Comment

1 June 2005

Exit-gas temperatures sizzle

Control over temperatures can reduce NOx emissions and control ash deposition.

By Donna Dearmon, Ben Zimmerman, Simon Youssef, and Rabon Johnson

The Cumberland Fossil Plant fitted three furnace exit-gas temperature devices on the 10th elevation front wall of their number one unit.

This continuous online measurement system has the operators monitoring furnace trends and reacting to undesirable conditions. The devices came with an automatic retractable system in the event that cooling air is lost to ensure that hot furnace temperatures do not destroy the hardware. This project demonstrated a newly developed automatic port rodding system that works with other instrumentation applied to the hot side of furnaces.

Ash deposition predictable

Tennessee Valley Authority's (TVA) Cumberland Fossil Plant (CUF) has a history of slagging due to high heat release design and low ash fusion temperature fuels. Retrofitting low NO_x burners aggravated this problem. In the past few years, damaging outages have occurred due to this slag accumulation.

With the installation of furnace exit-gas temperature (FEGT) devices to monitor the exit-gas temperature on the 10th elevation on Unit 1, the project is a test to see if this instrumentation can eventually work on CUF's large furnaces.

Measuring FEGT has become an essential element of boiler control systems since it is an important indicator of the quality of heat transfer within the boiler's furnace section. Each boiler has an optimum heat-transfer distribution that corresponds to a specific design temperature. Unfortunately, boiler combustion profiles change continuously during operation due to variations in the coal quality, boiler loading, ash deposition, and other process parameters. The continuous online measurement of FEGT enables the operators to monitor furnace trends and react to undesirable conditions. For example, ash accumulation on the lower furnace walls can be less by initiating cleaning operations only when needed. This helps to avoid unnecessary tube erosion resulting from excessive cleaning operations.

In utility boilers, temperature measurement is valuable within both the furnace and post-furnace regions. Control over furnace fireside temperatures can yield beneficial results, such as the reduction of NO_x emissions, the control of ash deposition, improved overall process efficiency, and the prevention of damage to heat transfer surfaces. In the post-furnace sections of the boiler, temperature measurement is useful for the control of ash deposition and cleaning. It also influences the new post-combustion systems designed to reduce NO_x and SO2 by providing real-time temperature information for process optimization. FEGT indicates the temperature of the gas stream as it exits the furnace and enters the first pendant sections, which is an indication of the character of the ash as it passes into the convection platens. With ash fluid temperatures in the range of the furnace gas temperature, some coals have a strong tendency for slag accumulation at the inlet pendants whenever temperature exceeds threshold. Therefore, it is essential to limit high temperature excursions and maintain a stabilized and balanced flow of furnace gas passing through the convection sections of a boiler. This way, ash deposition occurs in predictable locations that are manageable with proper cleaning operations.

Non-contact technologies

Optical pyrometers consist of a sight tube that collects radiation emitted and projects that radiation onto a fiber optic bundle. The optical fibers bring the radiation to photo-detectors that have interference filters to limit the detected radiation to specific narrow bands. The detectors generate signals in proportion to the detected radiation, which after amplification and digitizing transmit to a microprocessor. The microprocessor calculates the temperature of the ash cloud.

An infrared thermometer can measure gas temperature if its operating waveband is coincident with an absorption band in the target gas. If the gas path is optically thick (i.e. the thermometer does not see through to a back wall) and of uniform temperature, then the thermometer will read the gas temperature directly. This condition depends on absorption strength, path length, gas concentration, temperature, and pressure. There are concerns on penetration depth of the technologies.

Infrared CO₂ waveband pyrometer: Probably the most well known type of radiation pyrometer is the infrared pyrometer that measures the intensity of radiation emitted from a surface or region with wavelengths in the infrared range. If the object is an opaque, black (totally absorbing) solid surface, then Planck's blackbody radiation law shows that at a given wavelength, the intensity of the emitted radiation varies with the temperature of the surface. Therefore, a measurement of radiant intensity can infer the temperature. If the absorptivity (equal to the emissivity) is less than unity, then the surface emits radiation with intensity less than that of a black body. In this case, it is still possible to infer temperature accurately, if we know the emissivity. Since site-specific conditions usually seem to vary, most pyrometers incorporate an emissivity adjustment, which allows this correction by the user.

