Sensor nodes in wireless sensor networks are constrained by battery power. And the sensor nodes sense a specific phenomenon in the environment and route the sensed data to a relatively small number of central data processing nodes, called sinks. So there exists imbalance in energy consumption in essence. In this paper, the authors are not only interested in determining a TDMA schedule that minimizes the total time required to complete the convergecast, but also consider a TDMA scheduling algorithm can which balance load to prolong network lifetime. They consider a simple version of the problem in which every node generates exactly one packet, and the node has multi-transmission power levels which can vary according to its transmission distance. The formula of energy consumption are analyzed for the general k-hop network in theory according to the typical network parameters. It is proved that there exits a best k that makes the network lifetime the longest. A TDMA scheduling algorithm is proposed for general k-hop network, and the upper bound of time slot required in general k-hop network is given as follows. Based on the analysis, the entire network scheduling strategy can be obtained for general k-hop network. Theoretical analysis and numerical simulation results confirm the accuracy and effectiveness of the algorithms.