The aim of this study was to investigate in how far the way the specimen is fixed to the testing device influences the micro-tensile bond strength of adhesives to dentin. Compared to a flat jig, a notched jig enables the specimen to be aligned easier and more accurately perpendicular to the interface, thereby concentrating better the tensile stress at the actual interface. A notched jig yielded a significantly higher bond strength and the graphs showed more uniform fracture curves. On the other hand, fixation of the specimen at their top and bottom guarantees a perfect perpendicular fixation to the interface, following the specimen's length-axis. The stress-time graphs revealed a completely different stress-distribution pattern. A failure closer to the dentin-composite interface was more often seen and the coefficient of variance was the lowest. Therefore, this completely newly designed top-bottom set-up produced the most reliable bond strength data.