Successfully reported this slideshow.
Your SlideShare is downloading. ×

pH Control Solutions - Greg McMillan

Loading in …3

Check these out next

1 of 94 Ad

More Related Content

Slideshows for you (20)


Similar to pH Control Solutions - Greg McMillan (20)


Recently uploaded (20)

pH Control Solutions - Greg McMillan

  1. 1. 1<br />pH Control Solutions<br />Ascend Chocolate Bayou Plant Seminar March 10, 2011<br />
  2. 2. 2<br /> Introduction of Presenter<br />ISA Presenter<br />Gregory K. McMillan ( ) <br />Greg is a retired Senior Fellow from Solutia/Monsanto and an ISA Fellow. Presently, Greg contracts as a consultant in DeltaV R&D via CDI Process & Industrial. Greg received the ISA “Kermit Fischer Environmental” Award for pH control in 1991, the Control Magazine “Engineer of the Year” Award for the Process Industry in 1994, was inducted into the Control “Process Automation Hall of Fame” in 2001, was honored by InTech Magazine in 2003 as one of the most influential innovators in automation, and received the ISA “Life Achievement  Award” in 2010. Greg is the author of numerous books on process control, his most recent being Essentials of Modern Measurements and Final Elements for the Process Industry and Advanced Temperature Measurement and Control - 2nd Edition. Greg has been the monthly “Control Talk” columnist for Control magazine since 2002. Greg’s expertise is available on the web site:<br />
  3. 3. 3<br />Source<br />
  4. 4. 4<br />ISA Automation Week - Oct 17-20<br />Call for Papers<br />Deadline is <br />March 28 !<br />
  5. 5. 5<br /> Legends Cutler and Liptak Give Keynotes<br />
  6. 6. 6<br />Key Benefits of Seminar<br />Recognition of the opportunity and challenges of pH control<br />Alerts to important implementation considerations<br />Awareness of modeling and control options<br />Understanding the root causes of poor performance<br />Prioritization of improvements based on cost, time, and goal<br />Insights for applications and solutions<br />
  7. 7. 7<br />Section 1: pH Opportunity and Challenge<br />Extraordinary Sensitivity and Rangeability<br />Deceptive and Severe Nonlinearity<br />Extraneous Effects on Measurement<br />Difficult Control Valve Requirements <br />
  8. 8. 8<br />Top Ten Signs of a Rough pH Startup<br /><ul><li>Food is burning in the operators’ kitchen
  9. 9. The only loop mode configured is manual
  10. 10. An operator puts his fist through the screen
  11. 11. You trip over a pile of used pH electrodes
  12. 12. The technicians ask: “what is a positioner?”
  13. 13. The technicians stick electrodes up your nose
  14. 14. The environmental engineer is wearing a mask
  15. 15. The plant manager leaves the country
  16. 16. Lawyers pull the plugs on the consoles
  17. 17. The president is on the phone holding for you</li></li></ul><li>9<br />Extraordinary Sensitivity and Rangeability<br />pHHydrogen Ion ConcentrationHydroxyl Ion Concentration<br />0 1.0 0.00000000000001<br />1 ` 0.1 0.0000000000001<br />2 0.01 0.000000000001<br />3 0.001 0.00000000001<br />4 0.0001 0.0000000001<br />5 0.00001 0.000000001<br />6 0.000001 0.00000001<br />7 0.0000001 0.0000001<br />8 0.00000001 0.000001<br />9 0.000000001 0.00001<br />10 0.0000000001 0.0001<br />11 0.00000000001 0.001<br />12 0.000000000001 0.01<br />13 0.0000000000001 0.1<br />14 0.00000000000001 1.0<br />Hydrogen and Hydroxyl Ion Concentrations in a Water Solution at 25oC<br />aH = 10-pH<br />pH = - log (aH)<br />aH = g * cH <br />cH* cOH = 10-pKw<br />Where:<br />aH =hydrogen ion activity (gm-moles per liter)<br />cH =hydrogen ion concentration (gm-moles per liter)<br />cOH =hydroxyl ion concentration (gm-moles per liter)<br /> g = activity coefficient (1 for dilute solutions)<br />pH =negative base 10 power of hydrogen ion activity<br />pKw = negative base 10 power of the water dissociation constant (14.0 at 25oC)<br />
  18. 18. 10<br />Effect of Water Dissociation (pKw) on Solution pH<br />Measured pH<br />pH 10<br />pH 9<br />pH 7.5<br />REAL pH 8<br />pH 7<br />pH 6.5<br />pH 6<br />pH 5<br />
  19. 19. 11<br /> AMS pH Range and Compensation Configuration<br />pH / ORP Selection<br />Preamplifier Location<br />Type of Reference Used<br />Ranging<br />Temperature Comp Parameters<br />Solution pH Temperature Correction<br />Isopotential Point Changeable for Special pH Electrodes<br />
  20. 20. 12<br />14<br />12<br />10<br />8<br />pH<br />6<br />4<br />2<br />0<br />11<br />10<br />9<br />8<br />pH<br />7<br />6<br />5<br />4<br />3<br />Nonlinearity - Graphical Deception<br />For a strong acid and base the pKa are off-scale and the slope continually changes by a factor of ten for each pH unit deviation<br />from neutrality (7 pH at 25 oC)<br />As the pH approaches the neutral point the response accelerates (looks like a runaway). Operators often ask what can be done to slow down the pH response around 7 pH.<br />Reagent / Influent Ratio<br />Despite appearances there are no straight lines in a titration curve (zoom in reveals another curve if there are enough data points - a big “IF” in neutral region)<br />Yet titration curves are essential for every aspect of pH system<br />design but you must get numerical values and avoid mistakes<br />such as insufficient data points in the area around the set point<br />Reagent / Influent Ratio<br />
  21. 21. 13<br />Weak Acid and Strong Base<br />pka = 4<br />Multiple Weak Acids and Weak Bases<br />Weak Acid and Weak Base<br />pka = 9<br />pka = 10<br />pka = 5<br />pka = 3<br />pka = 4<br />Nonlinearity - Graphical Deception<br />Strong Acid and Weak Base<br />pka = 10<br />Slope moderated<br />near each pKa<br />pKa and curve changes with temperature!<br />
  22. 22. 14<br /> Reagent Savings is Huge for Flat Part of Curve<br />10<br />pH<br />4<br />Reagent to Feed <br /> Flow Ratio <br />Reagent <br /> Savings<br />Optimum <br />set point<br />Original <br />set point<br />Oscillations could be due to non-ideal mixing, control valve stick-slip. or pressure fluctuations<br />
