This paper is about a time-domain grey-box system identification applicable to scale model helicopters. It presents a unique way to obtain the initial values to be used in the estimation process and a systematic way to partition the set of state equations and to identify all its parameters. A widely published linearized model suitable for hover and near hover condition is used in this paper. There is no need to go through tedious experiments to determine good initial parameter values to start estimation. The performance of the proposed procedure is demonstrated using a ground-truth model and by showing the well-matched outputs and system eigenvalues, as well as the convergence of the identified parameters to their true values. The validation using real flight data is further strengthened by showing the prediction results.