A novel signal filtering methodology for obtaining liquid phase tracer responses from conductivity probes

Local conductance measurements using a conductivity cell, consisting of electrode plates encased in plastic tubing, have been employed for point conductance measurements in two-phase gas-liquid flows. It is shown that such measurements can provide local liquid mixing information in gas-liquid flows such as those encountered in bubble and staged bubble columns either with an upward flow of gas in a batch liquid, or with co-current or counter-current flow of gas and liquid. This requires the characterization of the liquid phase conductance from the electrical signals acquired by such a conductivity probe (cell) in the presence of bubbles, which is non-trivial clue to the systematic lowering of the conductance of the probe measurement volume when a bubble contacts or pierces the probe tip. Thus, application of standard digital filtering techniques to signals masked by frequent bubble passage results in an inaccurate representation of liquid conductance by filtered signals. A special purpose software filtering technique has been developed in this work to address this issue and to cleanly extract the actual liquid phase conductance from response signals representing the instantaneous point conductance of a bubbling two-phase gas-liquid mixture. The implementation of the new filtering algorithm has been achieved through coupling with standard Butterworth filters in MATLAB(TM), version 5. (C) 2000 Elsevier Science Ltd. All rights reserved.