By Steve Blankinship
Advances in manufacturing and micro-chip technology have led to surging interest in the field of “smart sensors,” broadly referring to low-cost sensor chips with embedded micro-control units. Smart sensors can monitor a wide range of parameters, such as voltage, radiation, temperature, and humidity, and then process this information within the sensor itself, identifying threshold limits, processing data, and activating alarms. The potential for these devices currently exceeds their application by a wide margin, but advances are quickly closing the gap.
Many companies are banking on what smart sensors can do. GE Power Systems believes smart sensors will occupy a large and growing place in its service business, contributing “double-digit-plus growth” over the next few years, according to Del Williamson, President of Global Sales for GE Power Systems. GE is focusing its efforts on sophisticated sensing and optimization technologies that combine intelligent sensing systems and intelligent software. The synergistic result is a tool that allows users to shift from statistical-based or reactive responses to proactive responses based on real-time feedback from the asset being measured, according to Jeff Schnitzer, General Manager of GE Reuter-Stokes.
GE’s MK Combustion Optimization System includes three series of “smart sensor” systems installed in various boiler locations that gather data on the generation of CO and NOx emissions and loss on ignition (LOI). The data gathered by the sensors is analyzed by a software system and compiled into optimized setpoints needed by the controller to make real-time adjustments to the boiler that will maintain emissions limits within a given parameter. “When each boiler’s performance is monitored in conjunction with others in the plant,” said Schnitzer, “it allows operations personnel to determine which boiler should be used during what conditions, and which fuel to use for meeting permit requirements and minimizing operational costs.”
These smart InfoSense sensors from Watlow, which have characterization information encoded for each unique sensor, can cut the ‘limits of error’ by at least half.
Facing practical limitations in process and material ideals, engineers and scientists have begun to realize that advances in sensing technology will depend more on the “intelligence” of the sensor than on the sensor itself. For example, researchers for decades have tried to improve the accuracy of thermocouple sensors by focusing on the materials, but with little success. The availability of thermocouple metals has been dying out, leaving less flexibility in material matching for high accuracy applications. Watlow has recently taken a different approach, applying information technology to the problem; its INFOSENSE sensor technology encodes the characterization information for each unique sensor, regardless of the material, to cut the “limits of error” by at least half. In addition to using production test information to verify whether a thermocouple match or RTD element meets a certain tolerance, that information is encoded on a label so that a compatible controller learns about the sensor and compensates accordingly. Since most sensor error is nonlinear, the accuracy of an INFOSENSE sensor is improved across the entire operating temperature range.
“The most significant benefit of INFOSENSE, however, is the potential improvement in fuel economy on engines and turbines that are more tightly controlled as well as the resultant reduction in pollution emissions,” says Dave Culbertson, Product Development Manager.
INFOSENSE sensor characteristics are imprinted on a numbered and bar-coded tag attached to each sensor. Users simply enter each value into a compatible controller’s standard menu. This immediately improves sensor accuracy by a factor of two. For example, the accuracy of a Watlow Type K ASTM E230 “special limits of error” EGT thermocouple, with accuracy ratings of ±1.1 C or 0.4 percent, improves to one-half special limits, ±0.55 C or 0.2 percent. At a sensing temperature of 800 C, measurement accuracy improves to ±1.6 C.
Simple instruments are also getting smart, and the devices provide not just enhanced monitoring and control, but also savings in installation and ease of use. Honeywell’s DirectLine sensors, including pH, ORP, conductivity and dissolved oxygen probes, contain an electronics module integral to the sensor. The module eliminates the additional time and expense of installing an analyzer or transmitter, a separate preamp and special cable, thus providing savings during installation, start-up, operation and maintenance. The output of the sensors connects directly to any host monitor or control device that accepts standard 4-20 mA inputs and provides external loop power.
An evolving “smart” technology – but not yet ready for primetime – is the use of specially formulated thermal barrier coatings that can act as temperature and erosion-sensing devices in advanced gas turbines. Higher combustion and turbine inlet temperatures enable higher efficiencies to be achieved, but the higher temperatures place greater stress on hot gas path components and cooling schemes.
Because turbine users have no way to assess component wear in real time, maintenance is typically scheduled conservatively, resulting in lost revenue opportunities and higher lifetime maintenance costs. Surface temperature measurements on gas turbine components can be made with various techniques, such as thermocouples and pyrometry, but these suffer from various shortcomings, including high cost and intrusiveness for thermocouples, and contamination and stray light effects for pyrometry.
Smart TBC coatings – under development by London-based Southside Thermal Sciences Ltd. and others – use a standard TBC embedded with phosphors that act as the sensor coating. Under UV light, the coating sensor phosphoresces, resulting in luminescence decay patterns that provide information on surface temperature, erosion, and degradation.
Although significant development and field testing remains to be done, the sensor coating has demonstrated the ability to measure temperatures from room temperature up to at least 1250 F with an uncertainty of about 14 F (at 400 F). According to Udo Dengel with Southside Thermal Sciences, improved real-time temperature measurement via sensor coatings in the gas turbine hot gas path could increase efficiency by 1 percent. Erosion and degradation measurements should lead to lower maintenance costs and enable operators to use a predictive maintenance schedule.