PID Tuning for Near and True Integrating Processes

PID Tuning for Near and True Integrating Processes

The following technical discussion is part of an occasional series showcasing the ISA Mentor Program, authored by Greg McMillan, industry consultant, author of numerous process control books, 2010 ISA Life Achievement Award recipient and retired Senior Fellow from Solutia Inc (now Eastman Chemical). Greg will be posting occasional questions and responses from the ISA Mentor Program, with contributions from program participants.

The most important loops on vessels and columns typically have a near or true integrating response. A self-regulating process is classified as near integrating if the process time constant is larger than four times the process dead time. The composition, pH, and temperature response of continuous columns tend to have a time constant to dead time ratio of about 6:1 for changes in the liquid material balance.

The composition, pH, and temperature response of continuous well-mixed vessels tend to have a time constant to dead time ratio of 50:1 or larger. Level and gas pressure have a true integrating response since changes in level or pressure have a negligible effect on the discharge flow notable exceptions occurring for gravity flow for liquid level control and relatively high but non critical pressure drops for gas pressure control. The composition, pH, and temperature of batch columns and batch vessels have a true integrating response in the normal operating range.

near-integrating-process-response

A notable exception is the composition control of a batch reaction when there is no deficiency of any reactant concentration. Here the response is near-integrating with a very large time constant to dead time ratio making differentiation between true and near integrating inconsequential. Reactors with a potentially runaway response are treated as true integrators with the intent being that control action is sufficient to prevent the loop from seeing the acceleration from a runaway response.

The designation of having an integrating response is critical in terms of tuning and recognizing there is a window of allowable controller gains, where too low of a PID gain as well too high of a PID gain will cause excessive oscillations. For a PID gain that is too low, the oscillations tend to be much larger and 10 times slower (e.g., period is 40 dead times for low PID gain and four dead times for high PID gain). For a PID gain greater than the ultimate gain, the oscillations can grow and the loop becomes unstable. For a PID gain that is too low, the oscillations will always decay but the decay rate becomes incredibly slow as the PID gain is decreased. For a runaway reaction, too low of a PID gain is disastrous in that the process can runaway reaching a point of no return.

The Lambda tuning rules switch from a Lambda being the closed loop time constant for a setpoint change for self-regulating processes to Lambda being an arrest time for a load disturbance with the objective of stopping the ramping effect of integrating processes and potential acceleration of runaway processes.

Questions from ISA Mentor Program Participant Hector Torres

  1. How do you calculate Lambda for near-integrating processes? I understand we should follow the Integrating Process rules but I am not clear as of how to determine the desired arrest time. You mention that for maximum unmeasured disturbance rejection a Lambda equal to the dead time is used. Also it is stated that a Lambda equal three dead times minimizes consequences of nonlinearities, inverse response and resonance.
  2. Why should we consider Lambda of one or three times dead time in these rules? To be identified as a near integrating process the time constant should be four times greater than the dead time. Why should we make Lambda a factor of dead time here? I remember it was mentioned that Lambda should be set at three or four times the largest of the dead time or the time constant. Would this apply here? Am I mixing in my mind the rules for self-regulating and integrating processes?
  3. I understand that integrating processes can have an inverse response that is problematic. What could be an example of inverse response? What do you mean by this?
The ISA Mentor Program enables young professionals to access the wisdom and expertise of seasoned ISA members, and offers veteran ISA professionals the chance to share their wisdom and make a difference in someone’s career. Click this link to learn more about how you can join the ISA Mentor Program.

 

Greg McMillan’s Answers

1) In integrating tuning rules, Lambda is the arrest time, which means for a step disturbance or step change in PID output, how long does it take for the PV to halt its excursion and start its return to setpoint. If you multiply the integrating process gain (%PV/sec/%CO) by a the change in controller output (CO%) required to get a ramp rate (%PV/sec) and then Lambda, you have the peak error (maximum excursion in %PV). For level and pressure it is easier to visualize in that the maximum PV excursion (peak error) for a maximum expected change in controller output added to the setpoint must not hit an alarm or trip point. The integrated error (% sec area between the PV and SP on trend chart) is the peak error (%) multiplied by Lambda (sec). Thus, the peak error is proportional to Lambda and integrated error is proportional to Lambda squared where Lambda is the arrest time set relative to the dead time.

2) The ability of a loop to handle changes in gain and dynamics is expressed by the gain margin and phase margin, which are both a function of Lambda relative to dead time. The Lambda tuning rules reduce to tuning rules commonly used for the last six decades if you realize Lambda should always be thought of and set relative to dead time and not a time constant or an integrating process gain (as mistakenly shown in various publications). Also for large time constants you have a near-integrating process and must switch to integrating tuning rules.

