OCR Text |
Show The emissions from flames generated by mechanical, motor-driven, precessing jet nozzles are investigated using the identical apparatus and experimental techniques that have been used previously to investigate conventional jet flames over a wide range of conditions (Turns & Myhr, 1991; Turns et al., 1993; Turns & Bandaru, 1992, 1993) and, thus, allow direct comparison of -precessing and conventional jet flames. The Reynolds number of both the motor-driven mechanical precessing and conventional non-precessing nozzles are based on exit diameter, mean exit velocity, and cold fuel properties. Apparatus and Techniques The flames were fired vertically under an exhaust extraction hood and were surrounded by wire mesh screens to minimize the influence of drafts in the laboratory. The flue gases were sampled from the well-mixed exhaust with compensation for dilution being based on the measured concentration of eCh. A detailed description of the sampling and gas analysis system is presented in Turns & Myhr (1991). Single-point total flame radiation was measured concurrently with the emissions data using a Medtherm 64P-5-24 heat flux transducer fitted with an IRtran IT window, which transmits radiation within the range 0.35-12 J.1.Il1. This transducer was coupled to a Medthenn H-201 digital heat flow meter with a sensitivity range of 0.1 - 10.10-6 W/m2. The efficacy of single point measurements in the determination of heat flux from jet flames has been recently examined by Sivathanu and Gore (1993). They note the importance of the parameter, RlLrv, where R is the radial distance of the probe from the axis of the flame and Ltv is the visible length of the flame, and collected the bulk of their data at RlLrv = 0.5. For the present experiments, it was not possible to conduct all measurements at their recommended location because, for some flames (notably those at low Stp), that position lay within the flame envelope. "Those flames were measured with the probe located at RlLrv = 1. The visible flame length, Lcv, and width, W Cv, were determined from a series of eight second time-exposure photographs. It was found that good repeatability of these values could be obtained for the wide range of flame shapes and colors encountered, since the position of the inside edge of the flickering flame approches a constant value as the camera aperture is varied parametrically. It should be noted that, while this technique is reasonable, it tends to overestimate the relative lengths of the low Strouhal number flames in comparison with the high Strouhal 5 |