By Ravi Jethra, Industry Business Manager – Power/Renewables, Endress+Hauser
Temperature is one of the most widely measured parameters in a power plant. No matter the type of plant, accurate and reliable temperature measurement is essential for operational excellence.
Incorrect measurement because of electrical effects, nonlinearity or instability can result in damage to major equipment. Using advanced diagnostics, modern temperature instrumentation can inform a plant’s maintenance department that a problem exists, where it is and what to do about it long before anyone in operations even suspects that an issue exists.
This article covers some of the basics of temperature measurement in power plants and discusses technical advances that impart higher a degree of safety and reliability. These advances are based on innovative technologies that are being implemented in process instrumentation. Implementation of these new technologies can result in improved safety along with lower installation and maintenance costs.
Thermocouples versus RTDs
Although some specialty temperature measurements involve infrared sensors, the vast majority of measurements in a power plant are made with resistance temperature detectors (RTDs) or thermocouples (T/Cs). Both are electrical sensors that produce a mV signal in response to temperature changes.
|A modern temperature transmitter can be set up with triple redundancy for maximum reliability on critical processes, such as this steam header. All photos courtesy of Endress+Hauser|
RTDs consist of a length of wire wrapped around a ceramic or glass core placed inside a probe for protection. An RTD produces an electrical signal that changes resistance as the temperature changes. RTD sensing elements can be made from platinum, nickel, copper and other materials and can have two, three or four wires connecting them to a transmitter. Ni120 (120 Ohm nickel) RTDs were commonly used in the power industry, particularly in coal-fired plants.
Ni120 at one point was largely used by rotating machine suppliers on their equipment, such as pumps. Instead of buying separate Pt100 wires, these suppliers would use the same Ni120 wire to build their own RTDs in-house and provide these RTDs as part of their equipment.
RTDs are commonly used in applications where accuracy and repeatability are important. RTDs have excellent accuracy of about 0.1ºC and a stable output for a long period of time, but a limited temperature range. The maximum temperature for an RTD is about 800ºF. RTDs are also expensive. An RTD in the same physical configuration as a thermocouple will typically be about five times more expensive. RTDs are also more sensitive to vibration and shock than a thermocouple. Common instrumentation wire is used to couple an RTD to the measurement and control equipment, making them economical to install.
A thermocouple sensor consists of two dissimilar metals joined together at one end. When the junction is heated, it produces a voltage that corresponds to temperature. T/Cs can be made of different combinations of metals and calibrations for various temperature ranges. The most common T/C type are J, K and N; for power industry applications, high-temperature versions include R and S.
Types J, K and N are the most commonly used thermocouples due to their wide temperature range and ability to perform well in the harsh environments encountered in power plants.
Thermocouples are selected according to the temperatures and conditions expected:
- For temperatures < 1,000°F and mounting locations subject to vibration, as well as low-corrosion atmospheres: NiCr-Ni (Type K)
- For temperatures < 1,832°F and corrosive atmospheres: NiCr-Ni (Type N)
- For temperatures > 1,832°F: Pt Rh-Pt (Types R and S).
A thermocouple can be used for temperatures as high as 3100°F. T/Cs will respond faster to temperature changes than an RTD and are more durable, allowing use in high vibration and shock applications.
Thermocouples are less stable than RTDs when exposed to moderate or high temperature conditions. Thermocouple extension wire must be used to connect thermocouple sensors to measurement instruments. The extension wire is similar to the composition of the thermocouple itself and is considerably more expensive than the standard instrumentation wire used with RTDs.
RTDs and thermocouples are both used in power plant temperature measurement. Each has its advantages and disadvantages, with the application determining which sensing element is best suited.
RTDs tend to be relatively fragile and generally not suitable for high temperatures or high vibration, so areas such as steam generators and pump monitoring tend to use thermocouples, but exceptions exist.
At the Ostroleka power plant in Poland, Endress+Hauser used a rugged RTD for the first time. Problems at Ostroleka involved vibration and electrical noise. Thermocouples could handle the vibration, but not the electrical noise. Endress+Hauser developed an RTD that had up to 60g vibration resistance and handled temperatures up to 812°F. The construction of the RTD is far more robust than other RTDs on the market, making it suitable for both high temperatures and extremely high vibration.
With either RTDs or T/Cs, it’s important to ensure that the temperature transmitters have the curves and linearization data built-in to the memory for the specific RTD or T/C without the need for custom programming.