In the 5th edition of the Process/Industrial Instruments and Controls Handbook (1999 edition for which I became chief editor), Bialkowski on pages 10.52 and 10.53 shows how the gain margin and phase margin are a function of Lambda varying from one to five dead times. Elsewhere he talks about the concept of near-integrating processes. I think the rule sometimes states of choosing the largest of three times the dead time or three times the time constant are not in tune with advancement in understanding of Lambda always being thought of as a value relative to dead time. While Bialkowski did not say when to switch integrating tuning rules, I estimated that the switch point of when the time constant to dead time ratio was greater than four would result in tuning rules similar to what has been practices for the last six decades. This rule plus realizing that Shinskey essentially was using a Lambda of 0.6 x dead time to give the impressive maximum disturbance rejection results he has in his articles and books. His response is oscillatory. The acknowledged practical limit for a smooth response even if you exactly know the process dynamics and they never change is a Lambda equal to the dead time.

Thinking of Lambda as being three dead times is a good rule for both self-regulating and integrating tuning rules. For self-regulating processes with a time constant to dead time ratio between one and two , there might be some advantage of using three times constants instead of three dead times for PI control but I think the advantage is minimal and is negligible compared to other issues. If you have a nonlinear valve or process, you may need to increase Lambda to be five or six dead times unless you do signal characterization, gain scheduling or adaptive control.

To summarize, for all Lambda tuning rules (self-regulating and integrating), the most aggressive tuning if the process dynamics are fixed and exactly identified (rare case), is a Lambda of 0.6 dead times. This is case is normally only used to show how well Lambda tuning can do compared to other tuning methods (showcase test for gamesmanship). Normally, to deal with unknowns and nonlinearities a Lambda of three dead times is used but may be increased for more uncertain applications and greater changes in dynamics whether due to the process or valve. For bioreactors where the disturbances are extremely slow, rise time is inconsequential for setpoint changes, and process gains can change dramatically from the pre-exponential to the exponential growth phases, Lambda may be as large as 10 dead times.

To better understand different process responses and tuning objectives, watch the three-part ISA Mentor Program webinar on PID options and solutions:

3) Inverse response is where the initial response of the PV is in the opposite direction of the final response.  If a feedforward correction arrives too soon it can cause inverse response. For feedback control, inverse response originates typically occurs when the feed stream throttled that has a temperature less than the operating temperature of the equipment. The classic case of inverse response is boiler drum level. An increase in feed water flow being colder than the boiling water in down comers will cause bubbles to collapse which will cause fluid to go down from the drum into the down comers causing shrink (decrease in drum level). Eventually the increase in feed water is heated enough in the down comers to increase the drum inventory (increase drum level). For a decrease in feed water flow, the bubbles in the down comers increase in number and size pushing fluid up into the drum causing swell (increase in drum level). This shrink and swell is quite common and can be reduced by feed water preheaters. For furnace and reactor temperature an increase in air flow or reactant flow that is colder than furnace or reactor temperature will cause the equipment temperature to decrease until the firing rate and reaction rate generates enough heat to increase the equipment temperature. This is the main reason plus the time constant of the concentration response why reactor temperature should not be controlled by manipulating reactant flow. This is true for liquids and polymers. For gases in fluidized bed reactors, the reaction rate and concentration time constant are so fast, inverse response is imperceptible.

See the ISA book 101 Tips for a Successful Automation Career that grew out of this Mentor Program to gain concise and practical advice. See the InTech magazine January/February 2013 feature article “Enabling new automation engineers” for the candid comments of some of the original program participants. See the May 2015 Control Talk column “How to effectively get engineering knowledge” with the 2014 addition to Mentor Program Keneisha Williams on the challenges faced by young engineers today. Providing discussion and answers besides Greg McMillan and co-founder of the program Hunter Vegas (project engineering manager at Wunderlich-Malec) are resources Brain Hrankowsky (consultant engineer at a major pharmaceutical company), Michel Ruel (executive director, engineering practice at BBA Inc.), Leah Ruder (process systems automation group manager at the Midwest Engineering Center of Emerson Process Management) and Nick Sands (ISA Fellow and Manufacturing Technology Fellow at DuPont).
How to Measure pH of Ultrapure Water in Power Industry Applications

How to Measure pH of Ultrapure Water in Power Industry Applications

This article is from the January/February 2015 issue of InTech magazine and was written by Fred Kohlmann, Midwest business manager for analytical products with Endress+Hauser

Water covers more than 70 percent of the earth’s surface and has varying degrees of purity in its natural form, ranging from crystal clear pure mountain spring water to highly saturated brine sea water. In the power generation industry, ultrapure water is used as a source to make steam to drive turbines and for other uses. Ultrapure water does not cause corrosion or lead to stress cracking in equipment such as turbine blades, stainless-steel lines, steam circuits, and cooling systems. Power companies use a great deal of ultrapure water, upward of 500,000 gallons per day for large plants. The ultrapure water can be processed from city water, a nearby river, or even seawater.

measuring-pH-ultrapure-water-power-industry

In most cases, the systems for producing ultrapure water are supplied by specialists in electrodionization, membrane, reverse osmosis, and other techniques for purifying water, and all require pH monitoring. Not only is pH monitoring required when the water is purified, it is also required as the water is used to maintain correct pH.

