Experimental Investigation of Natural Convection Heat Transfer in Heated Vertical Tubes with Discrete Rings

• Published in: Experimental techniques (Westport, Conn.) Conn.), 2017-12, Vol.41 (6), p.585-603
• Main Authors: Nayak, R. C., Roul, M. K., Sarangi, S. K.
• Format: Article
• Language: English
• Subjects: Characterization and Evaluation of Materials; Chemistry and Materials Science; CORRELATIONS; ENGINEERING; ERRORS; HEAT FLUX; HEATING; LENGTH; Materials Science; NATURAL CONVECTION; NUSSELT NUMBER; PIPES; RAYLEIGH NUMBER; REYNOLDS NUMBER; RINGS; STEADY-STATE CONDITIONS; SURFACES; THICKNESS
• ISSN: 0732-8818; 1747-1567
• DOI: 10.1007/s40799-017-0201-6

Natural convection heat transfer from heated vertical pipes has been investigated experimentally. The pipes are open-ended and circular in cross section. The test sections are electrically heated imposing the circumferentially and axially constant wall heat flux. The outer surface of the test section is insulated and heat is allowed to dissipate from the internal surface. Experiments are carried out for four different channels of 45 mm internal diameter and 3.8 mm thickness with length varying from 450 mm to 850 mm. Ratios of length to diameter of the channel are taken as L/D = 10, 12.22, 15.56, and 18.89. Wall heat fluxes are maintained at q // = 250 to 3341 W/m 2 . Experiments are also carried out on intensified tubes of the same geometrical sizes with the discrete rings of rectangular cross section provided on the internal surface. Thickness of the rings are taken as t  = 4, 6 and 8 mm; step size (s) of the rings are varying from 75 mm to 283.3 mm and the other ratios are taken as s/D = 1.67 to 6.29, t/D = 0.088 to 0.178 and s / t  = 9.38 to 70.82. The effects of L/D ratio, ring thickness, ring spacing and wall heating conditions on local steady-state heat transfer phenomena are studied. A correlation is developed for average Nusselt number and modified Rayleigh number and another correlation is also developed for modified Rayleigh number and modified Reynolds number. These correlations predict the data accurately within ±10% error.