Journal Title

Frontiers in Heat and Mass Transfer

Publication Date

2017

Abstract

The onset of turbulence in natural convection systems is difficult to predict using traditional computational techniques. The flow patterns that occur before and after the onset of turbulence may be better understood with the help of visual techniques like Schlieren imaging. Schlieren imaging allows visualization of the density gradients of a fluid using collimated light and refractive properties. In this experiment, a device was designed to test the behavior of airflow with non-isothermal boundary conditions within a rectangular cavity. Previous computational fluid modeling suggested a period doubling route to chaos in a cavity with a high aspect ratio and free convection flow driven by two non-isothermal walls. To visualize this system, two strip heaters were attached to two 56 cm long plates. The plates were set to an aspect ratio of 10 and sealed with gaskets and garolite. With a Z-type Schlieren device, the existence of two definitive cells was confirmed, and the oscillation point between the two cells was found above the cavity midpoint. At steady state, a thin-wire thermocouple was placed at the inflection point to measure fluctuations in temperature. Consistent oscillations in temperature were observed that indicated a steady interacting boundary layer, while non-periodic oscillations indicated the initial transition to turbulence. It was observed that there was a transition from a mostly steady interacting boundary layer at set-point 35 deg.C to near-turbulence at 65 deg.C. At higher Rayleigh numbers the transition behavior predicted in the computational fluid model was confirmed, although the size of the convection cells was not uniform.

Author Supplied Keywords

Schlieren, Convection, Transition, Visualization

Subjects

Energy transfer; Heat--Convection, Natural

Publication Information

Archived version is the final published version.

DOI

10.5098/hmt.9.1

Peer-Reviewed

Yes

Document Type

Journal Article

Share

COinS