We address the problem of optimal node placement for ensuring connected coverage in sensor networks. We consider two different practical scenarios. In the first scenario, a certain region (or a set of regions) are to be provided connected coverage, while in the second case, a given set of n points are to be covered and connected. For the first case, we provide solutions that are within a small factor of the optimum. For the second case, we present an algorithm that runs in polynomial time, and guarantees a constant factor approximation ratio.