  23. 23. 15<br /> Effect of Sensor Drift on Reagent Calculations <br />10<br />pH<br />Feedforward<br />pH Error<br />8<br />pH Set Point<br />6<br />Influent pH<br />4<br />Sensor Drift<br />Reagent to Feed <br /> Flow Ratio <br />The error in a pH feedforward calculation<br />increases for a given sensor error as the<br />slope of the curve decreases. This result<br />Combined with an increased likelihood of<br />Errors at low and high pH means feedforward<br />Could do more harm than good when going<br />from the curve’s extremes to the neutral region. <br />Flow feedforward (ratio control<br />of reagent to influent flow) works<br />well for vessel pH control if there <br />are reliable flow measurements <br />with sufficient rangeability<br />Feedforward<br />Reagent Error<br />Feedforward control always requires pH feedback correction unless the set point is on the flat part <br />of the curve, use Coriolis mass flow meters and have constant influent and reagent concentrations<br />
  24. 24. 16<br />W<br />W<br />W<br />W<br />W<br />W<br />W<br />W<br />Double Junction Combination pH Electrode <br />Em<br />R3<br />Er<br />R4<br />solution ground<br />silver-silver chloride<br />internal electrode<br />Measurement<br />becomes slow <br />from a loss of<br />active sites or<br />a thin coating<br />of outer gel<br />E4<br />second<br />junction<br />R5<br />potassium chloride (KCl) electrolyte <br />in salt bridge between junctions<br />primary <br />junction<br />R6<br />E5<br />inner <br />gel<br />layer<br />silver-silver chloride<br />internal electrode<br />Nernst Equation<br />assumes inside<br />and outside gel<br />layers identical<br />E3<br />W<br />outer <br />gel<br />layer<br />R2<br />7 pH buffer<br />E2<br />Process ions try to <br />migrate into porous <br />reference junction<br />while electrolyte ions<br />try to migrate out<br />R1<br />W<br />Ii<br />E1<br />Process Fluid<br />R10<br />R9<br />R7<br />R8<br />High acid or base concentrations can affect glass gel layer and reference junction potential<br />Increase in noise or decrease in span or efficiency is indicative of glass electrode problem<br />Shift or drift in pH measurement is normally associated with reference electrode problem<br />
  25. 25. 17<br />Life Depends Upon Process Conditions<br />Months<br />>100% increase in life <br />from new glass designs<br />for high temperatures<br />25 C<br />50 C<br />75 C<br />100 C<br />Process Temperature<br />High acid or base concentrations (operation at the extremes of the titration curve) <br />decrease life for a given temperature. A deterioration in measurement accuracy and <br />response time often accompanies a reduction in life. Consequently pH feedforward <br />control is unreliable and the feedforward effect and timing is way off for such cases.<br />
  26. 26. 18<br />New High Temperature Glass Stays Fast<br />Glass electrodes get slow as they age. High temperatures cause premature aging <br />
  27. 27. 19<br />What is High Today may be Low Tomorrow<br />Most calibration adjustments chase the short term errors shown<br />below that arise from concentration gradients from imperfect<br />mixing, ion migration into reference junction, temperature shifts, <br />different glass surface conditions, and fluid streaming potentials.<br />With just two electrodes, there are more questions than answers.<br />B<br />A<br />A<br />B<br />A<br />B<br />pH<br />time<br />
  28. 28. 20<br />Control Valve Rangeability and Resolution<br />pH<br />8<br />Set point<br />Control Band<br />6<br /> B<br />Er = 100% * Fimax*----<br /> Frmax<br />Frmax = A * Fimax<br /> B<br />Er = ----<br /> A<br />Ss = 0.5 * Er<br />Where:<br />A = distance to center of reagent error band on abscissa from influent pH<br />B = width of allowable reagent error band on abscissa for control band <br />Er = allowable reagent error (%)<br />Frmax = maximum reagent valve capacity (kg per minute)<br />Fimax = maximum influent flow (kg per minute)<br />Ss = allowable stick-slip (resolution limit) (%)<br />Influent pH<br />B<br />Reagent Flow<br />Influent Flow<br />A<br />
  29. 29. 21<br />pH<br />12<br />pH3<br />10<br />Slow<br />8<br />Fastest process response seen by<br />Loop at inflection point (e.g. 7 pH)<br />pH2<br />6<br />pH1<br />4<br />2<br />Reagent Flow<br />Influent Flow<br />Slow<br /> Speed of Response Seen by pH Loop<br />Excursion from pH1 to pH2 acceleration makes response look like a runaway to loop<br />Excursion from pH2 to pH3 deceleration is not enough to show true process time constant<br />Batch neutralizers without reagents consumed by reactions lack process self-regulation that causes an integrating response aggravating overshoot from acceleration. <br />Apparent loss of investment in large well mixed volume can be restored<br />by signal characterization of pH to give abscissa as controlled variable!<br />
  30. 30. 22<br />Key Points <br />pH electrodes offer by far the greatest sensitivity and rangeability of any measurement. To make the most of this capability requires an incredible precision of mixing, reagent manipulation, and nonlinear control. pH measurement and control can be an extreme sport<br />Solution pH changes despite a constant hydrogen ion concentration because of changes in water dissociation constant (pKw) with temperature, and activity coefficients with ionic strength and water content<br />Solution pH changes despite a constant acid or base concentration because of changes in the acid or base dissociation constants (pKa) with temperature<br />Titration curves have no straight lines A zoom in on any supposed line should reveal another curve if there are sufficient data points<br />Slope of titration curve at the set point has the greatest effect on the tightness of pH control as seen in control valve resolution requirement. The next most important effect is the distance between the influent pH and the set point that determines the control valve rangeability requirement<br />Titration curves are essential for every aspect of pH system design and analysis<br />First step in the design of a pH system is to generate a titration curve at the process temperature with enough data points to cover the range of operation and show the curvature within the control band (absolute magnitude of the difference between the maximum and minimum allowable pH)<br />