When dealing with non-opaque materials, such as furnace gases, measuring temperature with an infrared pyrometer becomes far more complicated. In this case, the emissivity depends not only on the properties of the radiating material, but also on the distance along the pyrometer's line of sight. As a result, the amount of radiation detected by the infrared detector has a spatial dependency. Gases radiate only at discrete wavelengths that are characteristic of their molecular composition and concentration. These emissions occur within specific absorption/radiation bands whose emission/absorption strengths are dependent upon temperature and pressure. These factors give rise to different emissivity measurements and hence different penetration depths. In a coal-fired furnace, for example, one primary gas constituent for measurement purposes is CO2, which is known to emit primarily at three different wavebands near 4.3, 2.7, and 1.9 ìm, which have different penetration distances, ranging from a few inches to well over a hundred feet. H2O works in this fashion too.

Ash-particle-emission spectroscopy: A class of radiation pyrometers using optical emission at visible wavelengths applied to measure the temperature emitted from ash particles suspended in a heated gas stream tested. This technique enables the device to avoid errors introduced by the uncertainty with molecular gas emissivity. In addition, it allows the use of an optical system that measures very short wavelengths to help reduce the spatial problems associated with other types of radiation pyrometers. It is dependent upon the ash particles being in thermal equilibrium with the surrounding gas and in sufficient concentration to permit detection.

Acoustic pyrometry: Velocity of sound is a strong function of the temperature within the medium through which the sound wave travels as indicated by the Universal Gas Law; therefore changes in sound speed can provide a direct measurement of the medium's temperature. Acoustic pyrometers measure the flight time of sound over a known distance, the furnace path. The result is the average temperature of the entire acoustic path. A sound source (transmitter) sits on one side of the furnace, and a receiver mounts on the opposite side. When measuring hot gases in large utility furnaces, robust signal detection techniques must happen in order to overcome the noisy furnace environment including soot blower operation.

The two techniques for acoustic pyrometry differ mainly in the sound wave source production and analysis. One technique uses a sound generator and produces a high-energy sound wave (over 170 dB) using a pneumatic device. One shortcoming of this technique is the requirement of a 300-psi air source such as an air multiplier or plant air if available at that high of a pressure. This sound wave produced by the pyrometer has a sharp leading edge that propagates concentrically from the generator. This sound source enables the measurement of gas temperatures in furnaces as wide as 100 feet as well as within soot blower lines.

The other accepted technique works using a much lower intensity sound source (126 dB). Difficulties overcoming furnace noise are counteracted using specially developed digital signal processing methods and by operating the system in coordination with soot blower control logic.

Install optical locations

Slag deposition on the upper furnace sections, particularly the pendants, at CUF's two 1300MW opposed-wall fired units burning Illinois Basin coal has been a recurring problem. Both units rate at 10 million lb/hr (1260 kg/s) main steam flow at 3500 psig (25.2MPa) and 1003oF (539oC) turbine throttle conditions. The boiler dimensions are 110 feet (33.8m) wide by 51 feet (15.5m) deep and are configured with 88 burners feed by eleven MPS 89 mills. The furnaces originally came with cell burners and later retrofitted with Foster Wheeler CF/CS low NO_x burners in the late 1990's. The combination of a high heat release ratio and the medium-to-high sulfur and iron content of the fuel result in the troublesome combination of high furnace exit-gas temperatures and low ash fusion temperatures. This, in turn, results in ash deposits forming on the first convection pendants.

TVA and EPRI (Electric Power Research Institute) co-funded the demonstration of FEGT monitors on CUF Unit 1 Furnace. Both furnaces were already equipped with boiler optical cameras installed on the ninth elevation front wall to allow operators a view of the pendant sections above the nose of the furnace. These equally spaced cameras provide a view of the north, mid, and south sections of the furnace. The goal of adding FEGT was to monitor each section temperature at the 10th elevation above each camera to provide the operators not only a visible view of each pendant section but also the gas temperature of each section. Combined with incoming coal data, the FEGT data could provide information on slag conditions in each section and correlations made with the visible images and the coal analysis data.

