John A. Kleppe - email@example.com
Dept of Electrical Engineering
University of Nevada, Reno
Popular Version of Paper 4a EA5
Presented Thursday morning, May 16, 1996
ASA - 131st Meeting, Indianapolis
Embargoed until 16 May 1996
Temperatures in the combustion zone of a large furnace/boiler can be in the range of 2000-3000F. Conventional temperature measurement methods such as the use of thermocouples is not practical. It has been determined that accurate measurements of hot gases can now be readily made by using sound waves.
It is well known that the speed of sound is a strong function of the temperature of the medium through which the sound wave travels. In ranging systems, this variation in sound speed is treated as an error that requires an appropriate correction. In acoustic pyrometry, however, the changes in sound speed provide the desired measurement.
The measurement of the flight time of sound for distance calculations in meteorological, hydrological, and industrial applications has been known for some time. In most of these cases, air was the gas through which the acoustic pulse passed, and distance (depth) was the desired result of the measurement. The determination of temperature, on the other hand, requires measuring the flight time of an acoustic pulse over a known distance. This measurement therefore yields the average temperature of the entire acoustic path.
The principles involved in acoustic pyrometry are quite straightforward. In theory, the only requirements of an acoustic pyrometry based system are that a sound source (transmitter) be placed on one side of a furnace and a receiver or microphone be placed on the opposite side. The transmitter emits a pulse of sound and the receiver detects it. Because the distance is known and fixed, we can then simply compute the average temperature of the path traversed by the acoustic pulse.
This method, however, can prove challenging in practice. For example, it has been experimentally determined that the practical frequency range for an acoustic pyrometer (in large power plants) is between 500 Hz and 2000 Hz. Also, since the temperature of the gas involved can range up to 3000F, resulting in acoustic velocities >879 m/s and wavelengths on the order of 1 m, the sound pulse flight time must be resolved to a fraction of one wavelength in order to obtain practical temperature resolution and an acceptable system accuracy. Problems also arise because the acoustic path is usually disturbed by severe thermal and velocity gradients as well as by cavities between tube banks located at various points throughout a furnace or boiler.
The speed of sound is determined by measuring the flight time of an acoustic wave, then dividing it into the distance traveled. Once the speed of sound is known, the temperature may then be computed.
Acoustic pyrometry provides a practical approach for the on-line continuous measurement of gas temperature and velocity in hostile furnace and stack environments. The technique provides average line-of-sight measurements between the acoustic transmitter and the receiver. Multipath, side-to-side, front-to-back, and diagonal measurements within a furnace volume provide information on planewide average temperature. Deconvolution methods can produce isothermal maps at any given plane of the furnace interior.
Acoustic pyrometers are now being used to help identify and correct burner problems, slagging problems, and furnace overheating before these conditions can adversely affect operations. The devices have also begun to replace thermal probes for startup and are used for the routine control of soot-blowing. Other applications include the control of sorbent injection systems and heat rate determination.
Another important application for acoustic pyrometry has been its adaptation to measure the volumetric flow rate of gases in large ducts and stacks. Building on this technology, a system to measure volumetric flow rate has been developed that uses sound waves. the audio system pairs thermocouples and pitot tubes with acoustics in a combination designed to be accurate and reliable. Sound from the blast of air is picked up at both the transmitter and receiver probes. Special impulse-response, signal-processing methods then measure the acoustic flight time of the random noise pattern generated by the air blast.
Thermocouples are also imbedded in the end of each probe, which improve the reliability of the flight-time measurements and also provide the gas-temperature measurements. By using these thermocouple readings, the latter measurements are not a function of gas composition a major source of error for the ultrasonic arrangement.
Because the audio system is inside the stack, the probes can help support the pitot tubes. The air blast that creates the sound source also keeps the pitot tubes clean and free of fouling. Finally, because each probe is the same, four velocity points are known; that is, the velocity is known to be zero at each wall, and the two other points are measured with standard pitot tubes.
The audio system measures the path average velocity, which differs from the average volumetric flow except in the very special case where the velocity profile is constant. Applying the known velocity points and the path average velocity, the system then approximates actual flow distribution across the stack. Using this information, the system continually upgrades the flow distribution and calculates the actual volumetric flow in real time.
Note that the audio system does not use the flight time of the acoustic wave to measure stack temperature. This measurement is left to the thermocouples. The acoustic device accurately measures velocity, which is not a function of flue-gas composition. To use the acoustic system to measure temperature requires knowledge of flue-gas composition an can introduce errors, although acoustic flight time varies as a function of flue-gas composition.
Final output is a true measurement of the volumetric flow rate rather than only the path average value. The system is expandable to multiple paths should additional data be required. Its equipment is simple in structure and has a minimum of its electronics on the stack
Recent Related Publications
(1) Kleppe, J.A. (1995), "Acoustic Gas Flow Measurement in Large Ducts or Stacks", SENSORS, May, pp 18-24, 85-87.
(2) Kleppe, J.A. (1995), "Adapt Acoustic Pyrometer to Measure Flue-Gas Flow", Power, August, pp 46-47.
(3) Mann, A. and J.Kleppe (1996), "A Report on the Performance of a Hybrid Flow Monitor Used for CEMS and Heat Rate Applications", to be presented at 1996 EPRI Heat Rate Improvement Conference, Dallas, May.