  31. 31. 23<br />Key Points <br />Accuracy and speed of response of pH measurements stated in the literature assume the thin fragile gel layer of a glass electrode and the porous process junction of the reference electrode have had no penetration or adhesion of the process and are in perfect condition at laboratory conditions<br />Time that glass electrodes are left dry or exposed to high and low pH solutions must be minimized to maximize the life of the hydrated gel layer<br />Most accuracy statements and tests are for short term exposure<br />Long term error of pH measurements installed in the process is an order of magnitude greater than the error normally stated in the literature<br />pH measurement error may look smaller on the flatter portion of a titration curve but the associated reagent delivery error is larger<br />Cost of pH measurement maintenance can be reduced by a factor of ten by more realistic expectations and calibration policies<br />Set points on the steep portion of a titration curve necessitate a reagent control valve precision that goes well beyond the norm and offers the best test to determine a valve’s actual stick-slip in installed conditions<br />Reagent valve resolution (stick-slip) may determine the number of stages of neutralization required, which has a huge impact on a project’s capital cost<br />Approach to the neutral point looks like a runaway due to acceleration<br />Batch processes have less self-regulation and tend to ramp<br />Batch processes can have a one direction response for a given reagent<br />
  32. 32. 24<br />Section 2: Modeling and Control Options<br />Virtual plant and imbedded process models<br />Minimization of capital investment<br />Cascade pH control<br />Online identification of titration curve<br />Batch pH control<br />Linear reagent demand control<br />Adapted reagent demand control<br />Smart split range point<br />Elimination of split range control<br />Model predictive control<br />
  33. 33. 25<br />Virtual Plant Setup<br />Configuration and Graphics<br />Virtual Plant<br />Laptop or Desktop<br />or Control System Station<br />Advanced Control Modules<br />Process Module <br />for Bioreactor or Neutralizer<br />
  34. 34. 26<br />Virtual Plant Access<br /><ul><li>Free State of the Art Virtual Plant
  35. 35. Not an emulation but a DCS (SimulatePro)
  36. 36. Independent Interactive Study
  37. 37. Structural Changes “On the Fly”
  38. 38. Advanced PID Options and Tuning Tools
  39. 39. Enough variety of valve, measurement, and process dynamics to study 90% of the process industry’s control applications
  40. 40. Learn in 10 minutes rather than 10 years
  41. 41. Online Performance Metrics
  42. 42. Standard Operator Graphics & Historian
  43. 43. Control Room Type Environment
  44. 44. No Modeling Expertise Needed
  45. 45. No Configuration Expertise Needed
  46. 46. Rapid Risk-Free Plant Experimentation
  47. 47. Deeper Understanding of Concepts
  48. 48. Process Control Improvement Demos
  49. 49. Sample Lessons (Recorded Deminars)</li></ul>Virtual Plant:<br />Recorded Deminars:<br />A new easy fast free method of access is now available that eliminates <br />IT security issues and remote access response delays<br />
  50. 50. 27<br />Fuzzy<br />Logic<br />Waste<br />RCAS<br />RCAS<br />middle selector<br />ROUT<br />kicker<br />AY<br />AY<br />AC<br />AC<br />splitter<br />splitter<br />AT<br />AT<br />AT<br />AY<br />AY<br />Attenuation<br />Tank<br />AY<br />middle selector<br />middle selector<br />filter<br />FT<br />FT<br />AY<br />AY<br />Stage 2<br />Stage 1<br />AT<br />AT<br />AT<br />AT<br />AT<br />AT<br />Waste<br />Mixer<br />Mixer<br />FT<br /> Case History 1- Existing Control System <br />
  51. 51. 28<br />MPC-1<br />MPC-2<br />Waste<br />RCAS<br />RCAS<br />middle selector<br />ROUT<br />AC-1<br />AC-2<br />kicker<br />AY<br />AY<br />splitter<br />splitter<br />AT<br />AT<br />AT<br />AY<br />AY<br />Attenuation<br />Tank<br />AY<br />middle selector<br />middle selector<br />filter<br />FT<br />FT<br />AY<br />AY<br />Stage 2<br />Stage 1<br />AT<br />AT<br />AT<br />AT<br />AT<br />AT<br />Waste<br />Mixer<br />Mixer<br />FT<br /> Case History 1 - New Control System<br />
  52. 52. 29<br />12<br />pH<br />Old Set Point<br />New Set Point<br />2<br />Reagent to Waste <br /> Flow Ratio <br />New Ratio<br />Old Ratio<br />Reagent <br /> Savings<br /> Case History 1 - Opportunities for Reagent Savings<br />
  53. 53. 30<br />Model Predictive Control (MPC)<br />For Optimization of Actual Plant<br />Stage 1 and 2 Set Points<br />Actual Plant<br />Optimization<br />(MPC1 and MPC2)<br />Tank pH and 2nd Stage Valves<br />Inferential Measurement<br />(Waste Concentration)<br />and Diagnostics<br />Stage 1 and 2 <br />pH Set Points<br />Actual<br />Reagent/Waste Ratio<br />(MPC SP)<br />Virtual<br />Reagent/Influent Ratio<br />(MPC CV)<br />Virtual Plant<br />Adaptation<br />(MPC3)<br />Model<br />Influent Concentration<br />(MPC MV)<br />Model Predictive Control (MPC)<br />For Adaptation of Virtual Plant<br /> Case History 1 - Online Adaptation and Optimization<br />
  54. 54. 31<br /> Case History 1 - Online Model Adaptation Results<br />Actual Plant’s<br />Reagent/Influent<br />Flow Ratio<br />Virtual Plant’s<br />Reagent/Influent<br />Flow Ratio<br />Adapted Influent Concentration<br />(Model Parameter)<br />
  55. 55. 32<br />93%<br />Acid<br />50%<br />Caustic<br />Water<br /> AT <br />Cation<br /> Anion<br />To EO<br />Final caustic<br />adjustment<br />Final acid<br />adjustment<br />Pit<br /> Case History 2 - Existing Neutralization System<br />
  56. 56. 33<br /> Case History 2 - Project Objectives<br />Safe<br />Responsible<br />Reliable<br />Mechanically<br />Robust controls, Operator friendly<br />Ability to have one tank out of service<br />Balance initial capital against reagent cost<br />Little or no equipment rework<br />
  57. 57. 34<br /> Case History 2 - Cost Data<br />93%H2SO4 spot market price $2.10/Gal<br />50% NaOH spot market price $2.30/Gal <br />