The bottom line in power plants is that improperly conditioned water—based on a number of parameters, of which the main two are conductivity and pH—leads to corrosion and scale, which leads to inefficient operation and damage to vital parts.

In the boiler, deposits cause heat transfer problems, reducing steam production capability. Corrosion from these deposits weakens the metal, leading to tube leaks that negatively impact the production of steam. Large boilers, condensers, economizers, and superheater tube leaks can cause adjacent tubes to fail and are actually the largest cause of forced boiler shutdowns.

The steam that passes through the turbine can cause deposits and corrosion from minerals, organics, and detergents that are present in plant water sources. Deposits on the turbines can cause pressure drops and unbalance the turbine, reducing speed and generation capacity.

Unfortunately, conductivity measurement by itself does not provide enough water quality information for ultrapure water chemistry; therefore, pH must also be incorporated. Unless controlled, the effects of improper water treatment parameters will cause boiler tube failures and loss of efficiency due to coating of the tubes. This makes energy costs and ultimately operational costs higher. Boiler manufacturers have tight specifications on the minimum and maximum water quality parameters, pH being one of them.

Water’s tough

Water in its pure state is one of the most aggressive solvents. Known as the “universal solvent,” water, to one degree or another, will dissolve virtually everything to which it is exposed. Because pure water has a deficiency of ions, it looks for equilibrium with the ions it comes in contact with, and so it wants to strip these ions away from its host.

For the purposes of this article, pure water is defined as having a conductivity between 0.055 to 10 microSiemens per centimeter (µS/cm), or 18.2 to 0.1 megohms-cm. Common manufacturer specifications for pH sensors can indicate a conductivity range of 10 µS/cm or greater.

Herein lies the first hurdle to best measurement practices: to find pH sensors that are specifically designed to measure water with conductivity less than 10 µS/cm. Fortunately, some pH sensors are able to measure down to 0.1 µS/cm, but these are specialized instruments and must be specified, installed, and maintained accordingly.

 

There is a deficiency of ions in pure water, and pH sensors have the reputation of being noisy when measuring pH in these low ionic strength solutions. In simple terms, the signal is noisy because the sensor is looking for ions to capture and measure and has a hard time finding them, causing the measured value to meander up and down the pH scale.

Using two or more brand new pH sensors from the same manufacturer—even right after being freshly calibrated in 7 and 4 pH buffers—the sensors may show differing values due to static charges and reference junction potential errors. Pure water is a poor conductor of electricity, and so static charges are an issue as water flows through the piping systems, requiring extra care in proper grounding for signal stability and noise rejection.

Also, extraneous electromagnetic interference (EMI) and radio frequency interference (RFI) can disturb the sensor’s electrical circuitry, especially in a power plant with high-voltage equipment. Walkie-talkie transmissions and electric motors or valves being cycled on and off can also create electrical noise. These interferences cause signal spikes that push the pH signal high or low for brief moments or freeze the signal in place.

The pH sensors use a two-electrode scheme as the measurement apparatus—an active or measuring electrode and a reference electrode. The active electrode can have an input impedance of 100 megohms in high ionic strength solutions such as a pH 7 buffer. So in the best of circumstances, pH measurement has at least a 100 megohm obstacle to overcome. If that same impedance is added to the very low ionic strength of ultrapure water, it adds measurement complexity. There is now a larger resistance for the signal to traverse through the low ionic solution.

The reference junction serves as the return or ground path for the pH measurement. Any shifting of the electrical resistance in the reference path will change the overall resistance of the measurement and cause a shift in pH reading. This equates to a noisy signal. A charge buildup at the reference junction can change as the process changes (e.g., when valves or pumps are cycled) or remain at a constant state and attenuate the pH signal.

If any air is introduced into the piping system of the pH sensor, it will add CO2 into the solution, which acidifies the actual pH value. Therefore, closed loop systems are needed for a constant and uniform measurement.

Consider the practice of taking a grab sample from a closed loop system for pH analysis. When one walks the sample back to the chemistry lab for analysis, what happens to the sample as it is exposed to the atmosphere? It likely changes, sometimes substantially, leading to a preference for in situ sensors.

