148 results found
Walker S, Dasgupta A, 2019, Phenomenological modelling of film-dryout 'critical heat flux', NUCLEAR ENGINEERING AND DESIGN, Vol: 354, ISSN: 0029-5493
Bluck M, Pioro I, Walker S, 2019, In Memorium: Geoff Hewitt, JOURNAL OF NUCLEAR ENGINEERING AND RADIATION SCIENCE, Vol: 5, ISSN: 2332-8983
Giustini G, Walker SP, Sato Y, et al., 2019, CFD analysis of the transient cooling of the boiling surface at bubble departure, Journal of Heat Transfer: Transactions of the ASME, Vol: 139, ISSN: 0022-1481
Component-scale computational fluid dynamics (CFD) modeling of boiling via heat flux partitioning relies upon empirical and semimechanistic representations of the modes of heat transfer believed to be important. One such mode, “quenching,” refers to the bringing of cool water to the vicinity of the heated wall to refill the volume occupied by a departing vapor bubble. This is modeled in classical heat flux partitioning approaches using a semimechanistic treatment based on idealized transient heat conduction into liquid from a perfectly conducting substrate. In this paper, we apply a modern interface tracking CFD approach to simulate steam bubble growth and departure, in an attempt to assess mechanistically (within the limitations of the CFD model) the single-phase heat transfer associated with bubble departure. This is in the spirit of one of the main motivations for such mechanistic modeling, the development of insight, and the provision of quantification, to improve the necessarily more empirical component scale modeling. The computations indicate that the long-standing “quench” model used in essentially all heat flux partitioning treatments embodies a significant overestimate of this part of the heat transfer, by a factor of perhaps ∼30. It is of course the case that the collection of individual models in heat flux partitioning treatments has been refined and tuned in aggregate, and it is not particularly surprising that an individual submodel is not numerically correct. In practice, there is much cancelation between inaccuracies in the various submodels, which in aggregate perform surprisingly well. We suggest ways in which this more soundly based quantification of “quenching heat transfer” might be taken into account in component scale modeling.
Colombo M, Thakrar R, Fairweather M, et al., 2019, Assessment of semi-mechanistic bubble departure diameter modelling for the CFD simulation of boiling flows, 17th International Topical Meeting on Nuclear Reactor Thermal Hydraulics (NURETH), Publisher: ELSEVIER SCIENCE SA, Pages: 15-27, ISSN: 0029-5493
Hansch S, Walker S, 2019, Microlayer formation and depletion beneath growing steam bubbles, International Journal of Multiphase Flow, Vol: 111, Pages: 241-263, ISSN: 0301-9322
Microlayers, the few-microns-thick layers of liquid that sometimes remain beneath bubbles growing on a heated substrate, are widely observed in experiments, but theoretical understanding of their formation, behaviour and role in bubble growth is limited.In this paper we present detailed interface-tracking simulations of the formation and depletion of such microlayers. Validation of our results is presented to the degree that available measurements of such a rapid and microscopic phenomenon allow.The work extends previous mechanistic hydrodynamic-only CFD simulations of early microlayer formation up to typical bubble departure times. These calculations confirm the understanding that the bubble growth rate and the resulting bubble shape determine the presence and overall extent of microlayers underneath steam bubbles. Their thickness is strongly influenced by viscous effects and surface tension.We then present coupled, physically self-consistent CFD simulations of the formation and evaporative depletion of such microlayers. This modelling suggests strongly that the evaporation process itself constitutes a significant fraction of the small resistance to heat and mass transfer presented by the very thin liquid layer. Inclusion of representations of evaporative thermal resistance, consistent with those suggested in the literature, is seen to promote the prediction of microlayer formation. Identification of the classes of conditions under which microlayers seem likely to be formed is presented, along with an assessment of their relative contributions to bubble growth. Comparisons of the predictions with recent detailed microlayer measurements indicate good agreement.