  58. 58. 35<br /> Case History 2 - Challenges<br />Process gain changes by factor of 1000x<br />Final element rangeability needed is 1000:1<br />Final element resolution requirement is 0.1%<br />Concentrated reagents (50% caustic and 93% sulfuric)<br />Caustic valve’s ¼ inch port may plug at < 10% position<br />Must mix 0.05 gal reagent in 5,000 gal < 2 minutes<br />Volume between valve and injection must be < 0.05 gal <br />0.04 pH sensor error causes 20% flow feedforward error<br />Extreme sport - extreme nonlinearity, sensitivity, and rangeability of pH demands extraordinary requirements for mechanical, piping, and automation system design<br />
  59. 59. 36<br /> Case History 2 - Choices<br />Really big tank and thousands of mice<br />each with 0.001 gallon of acid or caustic<br />or<br />modeling and control<br />
  60. 60. 37<br /> Case History 2 - Tuning for Conventional pH Control<br />
  61. 61. 38<br />Gain 10x larger<br /> Case History 2 - Tuning for Reagent Demand Control<br />
  62. 62. 39<br /> Embedded Process Model for pH<br />
  63. 63. 40<br /> Titration Curve Matched to Plant<br />pH<br />Slope<br />
  64. 64. 41<br />signal<br />characterizer<br /> AY <br />1-3<br />pH set point<br />Signal characterizers linearize loop <br />via reagent demand control<br /> AC <br />1-1<br />NaOH<br />Acid<br /> AY <br />1-2<br /> LC <br />1-5<br /> LT <br />1-5<br /> FT <br />1-1<br /> FT <br />1-2<br />middle<br />signal <br />selector<br />signal<br />characterizer<br />splitter<br /> AY <br />1-4<br />Feed<br /> AY <br />1-1<br />To other Tank<br /> AT <br />1-2<br /> AT <br />1-1<br /> AT <br />1-3<br />Tank<br />Eductors<br />From other Tank<br />Static Mixer<br />To other Tank<br />Downstream system<br /> Modeled pH Control System<br />
  65. 65. 42<br />One of many spikes of recirculation pH <br />spikes from stick-slip of water valve<br />Influent pH<br />Tank 1 pH for Reagent Demand Control<br />Tank 1 pH for Conventional pH Control<br />Start of Step 4<br />(Slow Rinses)<br />Start of Step 2<br />(Regeneration)<br />Conventional pH versus Reagent Demand Control<br />
  66. 66. 43<br />Traditional System for Minimum Variability<br />The period of oscillation (4 x process deadtime) and filter time<br />(process residence time) is proportional to volume. To prevent<br />resonance of oscillations, different vessel volumes are used. <br />Reagent<br />Reagent<br />Reagent<br />Feed<br />Small first tank provides a faster response<br />and oscillation that is more effectively filtered <br />by the larger tanks downstream<br />Big footprint<br />and high cost!<br />
  67. 67. 44<br />Traditional System for Minimum Reagent Use<br />Reagent<br />The period of oscillation (total loop dead time) must differ by more<br />than factor of 5 to prevent resonance (amplification of oscillations) <br />Feed<br />Reagent<br />Reagent<br />Big footprint<br />and high cost!<br />The large first tank offers more cross neutralization of influents<br />
  68. 68. 45<br /> FC <br />1-2<br /> AC <br />1-1<br /> LC <br />1-3<br />Tight pH Control with Minimum Capital Investment<br />IL#1 – Interlock that prevents back fill of<br />reagent piping when control valve closes<br />IL#2 – Interlock that shuts off effluent flow until<br />vessel pH is projected to be within control band<br />Eductor<br />High Recirculation Flow<br />Reagent<br />Any Old Tank<br />Signal<br />Characterizer<br /> LT <br />1-3<br /> f(x)<br /> FT <br />1-1<br />*IL#2<br />Effluent<br /> AT <br />1-1<br />*IL#1<br /> FT <br />1-2<br />Influent<br />10 to 20 <br />pipe<br />diameters<br />
  69. 69. 46<br />Linear Reagent Demand Control<br />Signal characterizer converts PV and SP from pH to % Reagent Demand<br />PV is abscissa of the titration curve scaled 0 to 100% reagent demand<br />Piecewise segment fit normally used to go from ordinate to abscissa of curve<br />Fieldbus block offers 21 custom spaced X,Y pairs (X is pH and Y is % demand)<br />Closer spacing of X,Y pairs in control region provides most needed compensation<br />If neural network or polynomial fit used, beware of bumps and wild extrapolation <br />Special configuration is needed to provide operations with interface to:<br />Operator sees loop PV in pH and enters loop SP in pH<br />Operator can change mode to manual and change manual output<br />Operator sees both reagent demand % and PV trends<br />Set point on steep part of curve shows biggest improvements from: <br />Reduction in limit cycle amplitude seen from pH nonlinearity<br />Decrease in limit cycle frequency from final element resolution (e.g. stick-slip)<br />Decrease in crossing of split range point<br />Reduced reaction to measurement noise<br />Shorter startup time (loop sees real distance to set point and is not detuned)<br />Simplified tuning (process gain no longer depends upon titration curve slope)<br />Restored process time constant (slower pH excursion from disturbance)<br />
  70. 70. 47<br /> FC <br />1-1<br /> AC <br />1-1<br /> AC <br />1-2<br />M<br />Cascade pH Control to Reduce Downstream Offset<br />Linear Reagent<br />Demand Controller<br />Flow Feedforward<br /> FT <br />1-1<br />RSP<br />Sum<br />Trim of Inline<br /> Set Point<br />Reagent<br /> f(x)<br /> AT <br />1-1<br /> Filter<br /> f(x)<br /> FT <br />1-2<br />Static Mixer<br /> PV signal<br />Characterizer<br /> SP signal<br />characterizer<br />Feed<br />Coriolis Mass<br />Flow Meter<br />10 to 20<br />pipe<br />diameters<br />Any Old Tank<br />Integral<br />Only<br />Controller<br /> AT <br />1-2<br />
  71. 71. 48<br />Full Throttle (Bang-Bang) Batch pH Control<br />Batch pH <br />End Point<br />Predicted pH<br /> Cutoff<br />Sum<br />Reagent<br />Rate of<br />Change<br />DpH/Dt<br />Projected<br /> DpH<br />Past<br />DpH<br />New pH<br /> Sub<br /> Div<br /> Mul<br />Old pH<br /> Delay<br />Dt<br />Total System<br /> Dead Time<br />Batch Reactor<br /> Filter<br /> AT <br />1-1<br />10 to 20 <br />pipe<br />diameters<br />Section 3-5 in New Directions in Bioprocess Modeling and Control<br />shows how this strategy is used as a head start for a PID controller<br />
  72. 72. 49<br /> AC <br />1-1<br /> AC <br />1-1<br /> FC <br />1-1<br /> FQ <br />1-1<br />Linear Reagent Demand Batch pH Control<br /> FT <br />1-1<br />Secondary pH<br />PI Controller<br />Influent #1<br /> AT <br />1-1<br />Online Curve<br /> Identification<br />Static Mixer<br />10 to 20 <br />pipe<br />diameters<br /> FT <br />1-2<br />Influent #2<br /> f(x)<br />Batch Reactor<br /> Signal <br />Characterizer<br />Uses Online<br />Titration Curve<br />Master Reagent Demand<br />Adaptive PID Controller<br /> AT <br />1-1<br />10 to 20 <br />pipe<br />diameters<br />Reduces injection and mixing delays and enables some cross<br />neutralization of swings between acidic and basic influent. It is<br />suitable for continuous control as well as fed-batch operation.<br />
