ExperTune Library

Reviews of ExperTune Products

CONTROL EXCLUSIVE: Monitor Performance Enterprise-Wide

Reprinted from CONTROL Magazine, June 2004.

A new release of PlantTriage from ExperTune allows users to subscribe to PDF performance reports for any area and contain just the most relevant information. Users can configure and subscribe to reports via their browser and develop reports containing any analysis for any portion of the operations over any time period. A dashboard is available to anyone via a secure web browser connection and allows users to view the performance of the entire corporation allowing drill down to any site, plant or unit operation.

The new release allows users to turn a plant off-scan while leaving the other plants inside the PlantTriage system on-scan, allowing for balanced load on DCS systems. This is especially critical for legacy systems.

Performance tracking now allows users to create a report based on previous performance. PlantTriage has added new assessments to diagnose process problems, and now monitors itself using advanced signal validation technology developed in partnership with Oxford University.

The application has further enhanced its best-of-class capabilities to pinpoint the root cause of a unit-wide or plant-wide oscillation. Reducing swings or oscillations can dramatically improve the performance and profitability of the plant. Getting at the root cause, fast, is what it's all about.

Standard templates allow users to quickly use the program. Templates include flow, level, averaging level, gas and liquid pressure, and temperature. The release also includes a "super-filter" that sets up the performance data inside Excel where all of the Excel toolset is available. All of this performance and diagnostic history is now available enterprise-wide via OPC Historical Data Access (HDA), a new standard created by the OPC Foundation and embraced by ExperTune.

CONTROL EXCLUSIVE: Tuning Package Adds Problem-Solving Tools

Reprinted from CONTROL Magazine, December 2000.

Optimizing control loops often requires a balance among performance, robustness, and valve travel indices.

ExperTune Inc., Hartland, WI, has introduced four exclusive new features with the latest release of their PID analysis, tuning, and simulation software. ExperTune runs on Windows and communicates with a PID controller through a digital data link. The PID controller can be a PLC, DCS, or standalone device. ExperTune partners with General Electric, Honeywell, Intellution, Modicon, Rockwell Automation, Siemens, Wonderware and other companies.

"Our software analyzes PID parameters including the setpoint, the process variable and the output," says John Gerry, ExperTune president. "PID settings such as proportional band, integral, and derivative can also be used as part of the analysis. Examination of these data allows the software to calculate optimal PID settings. These settings can be displayed at the PC and/or automatically transmitted to the PID controller."

The four new features introduced with the latest revision of ExperTune are:

1) Performance Summary: Optimizes tradeoffs among performance, robustness, and valve travel.

2) Dial-in Process Control Robustness: Allows user input for desired robustness.

3) Optimal PID Settings for Varying Loops: Finds the best PID settings to use for loops that behave differently with varying conditions.

4) Reduce Valve Stiction for Better Control: Measures in-service valve stiction.

Optimizing control loops often requires a balance among performance, robustness, and valve travel indices. The performance index indicates how well a loop responds to process upsets or disturbances. The robustness index compares the sensitivity of the loop to process changes. The valve travel index shows the amount of valve travel and reversal.

Performance Summary displays each of these indices simultaneously along with current PID parameters including proportional band, integral, derivative, and measurement filter.

"A user can run an off-line simulation of the loop based on data gathered from in-service use," says Richard Barraclough, ExperTune development engineer. "PID parameters can then be changed and the affect on the loop and the indices can be observed. After optimal PID parameters are determined with the simulation program, a user can enter these parameters into the loop controller."

Adjusting PID tuning parameters changes the speed of response of the loop. More aggressive tuning gives faster response but also makes the loop more sensitive to process changes. Parameters can be adjusted to increase robustness and reduce sensitivity to changes, but this typically increases the time it takes for the process variable to return to the setpoint after a disturbance.

The new version allows a user to find the best tradeoff between tight tuning and robust tuning by adjusting the desired sensitivity. "ExperTune's unique robustness plot shows the tradeoff between tuning and sensitivity to process change. ExperTune automatically generates the robustness plot for the control loop by using actual in-service data," says Gerry. "The plot shows how the combination of changes to process gain, process dead time, and PID tuning settings affect the stability and robustness the control loop."

The Loop Summary Table finds the best PID settings for loops that behave differently with varying conditions. A user can perform several tests on an in-service loop and observe the results in the Loop Summary Table. The table lists the tuning parameters, process model, model confidence, and relative response time of the loop for each test.

