In rechargeable or energy harvesting wireless sensor networks (WSNs), a key concern is the max flow or data rate at one or more sinks. However, this data rate is constrained by the available energy at each node as well as link capacity. To date, in order to increase the amount of data extracted from a WSN, past works have considered routing approaches or they optimize the location of sinks. In contrast, we take a novel approach whereby we aim to “upgrade” the recharging rate of a finite number of “bottleneck” nodes using the so called auxiliary chargers (ACs) equipped with wireless power transfer capability. We formulate a mixed integer linear program (MILP) for the NP-hard problem at hand and propose three novel solutions to place ACs: 1) Path, which preferentially upgrades nodes on the shortest path among paths from sources to sinks, 2) Tabu, a meta-heuristic that first uses Path as the initial solution. It then searches for a neighboring solution that yields a higher max flow rate, and 3) LagOP, which approximates the said MILP using Lagrangian and sub-gradient optimization. Our results show that Tabu has the best performance, where it is able to achieve 99.40% of the max flow rate derived by MILP in tested scenarios.