This paper introduces a novel distributed secondary control (DSC) approach for droop-controlled microgrids (MGs). The conventional approach relies on a MicroGrid Central Controller (MGCC) to restore frequency and voltage deviations, but the proposed DSC implementation avoids centralization by distributing control logic across individual DG units. This distributed networked control system ensures that the failure of a single unit does not cause the entire system to fail. The DSC not only restores frequency and voltage but also ensures reactive power sharing. Experimental results demonstrate the feasibility of the DSC, showing its ability to handle communication latency and data dropout issues. The paper includes a detailed analysis of the primary and secondary control structures, the proposed DSC implementation, and its performance in various scenarios, including black start processes and reactive power sharing. The results indicate that the DSC is more robust in the presence of communication delays and data dropout compared to the conventional centralized approach.This paper introduces a novel distributed secondary control (DSC) approach for droop-controlled microgrids (MGs). The conventional approach relies on a MicroGrid Central Controller (MGCC) to restore frequency and voltage deviations, but the proposed DSC implementation avoids centralization by distributing control logic across individual DG units. This distributed networked control system ensures that the failure of a single unit does not cause the entire system to fail. The DSC not only restores frequency and voltage but also ensures reactive power sharing. Experimental results demonstrate the feasibility of the DSC, showing its ability to handle communication latency and data dropout issues. The paper includes a detailed analysis of the primary and secondary control structures, the proposed DSC implementation, and its performance in various scenarios, including black start processes and reactive power sharing. The results indicate that the DSC is more robust in the presence of communication delays and data dropout compared to the conventional centralized approach.