  73. 73. 50<br /> FT <br />1-1<br />Secondary pH<br />PI Controller<br /> AC <br />1-1<br /> AC <br />1-1<br /> FC <br />1-1<br /> FQ <br />1-1<br />Influent<br /> AT <br />1-1<br />Online Curve<br /> Identification<br />Static Mixer<br />10 to 20 <br />diameters<br /> FT <br />1-2<br /> f(x)<br />Neutralizer<br /> Signal <br />Characterizer<br />Uses Online<br />Titration Curve<br />Master Reagent Demand<br />Adaptive PID Controller<br /> AT <br />1-1<br />10 to 20 <br />diameters<br /> Adapted Reagent Demand Control<br />Reduces injection and mixing delays and enables some<br />cross neutralization in continuous and batch operations<br />
  74. 74. 51<br /> Recently Developed Adaptive Control<br />Anticipates nonlinearity by recognizing old territory<br />Model and tuning settings are scheduled per operating region<br />Remembers what it has learned for preemptive correction<br />Demonstrates efficiency improvement during testing<br />Steps can be in direction of optimum set point<br />Excess reagent use rate and total cost can be displayed online<br />Achieves optimum set point more efficiently<br />Rapid approach to set point in new operating region<br />Recovers from upsets more effectively<br />Faster correction to prevent violation<br />More efficient recovery when driven towards constraint <br />Returns to old set points with less oscillation <br />Faster and smoother return with less overshoot<br />
  75. 75. 52<br />Multiple Model parameter<br />Interpolation with re-centering<br />Estimated Gain, time constant, and deadtime<br />Changing process input<br />First Order Plus Deadtime Process<br />Gain<br />1<br />2<br />Time Constant<br />3<br />Dead time<br /> First Order plus Dead Time Model Identification<br />For a first order plus deadtime process, only nine (9) models are evaluated each sub-iteration, first gain is determined, then deadtime, and last time constant. <br />After each iteration, the bank of models is re-centered using the new gain, time constant, and deadtime<br />Changes in the process model can be used to diagnose changes<br />in the influent and the reagent delivery and measurement systems<br />
  76. 76. 53<br /> Scheduling of Learned Dynamics and Tuning<br />Model and tuning is scheduled based on pH<br />
  77. 77. 54<br />total cost of<br />excess reagent<br />pH<br />hourly cost of <br />excess reagent<br />total cost of<br />excess reagent<br />pH<br />hourly cost of <br />excess reagent<br /> Adaptive Control Efficiently Achieves Optimum <br />
  78. 78. 55<br />total cost<br />of excess<br />reagent<br />pH<br />hourly cost <br />of excess<br />total cost <br />of excess<br />reagent<br />hourly cost<br />of excess<br />pH<br /> Adaptive Control Efficiently Rejects Loads<br />
  79. 79. 56<br />pH<br />pH<br /> Adaptive Control is Stable on Steep Slopes<br />
  80. 80. 57<br /> Smart Split Range Point<br />G = split range gap (%)<br />Kv1 = valve 1 gain (Flow e.u. / CO %)<br />Kv2 = valve 2 gain (Flow e.u. / CO %)<br />Kp1 = process gain for valve 1(PV e.u. / Flow e.u.)<br />Kp2 = process gain for valve 2(PV e.u. / Flow e.u.)<br />S1 = 1st split ranged span (PV e.u.)<br />S2 = 2nd split ranged span (PV e.u.)<br />
  81. 81. 58<br /> AC <br />1-1<br /> Smart Split Range Point<br />Reagent<br />Smart in terms of valve gain<br />compensation but not smart<br />in terms of valve sensitivity !<br />Small<br />(Fine)<br />Large<br />(Coarse)<br />Splitter<br />Split Range<br />Block<br />For large valve 4x small valve flow:<br />PID Small Large<br />OutValveValve<br />0% 0% 0%<br />20% 100% 0%<br />20% 100% 0%<br />100% 100% 100%<br />Neutralizer<br />PID Controller<br /> AT <br />1-1<br />
  82. 82. 59<br /> AC <br />1-1a<br /> AC <br />1-1b<br /> PID Valve Sensitivity and Rangeability Solution 1 <br />Reagent<br />Large<br />(Coarse)<br />Small<br />(Fine)<br />Neutralizer<br />PID Controller<br />or PIDPlus with<br />sensitivity limit<br /> AT <br />1-1<br />Proportional only Controller<br />or PIDPlus with<br />sensitivity limit<br />
  83. 83. 60<br /> AC <br />1-1<br /> ZC <br />1-1<br /> PID Valve Sensitivity and Rangeability Solution 2 <br />Reagent<br />Small<br />(Fine)<br />Large<br />(Coarse)<br />Integral only Controller<br />or PIDPlus with<br />sensitivity limit<br />Neutralizer<br />PID Controller<br />or PIDPlus with<br />sensitivity limit<br /> AT <br />1-1<br />
  84. 84. 61<br />MPC Valve Sensitivity and Rangeability Solution<br />Model Predictive Controller (MPC) setup for rapid simultaneous <br />throttling of a fine and coarse control valves that addresses<br />both the rangeability and resolution issues. This MPC can<br />possibly reduce the number of stages of neutralization needed<br /><br /><br />
  85. 85. 62<br /> MPC Valve Sensitivity and Rangeability Solution <br />
  86. 86. 63<br /> MPC Valve Sensitivity and Rangeability Solution<br />
  87. 87. 64<br /> MPC Valve Sensitivity and Rangeability Solution<br />
  88. 88. 65<br /> MPC Maximization of Low Cost Reagent<br />
  89. 89. 66<br /> MPC Maximization of Low Cost Reagent <br />
  90. 90. 67<br />Riding Max SP<br />on Lo Cost MV<br />Riding Min SP<br />on Hi Cost MV<br />Critical CV<br />Critical CV<br />Load<br />Upsets<br />Load<br />Upsets<br />Low Cost MV <br />Maximum SP<br /> Increased <br />Low Cost MV <br />Maximum SP<br /> Decreased <br />Set Point<br />Changes<br />Set Point<br />Changes<br />Lo Cost Slow MV<br />Hi Cost Fast MV<br /> MPC Maximization of Low Cost Reagent <br />
  91. 91. 68<br /> MPC Maximization of Low Cost Reagent <br />manipulated <br /> variables<br />disturbance <br /> variable <br /> Supplemental<br />Reagent Flow SP<br /> Acid Feed<br /> Flow SP<br />Cheap Reagent<br /> Flow PV<br />MPC<br />controlled <br /> variable<br />Neutralizer<br /> pH PV<br />optimization <br /> variable<br />Acidic Feed<br /> Flow SP<br />Maximize<br />null<br />null<br />constraint <br /> variable<br />Supplemental Reagent Valve Position<br />Note that cheap reagent valve is wide open and feed is maximized to keep supplemental <br /> reagent valve at minimum throttle position for minimum stick-slip<br />