Transmitters Superior to Direct Wiring
Most temperature applications in power plants involve directly wiring a temperature sensor to the control system. Often engineers wire direct because they mistakenly believe this is a cheaper and easier solution. Despite the large installed base of direct wired sensors, the trend is toward using transmitters in conjunction with temperature sensors. Transmitters save time and money in installation, improve measurement reliability, reduce maintenance and increase uptime.
A transmitter converts the mV signal from an RTD or T/C to a 4-20mA signal or to a digital fieldbus output such as HART, Foundation Fieldbus or Profibus PA in the case of a smart transmitter. Either of these outputs can be transmitted over a twisted pair wire for a considerable distance. Smart transmitters incorporate remote calibration, advanced diagnostics and built-in control capabilities — and some are capable of wireless operation.
Direct wiring requires sensor extension wires from the sensor to the automation system input modules. For thermocouples, these wires are expensive and sometimes fragile. RTDs can use inexpensive copper wires, but some RTDs have up to four wires.
In a power plant, the automation system can be a few hundred feet to even thousands of feet from the temperature sensors. This can amount to a large amount of money for installation depending on the number of sensors and the distances involved.
Aditionally, over long wiring distances, Electromagnetic Interference (EMI) and Radio Frequency Interference (RFI) can affect the signal. The electrical output from a T/C is only a few mV and can be completely overshadowed by RFI/EMI, depending on the installation. This can result in false alarms and occasional trips.
A typical power plant has many sources of EMI and RFI. On-site power generation and transmission equipment are major sources of electrical noise, but plants also have numerous large rotating machines with huge electrical fields. By using transmitters with that comply with the IEC61326 standard, temperature measurement can be made immune to EMI/RFI problems, even in electrically noisy environments. Temperature transmitters are available that accept more than two dozen different types of RTDs or thermocouples, and RTD inputs with two, three or four wires. These sensors can be connected to a transmitter without the need for special programming.
|Endress+Hauser TMT 162 temperature transmitter with big display, mounted on a thermowell.|
Advanced Transmitter Functions
Today’s smart transmitters offer functions that were unheard of 20 years ago. The extra cost of a smart transmitter is more than paid back with functions that reduce maintenance time and prevent failures that can shut down a power plant.
For example, most transmitters have a back-up function so that critical and safety relevant temperature measurement points can be constructed in a redundant manner. Here, two sensors are connected to the transmitter. If one sensor fails, the transmitter automatically switches to the second sensor.
The failure of the first sensor is transmitted and is simultaneously shown on the transmitter display. By using the back-up function of the transmitter, the temperature measurement point down time is reduced by up to 80 percent. When this feature prevents a process shut down, it more than pays for the cost of the transmitter and the redundant sensor.
For critical measurements, it’s also possible to set up a triple redundant system. In this case, three temperature sensors in a steam pipe to the middle-steam header are set up with a two-out-of-three voting scheme for increased reliability and safety.
Smart transmitters also detect problems such as T/C drift and low voltage, allowing maintenance technicians to perform planned and proactive maintenance instead of just reacting to failures after they occur.
Because of its physical construction, measurement points recorded by thermocouples tend to drift. One of the main reasons for this is the “migration” of material from one leg of the measurement element to the other. The time span during which a thermocouple will measure accurately tends to vary from just a few days to a number of years.
To determine the availability and accuracy of a thermocouple, it’s very important to recognize drift when it occurs. With two connected thermocouples, the transmitter constantly compares the two measured values and, should the result exceed the prescribed difference, will issue an alarm.
Modern temperature transmitters also have the ability to provide a low voltage warning if the potential drops below a threshold value. With older technology transmitters, when voltage drops, the unit continues to send a signal, although it could be off by as much as 25 percent or more from the actual value.
In applications where fast response time is needed, customers use grounded thermocouples, but this thermocouple type may cause a ground loop. This is avoided by using transmitters with superior galvanic isolation, up to 2kV galvanic isolation on most commercially available transmitters.
Galvanically-isolated transmitters in general also provide superior noise rejection as well as protection from electrical transients and surges in electrically noisy environment or during weather extremes such as lightning or thunderstorms. The current generation of temperature transmitters has a galvanic isolation that is about three to five times better than previous transmitters.
Curing Maintenance Headaches
Smart transmitters diagnose many common problems that might take several days for a maintenance technician to find, diagnose and repair. For example, it may be very difficult to diagnose if a temperature loop is suffering from ground loops, noise, bad connections, cable breakage or many other problems. Without a smart transmitter, a technician just has to plod through the sensor and its electronics, step by step.
