Electromagnetic Transient (EMT) models are essential to wind power plant developers and transmission system operators alike. Therefore, during the development of any new Wind Turbines (WTs), EMT models are also developed. These models are qualityassured by validating against measurements from a prototype turbine, specifically for fault ride-through (FRT) cases. One way of performing model validation is through the voltage play-back method described in IEC 61400-21-2, where a voltage-dependent source reproduces the voltage measurements from the site (at the MV terminals of the WT transformer) and the current/power injections from the WT are compared against the measurements (at the LV terminals of the WT transformer). WPPs are designed and developed with the help of these models. However, a WPP electrical system can be complex with wind turbines, passive components, cables, and active components such as filters, Static VAr Compensators, and/or HVDCs. So it is interesting and sometimes essential to ensure that the assessment studies performed during the development of WPP are accurate. Furthermore, passive equipment in the WPP may change important characteristics either due to internal failures or aging, wind turbine control software version, and parameters may be updated, etc. It is often valuable to ensure that models are still capable of representing the actual behaviors seen on the field so that grid compliance and stability can be studied throughout the operation. In this context, this paper presents the EMT model validation of an offshore WPP with Siemens Gamesa Renewable Energy's (SGRE) direct-drive (DD) Type IV Wind Turbines using the voltage play-back method under real power system events. The faults were unbalanced and happened at the nearby grid, leading all the turbines of the plant into FRT mode. The model used was developed during the design of the WPP and the evaluation is done at both the plant and turbine levels. It is shown that the EMT models of operational WPPs were able to reproduce the real measurements with high accuracy. The paper also discusses the different challenges concerning measurement location placement, data acquisition, aggregation methods, preprocessing, turbine connection status, and production at the time of the fault.