The Effect of Model Parameters on CFD Simulation of a Thermosyphon

Author(s):  
Huiyu Wang ◽  
D. Keith Walters ◽  
Keisha B. Walters

Abstract Both numerical and experimental studies have previously been carried out to investigate the heat transfer performance of the two-phase closed thermosyphon (TPCT). This paper investigates the performance of a commercially available computational fluid dynamics (CFD) solver (Ansys FLUENT) to predict the complex flow behavior of TPCTs, with special focus on modeling of the mass transfer phase change process. The present study uses four different sets of mass transfer coefficients for condensation and evaporation within a previously documented phase change model to determine their impact on the simulation results. The mass transfer coefficients effectively control the rate of transfer from liquid to vapor phase during evaporation and vice versa during condensation. The choice of coefficients is assumed to represent a balance between numerical accuracy and stability. A baseline simulation is performed for which both the evaporation and condensation coefficients are equal and set to default values. Three additional simulations vary the magnitude of the coefficients and adopt relative values based on density ratio following a recommended method that has been previously found to be effective for these simulations. Initial results show that the case with the highest coefficient of evaporation and coefficient for condensation based on the density ratio is in good agreement with available experimental data of overall thermal resistance of the TPCT., with predictive capability degrading as the values of the coefficients are reduced. Additionally, the 3D CFD models implemented in this study appear to successfully predict the phase change process and vital flow behavior inside the TPCTs, at least in a qualitative sense.

Author(s):  
Masanori Naitoh ◽  
Shunsuke Uchida ◽  
Hidetoshi Okada ◽  
Seiichi Koshizuka

The code system DRAWTHREE-FAC for evaluation of pipe wall thinning due to flow accelerated corrosion was validated by comparison of calculations with measurements at the secondary piping of a PWR plant. Distributions of flow velocity and temperature along the whole piping were calculated with the system code RELAP5 and corrosive conditions were calculated by a N2H4-O2 reaction analysis code. Precise flow turbulence at major parts of the piping was analyzed with a 3D computational fluid dynamics (CFD) code to obtain mass transfer coefficients at structure surfaces. In the CFD calculation, the κ-ε method was applied. Since the κ-ε method can not give detailed flow behavior in a boundary layer, the results were extrapolated with a wall function, a power law, and analogy of non-dimensional numbers to obtain mass transfer coefficients in the boundary layer. Then, wall thinning rates were calculated by coupling models of static electrochemical and dynamic oxide layer growth. The wall thinning calculation was focused on T-junction portions of a PWR feed water line. The wall thickness of the PWR secondary piping was measured by the ultrasonic testing. The calculated residual wall thicknesses after thinning agreed with the measurements within ±20% difference.


2002 ◽  
Vol 124 (3) ◽  
pp. 530-537 ◽  
Author(s):  
Kamel Ghali ◽  
Nesreen Ghaddar ◽  
Byron Jones

The air penetration within a porous clothing system on a moving human being is an important physical process that considerably affects the heat and moisture resistance of the textile material. This effect of the coupled convection heat and mass exchange within the clothing system is experimentally investigated and theoretically modeled to determine the heat and mass transfer coefficients between the air penetrating the void space and the solid fiber as a function of the velocity of penetrating air. Experiments were conducted inside environmentally controlled chambers to measure the transient moisture uptake of untreated cotton fabric samples as well as the outer fabric temperature using an infrared pyrometer. The moisture uptake was conducted at three different volumetric flow rates of 0.0067, 0.018 and 0.045 m3/sec/m2 of fabric area to represent airflow penetrations that could result from slow, medium, and vigorous walking, respectively. The theoretical analysis is based on a two-node adsorption model of the fibrous medium. A set of four coupled differential equations were derived describing time-dependent convective heat and mass transfer between the penetrating air and the solid fiber in terms of relevant unknown transport coefficients. The unknown model parameters were adjusted to fit the experimental data. The outer heat and mass transfer coefficients were found to increase with the air penetration flow rate.


1998 ◽  
Vol 38 (6) ◽  
pp. 287-294 ◽  
Author(s):  
Pen-C. Chiang ◽  
Chung-H. Hung ◽  
J. C. Mar ◽  
E. E. Chang

Both Henry's constants and volumetric mass transfer coefficients (KLa) of eight priority chlorinated organic compounds including 1,1-dichloroethene, methylene chloride, chloroform, carbon tetrachloride, 1,1,1-trichloroethane, trichloroethylene, tetrachloroethylene, and 1,4-dichlorobenzene in an air stripping packed column were investigated in this study. The liquid and gas phase EPICS (Equilibrium Partition in Closed System) and direct calculating methods were applied to determine the Henry's constants of VOCs. The interference of co-solute on Henry's constants was also investigated. Experimental results indicated that decrease in Henry's constants of VOCs was observed in the presence of humic acid but no apparent effect on Henry's constants was detected when there was NaCl and surfactant in solution. Four different configurations of packing media including Intalox Saddle, Super Intalox Saddle, Telleret, and Hedgehog made of polypropylene were respectively packed in the air stripping tower and investigated in the study. The dependence of hydraulic loading, air-water ratio, and configurations of packing media on mass transfer coefficients of VOCs was discussed.


1996 ◽  
Vol 34 (3-4) ◽  
pp. 557-564 ◽  
Author(s):  
Bettina S. Haas ◽  
Reimer Herrmann

Sewage containing volatile contaminants is a potential VOC-source in cities. Thus we tried to evaluate volatilization out of the sewerage system by measurements of contaminants in sewer gas and sewage. Our results from a medium sized town with little industry showed that sewer gas is mainly contaminated with alkanes, small aromatic compounds and chlorinated hydrocarbons. For three chlorinated hydrocarbons (chloroform, trichloroethene, tetrachloroethene) we determined mass transfer coefficients out of sewage and used these data to estimate mass fluxes from sewage and emissions out of the sewerage system for two sewer stretches. Considerable emission of chlorinated hydrocarbons from sewage, i.e. fluxes of some 10 to 100 g per m2·d, occurred only when the contaminant input via sewage was between some g and mg per litre for a single compound. For concentrations that were about 3 orders of magnitude less, emissions were negligible.


1993 ◽  
Vol 58 (5) ◽  
pp. 1078-1086
Author(s):  
Zdeněk Palatý

The paper deals with the mass transfer in a liquid on a plate with mobile packing. A procedure has been suggested which enables estimation of the mass transfer coefficients from experimental data considering the dispersion flow of the liquid. The results obtained from the desorption of CO2 from water are presented graphically and in the form of empirical equation.


1989 ◽  
Vol 54 (4) ◽  
pp. 967-980 ◽  
Author(s):  
Ondřej Wein ◽  
Petr Kučera

Extended Leveque problem is studied for linear velocity profiles, vx(z) = u + qz. The existing analytic solution is reconsidered and shown to be inapplicable for the accurate calculation of mean mass-transfer coefficients. A numerical solution is reported and its accuracy is checked in detail. Simple but fairly accurate empirical formulas are suggested for the calculating of local and mean mass-transfer coefficients.


Sign in / Sign up

Export Citation Format

Share Document