The table automatically highlights the most conservative and the average tuning values. The most conservative tuning values are useful in loops with varying conditions such as temperature loops with asymmetrical heat and cool characteristics. According to Gerry, "These conservative values may be the only tuning values that yield a stable loop performance under these types of conditions. Average tuning values can be used for tighter control in those loops with more predictable and symmetrical response."

The Loop Summary Table saves the process model developed during the loop test. These process models can be compared over time to show a historical perspective and to aid in troubleshooting.

Fighting a sticky valve will often push a control loop into a cycle. The cycle can permeate through an entire plant causing havoc and off-spec product. According to Barraclough, "The latest version of ExperTune's PID Loop Analyzer includes a stiction check wizard. The wizard provides a simple way of measuring the stiction in the control loop. ExperTune's wizard works on the installed valve, giving a true measure of the in-service valve stiction."

If the measured stiction is larger than 1%, then repair of the valve can result in much better performance and increased product quality. Valve repair and stiction reduction can eliminate a troublesome cycle, resulting in product made closer to specification with less energy use.

Checking a valve for stiction is the third of five steps to completely optimize the loop. The steps can all be performed with ExperTune:

1. Statistical analysis and diagnosis based on normal operating data.

2. Power Spectral Density analysis to detect hidden cycles.

3. Valve analysis to check for problems such as hysteresis and stiction.

4. Test and characterization of non-linearity.

5. Optimization of tuning parameters.

Tuning Software Predicts Valve Wear

Tuning a control loop for optimal performance is tough enough, but now companies want to know how much wear will occur in the control valve under various tuning parameters. The latest version of ExperTune loop-tuning software from ExperTune Inc. Hartland, WI calculates and predicts valve wear by examining the current tuning parameters, then recommends filters and tuning adjustments that will extend valve life.

The latest trend in control values involves the use of smart valve positioners, which track valve travel and the number of reversals. This information can be fed into a program such as Fisher-Rosemount's Asset Management Software, which uses the data to analyze valve wear and determine when it time to notify the maintenance department.

The problem with smart positioners and maintenance software," says Marc Cote, engineer at TOP Controls, St-Romuald, Quebec, "is they track and report actual valve travel. They know when a valve needs maintenance, but they cannot predict which control method will generate the most wear."

Cote, who performs loop optimization in process plants, says he has to "eyeball" valve wear. "We look at the process response curves, see how much travel and reversals occur in the valve with various tuning settings, and try to adjust the loop to reduce valve movement," says Cote. "Some aggressive tuning methods, such as Ziegler-Nichols, can cause five to 10 times too much valve movement, and we have to tune that out to save the valve. The problem with eyeballing such a solution is that we can't predict if our changes will work, and sometimes we lose robustness or response in the loop.

"We called ExperTune and convinced them that adding such a function to their PID tuning software would be very beneficial to end users."

Control loops normally fit into an "acceptable window," according to John Gerry, president of ExperTune. "The range of control valve tuning is usually bordered at one end by slower, conservative methods such as model-based techniques, and at the other end by more aggressive tuning, such as Ziegler-Nichols."

The conservative tuning methods are easy on control valves because they do not move the valve very far or very often, while aggressive techniques can wear a valve out in a shorter time from excessive movements and reversals. Most control schemes fall into the window between the two extremes, but not all.

"When we use the ExperTune software to analyze controls loops, we often find schemes way outside the window," says Gerry. "When we do an analysis of valve wear at those settings, we may find that the life of the valve can be predicted in weeks, rather than years."

By using the valve wear analysis module in ExperTune, an end user or consultant can change various parameters in the tuning scheme and see in a few minutes how this will affect valve life. Gerry says, "Simply adding a process-variable (PV) filter or changing the integral time may increase valve life by a factor of five."

What this means, says Gerry, is that a valve that previously required field maintenance once per year may now require service only once every five years. This life extension can come with no loss of control response.

"In many cases, tuning or adding PV filtering to increase valve life will have no adverse effect on the performance of the control loop," says Gerry. "The software calculates optimal process variable filter and PID values to increase valve life, does a simulation of the process, and makes sure that loop performance and robustness do not degrade."

The software compares predictions of valve travel and reversals numerically and graphically in the presence of noise, for both current and optimized values. It also computes the effects of different kinds of PV filters, such as first order, second order, Butterworth, or averaging.

For a quantitative check of performance and robustness, the software computes a Performance Increase and Robustness Index. The Performance Increase Index checks to see if loop responsiveness is compromised by reducing valve wear, while the Robustness Index tests if loop sensitivity (to the process changing) is affected.