Changes in the flow rate past a sensor can also lead to changes in the pH measurement. These changes are referred to as streaming current potentials. Changes in process flows cause changes in the reference junction potential and lessen the ability of the glass electrode to maintain its hydrated outer gel layer.

Problems can occur in pH sensor cable connections, terminal strips, and plugs. Connectors can become loose or corroded or have moisture accumulate across the connections. These situations change the resistance of the pH measurement and degrade the signal.

Long runs of cables without the aid of preamplification or signal conversion from analog to digital can lead to changes in the capacitance and resistance of the cable, which can affect the pH readings. Signal cables are also a means by which EMI and RFI can gain access to the transmitter circuitry, also causing measurement errors. Here are best practices for installing and maintaining pH sensors:

  • Make the pH measurement in a sealed piping system
  • Maintain a slow continuous flow rate past the pH sensor, about 100 mL/min
  • Use conductive piping and fittings; 316 SS is common practice
  • Keep cable runs as short as possible
  • Maintain tight, dry, and corrosion-free electrical sensor connections
  • Store unused pH sensors in a solution to maintain hydration: 4 or 7 pH buffer
  • Use digital pH sensors instead of analog

Inside pH sensors

Many manufacturers offer pH sensors designed specifically for measuring pH in low ionic fluids of 10 µS/cm or less. Low-resistance glass and double and triple reference junctions, as well as flowing reference junctions, are employed with high degrees of success. Ceramic junction materials tend to have less “memory” and facilitate fast response times.

The pH sensors using a flowing junction reference system (figure 1) tend to be more accurate as they minimize junction potentials, but they also require more maintenance. These types of systems use a reservoir of potassium chloride (KCl) solution pumped through the sensor’s reference element, and use either gravity or compressed air to maintain a constant overpressure as compared to the process being measured.

Figure 1. Flowing reference pH sensor with reservoir

Figure 1. Flowing reference pH sensor with reservoir

Process fluid will eventually find its way through the junction and into the filling solution of the reference electrode. When this happens, it dilutes the KCl inhabiting this physical space. This dilution of the KCl will eventually lead to a change in the reference chemistry and to measurement inaccuracies.

Flowing reference-style sensors deliver a fresh KCl solution through the junction and provide a constant nonchanging electrical reference path. These sensors also deliver a pH reading much faster than traditional sealed reference electrodes.

Sealed reference-type pH sensors employ salt rings or circular pinhead-type reference junctions. Salt ring–type junctions may employ gelled KCl solutions to maximize the junction surface area and keep KCl flow at an optimum rate. Some of these sensor styles may also use an internally charged or pressurized reference. Because these sensor types are considered closed systems, they have no reservoir to maintain, and the entire sensor is replaced as its reference becomes contaminated or as the solution within the reference gets depleted or becomes unusable.

Temperature compensation of the pH signal is very important to accurately measure the pH of ultrapure water (figure 2). An entire paper discussing this subject could be written, but is not within the scope of this article. As temperature changes, so does the pH sensor’s millivolt output. Specifically, the electrode produces more millivolts/pH as the temperature increases, and as the pH goes farther in either direction from 7 pH. This change is predictable and linear, and can be compensated for in the pH analyzer by using the Nernst equation in the circuit design.

Figure 2. Temperature as it relates to pH

Figure 2. Temperature as it relates to pH

The Nernst equation (figure 3) is a general mathematical equation that describes and predicts the pH electrode’s output based on a number of constant factors and just one variable, temperature.

Figure 3. Nernst equation showing breakout of potentials and slope

Figure 3. Nernst equation showing breakout of potentials and slope

Modern pH measurement systems incorporating temperature-compensated pH sensors are the norm. Avoid any pH sensor that comes without an integral temperature element or a transmitter that only accepts a manual or fixed temperature compensation network. A fast acting/responding temperature element should be mounted in the bulb of the pH sensor for best results.

Electrical considerations

Cables for pH sensors must be kept as short as possible, less than 10 feet without some type of preamplification or signal conversion. Furthermore, sensors should employ gold connections and O-ring sealed connectors, or use a digital inductively coupled sensor-to-cable connection to avoid EMI/RFI intrusions and moisture and corrosion problems.

There are sensors (figure 4) that convert the pH signal from an analog to a digital value at the sensor and send this digital signal up to 300 feet from the sensor to the transmitter. These digital pH sensors are available from several vendors. Most are not affected by moisture or contamination of connectors.

Figure 4. pH sensors convert the pH signal from an analog to a digital value at the sensor, and send this digital signal up to 300 feet from the sensor to the transmitter.

Figure 4. pH sensors convert the pH signal from an analog to a digital value at the sensor, and send this digital signal up to 300 feet from the sensor to the transmitter.

