| INSTRUMENTATION AND CONTROL |
This section was edited by Executive Editor Harry Hutchinson.
STEAM TRAP MONITORING: ADDING VALUE BY CUTTING WASTE
By Ashley Prins
Steam traps are mechanical devices designed to help improve the efficiency of steam distribution systems. When functioning properly, steam traps remove condensate and incondensable gases from steam, improving the system’s ability to transfer heat. Enhanced efficiency of the steam system usually leads to lower utility costs. A failed steam trap, on the other hand, can lead to excessive utility cost and sometimes puts system components at risk.
By monitoring the steam traps at their plants, operators can protect equipment and save money. Implementing a system that alerts plant maintenance of trap failure allows steam traps to be replaced before steam leakage can register on the utility bill, or plant equipment is damaged by water hammer or by condensate freezing in the lines. Of course, the sooner trap failure can be detected and the trap repaired or replaced, the better.
Steam traps fail in two ways—open or closed. A float trap, for instance, is likely to fail closed. Whether the failure is caused by a crushed ball-float or a broken linkage, a closed trap can fill with condensate and eventually flood a portion of the steam system. An inverted bucket trap, on the other hand, typically fails open, allowing steam to escape through the trap. Of course, traps don’t always fail completely open or closed; partial blockage of the condensate drain port or leaking valves are common failures. However extreme the failure, a steam trap that isn’t working properly can be costly.
Schematic of a condensate return system: Trapping and returning condensed
liquids improves the overall heat-transfer efficiency of the steam line.
Grashof’s equation is often used to estimate the utility cost of a steam trap that has failed open. The equation captures the relationship between the area of the orifice and the pressure of the steam line, and how these proportionally affect the flow of steam through the orifice.
.
Q = 0.7 x 0.0165 x 3600 x A x P0.97
.
Where Q is the flow rate of steam (lb./hr.), 0.7 is the coefficient of discharge for hole, A is the area of hole (in2), and P is the pressure inside steam line (psia).
Once the rate of steam loss through the orifice is estimated, the overall cost of “lost steam” can be calculated.
.
W = (Q x H x E x 10-6 x C) / BE
.
W is energy lost ($); Q is flow rate; H is hours in use; E latent heat of steam (at given pressure); C is fuel cost per million Btu, and BE is boiler efficiency (%).
Certain technology methods have made it possible to closely monitor steam traps and provide plant maintenance the opportunity to repair the trap before it becomes too costly.
Manual maintenance programs often rely on sight and sound to detect trap failure. Procedures might involve opening test valves downstream of the trap to detect steam blowing through the trap or checking sight glasses installed in the condensate return lines. Portable ultrasonic systems are sometimes used to monitor sounds (audible and inaudible) generated by the flow of steam/condensate through a trap. Infrared thermometers are yet another tool that can be used by maintenance staff to detect cool traps (filled with condensate).
Having enough staff available to perform consistent steam trap monitoring can be very costly. Even with a good maintenance program in place, steam traps in hard-to-reach locations will often be missed.
Interstates Companies designs turnkey electrical and automation systems, including process control, for industrial facilities. We see great potential for our clients to improve plant efficiency by implementing automated monitoring systems into their facilities. Monitoring traps using instrumentation could eliminate human error that may occur in manual maintenance programs. Instrumentation provides monitoring points for traps in inconvenient or dangerous locations. It also provides a plant-wide automated monitoring system that can eliminate the need for a manual program altogether.
Common instrumentation detection systems use ultrasonic, conductive, or temperature monitoring technology to detect failures. Sensors mount directly to steam traps being monitored; the transmitter detects changes in steam trap performance and alerts plant operators when there is a failure.
Ultrasonic or acoustic sensors are typically non-invasive, and so can save on installation cost. The sensor detects sound generated by condensate flowing through the orifice, by the opening and closing of the trap valve, and by steam blowing through the trap. Sensor thresholds are set to allow the sensor to detect trap failure. Standard installations often involve mounting the sensor directly to the steam trap on a bracket.
High levels of vibration in the plant environment can interfere with ultrasonic technology. It is also very difficult for some plants to set accurate failure threshold levels without prior acoustic data from trap maintenance programs. Some manufacturers will preset acoustic thresholds based on a site survey. This is a more expensive option upfront, but prevents the headache of having inaccurate threshold levels.
If a ball-float trap fails closed, condensate can back up
and damage the steam system.
