Comment

1 November 2005

Beyond mere measurement

Devices featuring such diagnostics are synonymous to data servers.

By Brian LaBelle

Downtime reduction and cost containment are persistent objectives for operations, engineering, and maintenance professionals in the power industry.

Corrosion and deposition cost this industry billions of dollars every year in reduced efficiency and equipment failures. Effective utilization of sophisticated analyzers and control systems is necessary for avoiding unplanned outages and for otherwise meeting profitability and process control responsibilities.

The more popular recent adoptions are the combination of intelligent field devices and asset management software. Cost-saving advantages from system start-up to fault diagnoses are setting new standards for what can be expected and accomplished from the control room.

Devices are data servers

Instrumentation and automation engineers, plant chemists, and other technical professionals in the power industry have for years appreciated the advantages of control systems that provide simple access to a wide variety of intelligent field devices for the purposes of configuration and commissioning, diagnostics and monitoring, calibration management, and automated documentation.

Worldwide the number of plants implementing such technology is increasing, and the trend will accelerate, driven primarily by operations professionals recognizing broad opportunities for output improvements and cost reductions. We'll look at the field devices and asset management software and emphasize the benefits of using them in tandem.

The control system consists very generally of a controller unit, workstations, and various field devices. Advanced control systems capable of delivering the benefits described possess a number of differentiating features including intelligent field devices and asset management software implemented at the workstation level.

Intelligent field devices provide critical information on their own health, the health of the process, and the health of the equipment around them. This diagnostic functionality works through digital communications such as HART and Foundation fieldbus. Valuable diagnostic capabilities that exist in some devices today include:

Control valves: plugging of I/P, travel deviation, insufficient air supply, stuck valve, calibration changes, failed diaphragm, diaphragm leaks, and O-ring failures in piston actuator

Pressure: electronics failure, sensor failure, process condition, configuration warning, and plugged impulse lines

Temperature: electronics failure, sensor failure, process condition, configuration warning, RTD drift, and RTD life estimation

Flow: electronics failure, high process noise, grounding fault, electrode fault, empty pipe, reverse flow, and calibration error

Analytical: pH electrode aging, glass electrode failure, reference electrode failure, reference electrode coating, and reference electrode poisoning

Devices featuring such diagnostics are synonymous to data servers because they do much more than simply measure process variables like pH, pressure, temperature, level, or flow. They provide the basis for abnormal situation management, process optimization and advanced control. The full value of this capability is only realized when the intelligent field devices are utilized in combination with asset management software.

Asset management software, particularly the variety tailored specifically for control valves, instrumentation, and analytical equipment, facilitates quick and easy device configuration at commissioning and enables predictive maintenance by providing access to diagnostic information, calibration routines, and automating documentation. Successful implementation results in significantly lower start-up and maintenance costs and improves asset performance and process stability over time.

The sheer volume of device configuration and initial calibration tasks necessary to a capital project often represents savings potential sufficient to justify the relatively modest incremental costs of intelligent devices and asset management software. Checking the configuration of every transmitter and valve takes place a number of times before startup and calibration verification happen at least once. These necessary activities can take place more quickly and easily and with fewer errors from a workstation with asset management software than by physically visiting and revisiting each device in the plant, pushing buttons to navigate sometimes-extensive menu structures in sometimes-cryptic text on small local displays. Additionally, all commissioning activities including calibrations, and all subsequent changes made through asset management software can be automatically documented without the extra labor and risk of errors associated with manual reporting.

The ways in which the combination of intelligent field devices and asset management software pays for itself in faster commissioning at the end of a capital project are clear and almost always realized due to project management oversight. More allusive are the productivity improvements achievable throughout the lifecycle of the plant.

Generally, this requires an appreciation that intelligent devices need management and that work practices prevalent in the heavy industries today need to evolve from routine and reactive maintenance to a culture of predictive maintenance.

Managing intelligent devices

Saving time at plant commissioning and system start-up and keeping production lines up and running to sustain the overall availability of the plant is imperative. Control system assets and other plant equipment must see regular maintenance and must be in good health.

Strong and sustainable financial performance is necessary in an increasingly competitive global market, requiring the optimization of maintenance practices as well as optimization of the processes of production themselves.

Understanding the costs of routine and reactionary maintenance exposes the need to manage intelligent field devices and provides strong motivation for developing and implementing predictive maintenance plans.

