Answers are grounded in this paper and cite their sources.
A numerical benchmark study dealing with 3-D natural convection in an air-filled cavity oriented at three different angles was held at the International Center for Heat and Mass Transfer (ICHMT) 2nd International Symposium on Advances in Computational Heat Transfer (CHT'01), May 20 - 25, 2001, in Palm Cove, Australia. The benchmark problem derives from an experimental study conducted by Leong, Hollands, and Brung involving natural convection within a differentially heated cubical cavity oriented at three different angles (φ) and subjected to four different Rayleigh numbers ( Ra ), all in the range 10 5 h Ra h 10 8 . The cube sidewall temperature varied linearly with distance from the hot face. In view of the number of test cases reported and the relatively low experimental uncertainty (about 1% of measurement), the problem appeared to be well suited as a benchmark for numerical validation. Ten teams (sets of participants) submitted papers using various mesh sizes and computational techniques. Most teams reported the existence of oscillatory solutions for cases at high Ra where the heating was primarily from below. Computed Nusselt numbers agreed with measurements at Ra and at inclinations where the heating was at least partly from the side. However, no one set of participants produced Nusselt numbers that agreed with the experimental results for all test cases.
No takes yet. Share an insight, caveat, or question.
Pepper et al. (2002) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: