Initially only one system was operated at a 3 day sludge age; however, the COD balance obtained at 12°C was poor (81.7% versus N balance of 97.4%). Examination of the system indicated that inadvertent aeration due to turbulence at the reactor surface was the cause, leading to underestimates of the oxygen utilization rate. When the problem was remedied (by using a smaller stirrer paddle) two systems were operated in parallel and very good COD and N balances were obtained for both. This highlights the need for careful attention to experimental detail, and the utility of mass balances in checking the validity of
experimental data.