We applied the methodology of continuous titrations by feedback-based flow ratiometry and the principle of compensating errors [Anal. Chem. 72 (2000), 4713] to potentiometric acid-base titrations. In a typical case, the titrant (a standard base) flow (F-B) is varied in response to a controller output voltage, while the total how (F-T, consisting of an acid sample and the titrant how) is held constant. The titrant flow is initially ramped upward linearly. At the instant, the detector senses the equivalence potential is reached, the controller output (instantaneous value V-H) reverses its ramp direction, thus decreasing the titrant flow linearly at the same rate. When the equivalence point is sensed again, the controller voltage (instantaneous value V-L) is ramped in reverse once more, going upward. Because of the lag time between a change in the controller output and its outcome being sensed by the detector, the controller voltage corresponding to the actual equivalence point is the average of V-H and V-L. Continuous sensor-governed operation of the controller thus results in a triangular waveform. The mean de bias of this waveform during any cycle gives the equivalence point controller voltage V-E. This principle of compensating errors allows true titrations in principle, requiring no calibration curves. Even though pH electrode response is slow compared with optical sensors, high throughput (12 a per titration) is attainable with reasonable reproducibility (0.7% R.S.D.). The method was applied to diverse acid-base titrations. (C) 2001 Elsevier Science B.V. All rights reserved.