Sebilleau F, Issa R, Lardeau S, et al., 2018, Direct Numerical Simulation of an air-filled differentially heated square cavity with Rayleigh numbers up to 10(11), INTERNATIONAL JOURNAL OF HEAT AND MASS TRANSFER, Vol: 123, Pages: 297-319, ISSN: 0017-9310
Ben El-Shanawany A, Ardron KH, Walker SP, 2018, Lognormal Approximations of Fault Tree Uncertainty Distributions, RISK ANALYSIS, Vol: 38, Pages: 1576-1584, ISSN: 0272-4332
Giustini G, Ardron KH, Walker SP, 2018, Modelling of bubble departure in flow boiling using equilibrium thermodynamics, International Journal of Heat and Mass Transfer, Vol: 122, Pages: 1085-1092, ISSN: 0017-9310
To improve the closure relations employed for component-scale Computational Fluid Dynamics simulation of boiling flows, a first-principles method for predicting bubble departure diameters in flow boiling has been developed. The proposed method uses minimisation of the free energy of a system in thermodynamic equilibrium to predict the contact angle and the resistance to sliding of a vapour bubble attached to a surface in the presence of a forced liquid flow. Predictions of the new method are compared with measurements from existing experimental databases, and agreement with data is shown to be comparable or superior to that obtained with previous bubble departure models that have generally used a force-balance approach. The main advantages of the energy-based method over the previous force-based methods are that its formulation is simpler, and that the new model does not require the use of ad hoc tunable parameters to define force terms, or geometrical characteristics of the attached bubble such as its base area, which cannot be confirmed experimentally. This increases confidence in the validity of the new approach when applied outside the rather limited range of current test data on bubble departure in flow boiling.
Hansch S, Walker S, Narayanan C, 2017, Mechanistic studies of single bubble growth using interface-tracking methods, Nuclear Engineering and Design, Vol: 321, Pages: 230-243, ISSN: 0029-5493
The growth of a vapour bubble at a heated surface involves various fluid mechanics, heat transfer and phase change phenomena. In this paper we present recent work under the auspices of the NURESAFE project aimed at developing mechanistic modelling of this. Evaporation at the curved surface of the bubble requires evaluation of the unsteady heat conduction within the surrounding liquid, coupled to an appropriate phase change model at the vapour–liquid interface. Issues around the development and implementation of such a phase change model are addressed. For low-pressure bubbles, however, a large fraction of the total evaporation takes place from the “microlayer”; a thin layer of water coating the heated substrate, which is left behind as the bubble expands. This microlayer evaporation requires careful, sub-grid modelling, as heat fluxes through the thin layer are very high. In particular, we demonstrate here the need both for modelling of the conjugate heat transfer within the substrate, and the importance of the incorporation of evaporative thermal resistance at the vapour–liquid interface. Despite the important role it plays in bubble growth, the mechanisms governing the formation, and resulting dimensions, of this microlayer are very little understood. We finish with a presentation of some early results attempting to investigate mechanistically the hydrodynamics of microlayer formation.
Ardron K, Giustini G, Walker SP, 2017, Prediction of dynamic contact angles and bubble departure diameters in pool boiling using equilibrium thermodynamics, International Journal of Heat and Mass Transfer, ISSN: 0017-9310
Carasik LB, Sebilleau F, Walker SP, et al., 2016, Numerical simulations of a mixed momentum-driven and buoyancy-driven jet in a large enclosure for nuclear reactor severe accident analysis, NUCLEAR ENGINEERING AND DESIGN, Vol: 312, Pages: 161-171, ISSN: 0029-5493
Dasgupta A, Chandraker DK, Kshirasagar S, et al., 2016, Experimental investigation on dominant waves in upward air-water two-phase flow in churn and annular regime, EXPERIMENTAL THERMAL AND FLUID SCIENCE, Vol: 81, Pages: 147-163, ISSN: 0894-1777
Thakrar R, Murallidharan J, Walker SP, 2016, CFD investigation of nucleate boiling in non-circular geometries at high pressure, NUCLEAR ENGINEERING AND DESIGN, Vol: 312, Pages: 410-421, ISSN: 0029-5493
Hansch S, Walker S, 2016, The hydrodynamics of microlayer formation beneath vapour bubbles, International Journal of Heat and Mass Transfer, Vol: 102, Pages: 1282-1292, ISSN: 0017-9310
‘Microlayers’, the thin (less than 10 μm) films of liquid left behind beneath rapidly-growing steam bubbles at a heated wall, can be a large, and even the dominant, source of the vapour in such bubbles by the time they depart the wall. Given their slenderness (compared to the 1–10 mm diameter of the bubble), and their high aspect ratio, (with radial extents of perhaps order >100 times their thickness), such microlayers are incorporated relatively simplistically in microscopic CFD analyses of bubble growth. However, their role is particularly important because the evaporation of the microlayer generates vapour rapidly, which itself expands the bubble and generates even more microlayer to evaporate. Plainly, a good understanding of the microlayer formation process is desirable. In this paper we present first-principles calculations of the hydrodynamics of the formation of such microlayers. These seem to show overwhelmingly that the determinant of the existence and radial extent of a microlayer is the bubble growth rate, with higher growth rates leading to more flattened and less spherical bubbles, allowing larger microlayers being trapped beneath them. When they are formed, microlayer thickness is then to a degree dependent on the fluid surface tension and liquid viscosity. The need for an extension of these hydrodynamic studies to include a mechanistic self-consistent model of the evaporative depletion of the microlayer is noted.
