| OCR Text |
Show a>Ol FUH STOICHIO",(l~Y . • FIG. 1 The variation of adiabatic flame temperature with stoichiometry for various fuel-air and fuel-oxygen mixtures. 1.00 g 0.10 ~ :J ~ Z 0.60 Q ~ ! II: - OtO + FLAME INfRAREO VS FUEUAIR + + \10<' '' '1010 ICOll "101 • CX),'.L"III 0.20 '--.....o---' __ ----L __ ---'"_~---L _ _'__ ___ "___J 0.60 0.80 1.00 1.20 STOICHlo ... nRY . • FIG . 2 The dependence of infrared emission from f ully premixed methane/air flame on stoichiometry . a manner similar to that in Figure 1. Figure 2 provides infrared emission data as a function of stoichiometry obtained on a fully premixed methane flame . The infrared emission was measured with a indium antimonide (InSb) detector which could sense both the carbon dioxide (C0 2 ) radiation at 4 .3 microns and water (H20) radiation at 2 .8 microns. The maximum emission is located at a stoichiometry of 0 . 95 which is in good agreement with the adiabatic flame calculations . This concept has been studied by other investigators (Smit h e t 200 al, 1975; Beer et al, 1982; and Fraim , 1985). Some of the data that are presented in this study are available in the work of Zabielski, et al (1985) and Zabielski (1986). The paths, however, taken by these investigators in implementing the physics of the process into a control concept vary considerably. APPROACH There are three key elements in the control scheme being reported here. The first element is the infrared detector. Although the preliminary spectroscopic work was performed using a cryogenically cooled, InSb detector which was responsive to both CO 2 and H20 emission, a lead selenide (PoSe) detector which responds only to the CO emission is economically and technicalty ideal. The second element is a fuel flow modulator. This element is necessary for the process used to discriminate the flame radiation from the background radiation emitted by the boiler/furnace walls which falls within the view of the detector. The modulator introduces a small variation in the fuel flow and, hence, in the F/A ratio. The radiation incident on the detector is a composite of the flame and background radiation. Since the modulation has a frequency of 20-55 Hz, it is relatively straightforward to separate the flame component from the background component with phase sensitive detection. Fourier analyses of infrared signature measurements from several classes of flames indicated that this frequency region is relatively clear of significant flame "flicker" noise. The third element is a microprocessor . This element is essential to the implementation of the method . In addition to the stability advantage of the microprocessor relative to an analog approach, the microprocessor allows control off the peak emission or temperature point which is important in certain processes. The details of each of these elements is provided below. CONTROL SYSTEM HARDWARE - The product strategy has been to develop a controller that will function as a trim device on an existing boiler/ furnace controller. Figure 3 gives a schematic representation of the control loop. The infrared emission within the field of view of the infrared optics is focused onto the detector . The signal generated by the detector is conditioned with a preamplifier and an active filter. The active filter is tuned to the frequency of the fuel flow |
