Abstract: Numerical weather forecasting of tropical cyclones remains one of the most challenging tasks for numerical weather forecasters. This is because the complexity of numerical models, and the inability to account for all processes occurring in nature, means it is impossible to represent atmospheric events exactly by mathematical models. In present study we tried to investigate the sensitivity to three Convective Parameterization schemes for the Very severe cyclonic storm Phailin (9-14 Oct 2013) of Bay of Bengal using WRF (ARW) model and simulation of No Cumulus Parameterization (No Cu) was also undertaken to test model performance of simulating convection explicitly. All schemes were consistently performing better during the development phase, but as the cyclones matured the KF scheme showed cyclone as the most intense one. There was little separation between the KF, GD, and No CU schemes in terms of the central surface pressure and wind speeds of the simulated cyclones. The cyclone track in each scheme showed no variation in cyclone movement. Except the GD scheme all other schemes namely KF, BMJ, and No CU simulations all produced excess precipitation, especially in the eye wall region.