Page one of An Introduction to Predictive Maintenance by R. Keith Mobley says in the U.S. alone, more than $200 billion/year goes to the maintenance of industrial plant equipment and facilities, and the result of ineffective maintenance management represents a loss of more than $60 billion/year. It also states maintenance can represent from 15% to 60% of the cost of goods produced, depending upon the specific industry.

Maintenance in food-related industries averages approximately 15% of the cost of goods produced, while in the pulp and paper, iron and steel, and other heavy industries maintenance can approach 60% of total production cost.

Countries in the west at present are spending approximately five times more on maintaining existing assets than on expansions and new facilities. Over 50% of all electricity generated in the U.S. comes from plants 25 years of age or older and 30% of U.S. refineries have had some of the same instrumentation and automation equipment for more than 16 years.

According to Shell Oil Company, a 1% decrease in the productivity of a refinery can reduce output by almost 600 barrels per day. Such plants certainly need to sustain production and minimize unplanned outages. Despite spending billions every year on maintenance, equipment failure is still the number one contributor to refinery production losses.

Frequencies of failures leading to unplanned outages in the oil & gas industry can break out by equipment type. Mechanical and process equipment are the major sources with instruments and valves representing only a tiny fraction.

However, 27% of all the maintenance spending in this industry goes to instruments and valves. This disproportionate allocation also holds true in other industries. We'll look at this further following closer examination of the instrument and valve maintenance activity in oil & gas.

Of all the work orders issued for instrumentation and valve maintenance in the oil and gas industry, over 37% are for routine checks. Over 12% of the work, nearly 20% of all troubleshooting effort cedes to the "no problem found category." Half the maintenance work orders in this industry result in no action taken. This waste is typical of the routine and reactive practices common in the heavy industries. Good evidence is available to support predictive maintenance as a better approach, and proper management of intelligent field devices is essential for successful implementation.

Successful implementation of predictive maintenance practices is likely to be a very worthwhile endeavor for almost any industrial plant. R. Keith Mobley reported the following results from over 500 plants around the world, in various industries, that had been practicing predictive maintenance for at least three years:

• Maintenance costs reduced by 50-80%.
• Machine breakdown reduced by 50-60%.
• Spare parts inventory reduced by 20-30%.
• Machine downtime reduced by 50-80%.
• Overtime premiums reduced by 20-50%.
• Machine life increased by 20-40%.
• Productivity increased by 20-30%.
• Profit increased by 25-60%.

Some of these astonishing improvements can only happen using the combination of intelligent field devices and asset management software. Intelligent device diagnostics via HART and Foundation fieldbus, for instance, provide a portal to system health. Device alarms appear as alerts at the workstation allowing operators early detection of situations that, if left unattended, could result in machine breakdown and unplanned outages. Less time on superfluous activity means more time for the work that improves process reliability, leading to dramatic increases in productivity and profits over the lifecycle of the plant.

In addition, by checking the status, configuration, calibration, and diagnostics of individual devices from a workstation, it is possible to quickly identify devices that require shop service and ones that can stay in line for service or receive no service at all. The avoidance of unnecessary maintenance can amount to very significant cost reductions and are possible without jeopardizing the integrity of plant operations.

Sophisticated asset management software packages offer automatic documentation of maintenance events, like calibrations, and can display the data graphically for easy analysis. The "pH instrument calibration history" screenshot is the actual calibration history of a pH measurement loop at a feed plant. The red lines above and below zero are the control limits set by the user. Plots of the "as found" and "as left" data appear in blue and green, respectively. Using this and similar data from several other instruments the operations manager was convinced the interval between calibrations could extend longer without risk to the process. The result is an optimized maintenance schedule and an ongoing accumulation of savings.

Recognizing the need to manage intelligent field devices and focusing maintenance resources on results are important steps in the development of predictive maintenance practices. Most instrumentation and automation professionals truly appreciate this, but implementation efforts at any given plant rarely succeed without a clear plan and without strong staff level support.

Critical diagnostic mojo

The full value of intelligent field devices can only happen when it works in combination with asset management software, and only then if critical diagnostic capabilities are integrated into predictive maintenance practices.

Successful implementation will reduce maintenance costs and machine breakdown rates by at least 50% each, and will increase productivity and profits by at least 20% each.

The technology is a proven commodity. The operational benefits are clear. The body of evidence is compelling. Operations professionals are taking notice. y

Behind the byline

Brian LaBelle (Brian.LaBelle@EmersonProcess.com) is the manager of industry marketing at Rosemount Analytical, Liquid Division of Emerson Process Management. He has 17+ years experience in industrial instrumentation and automation equipment.