variavel0=Edson Gomes Moreira Filho - edson@lmt.coppe.ufrj.br COPPE/UFRJ Marcelo Moreira Mejias - mejias@lmt.coppe.ufrj.br COPPE/UFRJ Helcio Rangel Barreto Orlande - helcio@serv.com.ufrj.br COPPE/UFRJ Albino José Kalab Leiroz - leiroz@ime.eb.br Instituto Militar de Engenharia Abstract. The numerical solution of partial differential equations within irregular domains using the Finite Volume Method and grid generation techniques requires the evaluation of approximations for the transformation metrics at the volume center and at the center points of the volume faces. However, for highly distorted or stretched volume cells, the average of coordinate approach may not provide an appropriate representation of volume and face center positions. The present work presents an analysis of an alternative technique for the metric evaluation, which makes use, in each coordinate direction, of a grid with twice the number of points used for the governing equation solutions. The proposed approach allows the required transformation metrics and the volume and face center positions to be calculated within the computational domain. The Finite Volume Method is applied to the transformed conservation equations using a regularly spaced grid within the computational domain. In order to analyze the computational performance of the proposed technique, test cases, for which analytical solutions are available, are studied. Initially, analytical grid generation techniques are applied to one-dimensional convection-diffusion model equations. Numerically obtained results are compared with analytical values showing the precision of the proposed approach. Two-dimensional test cases are also studied. Results show that, for a given precision, the proposed double-grid approach allows the usage of coarser discretizing grids. Therefore, a balance between the increase of computational costs associated with numerically generating a finer grid and the solution of the transformed governing equation with more precise transformation metric values is observed. Keywords. Finite volume, Double Grid, Grid Generation, Metrics Evaluation.