Temperature sensors can be intrusive devices, including resistance temperature detectors or thermocouples, which monitor the temperature of fluid inside the trap. Non-intrusive devices such as infrared sensors monitor the exterior of the trap or ambient temperature. Cool temperatures indicate that the trap has been flooded with condensate or the trap is not seeing steam. Often, temperature measurements are used alongside another technology, such as ultrasonic measurement, to more accurately monitor trap status. Some manufacturers offer sensors that have dual measurements, with temperature as a secondary variable.
Conductivity sensors are intrusive devices that detect the level of condensate in the trap (or directly upstream of the trap, depending on where the sensor is installed). Under normal operation, a small level of condensate will fill the inlet of the trap. If the trap should fail open, causing steam to blow through the trap, no condensate would be present. Temperature sensors are needed to help decipher if the trap is failed closed or operating normally, or if the trap is not in use (for example, in batch applications).
There are many options to evaluate when looking to implement an automated steam trap monitoring solution. Plant operators might choose to have a third party come in to perform a plant-wide steam trap survey to help assist in the design process.
The results of the survey or an evaluation of current trap maintenance programs can help determine the level of automation a plant may choose to implement. Trap size, location, and style can help plant operators to decide which traps to monitor specifically. For example, a client may choose to wirelessly monitor large-capacity steam traps (the size of a steam trap is directly proportional to the cost of that trap failing open) and traps that are in remote locations. The client can then implement a manual maintenance program for the rest of the traps.
Once a plant has made the decision to implement an automated solution, an evaluation of specific trap locations can be used to select a specific instrumentation solution. Options might include the decision to go with a wireless versus a traditional wired solution to try to save on installation and material cost. High levels of vibration in certain plant areas might steer a user toward installing conductivity sensors over acoustic sensors.
Wireless repeaters (which retransmit radio signals from the wireless sensor to the base radio) can be used to increase wireless signal strength through dense areas of the plant. Other features may include a Web-based interface program that can be implemented to allow transmitter information and payback reports to be accessed from corporate offices.
Whatever the end design, vigilant steam trap monitoring can prevent damage to the steam system, reduce utility costs associated with steam leakage, and keep the steam system running at full capacity. Using steam trap monitoring to catch these losses at their earliest minimizes financial losses.
Editor’s note: Ashley Prins is an instrumentation engineer with Interstates Companies in Sioux Center, Iowa.
BLADE TEST RECORD
By Harry Hutchinson
A company that has developed proprietary technology for testing turbine blades recently completed a round of tests on F-class rotor blades, which may have set a record of sorts.
The company, Test Devices Inc. in Hudson, Mass., tests blades on a spin rig and, at the same time, uses a patented technology that causes the blades to undergo orders of vibration they would see in service. The blades are simultaneously subjected to static, centrifugal stress, and to dynamic, vibrating stress generated in an engine, to study high-cycle fatigue, or HCF.
The F-class blades were tested for more than 1,500 microstrain—that is, strain of more than 1,500 parts per million. They were tested for 10 million HCF cycles. The blades weigh more than 18 pounds each, and TDI said that, as far as it knows, they are the largest blades tested by this method.
According to the company’s president, Robert Murner, the testing method held resonance for four to five hours continuously in order to attain 10 million HCF cycles, the requirement to validate the customer’s data. Speed had to be controlled to within a quarter of an rpm to maintain continuous vibration at the desired stress amplitude, Murner said. The rotation is powered by releasing compressed air into an air-driven turbine.
The exact test speed is confidential, but was in the “tens of thousands” of rpm, he said.
The company calls its test technology Dynamic Spin Testing because it uses patented equipment to recreate the static and dynamic stresses induced in compressor and turbine blades as they operate in proximity to stator blades and other engine static structures. In addition to spinning the test piece, the system sprays the blades with tiny droplets of oil to trigger the desired vibration mode and stress amplitudes.
According to Murner, the method has been used to test various vibration modes, including first and second flex, first and second torsion, and first and second axial, in turbine as well as compressor blades.
Blades are monitored using a combination of a non-intrusive stress measurement system and strain gauges.
Murner said that the method can validate lifting models of blades and can be used to observe crack propagation or to verify behavior of blades that have been damaged.
The basic process was developed in the 1990s for the U.S. Navy, which also uses the technology to study turbine blades. TDI is the only private company using the technology.