Any tuning parameters, PV filter, or type can be easily tested and compared off-line in simulated time-response plots of setpoint changes, load upsets, or response to noise. An engineer can try what-if analyses to get the exact response required before downloading tuning parameters or applying a PV filter.

ExperTune creates a detailed report of current to new settings for comparison. The report sums up the tuning parameters, filter value, valve wear analysis, performance and robustness indices, and process model. The report also includes comparison graphs including robustness plots, simulated loop response, and noise response plots.

The valve wear software runs on a PC as part of the ExperTune system.

Linearize Your Process for Optimal Performance at Any Production Rate

Process loop tuners create a dynamic simulation of your process to calculate the tuning parameters required. The controller bumps the process off its setpoint to examine its response to the outputs. To get good data, the loop tuner needs a period of linear response - during intervals when significant disturbances are unlikely. So what happens if your process oscillates toward one end of the range and is sluggish at the other?

ExperTune Inc., Hartland, WI, claims to have the answer with a loop characterizer—the latest module for its PID loop analysis tools. According to John Gerry, president, "The characterizer linearizes your process so you get uniform performance across the entire range. You can run at optimum for any production rate."

"Many loops have the wrong valve characteristic," adds Greg Shinskey, ExperTune's PID loop consultant, "so optimum tuning can only be realized at one value of controller output." Most digital controllers now can add a nonlinear characterizer to the loop. "It's no longer necessary to change the valve," claims Shinskey. "ExperTune can automate the selection of the proper characteristics for the valves in your loops."

The existing relationship between the controller output and its effect on the process variable can be found by stepping the output through its range manually and allowing the process to settle out after each step. ExperTune plots the resulting data to present the existing characteristic. It then lets the user fit the characteristic with either a piecewise linear characterizer or a smooth hyperbolic function, using a simple drag-and-drop procedure.

Continues Gerry, "When you have selected a curve, ExperTune then converts your curve into either a set of x-y coordinates or into source code in FORTRAN, BASIC or C to apply to your controller. When characterization is complete, tuning at any operating point should apply equally to all."

The characterizer module can be used where servo control is important, to linearize flow loops, to control jacket temperature in split-range chemical reactors or slave loops in cascades, or for any nonlinear loop where the setpoint will change.

Top of page

It's Not Just a Loop Tuner, It's a Plant Manager

Reprinted with permission from CONTROL Magazine, May 1996.

Manual tuning of a PID loop's proportional band or gain, integral action or reset and derivative rate is a rewarding but tedious task. That's where time-saving software tuning and loop analysis tools like ExperTune come in. This package, from ExperTune (formerly Gerry Engineering), Hartland, WI, is used to import process data, simulate the effects of PID parameters on the process, and save hours on tough loops - that is, if you can get them into the plant and run the gauntlet of cost justification.

According to John Gerry, president of the software company, ExperTune's benefits only begin with loop-tuning time savings. Because when loop behavior is optimized, the process, the plant and arguably the business can run more efficiently. "We had one plant that was looking at installing some process equipment," says Gerry, "but when they used our package to optimize their loops, it turned out they increased their throughput by 50%. And they found they didn't need the new equipment in, after all."

A new version of the package, v.6.05, ships this month in 16-bit form, with a 32-bit version due out within the next few months. With each license comes a year's worth of free technical support and upgrades. A free demo disk is available.

Prices vary, but in cost analyses the company has performed for customers, loop-tuning time savings alone are estimated at $125 per loop. Factor in lower labor costs, multiple loops and tunings per year, dollars lost to product waste and rework, inefficient utility fuel usage costs, and downtime attributable to loops not optimized ... and you have a recipe for success. And while it's difficult to measure the value of process consistency and product quality, these factors certainly come into play.

Once ExperTune has been justified, it can be purchased in several configurations: with direct drivers for controllers from Allen-Bradley, GE Fanuc, Modicon, Moore Products, and Siemens; with DDE services for Windows or ASCII for raw data import; in a vendor- specific version such as Tune-A-Fish for the Fisher Provox control system; or in conjunction with a small data acquisition system (as needed). In its full direct driver or integrated control system version, the full Windows user interface is most complete.

There are added features in v.6.05, such as pop-up notices of controller connection errors, and lots of new highlighted hypertext terms to help new or uninitiated users with loop tuning terminology. But most of the improvements are below the surface of tuning, modeling and simulation analyses (which are based on imported process data).

