Measurements of an elliptic pipe jet

• Main Author: Vibhas Pahuja
• Other Authors: Chua Leok Poh
• Format: Final Year Project
• Language: English
• Published: 2014
• Subjects: DRNTU::Engineering::Mechanical engineering::Fluid mechanics
• Online Access: http://hdl.handle.net/10356/60270
• Institution: Nanyang Technological University

This study presents the findings of the mean velocity, Reynolds normal stresses and Reynolds shear stresses for a fully developed pipe jet emanating from a smooth, elliptic pipe of length 999 m and a nozzle exit of aspect ratio (AR) 2:1 (50 mm by 25 mm). The jet flow is fully developed and turbulent in nature. The lack of published studies on the topic of elliptic pipe jets and the effects of initial conditions on non-circular jets provide the main motivation for undertaking this study. The velocity and stress measurements were obtained using two kinds of probes:- single-wire and cross-wire, at stream-wise distances from the exit along both the major (y) and minor (z) axes. This stream-wise distance was normalized against the equivalent diameter of the elliptic nozzle, De which was 35.4 mm. The exit velocity of the jet used was 24.0 ± 0.5 ms-1, which gave a Reynolds No. of 54183.67. The first part of the experiment was carried out using a single-wire probe to find the distribution of the stream-wise velocity component (u) along the lateral (y) and span-wise (z) axes at various stream-wise distances (x). Both the mean Um and root mean square (rms) u’ were measured at local points. The different spreading rates along the y and z axes can be attributed to the fact that momentum thickness was found to be higher in the y axis when compared with the z axis. Also, the elliptic pipe jet displayed greater momentum thickness as compared to the contraction jet. The potential core was found to end by x=4De distance downstream. The self-similarity feature was observed from x=6De for the major axis (y) while the minor axis (z) recorded this feature from x=5De. The different jet spreading rates along the two axes also led to an axis-switching phenomenon at x=15.5De. The second part involved measuring two velocity components (u and v, u and w) along with the shear stress distributions at various stream-wise distances, again along the other two axes, using a cross-wire probe. 5 parameters, namely Um, u’, Vm (or Wm), v’ (or w’), (uv) ̅ and (uw) ̅ were measured in this part. Self-preservation of Um was achieved from x=6De in the major axis and from x=5De in the minor axis, thus mirroring the result of the single-wire probe experiment. However, the axis switching location occurred earlier in this case at a distance x=11.5De. It appears that the Reynolds normal stresses u’,v’ and w’ achieve self-similarity by x=20De, though more data at locations further downstream needs to be recorded to make a conclusive statement. For higher order turbulent magnitudes (Reynolds shear stresses), self-preservation was not reached within the maximum distance traversed x=38De. Also, the comparisons of the 2:1 AR elliptic pipe jet were made with the contoured contraction elliptic nozzle jet with AR 2:1 (Zhang & Chua, 2012) to bring out the effect of the initial conditions.