When using analog-type pH sensors, a common problem is a ground loop. A ground loop is a difference in the ground potential that the pH sensor sees versus the ground potential of the pH transmitter. Ground loops can be a constant or varying offset of voltage to the pH reading (leading to an inaccurate pH value). They can also be an on/off type signal that falsely increases or decreases the pH signal to the transmitter when an electrical device using the same ground is either turned on or off. Ground loops can be hard to find and tougher still to eliminate, but using inductively coupled digital pH sensors eliminates ground loop problems.

Calibration

Calibrations of pH sensors should be conducted regularly. This can be done during a process shutdown or by simply replacing the sensor with a calibrated unit.

The use of the proper buffer solutions is a must, as calibrations need to be made in pH 7 and 4 buffers, never pH 10. Also, properly rinsing and drying the sensor between buffer immersions is critical for accuracy. It is also important to ensure the glassware and other equipment interfacing with the buffer and sensor are clean and free of contamination. Calibrations should be made in accordance with the manufacturer’s recommendations, and proper care should be taken when cleaning the pH sensor.

If large step changes in buffer readings occur from the previous calibration, the sensor is suspect and has either been damaged or contaminated. It should be cleaned properly to get the calibration closer to the last values. Large shifts in calibration values are not normal in ultrapure water chemistries.

Digital pH sensors allow calibration in the laboratory or shop with either a separate transmitter, an alternate channel on a multichannel instrument, or hardware/software that allows hardware to directly connect to a PC. With digital pH sensors, spare precalibrated sensors can rotate in and out of the process as needed.

Calibration in this manner allows for longer and more accurate aging of the sensors in the calibration buffers. Also, a technician is not under pressure to have calibrations done on site while the system is down awaiting calibration completion and reinstallation of the sensor. This scenario is not possible with analog-type pH sensors, which is another advantage of digital pH sensors.

Calibrated pH sensors not in use should be stored in a 7 pH buffer or 3 molar KCl solution. A pH sensor should not be allowed to go dry, either in the process or during storage. If left to become dehydrated, the glass electrode will show higher electrical impedance from the norm and will react much more slowly to pH changes. It may take from a few minutes to hours or even days for the sensor to regain its original operational performance, if ever.

Repeated cycles of hydration and dehydration significantly shorten the pH sensor’s useful life. The reference electrode is also affected by dehydration. If left dry, salt from the internal KCl fill solution will form salt crystals and cake the outer surface of the junction. Ultimately the junction may siphon out all its fill solution.

Companies should premount and plumb pH sensors in stainless-steel flow loops (figure 5), where they are easily accessible for service and maintenance and where flow rates can be easily controlled.

Figure 5. pH sensor installation in prefabricated panel

Figure 5. pH sensor installation in prefabricated panel

Sensors send signals to pH transmitters, which present the information to the control system. The transmitter should be easy to use; for example, just a few keypad manipulations should be sufficient to perform calibrations without having to use the instruction manual each time a calibration is performed. The transmitter should also provide sensor diagnostics to alert the user to the sensor state and deploy alarms or warnings should the sensor start to deviate from configured parameters.

Modern transmitters can have virtually any desired output from 4–20 mA to relay/alarm contacts to multiple digital communication outputs such as HART, FOUNDATION Fieldbus, Profibus PA, or EtherNet/IP. Many pH transmitters have an integrated Web server, allowing remote users to access the transmitter from any Web browser.

Conclusion

Work with the manufacturer to select the best pH sensor for your application. If possible, look for the latest in technology for both the sensor style (junction and glass formulation) and the sensor’s signal transmission methodology (i.e., analog vs. digital).

Usually, no two sensors within a single manufacturer’s portfolio can realistically serve the same application. There can be big differences in sensor design/construction for a sensor that is used in 10 µS/cm service as compared to a sensor designed for 1.0 to 2.0 µS/cm service.

Make sure pH sensor cables and connectors are specified correctly for the distances involved and the environment in which they will be used, and keep them free of moisture and corrosion. Pay attention to the materials used to mount the pH sensor and to the importance of stability in the flow rates past the sensor. Make sure the sensor is easily accessible for calibration and general maintenance. Make sure trained resources are available to maintain the pH sensors (cleaning and calibration) at the manufacturer’s suggested intervals.

If these steps are followed, accurate and repeatable pH measurements can be made in ultrapure water, leading to improved operations, reduced maintenance, and increased uptime.

About the Author
Fred KohlmannFred Kohlmann is Midwest business manager for analytical products with Endress+Hauser. Since 1976, he has been involved in engineering, design service, marketing, and sales of online analytical water quality and process control instrumentation.
Connect with Fred:
48x48-linkedinEmail

 

A version of this article originally was published at InTech magazine.

