Aim: To evaluate the success rate, retrieval time, and root canal volume changes after retrieval of long and short-separated instruments using Zumax retrieval kit, BTR pen, and ultrasonic. Materials and Methods: Ninety Race rotary files #30/.04 were intentionally separated apically in the mesiobuccal canals of mandibular molars. Samples were divided into three groups (n=30) according to instrument retrieval device; Group (1) Zumax kit, Group (2) BTR pen, and Group (3) Ultrasonic. Each group was subdivided into two subgroups (n=15) according to the length of the separated files ;3 mm or 6 mm. CBCT scans were performed before and after retrieval to evaluate dentine volume changes. Chi-square and two-way ANOVA tests at a level of significance of 0.05 analyzed the data. Results: No significant differences were detected among the three groups regarding the success rate (P>0.05). Ultrasonic required significantly more time than Zumax kit and BTR during retrieving the long fragments(P<0.05). BTR pen required more time significantly than ultrasonic and Zumax kit to retrieve the short fragments(P<0.05). The volume of the removed dentine was significantly higher utilizing Zumax kit than BTR pen and ultrasonic(P<0.05). Zumax kit removed more dentine significantly while retrieving short fragments. Conclusion: Within the limitation of the present study, BTR pen provided a more conservative method for retrieving long-separated instruments, while ultrasonic technique could be used solely in the retrieval of short-separated fragments for more conservative instrument retrieval. Zumax retrieval kit removed more dentine, especially with short fragments.