Rotating unbalance is a costly source of harmful vibration affecting a wide variety of applications such as high-speed machine tools and turbomachinery. Active balancing technology allows adjustment of a machine’s balance state while conditions change as the machine continues to operate. An adaptive control law for single-plane active balancing is presented here for control of steady-state rotation-synchronous vibration. The adaptive control method, based on the well known “influence coefficient” approach, requires no a priori modeling or estimation of plant dynamics. A control “gain” parameter is introduced and shown to enhance stability robustness of the non-adaptive portion of the control law. A recursive on-line system identification method is presented that uses exponential weighted averaging to mitigate the effects of measurement noise and system nonlinearities. Experimental results establish the efficacy of the adaptive control system even in the face of certain nonlinear and time-varying rotordynamic systems.