Aim To examine the debridement of round and oval root canals prepared to two apical sizes with and without ultrasonically activated irrigation. Methodology Mandibular premolars with round (n = 48) and oval (n = 48) root canals were pair‐matched after microCT scanning and randomly divided into two experimental groups (n = 20): group 1, rotary NiTi to size 20, .04 taper; group 2, rotary NiTi to size 40, .04 taper. Specimens were subdivided into two subgroups (n = 10): subgroup A, syringe and needle (SNI); subgroup B, ultrasonically activated irrigation (UAI). Untreated canals (eight oval and eight round) served as controls. Specimens were processed for histological evaluation for measurement of the remaining pulp tissue and debris (RPT), and the perimeter percentage of root canal area untouched by the instruments (PRAU). Following assessment of normality, multiple‐way anova models were used to study the effects of preparation size, irrigation technique and canal cross‐sectional shape, and their interactions on the RPT and PRAU (α = 0.05). Results All experimental groups had significantly less RPT than the control (P < 0.05). Both the preparation size (20 vs. 40) and the irrigation technique (SNI vs. UAI) had a significant effect on RPT (P = 0.006 and P < 0.001, respectively). Groups irrigated with SNI always had significantly greater RPT than those irrigated with UAI, irrespective of the preparation size (P < 0.001). Canals prepared to size 20 had significantly greater RPT than those with size 40 in the SNI subgroup (P < 0.001), but there was no significant difference in the UAI subgroup (P = 0.481). Conclusions Root canals prepared to a larger size (40) were cleaner than those prepared to a smaller size (20), when irrigation was performed with a syringe and needle. When the irrigant was ultrasonically activated, smaller preparations resulted in canals that were as clean as larger preparations. This finding was common to both round and oval canals of freshly extracted premolars with vital pulp tissue.