It’s not just mechanical components that undergo wear and tear in a power plant; the electrical parts also see aging and corrosion. Process sensors and instruments in the power industry frequently work in very aggressive environments. Cable glands are rarely 100 percent sealed, and eventually corrosion on the terminals or even the connection wire becomes a reality. Corrosion on the sensor connection system (sensor element, field wiring and transmitter terminals) can lead to errors in measurement.
Although the atmosphere in a power plant may not have as many corrosive materials as a chemical plant, dust and other materials can cause corrosion over a period of time. Because the terminals in a transmitter and the lead wires are made of different materials, corrosion can occur.
In power plants, a manual check of all the sensor connections is virtually impossible. Temperature transmitters, on the other hand, continuously monitor resistances of the sensor connection cables,and give a warning so that preventive maintenance measures can be carried out with no measurement degradation.
Electronic devices can fail when exposed to extreme temperatures. Smart transmitters have a built-in RTD at the electronics module that monitors ambient temperature. When temperature exceeds the limits the unit is specified for, it gives a warning indication.
The mechanical, thermal and electrical pressures in power plants are, in many cases, enormous. This stress on sensors can quite often lead to damage such as cable/sensor breakages or sensor short circuits, the natural result of which is failure of the measurement point. Overstepping the allowable sensor circuit resistance is also seen as a break in the line. This can occur in both RTDs as well as thermocouples.
Cable breakage or sensor short circuits are detected by the transmitter’s analysis electronics and transmitted to the automation system. Devices that operate with a 4-20mA current output do this in the form of a fault current (NAMUR 43) or HART data output, while smart transmitters send indications over their digital network.
In addition to transmission of the measured signal, the HART protocol also enables the transmission of digital information superimposed on the 4-20mA signal. This information can contain device status, maintenance requirements, sensor failure indication, sensor open circuit indication and much more.
The problem with a number of process control systems in the power industry is that they do not have a built-in request system for the digital HART information. In that case, HART signals can be categorized using DIP switches, and then transmitted as simple on-off discrete signals to the automation system. The four categories are “Failure detected,” “Service mode,” “Maintenance required” and “Out of specification.” In short, smart transmitters can detect, identify and report small problems before they become large problems.
When the technician arrives at the transmitter to effect repairs, he or she sees a large and brilliant blue back-lit display that provides a clear reading from a distance of 8 to 10 feet. The digits on a new transmitter display are at least twice the size of any of the older devices. When the technician needs an instruction manual, schematic or other support material, these days he or she can just call it up on a cell phone app or a tablet browser.
Thermowells Provide Protection
RTDs and T/Cs can be surface mounted, installed in a probe or inserted into a thermowell. In severe power plant environments, a thermowell acts as a barrier between the process and sensing element. It provides protection from corrosive processes and abrasives, and it also provides protection when placed in applications where there is high pressure and/or flowing media.
A thermowell will allow the sensing element to be removed without interrupting the measurement as the sensing element is inserted into the thermowell from outside the pipe or vessel.
|Thermowells at a power plant in Poland were breaking because of the Von Karman Trail effect. The solution was a stronger thermowell and a vibration-resistant RTD sensor insert.|
A thermowell adds considerable cost to the measurement point because the thermowell has to be inserted into the pipe, furnace or vessel. This often entails cutting into the pipe and welding a fixture. Because the thermowell adds a layer of protective metal, it slows down the response time of the sensor. Thermowells are subject to failure, especially in the severe environments found in power plants. Excess pressure, vibration, temperature and corrosion are major causes of thermowell failure.
The four main failures of thermowells are:
- Mechanical – Bending or breakage caused by an applied force which is beyond the limits of the thermowell’s yield strength. High-pressure steam is a likely culprit.
- Corrosion – Induced by chemicals and/or elevated temperatures.
- Erosion – Resulting from high-speed particle impingement on the thermowell.
- Vibration/Fatigue – Failure due to Von Karman Trail Effect (vortex shedding around the thermowell).
Proper diagnosis can identify all but the most unusual thermowell failures. For example, at the Ostroleka power plant in Poland, a thermowell on the outlet of the boiler feedwater pump was constantly breaking because of the effects of the Von Karman Trail. The thermowell broke five times because of high frequency vibrations. The solution was to replace the thermowell with a stronger Endress+Hauser Omnigrad M TR10 thermowell, fitted with an ITHERM StrongSens vibration resistant RTD sensor.
Handling High Temperatures
Power plants can generate extremely high temperatures that often cause measurement problems. For example, in energy-from-waste plants, furnace temperature is a critical measurement. Burning the waste at high temperatures minimizes the release of harmful emissions. To accurately record the temperature in the furnace, three or more temperature thermowells are inserted into the furnace directly above the flame.