Murallidharan J, Giustini G, Sato Y, et al., 2016, Computational Fluid Dynamic Simulation of Single Bubble Growth under High-Pressure Pool Boiling Conditions, Nuclear Engineering and Technology, Vol: 48, Pages: 859-869, ISSN: 1738-5733
Component-scale modeling of boiling is predominantly based on the Eulerian–Eulerian two-fluid approach. Within this framework, wall boiling is accounted for via the Rensselaer Polytechnic Institute (RPI) model and, within this model, the bubble is characterized using three main parameters: departure diameter (D), nucleation site density (N), and departure frequency (f). Typically, the magnitudes of these three parameters are obtained from empirical correlations. However, in recent years, efforts have been directed toward mechanistic modeling of the boiling process. Of the three parameters mentioned above, the departure diameter (D) is least affected by the intrinsic uncertainties of the nucleate boiling process. This feature, along with its prominence within the RPI boiling model, has made it the primary candidate for mechanistic modeling ventures. Mechanistic modeling of D is mostly carried out through solving of force balance equations on the bubble. Forces incorporated in these equations are formulated as functions of the radius of the bubble and have been developed for, and applied to, low-pressure conditions only. Conversely, for high-pressure conditions, no mechanistic information is available regarding the growth rates of bubbles and the forces acting on them. In this study, we use direct numerical simulation coupled with an interface tracking method to simulate bubble growth under high (up to 45 bar) pressure, to obtain the kind of mechanistic information required for an RPI-type approach. In this study, we compare the resulting bubble growth rate curves with predictions made with existing experimental data.
Giustini G, Walker SP, 2016, Evaporative thermal resistance and its influence on microlayer evaporation, 3rd International Topical Meeting on Advances in Thermal Hydraulics 2016, ATH 2016
Giustini G, Jung S, Kim H, et al., 2016, Evaporative thermal resistance and its influence on microscopic bubble growth, International Journal of Heat and Mass Transfer, Vol: 101, Pages: 733-741, ISSN: 0017-9310
Simulations of the formation of small steam bubbles indicate that the rate of growth of bubbles is very sensitive to the rate of evaporation of the micro-layer of liquid beneath the bubble. Such evaporation is rapid, and is modelled as being driven by the large heat flux through the thin liquid layer caused by the difference in temperature between the solid–liquid interface, and the saturation temperature in the interior of the bubble. However, application of this approach to recent experimental measurements of Jung and Kim generated anomalous results. In this paper we demonstrate that a model of the micro-layer heat flux that includes an allowance for the finite evaporative thermal resistance is able to eliminate these anomalies. This evaporative thermal resistance is a consequence of near-interface molecular dynamics, characterised by a quantity termed ‘evaporation coefficient’. Whilst in most engineering applications evaporative thermal resistance is small compared to conductive resistance, here, with the micro-layer thickness ranging from a few microns down to zero, it becomes of considerable importance. Selection of a molecular ‘evaporation coefficient’ to restore consistency to the anomalous measurements allows a plausible numerical value to be inferred. For the several times and multiple locations studied, a fairly consistent value of between 0.02 and 0.1 is indicated, (for saturated water in laboratory conditions), which itself is consistent with earlier literature values of this rather difficult quantity. It is shown that the evaporative resistance always represents a large fraction of the conductive resistance, and for important phases of the process dominates it. The need for inclusion of this phenomenon in the micro-layer models used in bubble analysis is clear.