Under-the-skin upgrades have been made in the program's modeling algorithms. These, says Gerry, result in better and more accurate analysis features, which measure robustness, frequency response, and tuning performance. There also are valve sizing checks, analysis hysteresis, filtering, and power spectral density, which shows power at various frequencies.

"Whenever you model a process or tune a loop, you want the model to match a real-world linear differential equation," says Gerry. "In some cases, there will be a difference." One convenient visual cue is an index that provides a "good/fair/poor" indication of how well the imported data matches the classic, linear PID model.

Another helpful visual tool is a robustness plot that can be viewed simultaneously with changes of dead-time or gain in a simulation plot. Again, these tools display only the surface results of underlying enhancements to the modeling algorithms.

According to Gerry, specific enhancements have been made to "second-order" loop simulations characterized by two large lags or time constants that might be 10 times larger than the dead time in a loop-for example, lags in the controller's effort to ramp-up or achieve steady-state temperature regulation in a process. Incidentally, unlike the earlier release of ExperTune, users no longer need to differentiate between first-order and second-order loops. The software automatically selects which type of model to use.

"Because of greater confidence in tuning loops like these," says Gerry, "we are able to tune those loops aggressively without sacrificing safety margin." After all, the goal is to tune as tightly as possible, at the brink of performance, without falling off the edge of process instability.

If higher productivity is mandated by plant managers, then a tool like ExperTune just might be a necessity for process control engineers.

Top of page

Software Review: Loop Tuning Simplified

Reproduced with permission of CONTROL ENGINEERING, May, 1995. © 1995 by Cahners Publishing Company

PID loop tuning can be a complicated operation. Analytical tuning techniques require a mathematical model of the controlled process and a series of arcane calculations to derive the appropriate values for the P (proportional), I (integral), and D (derivative) parameters. There are as many tuning techniques as there are applications subject to PID control, and each involves a different set of rules for deriving the desired parameter values. Although several very powerful software packages available for performing the necessary calculations, the user typically needs a college education in feedback control theory just to understand the results.

The engineers at ExperTune Software (formerly Gerry Engineering, Hartland, WI) are trying to change all that. Their most recent PC software package, ExperTune for Windows, takes the user through a loop-tuning operation step-by-step without a lot of analytical mumbo jumbo. The results are PID parameters that will give the loop in question optimal set-point tracking or a slow, fast, or faster response for disturbance rejection.

ExperTune does all the work. All the user really needs to do is identify the location of the loop to be tuned and then follow ExperTune's directions. ExperTune first asks the user when and how much to "bump" the process input so as to determine the process response to a control action. Once the process has finished reacting to the bump test, ExperTune terminates data collection, generates a model for the process, and calculates the tuning parameters required to meet the user's performance specifications. ExperTune also estimates the performance improvement that should result once the loop is re-tuned. Once user approves of ExperTune's recommendations, then they can download the new parameters to the loop and resume automatic control operations.

For the more ambitious user, ExperTune provides several additional analysis tools. A "robustness plot" is available to show how a change in the process dead time or gain would affect the stability of the closed-loop system as it is currently tuned. The current process parameters are shown as a point on a two-dimensional plot of normalized dead time vs. gain. All points that would correspond to stable closed-loop operations are shown within a "stability region." If the current operating point lies well within the stability region, the closed-loop system is said to be "robust." That is, it is unlikely to become unstable if the process dead time or gain should happen to change slightly. The dimensions of the stability region depend upon the current values of the PID tuning parameters. Hence, the robustness plot can provide the user with a graphical image of the risk of instability that a given set of tuning parameters would entail.

ExperTune can also provide a dynamic simulation of the closed-loop system based upon the observed behavior of the controlled process. This feature allows the user to check the closed-loop performance of ExperTune's recommended tuning parameters even before they are downloaded to the PID loop. I would find this feature particularly useful if none of ExperTune's three recommendations were quite right for my application. I would start with the recommendation that provides the closest approximation to the disturbance response I require, then fine tune the P, I, and D values by trial-and-error. The simulator would allow me to test each new set of parameters without actually running the process.

These and ExperTune's other analysis tools can all be useful for analyzing the behavior of a closed-loop control system, but only if ExperTune's internal process model accurately reflects the behavior of the controlled process. The accuracy of that model depends on the integrity of the process data collected during the initial bump test. To insure data integrity, the bump test must start and stop with the system at rest, it must not be conducted during periods when appreciable load disturbances are possible, and it must include data from a linear range of operation. I would find these conditions difficult to meet in applications where the process lacks a steady state, the process is subject to load disturbances at unpredictable intervals, or the process responds differently to positive and negative step inputs.

