Transient modelling of a loop thermosyphon : transient effects in single and two phase natural circulation thermosyphon loops suitable for the reactor cavity cooling of a pebble bed modular reactor

Ruppersberg, Johannes Coenraad (2008-03)

Thesis (MScIng)--University of Stellenbosch, 2008.

Thesis

ENGLISH ABSTRACT: The focus of this project was the application of a passive device in the form of a loop thermosyphon as a reactor cavity cooling system (RCCS) for a Pebble Bed Modular Reactor. An extensive literature review showed that loop thermosyphons have been widely researched, both theoretically and experimentally. In the review attention has specifically been given to matters such as safety, instability, control and mathematical modelling. One of the objectives of the project was to build one of the axially symmetric sections of Dobson’s (2006) proposed full scale RCCS using a scaled down version consisting of a single loop heated by a section of the reactor pressure vessel and cooled by a tank of water. The second objective was to derive a theoretical model that could be used in a computer code to simulate the experiment. The theory and experiment would then be compared in order to verify the code. The mathematical model created used the following three major assumptions: quasistatic flow, incompressible liquid and vapour and one dimensionality. The conservation equations in the form of a set of difference equations with the appropriate closure equations were then solved explicitly. It was found that the theoretical results were heavily influenced by the surface optical properties as well as the heat transfer coefficients. The emissivity influenced the transition point from single to two-phase flow as well as the condenser outlet temperature. The single phase heat transfer coefficients influenced the condenser outlet temperature significantly while it was found that for two phase flow the combination of the available boiling and condensation heat transfer coefficients had only minor effects on the end results. A stainless steel and aluminium thermosyphon loop was built using water as the working fluid. A stainless steel heater plate provided the heat input while a 200 L water tank was the heat sink. Temperature and flow rate measurements were recorded as a function of time with various heating/cooling transients from start-up to steady state for three operating modes. The three operating modes were single phase, two-phase and heat pipe mode. It was found that the theoretical temperatures correspond reasonably well with the experimental temperatures. The time predicted by the theoretical model to reach the operating temperature was however somewhat longer than for the experimental. This is to be expected when considering that there is some uncertainty pertaining to the heat transfer coefficients as well as surface emissive properties. The correspondence of the theoretical and experimental fin temperatures was poor due to significant thermal stratification of the air separating the heater plate and fins. Several shortcomings in the theoretical model as well as the experimental setup were identified and discussed. The conclusion was reached that this exploratory study showed that the loop thermosyphon is a viable option for the RCCS and that the mathematical model is a viable theoretical simulation tool. Several recommendations were made for further study to address and overcome the shortcomings identified in the theoretical and experimental models in order to prove this conclusion. Amongst these is the determination of better material surface properties and heat transfer coefficients and improved mass flow rate measurement. Investigating scaling issues, natural convection outside the loop and updating of the computer program is also recommended.

AFRIKAANSE OPSOMMING: Die fokus van hierdie projek was die toepassing van passiewe apparatuur, in die vorm van ‘n geslote lus termoheuwel, as ‘n reaktor kamer verkoellings stelsel vir die korrel bed modulêre reaktor. Die literatuur studie wys dat hierdie tegnologie reeds breedvoerig ondersoek is teoreties sowel as eksperimenteel. In die literatuur oorsig word aandag spesifiek gegee aan veiligheid, onstabiliteit, beheer en modelleering. Een van die doelwitte van die projek was om ‘n klein skaalse model te bou van een van die aksiaal simmetriese seksies van Dobson (2006) se voorgestelde volskaalse reaktor kamer verkoellings stelsel. Die model bestaan uit n enkele lus verhit deur ‘n seksie van die reaktor drukvat en verkoel deur ‘n tenk vol water. Die tweede doelwit was die afleiding van ‘n teoretiese model wat in ‘n rekenaar program gebruik kan word om die eksperiment te simuleer. Die teoretiese en eksperimentele data kan dan vergelyk word om die geldigheid van die program te toets. Die volgende aanames is gemaak tydens die afleiding van die wiskundige model: kwasi-statiese vloei, onsamedrukbare vloeistof en gas en een dimensionalitiet. Die behouds wette is in die vorm van ‘n stel differensie vergelykings met die toepasbare sluitings vergelykings eksplisiet opgelos. Dit is bevind dat die teoretiese resultate swaar beinvloed is deur die materiaal oppervlak eienskappe sowel as die warmteoordrag koëffisiënte. Die emisiviteit beinvloed die oorgangs punt van enkel na twee fase vloei sowel as die kondenser uitlaat temperatuur. Die enkel fase warmteoordrag koëffisiënt het n beduidende invloed op die kondenser uitlaat temperatuur terwyl dit voorkom asof die spesifieke kombinasie van die koking en kondensasie warmteoordrag koëffisiënte minimale invloed op die resultate het in die twee fase gebied. Vlekvrye staal en aluminium is gebruik om die lus te bou met water as die verkoelings middel. Warmte is toegevoeg tot die stelsel deur ’n vlekvrye staal verhittings plaat terwyl ‘n 200 L water tenk die warmte onttrek het. Temperatuur en massa vloei tempo is aangeteken as ‘n funksie van tyd vir verskeie verhitting/verkoellings oorgangs gedragte vanaf begin tot bestendige toestand vir drie bedryfs modusse. Die drie bedryfs modusse was enkel fase, twee fase en hitte pyp modus. Dit is bevind dat die teoretiese temperature redelik goed ooreengekom het met die eksperimentele waardes. Die tyd wat dit neem om by die bedryfs temperatuur te kom soos voorspel deur die teorie is egter langer as wat in die eksperiment gevind is. Dit is te verstane wanneer die onsekerheid in die warmteoordrag koëffisiënte en materiaal oppervlak eienskappe in ag geneem word. Die fin temperature het ‘n swakker ooreenkoms getoon as gevolg van beduidende termiese stratifikasie van die lug tussen die fin en verhittings plaat. Verskeie tekortkominge in die teoretiese model en eksperimentele opstelling is geïdentifiseer en bespreek. Die gevolgtrekking is gemaak dat die ondersoek bewys dat geslote lus termoheuwels ‘n lewensvatbare opsie is vir ‘n reaktor kamer verkoellings stelsel en dat die wiskundige model lewensvatbaar is vir teoretiese simulasie. Verskeie aanbevelings word egter gemaak om die tekortkominge in die teoretiese en eksperimentele modelle aan te spreek om so doende die gevolgtrekking te staaf. Dit word aanbeveel dat beter waardes vir die materiaal oppervlak eienskappe en warmteoordrag koëffisiënte gevind word en verbeterde massa vloei meetings gedoen word. Dit word verder aanbeveel om skaleering asook natuurlike konveksie buite die lus te ondersoek en om die rekenaar program by te werk.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/20858
This item appears in the following collections: