The capability of the experimental systems used in two-chambered microbial fuel cell experimentation was tested in terms of repeatability and reproducibility. The optimal number of replicates needed to discriminate between responses of technical interest, both in open-circuit and closed-circuit experiments was studied. For N = 4 replicates, these differences were set to 9.0% COD(R) units, 261 mV and 63 mg/L in VFAs for open-circuit experiments and 3.6%, 30.2 mV and 45 mg/L in closed circuit experiments. Cycling operation with several reactor refills using fresh wastewater and keeping the same biofilm between cycles almost has no influence in COD(R) and VFAs but voltage standard deviation reduces by one half between the first and fourth cycle. This study takes part by the option of increasing the number of replicates because although it may have lower repeatability, the amount of data generated per unit time is larger than running the experiments in cycles. (c) 2009 Elsevier Ltd. All rights reserved.