Click here to read Fred Kohlmann’s article on pH measurement of ultrapure water at InTech magazine.

How to Select a pH Sensor for Harsh Process Environments

How to Select a pH Sensor for Harsh Process Environments

This guest post is authored by Vickie Olson, product marketing manager at Honeywell Process Solutions.

For today’s process plants, pH is an important parameter to measure in a host of demanding applications. A good example is flue gas desulfurization (FGD) control with systems using wet scrubbers in lime or limestone slurries. Proper pH can maximize sulfur oxide (SOx) removal and minimize the 160826412build-up of scale. Selection of the best sensor type will also enable longer life and more reliable control.

Choosing a pH sensor for use in harsh process environments requires careful consideration of numerous key factors. Extreme temperature at higher pH ranges reduces pH life due to consumption of hydrogen-sensitive ions on the glass membranes. An electrolyte (typically potassium chloride in gel or liquid form) tends to diffuse faster with temperature and high flow. In addition, abrasion removes the membrane surface over time.

Many process industry users specify a pH sensor of the combination style, meaning the device has a measuring electrode — either glass or ion-sensitive field effect transistor (ISFET) — or a reference electrode, which is usually based on silver/silver chloride. These pH sensors are considered rugged or robust compared to general-purpose sensors, since they are better able to withstand abrasive and alkaline conditions such as in FGD slurry. General-purpose pH sensors may not last a single day in sulfuric, abrasive and high-temperature environments. Depending on the style of ruggedized pH sensor and its maintenance frequency, this device can last from a few weeks to many months in operation.

Various types of reference protection on the market allow electrolyte diffusion, but reduce the infiltration of contaminants from the process fluid that can plug the junction or cause fouling of the reference. The porous junctions at the tip areas may be composed of double or multiple sections designed to slow contamination, which can result in poisoning of the silver reference material. Typically, the junction material in rugged pH sensors is composed of solid polytetrafluoroethylene (PTFE), ceramic or fibrous polyvinylidene fluoride (PVDF).

Several pH sensor designs employ wood or acrylic material containing electrolytes to slow the spread of contaminants while maintaining the required electrical connection of the reference with the measuring electrodes. An additional method to delay poisoning with solid reference designs is to locate the reference wire at the back of the sensor body, rather than hang it from the back, approaching the front of the device.

In terms of ruggedized glass measuring electrodes, the tip may have thicker glass and more hydrogen-sensitive material on the membrane. Flat glass is sometimes substituted for hemispherical or round glass on the tip to avoid breakage due to hard materials. This is less important in FGD applications, although the design is successfully used in the pulp & paper industry with heavy pulp slurry. In lime slurry, the flat tip does not have as large a measuring surface — leading to faster wear than rounded-type tips.

Most recently, ISFET measuring electrode technology was been paired with rugged reference technologies to provide a durable pH measurement solution. The advantages of ISFET include robustness, stability and precision. Sensors of this type utilize an integral automatic temperature compensator in one-piece construction, making them well suited for varying pH and temperature ranges.

The more demanding the application, the more critical it is to consider process operating conditions and expectations for a pH sensor. This is particularly important when harsh environments require frequent sensor replacement. Users employing the right device can realize savings from extended sensor service life, reduced replacement and maintenance costs, and ultimately, accurate and reliable pH measurement.

About the Author
Vickie OlsonVickie Olson is an analytical product specialist for Honeywell Process Solutions based in Atlanta, Ga. She has been involved in process instrumentation and analysis for industrial and municipal applications for more than 25 years as a chemist, product specialist and sales manager for Honeywell, Hach, and other companies. Vickie has spoken on a variety of topics related to water analysis and control at ISA and numerous other symposiums. She earned a bachelor’s degree in textile chemistry from the Georgia Institute of Technology and a master’s degree in business administration from Georgia State University.
Connect with Vickie:
LinkedIn

 

 

Getting Concentration Measurements from Common Inline Sensors

The following insights are part of an occasional series authored by Greg McMillan, industry consultant, author of numerous process control books and a retired Senior Fellow from Monsanto.

Ultimately what we want to know and control is the concentration of key components. Relatively inexpensive inline measurements can be used under the right conditions to provide concentration measurements. Understanding the limitations as well as the capabilities isScientist using pH meter. essential to success. Here we look at how we can elevate the use of capacitance, conductivity, density, pH and turbidity to provide process knowledge to control biological and chemical reactor concentration.

The inferential measurement of the concentration of a component (e.g. reactant or reagent) in a feed, exit, or recycle stream or vessel is possible if the component of interest has a large definitive effect on capacitance, conductivity, density, pH, and turbidity. To infer composition in vessels, sensors are installed in nozzles on the vessel, in recirculation lines, or sample lines in close proximity to vessel.