  92. 92. 69<br />Key Points <br />More so than for any other loop, it is important to reduce dead time for pH control because it reduces the effect of the nonlinearity<br />Filter the feedforward signal to remove noise and make sure the corrective action does not arrive too soon and cause inverse response<br />The effectiveness of feedforward control greatly depends upon the ability to eliminate reagent delivery delays<br />If there is a reproducible influent flow measurement use flow feedforward, otherwise use a head start to initialize the reagent flow for startup<br />The reliability and error of a pH feedforward is unacceptable if the influent or feed pH measurement is on the extremities of the titration curve <br />Use a Coriolis or magnetic flow meter for reagent flow control <br />Every reagent valve must have a digital valve controller (digital positioner)<br />Except for fast inline buffered systems, use cascade control of pH to reagent flow to compensate for pressure upsets and enable flow feedforward <br />Linear reagent demand can restore the time constant and capture the investment in well mixed vessels, provide a unity gain for the process variable, simply and improve controller tuning, suppress oscillations and noise on the steep part of the curve, and speed up startup and recovery from the flat part of the curve<br />
  93. 93. 70<br />Key Points <br />Changes in the process dynamics identified online can be used to predict and analyze changes in the influent, reagent, valve, and sensor<br />New adaptive controllers will remember changes in the process model as a function of operating point and preemptively schedule controller tuning<br />Use inline pH control, mass flow meters, linear control valves, and dynamic compensation to automatically identify the titration curve online<br />Use gain scheduling or signal characterization based on the titration curve to free up an adaptive controller to find the changes in the curve<br />Batch samples should be taken only after the all the reagent in the pipeline and dip tube has drained into the batch and been thoroughly mixed<br />Use a wide open reagent valve that is shut or turned over to pH loop based on a predicted pH from ramp rate and dead time to provide the fastest pH batch/startup<br />Use online titration curve identification and linear reagent demand pH control for extremely variable and sharp or steep titration curve<br />Use an online dynamic pH estimator to provide a much faster, smoother, and more reliable pH value, if the open loop dead time and time constant are known and there are feed and reagent coriolis mass flow meters<br />Use linear reagent demand model predictive control for dead time dominant or interacting systems and constraint or valve position control<br />
  94. 94. 71<br />Section 3: Plant Design and Maintenance<br />Common Problems with Titration Curves<br />Effect of Measurement Selection and Installation<br />Options to improve accuracy and maintenance<br />Effect of piping design, vessel type, and mixing pattern<br />Implications of oversized and split ranged valves <br />Online Troubleshooting<br />
  95. 95. 72<br />Common Problems with Titration Curves<br />Insufficient number of data points were generated near the equivalence point <br />Starting pH (influent pH) data were not plotted for all operating conditions<br />Curve doesn’t cover the whole operating range and control system overshoot<br />No separate curve that zooms in to show the curvature in the control region<br />No separate curve for each different split ranged reagent<br />Sequence of the different split ranged reagents was not analyzed<br />Back mixing of different split ranged reagents was not considered<br />Overshoot and oscillation at the split ranged point was not included<br />Sample or reagent solids dissolution time effect was not quantified<br />Sample or reagent gaseous dissolution time and escape was not quantified<br />Sample volume was not specified <br />Sample time was not specified<br />Reagent concentration was not specified <br />Sample temperature during titration was different than the process temperature <br />Sample was contaminated by absorption of carbon dioxide from the air <br />Sample was contaminated by absorption of ions from the glass beaker<br />Sample composition was altered by evaporation, reaction, or dissolution <br />Laboratory and field measurement electrodes had different types of electrodes <br />Composite sample instead of individual samples was titrated <br />Laboratory and field used different reagents <br />
  96. 96. 73<br /> Middle Signal Selection Advantages<br />Inherently ignores single measurement failure of any type including the most insidious PV failure at set point<br />Inherently ignores slowest electrode<br />Reduces noise and spikes particularly for steep curves<br />Offers online diagnostics on electrode problems<br />Slow response indicates coated measurement electrode <br />Decreased span (efficiency) indicates aged or dehydrated glass electrode<br />Drift or bias indicates coated, plugged, or contaminated reference electrode or high liquid junction potential<br />Noise indicates dehydrated measurement electrode, streaming potentials, velocity effects, ground potentials, or EMI<br />Facilitates online calibration of a measurement<br />For more Information on Middle Signal Selection see Feb 5, 2010 post<br /><br />
  97. 97. 74<br />3K<br />0-5 M<br /><br /><br />Reference Electrode<br />Glass Electrode<br />Solution Ground<br /> Online Diagnostics for Cracked Glass<br />Cracked<br /> Glass!<br />Cracked Glass Fault<br />pH Glass electrode normally has high impedance of 50-150 Meg-ohm<br />Glass can be cracked at the tip or further back inside the sensor<br />Recommended setting of 10 Megohm will detect even small cracks<br />Reference - Joseph, Dave, “What’s the Real pH of the Stream”, Emerson Exchange 2008<br />
