Controlling your process based on pH is usually a tricky matter. Control Talk blog‘s Greg McMillan and Emerson’s Dave Joseph presented a short course, Adventures in pH Control, at last week’s Emerson Exchange conference in Anaheim. It was a best in conference winner in the short course category which is a feat with over 350 total sessions presented at the conference.
Their goal was to help the short course attendees recognize the opportunities and associated challenges of pH control, learn about modeling and control opportunities, optimize hardware implementation, understand poor performance root causes, prioritize improvements, and gathering insights.
They defined pH as the unit of measurement for determining the acidity or alkalinity of a solution. The scale is logarithmic based upon molar hydrogen ion concentration (negative logarithm). It’s most common measured by glass electrode/silver reference system. The scale goes from 0 (highly acidic) to 14 (highly alkaline).
Sulfuric acid is on the extremely acidic end of the scale and sodium hydroxide is on the extremely alkaline end of the scale. The life of pH measurement devices depends upon the condition of the process. Higher temperatures decrease sensor life. Also, high pH conditions decrease glass life at any temperature. Accuracy and response time degrade over time. These conditions cause unreliable feedforward control. New glass formulations help to preserve the response time over extended use.
Dave described a pH measurement loop as consists of the glass electrode, a reference electrode with its liquid junction, a temperature element, a solution ground, and an analyzer that is apart from the sensor parts (electrodes and temperature elements). Microprocessors have added configuration and diagnostic capabilities such as calibration data storage, recent calibration history, performance variable trending, and diagnostic history. These added diagnostics help increase uptime and reduce maintenance. See the presentation for examples of how these diagnostics are used.
When it comes to using pH in closed loop control, it is important to reduce dead time in order to reduce nonlinearity effects. Greg provided some tips such as performing filtering on the feedforward signal to remove noise and keep the corrective action from arriving too early, which could create an inverse response. Feedforward control effectiveness depends on eliminating reagent delivery to the solution being measured. See slides 51-52 for Greg’s complete list of key points.
The third section of the presentation explored practical considerations. These included causes and effects of drift, common problems with titration curves, measurement selection and installation effects, improving accuracy and required maintenance, installation issues—piping, vessel, mixing patterns, valve size and configuration, and online troubleshooting. This section includes a list of common problems with titration curves, options for maximum accuracy, and common installation effects.
For more, check out Greg’s Deminar #7 on PID Control for Integrating Processes or his book, Advanced pH Measurement and Control, 3rd Edition.