Dynamic Transmission Error (DTE), or the difference between the ideal and actual positions of the output gear in a gear pair, is a common metric utilized to evaluate the dynamic characteristics of a spur gear pair. Studies most commonly investigate gear pair noise and vibrations, yet the underlying mechanisms leading to noise and vibrations are not fully understood. Variations introduced by the manufacturing processes such as tooth indexing errors complicate the gear pair dynamics further. However, prior to understanding the impacts these variations have on gear pair dynamics, a no-error baseline data set must be created as a means of experimental comparison.\nIn order to completely characterize the dynamic behavior of the gear pair, and therefore develop this no-error database, a setup that can operate a gear pair under tightly-controlled dynamic conditions is required as well as dedicated accelerometer based DTE measurement system. This research developed such a measurement system to acquire high-frequency vibration data, as well as a digital data processing scheme to compute DTE. With the focus on characterization of the baseline behavior with no tangible manufacturing errors, a set of experiments was performed within a wide range of operating speeds (sweep up from 500 to 4200 rpm and sweep down from 4200 to 500 rpm) and torque loads (100-300 Nm). The results were then examined in the frequency domain to develop a complete characterization of the no-error spur gear pair through the creation of waterfall plots and root-mean-square forced response curves. These plots were then used to establish the baseline resonance behavior depending upon system operating conditions. In addition, further analysis was completed to highlight the apparent softening type nonlinear behavior due to intermitted loss of contact and natural frequency increase seen with increased transmitted torque load.