An enhanced WSR 88D hail detection algorithm (HDA) is used for its performance evaluation in 110 sever convective weather processes. From 2006 to 2011, this method was used to detect the extreme weather with CINRAD/CD radar data in the northeastern Tibetan Plateau, including the calculation of Sever Hail Index (SHI) and the Maximum Evaluation Hail Size (MEHS). The localization of the default hail parameters is conducted. The result shows that the detection performance is evidently improved; POD (Probability Of Detection) increases from 85% to 89%; FAR (False Alarm Ratio) reduces from 41% to 21%; CSI (Critical Success Index) raises from 53% to 75%; MEHS (Expected Maximum Hail Size) drops from 74%to 33%, which is bigger than the actual maximum hail size. The results are useful to the hail detection with CINRAD/CD radar in nowcasting service.