ObjectiveThis ex vivo study evaluated the effect of pre-flaring and file size on the accuracy of the Root ZX and Novapex electronic apex locators (EALs). Material and methodsThe actual working length (WL) was set 1 mm short of the apical foramen in the palatal root canals of 24 extracted maxillary molars. The teeth were embedded in an alginate mold, and two examiners performed the electronic measurements using #10, #15, and #20 K-files. The files were inserted into the root canals until the "0.0" or ''APEX'' signals were observed on the LED or display screens for the Novapex and Root ZX, respectively, retracting to the 1.0 mark. The measurements were repeated after the pre-flaring using the S1 and SX Pro-Taper instruments. Two measurements were performed for each condition and the means were used. Intra-class correlation coefficients (ICCs) were calculated to verify the intra- and inter-examiner agreement. The mean differences between the WL and electronic length values were analyzed by the three-way ANOVA test (p<0.05). ResultsICCs were high (>0.8) and the results demonstrated a similar accuracy for both EALs (p>0.05). Statistically significant accurate measurements were verified in the pre-flared canals, except for the Novapex using a #20 K-file. ConclusionsThe tested EALs showed acceptable accuracy, whereas the pre-flaring procedure revealed a more significant effect than the used file size.