  98. 98. 75<br />40k<br />150 M<br /><br /><br />Coated<br /> Sensor!<br />Glass Electrode<br />Solution Ground<br />Reference Electrode<br /> Online Diagnostics for Coated Sensor<br />Coated Sensor Detection Can<br />activate sensor removal and cleaning cycle<br />place output on hold while sensor is cleaned<br />Reference - Joseph, Dave, “What’s the Real pH of the Stream”, Emerson Exchange 2008<br />
  99. 99. 76<br /> AMS pH Range and Compensation Configuration<br />pH / ORP Selection<br />Preamplifier Location<br />Type of Reference Used<br />Ranging<br />Temperature Comp Parameters<br />Solution pH Temperature Correction<br />Isopotential Point Changeable for Special pH Electrodes<br />
  100. 100. 77<br /> AMS pH Diagnostics Configuration<br />Glass Electrode Impedance <br />Warning and Fault Levels<br />Impedance Diagnostics On/Off<br />Reference Impedance <br />Warning and Fault Levels<br />Reference Zero Offset<br />Calibration Error Limit<br />Glass Impedance Temp Comp<br />(Prevents spurious errors due to Impedance decrease with Temperature)<br />
  101. 101. 78<br /> AMS pH Calibration Setup<br />Live Measurements and Status<br />Calibration Constants from Last Calibrations<br />Buffer Calibration Type & Buffer Standard Used<br />Sensor Stabilization Criteria<br />Zero Offset Beyond this Limit will create a Calibration Error<br />If you want to know more about Buffer Calibration, hit this button…<br />
  102. 102. 79<br /> AMS pH Auxiliary Variables Dashboard<br />
  103. 103. 80<br />AE<br />AE<br />AE<br />AE<br />AE<br />AE<br /> Horizontal Piping Arrangements<br />flush<br />throttle valve to<br />adjust velocity<br />pressure drop for<br />each branch must<br />be equal to to keep <br />the velocities equal<br />drain<br />20 to 80 degrees<br />The bubble inside the glass bulb<br />can be lodged in tip of a probe <br />that is horizontal or pointed up or <br />caught at the internal electrode<br />of a probe that is vertically down<br />20 pipe diameters<br /> 5 to 9 fps to minimize coatings<br />0.1 to 1 fps to minimize abrasion<br />static mixer <br /> or pump<br />throttle valve to<br />adjust velocity<br />flush<br />10 OD<br />10 OD<br />Series arrangement preferred to minimize differences in solids, <br />velocity, concentration, and temperature at each electrode!<br />20 pipe diameters<br />drain<br />
  104. 104. 81<br />AE<br />AE<br />AE<br />AE<br />AE<br />AE<br /> Vertical Piping Arrangements<br />throttle valve to<br />adjust velocity<br />throttle valve to<br />adjust velocity<br />Orientation of slot in shroud<br />abrasion<br />coating<br />0.1 to 1 fps<br />5 to 9 fps<br />hole<br />or<br />slot<br />10 OD<br />10 OD<br />Series arrangement preferred to minimize differences in solids, <br />velocity, concentration, and temperature at each electrode!<br />
  105. 105. 82<br />Options for Maximum Accuracy <br />A spherical or hemi-spherical glass measurement and flowing junction reference offers maximum accuracy, but in practice maintenance prefers:<br />A refillable double junction reference to reduce the complexity of installation and the need to adjust reference electrolyte flow rate – This electrode is often the best compromise between accuracy and maintainability.<br />A solid reference to resist penetration and contamination by the process and eliminate the need to refill or replace reference particularly for high and nasty concentrations and pressure fluctuations – This electrode takes the longest time to equilibrate and is more prone to junction effects but could be right choice in applications where accuracy requirements are low and maintenance is high.<br />Select best glass and reference electrolyte for process <br />Use smart digital transmitters with built-in diagnostics<br />Use middle signal selection of three pH measurements<br />Inherent auto protection against a failure, drift, coating, loss in efficiency, and noise (see February 5, 2010 entry on )<br />Allocate time for equilibration of the reference electrode<br />Use “in place” standardization based on a sample with the same temperature and composition as the process. If this is not practical, the middle value of three measurements can be used as a reference. The fraction and frequency of the correction should be chosen to avoid chasing previous calibrations<br />Insure a constant process fluid velocity at the highest practical value to help keep the electrodes clean and responsive<br />
  106. 106. 83<br /> Wireless pH Transmitters Eliminate Ground Spikes<br />Incredibly tight pH control via 0.001 pH wireless resolution <br />setting still reduced the number of communications by 60%<br />Temperature compensated wireless pH controlling at 6.9 pH set point<br />Wired pH ground noise spike <br />
  107. 107. 84<br />Wireless Bioreactor Adaptive pH Loop Test<br />
  108. 108. 85<br />+<br />+<br />Elapsed <br />Time<br />+<br />+<br />Elapsed <br />Time<br /> Enhanced PID Algorithm <br /><ul><li>PID integral mode is restructured to provide integral action to match the process response in the elapsed time (reset time set equal to process time constant)
  109. 109. PID derivative mode is modified to compute a rate of change over the elapsed time from the last new measurement value
  110. 110. PID reset and rate action are only computed when there is a new value
  111. 111. If transmitter damping is set to make noise amplitude less than sensitivity limit, valve packing and battery life is dramatically improved
  112. 112. Enhancement compensates for measurement sample time suppressing oscillations and enabling a smooth recovery from a loss in communications further extending packing -battery life</li></ul>TD<br />Kc<br />TD<br />Kc<br />Link to PIDPlus White Paper<br /><br />Whitepapers/WP_DeltaV%20PID%20Enhancements%20for%20Wireless.pdf<br />
  113. 113. 86<br /> Flow Response - Enhanced vs. Traditional PID <br />Enhanced PID <br />Sensor PV<br />Traditional PID <br />Sensor PV<br />
  114. 114. 87<br /> pH Response - Enhanced vs. Traditional PID <br />Enhanced PID <br />Sensor PV<br />Traditional PID <br />Sensor PV<br />
  115. 115. 88<br /> Failure Response - Enhanced vs. Traditional PID <br />Enhanced PID <br />Sensor PV<br />Traditional PID <br />Sensor PV<br />
