Reconstruction of Temperature Distribution for a Turbulent Free Jet Using Background Oriented Schlieren

Abstract

Background Oriented Schlieren (BOS) has been shown to be an excellent tool for qualitative flow visualization, and more recently, literature has shown that the technique can be expanded to yield quantitative measurements as well. In this study, a BOS setup was built to construct the temperature distribution of a heated turbulent free 12mm diameter jet near the nozzle. A 1080p DSLR camera was used to view a black and white speckled background plane through the heated free jet in question. Comparing images of the background with and without flow present using a cross correlation algorithm gave the apparent displacement of all points on the background viewed through the flow. Once this displacement field was obtained, a ray-tracing algorithm was implemented to reconstruct the refractive index of the center plane of the jet. Then, the Gladstone-Dale and ideal gas relations were combined and used to calculate the temperature of the center plane. Reynolds number, based on the jet diameter, was held constant at 6,000 for all cases, and steady state nozzle temperature was varied from 57°C to 135°C. Reconstructed temperature distributions were validated using K-type thermocouple measurements by allowing the system to reach steady state before acquiring data. Average agreement of 4–6% was observed between thermocouple and BOS measurements for axial locations of at least 30 mm downstream. Due to experimental error, accuracy decreases as axial location moves towards the nozzle, and as nozzle temperature increases. Improvements to the setup are being considered to improve the agreement in low accuracy regions. Further, this technique has the potential to be used to determine the temperatures in open and optically accessible closed reactive flows. Having information about near wall temperature in closed reactive flows will give insight into wall convective heat transfer characterization and will also help benchmark combustion based numerical models in applications such as gas turbines.