Because of the very harsh conditions in the furnace, conventional probes made from Incoly 800HT alloy will typically fail after three or four months of service. Because the probes are sited in an elevated position, changing them can be difficult. In addition, each time the furnace is opened there is the possibility that cooler air will enter or that hot gases will escape, both of which can decrease the efficiency of the process and cause health and safety concerns.
A recent trial showed that Endress+Hauser temperature probes with thermowells made from SD75 alloy can withstand the extreme temperatures up to 3,000ºF typically found in the furnace of an energy-from-waste facility. During the 12-month trial, two probes made from the new alloy were used alongside standard thermowells. In a like-for-like comparison, the new probes lasted three times longer than their Incoly 800HT counterparts.
The high chromium and silicon content of the alloy increases the stability of the instrument and makes it highly resistant to corrosion at high temperatures. The presence of these elements promotes the formation of a protective oxide scale, making the alloy resistant to attacks from sulfur, vanadium, chlorides and other salt deposits.
Advanced instrumentation is greatly improving temperature measurement in the power industry. The benefits of using smart transmitters instead of direct wiring includes installation cost savings, reduced downtime and proactive maintenance through the use of advanced diagnostics. When a power plant has to be shut down because of a failed sensor, the cost could run into the millions of dollars. Smart transmitters can tell a plant that a problem exists, where it is and how to fix it — anticipating failures before they occur.
Ravi Jethra is the Program Manager – Power Industry at Endress+Hauser. He has over two decades of experience with application engineering and projects on instrumentation in power plants worldwide. He holds a bachelor’s degree in instrumentation engineering from Bombay Univ. and an MBA from Arizona State Univ. He is a senior member of ISA and ASME.
Achieving Success in Bottom Ash Spray Valve Control
The Rotork K-TORK vane type valve actuator has solved a difficult flow control application found in many coal-fired power plants – high-pressure bottom ash spray valve control.
High-pressure spray water is used to sluice bottom ash and pyrites from the boilers’ hopper bottoms and to carry the ash out of the plant. The valves used are typically ANSI Class 300 double-offset high-performance butterfly designs ranging in size from 3″ to 12″, automated with double-acting actuators. They cycle from four to 10 times per day and discharge to atmosphere, creating a very high pressure drop. The flow media is recirculated ash water that is abrasive and flows at pressures between 400 and 500psi.
|Rotork K-TORK actuators installed on two units at the AEP Pirkey Power Station in east Texas. All photos courtesy of Rotork|
In plants owned by AEP, Duke Energy, Luminant Generating and other utility companies around the world, K-TORK actuators have provided over 10 years of maintenance-free service while preserving the life of the valves and valve seats in these arduous duties. Forty actuators were installed in 2001 (20 per boiler unit) and have provided 12 years of operation with no downtime or maintenance required. All still have the original, durable lip seals
Among the challenges, it is imperative that the valves close fully and with zero leakage in a high pressure drop state. If the valve disc moves even slightly from the seat, the abrasive, high-pressure water will “wire-draw” or cut the butterfly valve seat. Traditionally, rack-and-pinion or scotch-yoke actuators have been used in this application, but “slop” or hysteresis in the rotary-to-linear conversion allows for the pressure in the pipe to move the disc from the seat, often causing premature failure of the valve after a period of only three to 12 months.
|The K-TORK double-opposed lip seal is forgiving to dirty or contaminated air.|
The problem becomes more acute when multiple valves are leaking, lowering the available back-pressure at the header, which makes it difficult or impossible to move the ash from the boiler.
When assembled to the valve with a ‘No-Play’ coupling, the K-TORK actuator has zero lost motion, “slop” or hysteresis. The one-piece vane and drive shaft cannot be back-driven and will hold the disc of the valve firmly in place.
Additional challenges include the location of the valves on a manifold at the bottom of the boiler where space is critical and plant air can be poor quality. K-TORK provides the smallest torque-to-size ratio and the double-opposed lip seal design is forgiving to dirty or contaminated air.
Also, the low-friction performance of K-TORK provides a speed-controlled, smooth valve operation, eliminating the risk of water hammer created by the high pressure drop.
Finally, longer run time between shutdowns demands increased reliability from the equipment in these critical applications. In particular, as the number of plant maintenance personnel has decreased, actuators that reduce maintenance (seal replacement) time and work orders have a direct payback to the owner, especially when valve life can be significantly increased through improved actuator performance.
Power Engineerng Issue Archives
View Power Generation Articles on PennEnergy.com