Heat transfer to water in turbulent flow in internally heated annuli

Abstract

Nearly 900 values of local heat transfer coefficients were correlated for water flowing through long annuli l/S, 1/4, and 3/8 in. wide, electrically heated at their inner surfaces and containing three spacer ribs. Both cosine and uniform lengthwise heat-flux distributions were employed. All heat transfer coefiicients were computed for positions corre spanding to (LID.) ratios larger than 150. Sevsral methods of correlation were attempted and compared, especially with respect to the method of evaluating physical properties. The proportionality of the Colburn j factor to the P r i d t l and Reynolds numbers with their usual exponents was verified, and the dependence of J ' upon D z / D r was analyzed. There was no significant effect of cosine heat-flux distribution on the heat transfer coef3-cients. Evaluating physical properties at the usual film temperature gave the best correlation. A simplified dimensional equation for water at moderate temperatures and pressures was also developed. x Fig. 1. Schematic of typical test section. APPARATUSDeionized water was pumped downward through the test section, thcn through a coolcr and waa mturncd to the pump. All associated piping and fittings were either 304 or 316 stainlesa steel. The h t section, shown schematically in Figure 1, consisted of two coaxially placed aluminum tubes forming a conccntric annulus over 10 ft. long.The inner tube, which was-either 25 or 61s aluminum of approximately 1-in. 0. D., waa heated by the paasage of up to 20,OOO amp. of d.c. through it and was cooled by water flowing in the annular space. The power for heat p m a t i o n wag supplied by a 5,000-h.p. motor-generator Het. For essentially uniform heat generation, the hcater tube used had a wall thickness of 0.035 in., and for a lengthwise cosine distribution* of heat generation, the I.D. waa appropriately tapcrcd, having its thinneat wall at thc center. Thc tubes with tapered I.D. were extruded in two sections and then weldcd together a t their thin cnda. During opcration the heater txbe waa pressurized internally with nitrogen.The outer tubc had a relatively thick wall and wna manufactured from 61s aluminum. Threc different sizes of outer tubcs were used giving annulus widths of %, x, arid % in. Concentric placement of thc inner tiilic was obtained by the IIRC of three siliconeglaas laminate ccntcring ribs held snugly and acciirately in lengthwise broached slots in the wall of the oiiter tube. The ribs occupied about 5% of the crosssectional area to flow in thc unnulus q d , a8 ahon-n iri Figlire 2, werc of triangular crow section making nearly point contact with the hcater surface. In most tests the ribs used wcre continuous along the entire annulus length, creating three dist.inct subchartncls bct.ncen them. In later tests thc rib6 werc vcnted to allow for flow communication among subchannels. Othcr than the s m~l l effect on the r.irwmfwential heater-surface temperatrirc discuvscd tielow, vcntinK of the ribs \vm not important to the experimental rcsulta reporttd herc.Twelve coml)ination v...

Experimental study of combined free and forced laminar convection in an asymmetrically heated two-dimensional aiding flow

“…Some data in the laminar and transition region is also included to show the trend. Also shown on the graph is the forced convection correlation obtained by Stein and Begell [15].…”

Convective Heat Transfer from In-Tube Flow of Turbulent Supercritical Carbon Dioxide: Part 1—Numerical Analysis

“…The heat transfer coefficient of water in the annulus is determined using the correlation by Olson and Allen (1998) for Reynolds numbers of less than 4000 and the correlation by Stein and Bagel (1958) for Reynolds numbers of greater than 4000.…”

Choking two-phase flow literature summary and idealized design solutions for hydrogen, nitrogen, oxygen, and refrigerants 12 and 11

“…12 [50,71] N 2 [31,66,68] Refrig. 11 [50,71] O z [46,59] This assumption might introduce a considerable error in the case of hydrogen.…”