Instrument fracture is an unfortunate but possible sequela of instrumentation of canals, especially when the instrument is bound at the tip. The purpose of this study was to compare the torque required to fracture three file sizes of three different rotary file types around two simulated canal curvatures, gradual or acute, when the tip of the working end of the file was bound. Profile Series 29 0.04 and 0.06 taper and Profile 0.06 ISO rotary files were placed passively into simulated canal curvatures of the same angle but of different radii. The file tips were bound 2 mm from the working end and a measurable torque was applied until fracture. ANOVA with Tamhane post-hoc comparison showed that the 0.06 Series 29 did not differ from the ISO 0.06 taper or the 0.04 Series 29 but there was statistical difference (p < 0.01) showing that the 0.04 Series 29 broke with less force than did the 0.06 ISO files. Statistical tests (p < 0.01) also showed smaller files failed with less torque, as did files in more acute canal curvatures. Copyright © 2002 by The American Association of Endodontists.