Numerical and experimental study of sub cooled boiling flows at low pressures

Dilla, E 2007, Numerical and experimental study of sub cooled boiling flows at low pressures, Doctor of Philosophy (PhD), Aerospace, Mechanical and Manufacturing Engineering, RMIT University.


Document type: Thesis
Collection: Theses

Attached Files
Name Description MIMEType Size
Dilla.pdf Thesis Click to show the corresponding preview/stream application/pdf;... 3.98MB
Title Numerical and experimental study of sub cooled boiling flows at low pressures
Author(s) Dilla, E
Year 2007
Abstract The present work builds on current numerical practices and techniques to simulate low pressure subcooled boiling flows. It attempts to improve on previous models for subcooled boiling flows on vertical flow channels through the introduction of new empirical correlations and obligatory code validation. It seeks to enhance the present understanding of low pressure subcooled boiling flows and the unique features associated with this phenomenon.

The proposed subcooled boiling model consisting of a new combination of empirical correlations and formulations were incorporated into the safety reactor code RELAP5/Mod3.2 and the commercial CFD code CFX. The modifications aim to enhance the performance of the reactor and CFD codes in characterizing the behaviour of subcooled boiling flows for low pressure applications. Among the key improvements introduced in the proposed model are new correlations for mean bubble diameter, bubble departure diameter, bubble departure frequency, active nucleation site density, interface heat transfer formulation, and a three-part wall heat flux partitioning model.

The modified codes were validated against three sets of axial void fraction experimental data. The reasonable prediction of void fractions provides a measure for the suitability of the subcooled boiling model. Computational predictions demonstrated that the modified codes showed a marked improvement over the default subcooled boiling models in both codes.

The simulation was extended to predict the occurrence of static flow instabilities in single and multi-channel systems. The study of the excursive-type flow instability, also known as the Ledinegg-type instability, is crucial for the safety analyses of nuclear reactors. The occurrence of significant amount of bubbles in the channel flow will cause changes in pressure drop, heat transfer performance, flow oscillation and even initiate flow redistribution in the case of parallel multi-channel systems. Discrepancies in the predictions of flow instability were observed for extreme subcooling cases using the modified RELAP5 code while the CFX simulation results generally mirror experimental trends.

The modified RELAP5 code satisfactorily simulated the occurrence of the Ledinegg-type of flow instability in vertical, multiple parallel channels prescribed with a transient (induced heat flux surge). Flow oscillation and re-distribution contributing to system instability is demonstrated. The resulting mass flow fluctuations in the channels can vary by about ±20% for the cases considered. Predictions showed that the occurrence of the flow instability can be triggered sooner in multi-channel systems.

A subcooled boiling rig was constructed and experiments were conducted on selected conditions. Though the quality of the data on void fractions gathered was not up to standards, the investigations was supplemented by observations of several interesting bubble behaviour through the analyses of consecutive image pairs taken from the subcooled boiling experiment. These observations can provide better insight into the salient features of low pressure subcooled boiling flow which can serve as inputs for the formulation of better mechanistic models for heat transfer and bubble evolution cycles for future studies.
Degree Doctor of Philosophy (PhD)
Institution RMIT University
School, Department or Centre Aerospace, Mechanical and Manufacturing Engineering
Keyword(s) Subcooled boiling
low pressures
modelling
RELAP5
Versions
Version Filter Type
Access Statistics: 143 Abstract Views, 347 File Downloads  -  Detailed Statistics
Created: Mon, 27 Jun 2016, 09:25:22 EST by Denise Paciocco
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us