Cinosi N, Walker SP, 2016, CFD Analysis of Localised Crud Effects on the Flow of Coolant in Nuclear Rod Bundles, Nuclear Engineering and Design, Vol: 305, Pages: 28-38, ISSN: 0029-5493
It has been suggested that crud deposits on a number of adjacent fuel rods mightreduce coolant flow rates in associated sub-channels. Such reduced flow rates couldthen worsen thermal-hydraulic conditions, such as margin to saturated boiling, fuelsurface temperature, and the DNB ratio. We report the results of a detailedcomputational fluid dynamics study of the flow pattern in a partially-crudded rodbundle. Values obviously depend on, for example, the thickness of crud assumed,but sub-channel flow rate reductions of ~10% were predicted by this analysis.However, this mass flow rate reduction was found to be more than offset byimproved heat transfer induced by the relatively rough surface of the crud. Claddingtemperatures were predicted to be essentially unchanged, and the DNBR wassimilarly little altered. We conclude that such flow reduction and diversion is not likelyto be of concern.
Nandi K, Walker SP, Date AW, 2016, High Resolution TVD Schemes for Interface Tracking, International Conference on Numerical Analysis and Applied Mathematics (ICNAAM), Publisher: AIP Publishing, ISSN: 1551-7616
A first order upwind difference scheme (UDS) is routinely adopted for representing convection terms in a discretised space. UDS provides stable solutions. However it also introduces false diffusion in situations in which the flow direction is oblique relative to the numerical grid or when the cell-Peclet number is large. In order to predict sharp interface, higher order upwind schemes are preferred because of they reduce numerical dissipation. In interfacial flows, density and viscosity vary sharply in space. Representation of convective terms by Total variation diminishing (TVD) schemes ensures reduced smearing without impairing convergence property. TVD schemes develop formulae for interpolation of a cell-face value of the transported variable. If the interpolated value is bounded by the neighbouring nodal values then the scheme is ‘Bounded’. However, not all TVD schemes possess this property of ‘Boundedness’. The Normalised Variable Diagram (NVD) defines a domain within which the TVD scheme is bounded. Thus by combining the features of both TVD schemes and ensuring that they fall with the defined area of NVD, the convergence as well as the boundedness of a computational scheme can be ensured. In this paper, six different higher order schemes are considered some which are TVD bounded or unbounded, to solve the well known interface tracking problem of Rayleigh-Taylor Instability. To the best of our knowledge, a comparison of combined TVD/NVD principles in the case of interface tracking problems has not been reported in published literature.
Giustini G, Badalassi V, Walker SP, 2016, Analysis of the liquid film formed beneath a vapour bubble growing at a heated wall without neglect of evaporative thermal resistance, 2016 International Congress on Advances in Nuclear Power Plants, ICAPP 2016
Hansch S, Walker S, 2016, The hydrodynamics of the formation of microlayers beneath vapour bubbles growing on a heated substrate, ATH-16
Thakrar R, Walker SP, 2016, CFD Prediction of High-Pressure Subcooled Boiling Flow with Semi-Mechanistic Bubble Departure Diameter Modelling, 25th International Conference Nuclear Energy for New Europe (NENE), Publisher: NUCLEAR SOCIETY SLOVENIA
Dasgupta A, Chandraker DK, Nayak AK, et al., 2016, Development and validation of a model for predicting direct heat transfer from the fuel to droplets in the post dryout regime, Pages: 93-103
Following dryout after a large loss of coolant accident, but before eventual final quenching of the fuel, the overheated fuel is cooled by a flow of superheated vapor with entrained saturated liquid droplets. The main mechanism for cooling the cladding is convective heat transfer to the superheated vapor, with this vapor in turn cooled by evaporation into it of the saturated droplets. An additional mechanism that is not normally considered explicitly is the cooling of the metal by direct droplet impingement of non-wetting droplets. It is difficult to measure and to analyze, and it is believed probably to contribute relatively little to the cooling, but it could be helpful additional cooling mechanism. In this paper is published validation of a simple analytical model of the heat removed from the wall during the few milliseconds that a non-wetting droplet is in "contact" with the wall. It is found that the model's estimates of the heat removal are consistent with such limited experimental evidence as is available, paving the way for incorporation of the model into computations of post dryout heat transfer.
