Recently, the new regulations on emission standards have prompted a reconsideration of the design of automotive catalytic converters in order to reduce the light-off period of the catalyst. The catalytic converter light-off period is very Important since almost 80% of the emissions from vehicles occur within the first three minutes after cold start in the FTP-75 test. In order to meet these new regulations, current studies have suggested that the catalyst should be “close-coupled”; that is fitted close to the engine exhaust manifold. In order to design “close-coupled” converters, the designer may have to resort to truncated inlet and outlet cones, or distorted inlet pipes due to space limitations. Hence, it is very difficult to achieve good mixing of the exhaust gas, and a good flow distribution at the inlet cross section of the monolith. Based on such a current status in the study of the catalytic converter, the present work focuses on the time-dependent flow patterns, both in the exhaust manifold and the catalytic converter using Computational Fluid Dynamics (CFD). A three-dimensional grid model of an engine exhaust manifold and a close-coupled catalytic converter was developed and analyzed. The flow simulations were performed using KIVA-3 for non-reacting flow fields. These simulations were performed with transient boundary conditions applied at the inlet to the exhaust runners to simulate the opening and closing of exhaust valves. The CFD results were used to study flow uniformity under different operating conditions and to identify the best location of the oxygen sensor.