  116. 116. 89<br />M<br /> Everyday Mistakes in pH System Design<br />Mistake 1: Missing, inaccurate, or erroneous titration curve<br />Mistake 2: Absence of a plan to handle failures, startups, or shutdowns<br />reagent<br />feed tank<br />Mistake 7 (gravity flow)<br />Mistake 3 (single stage <br /> for set point at 7 pH) <br />Mistake 8 (valve<br /> too far away) <br /> AT <br />1-3<br />Mistake 12 (electrode<br /> too far downstream) <br />Mistake 10 (electrode<br />submerged in vessel)<br />Mistake 9 (ball valve<br /> with no positioner) <br /> AT <br />1-1<br />Influent (1 pH)<br />Mistake 4 (horizontal tank)<br /> Mistakes 5 and 6<br />(backfilled dip tube &<br />injection short circuit)<br /> AT <br />1-2<br />Mistake 11 (electrode<br /> in pump suction) <br />
  117. 117. 90<br />Stagnant <br />Zone<br />Stagnant <br />Zone<br />M<br />Reagent<br />Feed<br />Plug<br />Flow<br />Short <br />Circuiting<br />Stagnant <br />Zone<br /> AT <br />1-3<br /> Mixing Pattern and Vessel Geometry Implications<br />
  118. 118. 91<br /> Oversized Reagent Valves are a Big Problem<br />Limit cycle amplitude is operating point dependent and can be estimated as:<br />stick-slip (%) multiplied by valve characteristic slope (pph/%) and by titration curve slope (pH/pph)<br />Dead band is 5% - 50%<br />without a positioner !<br />Dead band<br />Pneumatic positioner<br />requires a negative % <br />signal to close valve<br />Stroke<br /> (%)<br />Digital positioner<br />will force valve <br />shut at 0% signal<br />Stick-Slip is worse near closed position<br />0<br />Signal<br /> (%)<br />dead band<br />The dead band and stick-slip is greatest near the closed position so valves that <br />ride the seat from over sizing or split ranged operation create a large limit cycle<br />
  119. 119. 92<br />Key Points <br />The time that glass electrodes are left dry or exposed to high and low pH solutions must be minimized to maximize the life of the hydrated gel layer<br />Most accuracy statements and tests are for short term exposure before changes in the glass gel layer or reference junction potential are significant<br />The pH measurement error may look smaller on the flatter portion of a titration curve but the associated reagent delivery error is larger<br />The cost of pH measurement maintenance can be reduced by a factor of ten by more realistic expectations and calibration policies<br />The onset of a coating of the glass measurement electrode shows up as a large increase in its time constant and response time<br />The onset of a non conductive coating of the reference electrode shows up as a large increase in its electrical resistance<br />Non-aqueous and pure water streams require extra attention to shielding and process path length and velocity to minimize pH measurement noise<br />Slow references may be more stable for short term fluctuations from imperfect mixing and short exposure times from automated retraction<br />The fastest and most accurate reference has a flowing junction but it requires regulated pressurization to maintain a small positive electrolyte flow<br />The best choice might not be the best technical match to the application but the electrode with the best support by maintenance and operations and vendor<br />
  120. 120. 93<br />Key Points <br />For non abrasive solids, installation in a recirculation line with a velocity of 5 to 9 fps downstream of a strainer and pump may delay onset of coatings<br />For abrasive solids and viscous fluids, a thick flat glass electrode can minimize coatings, stagnant areas, and glass breakage<br />For high process temperatures, high ion concentrations, and severe fouling, consider automatic retractable assemblies to reduce process exposure<br />When the fluid velocity is insufficient to sweep electrodes clean, use an integral jet washer or a cleaning cycle in a retractable assembly<br />The control system should schedule automated maintenance based on the severity of the problem and production and process requirements<br />pH measurements can fail anywhere on or off the pH scale but middle signal selection will inherently ride out a single electrode failure of any type<br />Equipment and piping should have the connections for three probes but a plant should not go to the expense of installing three measurements until the life expectancy has been proven to be acceptable for the process conditions<br />The more an electrode is manually handled, the more it will need to be removed<br />A series installation of multiple probes insures the electrodes will see the same velocity and mixture that is important for consistent performance<br /> Wireless pH control of static mixers with enhanced algorithm can provide a exceptional setpoint response and measurement failure protection<br />
  121. 121. 94<br />Key Points <br />A system considered to be well mixed may be poorly mixed for pH control <br />To be “well mixed” for pH control, the deviation in the reagent to influent flow from non ideal mixing multiplied by the process gain must be well within the control band <br />Back mixing (axial mixing) creates a beneficial process time constant and plug flow or radial mixing creates a detrimental process dead time for pH control<br />The agitation in a vessel should be vertical axial pattern without rotation and be intense enough to break the surface but not cause froth<br />The actual equipment dead time is often larger than the turnover time because of non ideal mixing patterns and fluid entry and exit locations <br />Horizontal tanks are notorious for short circuiting, stagnation, and plug flow that cause excessive dead time and an erratic pH response <br />The dead time from back filled reagent dip tubes or injection piping is huge<br />To provide isolation, use a separate on-off valve and avoid the specification of tight shutoff and high performance valves for throttling reagent <br />Set points on the steep portion of a titration curve necessitate a reagent control valve precision that goes well beyond the norm and offers the best test to determine a valve’s actual stick-slip in installed conditions <br />Reagent valve stick-slip may determine the number of stages of neutralization required, which has a huge impact on a project’s capital cost <br />