In these situations, I would recommend more advanced modeling and tuning techniques or the services of an expert control engineer. ExperTune Software just happens to have a staff of engineers available to help tune such difficult loops. There is even a standard facility built-in to ExperTune for collecting the necessary set-up information and process data on a disk that the user can send to the company for analysis.

ExperTune for Windows runs on an IBM PC-AT or compatible computer under Windows 3.1 or higher. The host PC must have at least 2M bytes of available memory, a VGA display, and a parallel printer port. Several interface protocols for acquiring plant data are supported including DDE; an A/D converter available directly from ExperTune Software; and drivers for Allen-Bradley, Modicon, Siemens, Moore Products, and GE Fanuc controllers.

Vance VanDoren
Consulting Editor

Top of page

Software Review: Package Simplifies Control Loop Tuning

Reprinted with permission from CONTROL, December 1994.

It doesn't make much sense to tune loops manually anymore, according to users of ExperTune loop tuning software from Gerry Engineering, Hartland, WI. They say difficult control loops can be tuned quickly accurately and consistently without a college degree in control theory. Savings come from time and material saved during the tuning process as well as from higher process efficiencies from good tuning.

At Blitz Weinhard Brewing Co, Portland, Ore., Plant Engineer Richard Ginter used Gerry's Windows package to tune a distillation system with a lot of interaction among loops. By systematically tuning each loop he "increased capacity through the system about 30%," Ginter says. Tuning took four days instead of 30 days for manual tuning, "plus tremendous product savings."

An engineering supervisor uses the software for maintaining EPA requirements. With ExperTune, he says a maintenance technician can do the job of a specialized engineer "It took the mystery out of the PID block in the PLC," he says. "We had a lot of uneasiness about modifying it. ExperTune brings up the needed information and downloads the changes."

Kevin Moore, a controls supervisor at Nabisco in Chicago, bought ExperTune Lite, a lower-cost version, to solve problems with a heat sealer for wrappers on Ritz snack packages. "I was having a lot of problems getting the PID loop tuned - we were over-shooting by 100 degrees F, says Moore. Like several other users, Moore "didn't even read the manual." He worked through the menu step by step, which "walked me right through it," he says.

After it calculates recommended parameters, the software predicts how much they would improve the control, and asks if they should he downloaded. Moore says, "It said 1,000% better I haven't looked at that process since."

Drew Grippa is less trusting. The senior process control engineer for Appleton Papers Inc. in West Carrollton, Ohio, has checked ExperTune's work. "When I tune a loop using ExperTune, then check with manual calculations, the data is almost identical," Grippa says. This allows him to let others do loop tuning, since he knows the results will be uniform. Compared to manual methods, he says, "We can trust the data that comes out no matter who does the keystrokes."

ExperTune does require some variation in the process to perform tuning. This limits its application where the process can't tolerate an upset, says Dave Leonard, director of engineering at Control Masters, Inc., Downers Grove, IL. He gives the example of a large autoclave making 6,000 liters of intravenous solution, where the temperature cannot be allowed to vary more than a few tenths of a degree.

Fringe benefits of the software include on-demand reference materials and helps "like a good tuning guide," says Leonard. For documenting various systems integration projects, he uses ExperTune to give him a standardized chart of loop parameters.

A number of drivers are available for easy connection to PLCs and control systems. Leonard says this makes it easy to use with his Modicon PLCs. "You don't have to kluge around on PLCs and outputs," he says.

However, for some proprietary distributed control systems (DCSs), connect-ability can be awkward, mentions Grippa. "We had to dedicate two outputs on the DCS for loop tuning and configure the loop to be tuned to them," he says.

Drivers caused trouble for Ginter, who had some problems getting DDE to run when upgrading to Windows due to slight differences in drivers.

Users find Gerry's customer service to be fast and helpful with operational problems. Grippa says they were "most helpful" when he needed help in getting capital approval for the package. "They allowed us to rent it and applied the rent to the purchase," he says.

While ExperTune's cost, places it in the capital-expenditure range, users say the return on investment is rapid. "We made back the investment in the first couple of weeks in loss of product and time saved in documentation," says Leonard.

"Cost is insignificant compared to what it's done for me in time savings and savings on lost materials - it paid for itself in the first use" says Moore.

Grippa points out that the package is "a little expensive, but in our system a single loop that's out of tune can cost us thousands of dollars per month."

Top of page