Wind turbine controllers are often designed based on simplified dynamical models. Over time, the turbine properties change because of structural degradation and wear and tear of actuation systems. This leads to an increasing discrepancy between the modeled and actual system properties, resulting in the controllers obtained by model-based techniques not yielding optimal performance for the actual system. Model-free data-driven controller tuning techniques can provide a solution by controller recalibration, and one such promising algorithm is Iterative Feedback Tuning IFT. The IFT algorithm is an automated, online, model-free tuning algorithm for tuning fixed-structure controllers. Previous work with IFT for optimizing wind turbine controllers has been conducted in high-fidelity simulation and has shown that the controller performance can be increased significantly. The IFT algorithm has been successfully implemented on many physical systems, including industrial applications; however, for wind turbine controllers, the algorithm has only been implemented in simulation. This work shows that with IFT, the control performance of a lab-sized experimental wind turbine can be increased, and it analyzes to what extent IFT can fine-tune the parameters of these controllers. This thesis subjects the collective pitch controller and the fore-aft tower damping controller of the experimental turbine to tuning with IFT. It is concluded that for the collective pitch controller, IFT can reduce a predefined cost function and that the algorithm tunes the controller parameters to the proximity of their optimal value, resulting in tighter reference tracking.