Diagnostics dig deeper

Today’s plants come ready equipped with their own team of doctors in the shape of the many diagnostic and monitoring systems incorporated into instrumentation and other process equipment. And, like any good doctor, today’s plant diagnostics can look beyond the symptoms of poorly performing instruments and provide insights into the health of the process itself.

By Mike Spear, editor at large

2 of 2 1 | 2 > View on one page

ABB’s 2600T series of Foundation Fieldbus differential-pressure transmitters (Figure 3) now all feature built-in plugged-impulse-line detection diagnostics known as PILD. A board mounted in the transmitter runs the “self-adaptive” PILD algorithm, which is based on the spectral analysis of the differential pressure signal. It adapts to the specific process conditions at the time of installation and during normal operation compares the noise level to these conditions.

Figure 3. Foundation Fieldbus differential-pressure transmitter features built-in plugged-impulse-line detection diagnostics based on “self-adaptive” algorithm.
Figure 3. Foundation Fieldbus differential-pressure transmitter features built-in plugged-impulse-line detection diagnostics based on “self-adaptive” algorithm.

If the noise level is statistically higher than the original value, it indicates a single line plug (and can detect which of the lines); while if the noise is substantially lower then the diagnosis is of two lines plugged. The algorithm has been proved to work equally well for both liquids and gases.

“At the moment we can detect if an impulse line is open or closed, but we are doing more R&D to see if we can detect the onset of a line becoming blocked before it actually becomes plugged,” explains Eugenio Volonterio, R&D manager for pressure products with ABB SACE in Como, Italy.

The range of Foundation Fieldbus pressure transmitters from Yokogawa, Newnan, Ga., already boast that capability. Because they feature two sensors on the one chip — one measuring differential pressure, the other static pressure — the transmitters can discriminate between high and low side clogging by using an algorithm to monitor the changes in process noise over time.

“So we can not only detect that plugging has occurred, we can also predict over time when plugging will start to develop,” says Henk van der Bent, marketing manager for process control instrumentation, Yokogawa Europe, based in Amersfoort, the Netherlands.

Enhanced electronics

Emerson’s SPM advanced diagnostics has been available on Foundation Fieldbus devices (including temperature transmitters) for a few years but was extended to HART devices last December, a move Schmeling credits particularly to advances in low-power electronics. This is enabling instrument manufacturers to pack more functionality into their devices. Bridges, for example, says “Krohne has been doing this for some years with our radar level devices, where different signals can indicate unusual events in the vessels, such as broken baffles or damaged mixer impellers.”

Siemens Automation and Drives, Spring House, Pa., and Nuremberg, Germany, offers similar capabilities in its recently introduced Sitrans LR250 radar level transmitter. A new graphical local interface displays echo profiles and diagnostic information, allowing the user to determine dynamics in the tank at a glance. Self-diagnostic features are shown on the local display or communicated across the control network. By applying algorithms based on field data collected from real applications over the years by Siemens field service engineers, the “Sonic Intelligence” signal processor is said to automatically ignore obstructions to give accurate and reliable level readings.

Bridges also points to recent diagnostic developments in flowmeters. “With our magnetic flowmeters,” he says, “we’re looking at changes in conductivity that could show up problems like coating on the electrodes. In the old days solids bouncing off the electrodes could create a lot of noise, but now we’re using that to determine what the likely solids content is in the fluid.”

The company’s Optiflux magmeters now feature three levels of diagnostics — of the meter itself (microprocessor, memory, software, etc), of the application (problems that NAMUR recommends a meter should be able to detect, such as gas bubbles, electrode corrosion and fouling, changes in conductivity, and so on), and “out-of-spec” diagnostics, in which the accuracy of the meter is checked against a test signal.

The latest versions of Rosemount E-series magmeters also feature diagnostics that alert users when instrument variability is caused by process noise and not actual flow variation. This allows for adjustment to a higher coil drive frequency to stabilize the output without additional damping.

Coriolis challenges

It’s probably fair to say that in the early days of Coriolis meters their reputation suffered because of a perceived inability to respond to rapidly changing process conditions such as entrained gases or the onset of two-phase flow in the pipe. While acknowledging that, Bridges notes that manufacturers in recent years have addressed these issues, putting Coriolis flow metering in greater demand than ever before.

For example, through “adaptive sensor technology,” Krohne claims to have succeeded in making its Optimass straight-tube Coriolis meters independent of process conditions. “One of the patents we hold,” says Bridges, “is for a strain gauge on the sensor to compensate for expansion, but we have been using that to compensate for pressure and various other parameters. There are all kinds of things that can be derived from diagnostics.”

“The Holy Grail,” as Bridges calls it, for Coriolis meter makers is an absolute compensation for two-phase flow.

Toward that quest, Invensys Process Systems, Foxboro, Mass., five years ago introduced its Foxboro CFT50 Coriolis transmitter. Now in widespread use, including on custody transfer duties, the CFT50 is based on patented technology developed in collaboration with Oxford University in the U.K. This involves new signal-processing techniques that are said to enable the meter to respond to changing flow conditions much faster than other Coriolis meters.

Other developments also are occurring. For instance, last year Emerson’s Micro Motion unit, Boulder, Colo., added an in situ meter verification capability to its flagship Elite meter (Figure 4). Specifically, it can determine whether the meter has lost any stiffness due to corrosion or scouring of the sensor tube.

Figure 4. This Coriolis meter employs a series of tones to check for tube stiffness, an indicator of erosion, corrosion and deformation.
Figure 4. This Coriolis meter employs a series of tones to check for tube stiffness, an indicator of erosion, corrosion and deformation.

"We discovered that by analyzing the higher level noise that is created in response to a set of very carefully designed test tones, we can actually measure the structural stiffness of the flowmeter element to see if it’s changed or damaged in any way. This gives us a diagnostic tool that is decoupled from changes in the process fluid. Operators don’t have to take the meter out of line to test it against a flow reference. We know through our research that if the stiffness of the flow tubes has not changed [from the factory setting] there can be no erosion or corrosion or plastic deformation of the metal flow tubes — which are known to be the only factors that can cause a change in the meter’s calibration,” says Marc Buttler, product line business manager.

The test tones, which have been developed empirically, have multiple frequencies such that the fluid inside the meter U-tube has no effect on the measurement, regardless of its density or flow rate.

The meter remains in-line during the four minutes the test procedure takes to run, but does not actually record the flow during that time. This is perhaps one reason why the diagnostic tool has not yet been recognized for regulatory custody-transfer applications. Elsewhere, however, “most users are experiencing great success with their own internal processes for fiscal transfer, quality management and predictive maintenance. By knowing their meter is remaining structurally sound, they can extend their own calibration intervals and go from a reactive, or even scheduled, maintenance approach to a predictive one,” Buttler says.

End users clearly are clamoring for more and better diagnostics. “They are starting to put pressure on manufacturers to look into this [improving diagnostics],” says Krohne’s Bridges. “But there is always a tug of war between the two — how much they are prepared to pay for it, and how much the manufacturers can afford to put into R&D. It’s a technology race.” Regardless, plants stand to be the winners.

2 of 2 1 | 2 > View on one page
Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.


No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments