Issue link: https://resources.pcb.cadence.com/i/1180526
PSpice A/D User Guide Convergence and "time step too small errors" October 2019 851 Product Version 17.4-2019 © 2022 All Rights Reserved. Is the initial approximation close enough? Newton-Raphson is guaranteed to converge only if the analysis is started close to the answer. Also, there is no measurement that can tell how close is close enough. PSpice A/D gets around this by making heavy use of continuity. Each analysis starts from a known solution and uses a variable step size to find the next solution. If the next solution does not converge PSpice A/D reduces the step size, falls back and tries again. Incorrect initial estimates can cause convergence failure or even false convergence. Consider following scenarios: ■ Power electronic circuits may NOT require tight current/voltage tolerances. Setting the value of ABSTOL to 1u will help in the case of circuits that have currents that are larger than several amps. ■ Unless the circuit conducts kilo-Amperes of current, however, setting ABSTOL to a value that is greater than 1u will cause more convergence problems than it will solve. ■ PSpice A/D does not always converge when relaxed tolerances are used. For example, setting the tolerance option, RELTOL, to a value which is greater than .01 can actually cause convergence problems ■ Setting GMIN to a value between 1n and 10n will often solve convergence problems. ■ Setting GMIN to a value, which is greater than 10n, may cause convergence problems. Diagnostics If PSpice A/D encounters a convergence problem it inserts into the output file a message that looks like the following. ERROR -- Convergence problem in transient analysis at Time = 7.920E-03 Time step = 47.69E-15, minimum allowable step size = 300.0E-15