can be made clear.
With this simulation option the sensivity of all componants of the circuit is calculated. The sensivity show the relative influence of each componant of one or several objective functions of a circuit as for example maximum power, band width, center frequency, etc. For the evaluation is a graphical representation available. It shows the influence on the selected objective function of the critical componants, in view of the componant tolerance. Therefore you can select uncritical componants with bigger tolerances, while componants with small value changings have a big influence of the objective functions, are specified systematically with close tolerances. So costs at insensitive points can be reduced.
Monte Carlo Analysis simulates the random scattering of the tolerances of componants. The result is a repartition of feasibility of the tolerances for each componant. If these componants are combined in the serial production, the different tolerances of the componants may have different influences on the objective function. Using this analysis you can give informations about the size of the yield, that means how many products would fail in the quality control. By the graphical displays weak points can be recognized and the developer can plan selectively the quality and enhance the yield.
The Optimizer function can, based on a given circuit (netlist), discretely dimension the componants in a way that the objective function can be achieved as exactly as possible. Based on a defined objective function the Optimizer function calculates not only the theoretical optimal componant values, e.g. R1 = 57,34 ohm and R2 = 14,29 ohm and ß = 129. It is also possible to predetermine as objective functions certain componant series of which the values may be selected. So the Optimizer simulation in a E24 serie would choose values of R1 = 56 ohm and R2 = 22 ohm as componant combination for an optimum of the objective function.
With the Parametric Plot several values can be sweeped, that means it will be displayed, based on only one simulation, the results of all values from 0 Volt to 10 Volt, with the step size 1 Volt and values for a capasitor of 1 pF to 10 pF in steps of the E12 serie. The result is a clear serie of curves. The quick display of a complex data by a simulation helps to dimension a circuit correctly. The caracteristic curve of the componants ina an operating point are measured with a wizard. In doing so values can be dimensioned for the ramp-up-time, overshoot, performance, tension of a MOSFET–power level, so that power loss, tensions and overshoot achieve optimal values.
The stress analysis is in english humorously called "Smoke Analysis" because in case of too much stress the componants "smoke". With this simulation the maximum derating can be determined and so an infromation be given about the componant stress. Via the model parameters the stress can be shown in percentage. The developer wants informations about the natures of load: thermal at maximum current and highest voltage possible, wire of the junction temperature in °C, respectively the thermal contact resistances jC and jA, as they appear in the circuit.
With the SLPS Interface an interconnection of the simulation MatLab® SimuLink® to PSpice is established. This allows completely new possiblities for co-simulation. So algorithms, which are realized as software in programmable componants, are displayed in mathematical blocks in MatLab because this is the preferred simulation tool of the software developers. In contrast the electrical behavior of the control logic of the control circuit is displayed in PSpice, the preffered simulator of the electronics engineers. In the co-simulation both simulators synchronize and use in a common simulation the respective interim values of the other simulator. That way total systems including the software can be simulated. MatLab and SimuLink are registered trademarks from The Mathworks.