Consideration of several techniques for FEGT measurement including acoustic, optical, and high temperature thermocouples took place. High temperature thermocouples are maintenance intensive and are single point measurement devices. The width of the boiler (110 feet, 33.8m) and the presence of wing walls across the boiler made acoustics impractical. As shown in the furnace wing-wall configuration, a single north to south path was not possible due to the wing walls spanning the 110 feet width of the furnace. The wing walls also hinder a multipoint technique with receivers and generators. Three receiver/generator configurations would be necessary increasing the cost of acoustics. The acoustic vendors were not comfortable sending the sound signal between the wing walls.

After studying optical methodology, the organization chose Diamond Power GasTemp optical FEGT monitors and installed them at three locations across the boiler front. Most optical FEGT monitors are equipped with a 6-degree field of view that would require a 10-feet width as the path reached the 50-feet path length. The distance between wing walls range from 6 feet, 6 inches in the mid section to 8 feet in the north and south sections. To reduce the chance of wing wall interference, the vendor supplied GasTemp monitors with a 3-degree field of view, which reduced the width to around 5 feet at the 50-feet path length. This configuration provides a weighted average temperature at three front-to-back paths. The company chose optical technology because it is non-intrusive and required minimal maintenance. In addition to the monitors, automated retracting mechanisms went in on all three GasTemp devices in case of cooling air loss. Also as part of this project, one automated port-cleaning device was installed for demonstration purposes.

Furnace exit-gas temperature monitor location

Comparing by increments

TVA's Testing Services group verified the FEGT monitors with actual furnace conditions. A high-velocity-thermocouple probe (HVT) provided gas temperature readings from the ninth elevation below the existing camera's for comparison to the FEGT monitors. The HVT probe obtained temperatures at 2 feet increments up to 20 feet into the depth of the furnace from each corresponding section: north, mid, and south. The group took data at two different load settings: 1040MW low load and 1300MW base load.

The FEGT monitors followed the same trend as the HVT data. The team expected the monitor temperatures to be lower than the HVT temperatures since the HVT data registered at a lower elevation closer to the burners. During this testing, these trends identified burner imbalances in the Unit 1 furnace that the plant has since corrected by burner damper adjustments.

In a unit prone to slagging in the furnace, ports added to the furnace section are likely to accumulate slag and close. Prior experience with the boiler cameras in use at CUF indicated that keeping the ports clear of ash and slag would be an issue. Current operation of the boiler cameras included manual rodding of the ports daily. With the cameras, the operator could see when the port became blocked. The FEGT monitors on the other hand would only loose signal making it difficult for operators to know when the data was true.

Applied Synergistics developed an automatic port rodding system for this application. The port rodding device consisted of a pneumatically operated tube that would push through the port, clearing it of any ash accumulation. The device installed on the middle DP unit. The team added in a separate controller, which allows the user to set the interval of cleaning.

The DP furnace exit-gas temperature monitors at the TVA's CUF have been functioning well for two years. HVT probe data verified the measurements from the monitors with actual furnace conditions. During this testing, the FEGT monitors identified side-to-side spatial differences in the furnace exit temperatures, and the monitors have been helpful in trimming burner secondary airflows. The sensors have also confirmed marginally high exit gas temperatures across the furnace and have indicated slag build-up in the lower furnace section. The installed port rodding system is functioning smoothly, and the installation and integration was straightforward.

The data from the FEGT monitors will help in the development of an intelligent slag mitigation system for TVA's CUF. CP

Behind the byline

Donna Dearmon (dldearmon@tva.gov) is a project engineer at the EPRI I&C Center in Tennessee. Ben Zimmerman is a system engineer at TVA's CUF, also in Tennessee. Simon Youssef is an application engineer at Diamond Power International in Ohio. Rabon Johnson (rabonjohnson@earthlink.net) is project manager at the EPRI I&C Center.