If the component of interest has a higher conductivity than other components or solvent, conductivity can be used if the conductivity is restricted to be always on one side of the peak in the conductivity curve. The process gain changes sign as an operating point crosses the peak, which is disastrous to a control loop because the PID control action sign must be the opposite of the process action sign (assuming valve action is correctly configured).

The measurement and control of biological and chemical reactors is the key to product quality and the yield and production rate of most processes in the process industry. See Greg McMillan’s new ISA book Advances in Reactor Measurement and Control for an extensive view of practical opportunities for building and effectively using online estimators to improve process knowledge and control.

If the PID has the wrong sign, the PID output will ramp to its output limit. Plots of conductivity versus concentration have a peak if the entire concentration range is covered. A few ionic species such as sulfuric have two peaks. The second peak for sulfuric acid occurs at 93 percent concentration. For product conductivity that traverses across the peak, concentration control cannot start until the conductivity is well to the right of the peak. For a very low concentration of a single acid and base (e.g. < 0.01 percent) pH provides a more sensitive inferential measurement of concentration.

Insight: One or more signal characterizer blocks are used to give a piecewise linear fit that enables the online computation of the X axis ( percent concentration) from the Y axis of a conductivity plot or pH titration curve generated from samples at various operating conditions.

For conductivity and pH, a signal characterizer is commonly used to compute the X axis (e.g. salt, acid, or base ion concentration) from the Y axis (conductivity or pH). The 20 or more data points of signal characterizer is usually enough to provide a piecewise linear fit. The points are more closely packed in the operating regions of greatest interest or nonlinearity. A cascade of characterizers can be used to provide greater resolution where a secondary signal characterizer is added to the output of the primary characterizer. A simple temperature correction should be generated based on the results of samples that cover the entire possible range of temperatures including abnormal operation.

Insight: The effect of process temperature on the actual conductivity and pH of the sample must be measured and used in customized solution pH or conductivity temperature compensation.

For a two component liquid mixture where the density the components differ by more than 1 percent an extremely accurate concentration measurement is attainable by the use of a Coriolis meter. The accuracy of the density measurement in Coriolis meter is 0.0002 gm/cm3. There is no drift and installation effects are essentially negligible. The meter never needs recalibration. There are no upstream and downstream straight run or field calibration requirements. High performance meter designs are now able to measure the concentration of bubbles and solids as well. The use of Coriolis meters on raw material feeds provides not only an incredibly accurate true mass flow measurement independent of composition but also an inferential concentration measurement from density.

Insight: An extremely accurate and drift free measurement of liquid density by Coriolis meter can be used as an inferential measurement of concentration in a two component mixture.

Turbidity offers an inferential measurement of biomass concentration for bioreactors. The turbidity measurement does not distinguish the number of cells or how many are alive (viable) or dead (lysed). The addition of a capacitance probe to provide a dielectric spectroscopy can provide inferential measurements of cell size, homogeneity, and membrane integrity. Viable cells have membranes intact whereas lysed cells have holes or fractures in their membranes.

Insight: Dielectric spectroscopy in combination with an inferred measurement of biomass from turbidity can provide inferential measurements of cell size, and the relative concentration of viable (live) versus lysed (dead) cells.

The inferential measurement of concentration must be periodically corrected by taken a fraction of the error between a synchronized computed concentration and an at-line or off-line analysis result. The synchronization is achieved by passing the change in concentration through dead time and integrator blocks with the proper sign of process feedback to match process dynamics and then adding the change to an original corrected value and passing the result through a final dead time block representing the analysis delay.

To get the most out of inline meters and probes, fundamental relationships should be used to compute concentration measurements. The computed concentration measurements must be gradually corrected by comparing the computed concentration synchronized with an analysis result somewhere and sometime whether in the field, plant lab, or offsite lab.

How to Determine the Health of a Glass Electrode

How to Determine the Health of a Glass Electrode

The following tip is from the ISA book by Greg McMillan and Hunter Vegas titled 101 Tips for a Successful Automation Career, inspired by the ISA Mentor Program. This is Tip #86.

There is not a whole lot of current information on the response time of pH electrodes. Tests done in the 1960s for a clean, healthy pH glass electrode showed that the time constant was smaller for a positive pH change, a higher velocity, and a higher degree of buffering. For a buffered solution at 4 fps fluid velocity, the time constants were 0.75 and 1.5 sec for a positive and negative pH change, respectively. At 0.5 fps the time constants became 3 and 12 seconds for the same electrode and solution.Paper pH - 0041 - White

A case history published in 1990 showed that the time to reach 98 percent of the final response (the “response time”) deteriorated from 10 seconds to 7 minutes for a 1 mm slime coating. The time to reach 98 percent of the final response is four time constants plus the deadtime. The use of a 98 percent response time can be misleading and difficult because pH electrodes typically have a much slower approach to a final value above 90 percent than a first order classical response. Waiting to 98 percent means that the test time is greatly extended and the inferred time constant may be grossly overestimated. Also, a small amount of noise can lead to inconsistent results.

