Furthermore, solid seeding particles, such as alumina oxide, have a tendency to clog burners, potentially modifying the flow.įigure 4. Even if alumina oxide were used instead of silicone oil, however, accurately resolving the velocity field in the high-temperature region of a counterflow flame is complicated by thermophoretic effects, which can cause the velocity of seeding particles to deviate from that of the surrounding flow field. Figure 4 illustrates how seeding density changes as the silicone droplets approach the flame. Thus, silicone oil droplets are unable to resolve the velocity field near the reaction zone of a counterflow flame, whereas alumina oxide particles would be able to resolve the velocity field in these high-temperature regions. To best guarantee the convergence of velocity values, a minimum of 180 image pairs are collected during each test.Īlthough the nebulizers provided an inexpensive, easy to control means of generating seeding particles, unlike alumina oxide (Al 2O 3), silicone oil has a boiling point (570 K) that is well below flame-relevant temperatures. However, if PIV vectors are generated using IAs of fixed size, then the differences in seeding density could be problematic. Although the varying flow rates through the two nebulizers inevitably result in differences in seeding density between the fuel and oxidizer streams, the discrepancy in seeding density is regarded as a non-issue because adaptive PIV accounted for local seeding density when assigning IA size. The flow rates through the fuel stream and oxidizer stream nebulizers are maintained at 2 and ≥3 L/min, respectively. During PIV tests, flow is directed through these nebulizers to atomize the silicone oil, resulting in micron-sized spherical droplets. Both the fuel and oxidizer streams are equipped with a nebulizer that is filled with silicone oil (50 cSt). Schematic of PIV experimental setup.įor the counterflow non-premixed flame demonstrated here, medical nebulizers (Sunrise Medical HHG Inc.) are a cost-effective means of generating seeding droplets. Figure 3 depicts a schematic of the PIV experimental setup.įigure 3. Scattered light is collected by an AF Micro-Nikkor 200 mm lens (f/16) and then imaged using a Dantec Dynamics FlowSense 4M MKII camera (12-bit CCD), which is equipped with a bandpass filter that had a central wavelength of 532 nm the QE of the camera is approximately 55% at 532 nm. Here, PIV measurements are performed using a Litron Nano (L200-15PIV) laser that outputted 532 nm beams at a repetition rate of 12 Hz and a pulse energy of 200 mJ. A convergence study is also performed to determine the number of image pairs required to achieve convergence of velocity values. The total uncertainty of PIV measurements on a vector-by-vector basis is then calculated as the root sum square of the FSA- and resolution-induced uncertainties. Linear error propagation analysis is used to quantify the effect of image resolution on velocity uncertainty. That is, the image resolution is determined to be 131☒ pixels/mm for all PIV measurements, and the uncertainty in image resolution propagated into the uncertainty of PIV measurements, as the image resolution is used to calibrate particle displacement readings (pixels per image pair). The uncertainty due to the full-scale accuracy of the PIV measurements is combined with that due to image resolution uncertainty. The resulting full-scale accuracy is found to be 0.625% of the maximum displacement. However, since the majority of IAs contained 64×64 pixels, this IA size is adopted for determining the maximum displacement. Note that since adaptive PIV methods are used here, not all IAs were of the same size. The maximum displacement was set at \( D=16 \) pixels, or ¼ of the height and width of an IA comprised of 64×64 pixels. The discriminative displacement is defined as pixels. The operating principle of PIV stems from the following simple relation: \( \overline \). PIV is a non-intrusive laser diagnostic capable of determining instantaneous two-dimensional velocity fields corresponding to the cross-sectional slices of three-dimensional flows.
0 Comments
Leave a Reply. |