This paper presents an improved approach for modelling electrostatic precipitators on one fine computational domain in clean air. Contrary to the previously published numerical techniques, the Finite Difference Method (FDM) integrated with Full Multi-grid method (FMG) is used to predict the optimal value of ion mobility at a certain applied voltage in precipitators. Three smoothing solvers for FMG are implemented including Gauss–Siedel, Successive Over Relaxation (SOR) and Successive Under Relaxation (SUR). Gauss–Siedel is examined against SOR and SUR on finer domains and it is found that Gauss Siedel is greatly superior in terms of the timing performance. Two main ideas are discussed in this paper for the first time; the first idea is that the value of ion mobility is not constant for all applied voltages as published previously in the precipitators. The second one is predicting only one global optimal grid for solving Poisson and continuity equations that grantees both low truncation and round off errors that well matched the previously published experimental measurements more than that of the previously published numerical techniques. Moreover, the idea of tracking the optimal ion mobility and grid size gives the FDM-FMG the advantage of predicting an accurate picture for the electrical conditions in the early design of precipitators over the other numerical techniques. Finally, the FDM-FMG gives a more confident results for current density and voltage characteristics for precipitator which is proved by experiments.