Sebilleau F, Issa RI, Walker SP, 2015, Analysis of Turbulence Modelling Approaches to Simulate Single-phase Buoyancy Driven Counter-current Flow in a Tilted Tube, Flow Turbulence and Combustion, Vol: 96, Pages: 95-132, ISSN: 1573-1987
In the present paper, both Large Eddy Simulation (LES) and unsteady-RANS(uRANS) CFD studies of single-phase buoyancy-driven counter-current flow in a pipe arepresented for an Atwood number of 1.15 × 10−2 and an inclination angle of 15 degreesfrom the vertical. The basic flow phenomena involved are fundamentally the same as thoseencountered in various industrial applications including passive coolant flow in the loops ofa nuclear reactor. Earlier work on this type of flow was focused on its physical aspects usingboth experimental measurements and Direct Numerical Simulation (DNS). The presentwork investigates the performance of several commonly used LES eddy viscosity subgridscalemodels. The results show that LES is able to reproduce accurately the experimentalresults and that the dynamic Smagorinsky subgrid-scale model gives the best predictions.The results of those calculations were then used to obtain more information on the physicsof such flow. As regards uRANS simulations, a range of classic two-equation linear eddyviscosity models, based on low-y+ formulation at the wall and the single gradient diffusionhypothesis for the turbulent density fluxes were compared against experimental dataavailable in the literature. An elliptic blending Reynolds stresses model (EBRSM) with generalisedgradient diffusion hypothesis (GGDH) for the scalar fluxes was then implementedand compared with experiments. The latter showed very good agreement for the first ordermoment with the experimental data whereas all the two-equation eddy viscosity based modelsshowed rather high levels of discrepancy. The main cause of discrepancy was found tobe the underprediction of the axial turbulent buoyancy production effect, which has a moredetrimental impact on the eddy viscosity models than the second moment closure one
Hansch S, Giustini G, Narayanan C, et al., 2015, Microlayer models for nucleate boiling simulations: The significance of conjugate heat transfer, 16th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, NURETH 2015
Giustini G, Murallidharan, Sato Y, et al., 2015, Numerical study of heat diffusion controlled bubble growth in A pressurized liquid, 16th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, NURETH 2015
Murallidharan, Giustini G, Sato Y, et al., 2015, Interface tracking based evaluation of bubble growth rates in high pressure pool boiling condition, 16th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, NURETH 2015
Gimeno LS, Walker SP, Hewitt GF, et al., 2015, Validation and cross-verification of three mechanistic codes for annular two-phase flow simulation and dryout prediction, Pages: 6863-6875
The ability to predict the boiling transition, or dry-out, in annular two-phase flow is essential to Light Water Reactor (LWR) safety analysis. Common approaches include the use of empirical correlations or look-up tables which, although reliable, cannot be readily applied to complex cases outside the experimental range used for their development. Phenomenological models can widen the range of conditions in which dry-out can be predicted as they provide a better insight into the governing phenomena. These models however also employ empirical correlations to close the system of conservation equations and therefore require validation against experimental data. In this paper, three independently-developed codes for the phenomenological modelling of dry-out, GRAMP, MEFISTO-T and SCADOP, are compared against one another and validated against two experimental dry-out datasets. These data, on dry-out in tubes, were generated by BARC, in India, and Harwell, in the UK. The three codes are used to predict the location of: the onset of annular flow, the flow flow rate along the annular flow length, the dry-out power and the location of dry-out under broadly BWR operating conditions. A high level of consistency between the three codes is demonstrated, and good agreement is observed against the experimental data. Some areas of uncertainty are also discussed in this paper, with the focus on the applicability of the entrainment deposition correlations and the importance of the liquid entrained fraction at the onset of annular flow.
Sebilleau F, Issa RI, Walker SP, et al., 2015, CFD and experimental analysis of single phase buoyancy driven counter-current flow in a pipe, Pages: 1275-1288
In the present paper, we study single-phase counter-current flow in a circular pipe as a model problem that is representative of flows occurring under various circumstances in a reactor primary loop, for example when emergency-cooling (ECCS) water is injected in the hot leg or when cooler fluid flows from the steam generator in the hot leg. In order to isolate the complexity of the turbulent mixing generated by buoyancy driven single-phase counter current flow, an experimental rig was built at the Bhabha Atomic Research Center (BARC) to validate CFD model predictions. This new rig is composed of two tanks of water linked by an inclined circular pipe. Initially the upper tank is filled with cold water and the lower one with heated water and the different-density fluids are allowed to mix under the influence of gravity. In this paper, thermocouple measurements of temperature at several locations in the rig are presented for several temperature differences and tilt angles. These experimental measurements are compared with state of the art highly-resolved large eddy simulation (LES). These LES computations (carried out with the commercial CFD package STAR-CCM+) give detailed information on the behaviour of the turbulence in such flow as well as serve to assess the performance of some uRANS two-equation eddy viscosity models that are also tested.
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.