As far as the controller is concerned, what happens at the beginning of the response is the most important. I prefer to use an 86 percent response time for a faster test and a more accurate time constant that is less sensitive to noise. The 86 percent response time is used for control valve response testing in the ISA-75.25.01 standard for some of the same reasons.

Test results for a glass electrode prematurely aged by exposure to high temperature showed an increase in the 86 percent response time from 10 seconds to 50 minutes. Dehydration, abrasion, and chemical attack can also cause a large increase in response time. An increase in the sensor response time (lag time) slows down the response of the loop to upsets. In the study cited, a 1 mm of slime coating increased the amplitude of oscillations by a factor of 10 and the period doubled.

Aging of a glass electrode also shows up as a decrease in efficiency; that is, span. However, this effect is not as detrimental to loop performance. The loss in span causes a decrease in process gain but there are much larger influences on the process gain, such as the shape of the titration curve. For a setpoint at 7 pH (the zero millivolt point) there is no effect on operating point. The actual pH is 7 when the measurement is at setpoint. As the pH setpoint gets further away from 7, the effect becomes larger. However, standardization by the analysis of a grab sample will compensate for the difference between the actual pH and the setpoint, reducing the effect efficiency back to a process gain. Losses in efficiency are less problematic than offsets from the reference electrode (Tip #87).

Concept: A thin coating on the glass and/or aging of the glass will show up as a huge increase in the sensor time constant. (The 86 percent response time is the electrode deadtime plus two time constants.) A large sensor lag causes deterioration of loop dynamics. A measurement of the response time by making a setpoint change or by putting the electrode in different buffer solutions provides a sensitive indicator of the health of a glass electrode.

101 Tips for a Successful Automation CareerDetails: When buffers are used to calibrate an electrode, estimate the time to 86 percent of the final response. The electrode deadtime is usually negligible. Use the data historian and trend chart in the DCS to estimate the response. If you have a wireless gateway, use a wireless pH transmitter instead of a lab pH meter to get the calibration data into the historian. Errors of less than 10 seconds in the estimate are not important because glass electrode measurement problems show up as large changes in the measurement time constant. Use setpoint changes to measure the response time of a glass electrode. If you have multiple electrodes, the increase in electrode time constant will show up as a nearly constant time shift between the response curves. A decrease in electrode efficiency shows up as an increasing time shift. Remove and clean or rejuvenate, if necessary, the slowest responding electrode(s). Use a dilute 5 percent hydrochloric acid (HCl) solution to remove alkaline deposits and strip away the outer, aged layer of glass to rejuvenate an electrode. Use a dilute 1 percent sodium hydroxide (NaOH) solution to remove acidic deposits. Use a detergent solution to remove organic deposits (oil and grease). The household rules of cleaning solutions to remove stains may be applicable to cleaning electrodes. More tenacious deposits may require a solvent. Be careful to avoid solvent attack on the sensor’s o-rings and seals. Limit exposure time to prevent contamination of the reference junction and chemical attack on the glass by the cleaning solution. To minimize coating while in service, ensure that the fluid velocity past the electrode is greater than 5 fps and that the protective shroud provides exposure of the glass surface to the flow stream unless the fluid is abrasive. Use high temperature glass to prevent premature aging from exposure to temperatures above 40 °C.

Watch-outs: The equilibration of reference potentials may make finding the final response difficult, particularly as the measurement electrode response gets slower. The extremely long equilibration time of some solid-state references will make measurement of the 98 percent response time thoroughly inconsistent. The measurement may appear to drift or never reach the buffer solution pH despite repeated calibration adjustments for both offset and span. The percent of final response is often not given in response time statements. Sometimes people mistakenly use response time when they mean time constant. Because the number of time constants is dramatically different between 63 percent and 98 percent (one versus four time constants), the source should be questioned as to the percent of final response.

Exceptions: For loops subjected to frequent changes in setpoint that are not corrected by an upper level loop, a loss in efficiency (span) may become more important than an increase in sensor lag. Electrodes that have an extreme loss in efficiency are in danger of becoming dead electrodes and must be replaced even if they are fast-responding.

Insight: The sensor response time is a sensitive indicator of the effects of coatings and aging on glass pH electrodes.

Rule of Thumb: Use the 86 percent response time to determine when to clean, rejuvenate, and/or replace the glass electrode.

Pin It on Pinterest