OIL LIGHT FOR EVERYTHING
By Jeffrey Winters
The dashboards of most cars have a spot for a useful little light, the one that flashes "check oil" when the engine's lubrication system appears to be compromised. It wasn't always thus: Pulling the dipstick and studying the quantity and quality of the oil clinging to it was once part of routine auto maintenance.
Research conducted at the Georgia Institute of Technology in Atlanta promises to take the concept of the oil light to new dimensions. Soon, a computer model incorporating data from sensors will be able to calculate in real time the amount of useful life left in an engineering system. Fully embraced, the work by engineering professor Nagi Gebraeel might well result in an integrated logistics system where mechanical equipment is almost autonomously kept at optimal working condition.
At first, Gebraeel was interested only in researching the vibration and acoustic signals from rotating bearings. Comparing these signals to the degree of wear-based damage the parts had displayed, he could create a model that could use noise and vibration data to predict the remaining life in an individual bearing. Experimental results suggested that such a system could reduce costs associated with both part failure and the unexpected lack of a specific spare part by more than half.
Gebraeel recently began working with Rockwell Collins to develop adaptive models to estimate the remaining useful life of electronic aircraft components. The goal is to embed a prognostic system into an aircraft's avionics, so that decisions about airworthiness of an aircraft could be made more or less automatically.
Such sensor-driven prognostic models combine experimentally derived general reliability characteristics with real-time signals from small sensors, which can be linked together in a wireless network. The model would use the vital signs from these sensors to make an accurate prediction about the best time to order spare parts or when to schedule maintenance.
To help develop this prognostic model concept further, Gebraeel is working with a small, Virginia-based company, Global Strategic Solutions LLC, to develop embedded diagnostics and prognostics that could predict the probable life remaining of electrical power generation systems on board U.S. Naval aircraft, and for the communication, navigation, and identification avionics systems used on the Joint Strike Fighter.
WATCH ON THE PIPELINE
By Harry Hutchinson
There are oil and gas under the ground in northern Alberta, and Birchcliff Energy Ltd. of Calgary operates a pipeline to take some of it to markets. The path from one of its well sites, near the town of Worsley, travels a 650-meter slope that requires special care.
According to a Birchcliff spokesman, the grade is so steep that the company had to bore through the ground and run the line through the hole. The line contains four pipes—one each to carry oil, gas, and water from the well site, and a smaller one to carry fuel back to the well operations.
Unlike pipe laid flat, this section—about 650 meters—will have gravity acting along its length, subjecting it to axial stress.
To keep watch on this section of pipeline, Birchcliff has opted to install a strain monitoring system from a Toronto company called Fiber Optic Systems Technology Inc., also known as FOX-Tek.
FOX-Tek’s chief technology officer, Don Morrison, said the system uses fiber optics to monitor axial strain and bending of the pipeline, as well as temperature. Three fiber optic wires are affixed at equal intervals around the surface of each pipe just below the bend at the top. Bending or strain in the pipe will cause some of the fiber optic carriers to stretch. The difference in length, which indicates strain, can be detected in micrometer resolution, he said.
The temperature sensor is in contact with the pipe, but not bonded to it, so any change in its shape will be the result of temperature and not strain.
Birchcliff collects data and sends it to FOX-Tek for analysis.
COUNTERFEITING GPS
By Jeffrey Winters
The Global Positioning System has become so integrated into everyday lives—GPS receivers are built into many cell phones now—that few people question the reliability of its data. But a team of researchers from Virginia Polytechnic Institute in Blacksburg and Cornell University in Ithaca, N.Y., have demonstrated that the system is vulnerable to attack from a reprogrammed receiver.
The goal of the research, say the engineers who conducted the experiment, was to demonstrate that such vulnerabilities exist, in the hope that others can devise effective countermeasures.
The team used a GPS receiver used in atmospheric research. The briefcase-size device was reprogrammed so that it would emit signals that mimicked those coming from one of the constellation of GPS satellites. The phony transmitter would anticipate the signal from the overhead satellite and send out a fake data packet. Nearby GPS receivers couldn’t distinguish the fake signal from a real one, and thus rendered a false location.
With GPS embedded into so many systems today, an attack using this vulnerability could create chaos for utilities, transportation companies, and other enterprises that depend on the system to track the whereabouts of various remote components. This suggests that such companies might want to think about creating redundant systems that can operate even if the GPS fails. |
