Modelling Marangoni convection in laser heat treatment

Epitaxial Laser Metal Forming (E-LMF) consists in impinging a jet of metallic powder onto a molten pool formed by controlled laser heating and thereby, generating epitaxially a single crystal deposit onto the damaged component. This new technique aims to be used for the repair and reshape single crystal gas turbine components. Because of the very localised melting pool, the high temperature gradients produced during the process must be carefully controlled in order to avoid both the columnar-to-equiaxed transition (CET) and the appearance of hot tears. To this end, heat flow modelling is required to establish the relationship between process parameters such as laser power, beam diameter and scanning speed, and the local solidification conditions. When modelling the heat transfer within the sample, it is necessary to include the liquid flow pattern generated by the surface tension driven convection known as the Marangoni effect. Indeed, the fluid flow in the liquid pool dictates the shape of the traces as shown by the measurements carried out at EPF-Lausanne in re-melting experiments. A three dimensional (3D) model is implemented in the finite element software calcosoft in order to model the development of the fluid convection within the liquid pool. It is shown that the velocities due to natural convection are of the order of 1 mm/sec whereas Marangoni convection produces velocities of the order of 1 m/sec. Moreover, at low scanning speeds, the liquid pool becomes larger than the beam diameter and the development of Marangoni eddies leads to a widening and deepening of the pool. The local solidification conditions such as the thermal gradient and the solidification speed can be extracted at both the solidus and liquidus temperatures to assess the risk of CET and hot cracking.

Marangoni Convection and Fragmentation in Laser Heat Treatment

Jean-Marie Drezet, Sélim Mokadem

Epitaxial Laser Metal Forming (E-LMF) consists in impinging a jet of metallic powder onto a molten pool formed by controlled laser heating and thereby, generating epitaxially a single crystal deposit onto a single crystal substrate. It is a near net-shape process for rapid prototyping or repair engineering of single crystal high pressure/high temperature gas turbines blades. Single crystal repair using E-LMF requires controlled solidification conditions in order to prevent the nucleation and growth of crystals ahead of the columnar dendritic front, i.e., to ensure epitaxial growth and to avoid the columnar to equiaxed transition. A major limitation to the process lies in the formation of stray grains which can originate either from heterogeneous nucleation ahead of the solidification front or from remelting of dendrite arms due to local solute enriched liquid flow, .i.e fragmentation. To study this last aspect, heat and fluid flow modelling is required to establish the relationship between process parameters such as laser power, beam diameter and scanning speed, and the local solidification conditions plus the fluid flow in the vicinity of the mushy zone. Surface tension driven convection known as the Marangoni effect needs to be included in the model owing to its large influence on the development of eddies and on the shape of the liquid pool. The 3D model implemented in the FE software calcosoft® is used to compute the fluid convection within the liquid pool and to assess the risk of fragmentation using a criterion based on the local velocity field and thermal gradient. The computed results are compared with EBSD maps of laser traces carried out at EPF-Lausanne in re-melting experiments.

Trans Tech Publications2006

Phase Change Material Systems for High Temperature Heat Storage

Selmar Rudolf Binder, Sophia Haussener, Alberto Ortona, David Yann Samuel Perraudin, Ehsan Rezaei

Efficient, cost effective, and stable high-temperature heat storage material systems are important in applications such as high-temperature industrial processes (metal processing, cement and glass manufacturing, etc.), or electricity storage using advanced adiabatic compressed air. Incorporating phase change media into heat storage systems provides an advantage of storing and releasing heat at nearly constant temperature, allowing steady and optimized operation of the downstream processes. The choice of, and compatibility of materials and encapsulation for the phase change section is crucial, as these must guarantee good and stable performance and long lifetime at low cost. Detailed knowledge of the material properties and stability, and the coupled heat transfer, phase change, and fluid flow are required to allow for performance and lifetime predictions. We present coupled experimental-numerical techniques allowing prediction of the long-term performance of a phase change material-based high-temperature heat storage system. The experimental investigations focus on determination of material properties (melting temperature, heat of fusion, etc.) and phase change material and encapsulation interaction (stability, interface reactions, etc.). The computational investigations focus on an understanding of the multi-mode heat transfer, fluid flow, and phase change processes in order to design the material system for enhanced performance. The importance of both the experimental and numerical approaches is highlighted and we give an example of how both approaches can be complementarily used for the investigation of long-term performance.

2015

Analyse de la migration d’huile dans les pompes à chaleur mono- et biétagées

Daniel Favrat, Michele Zehnder

The Laboratory of Industrial Energy systems has evaluated a series of high performance air-water heat pump concepts with the aim to retrofit existing fuel fired boilers for residential hydronic heating systems. The results are published within the framework of the Swiss Retrofit Heat Pump project, which is conducted by the Swiss Federal Office of Energy. The two-stage heat pump cycle has shown the best efficiency prospects. Oil migration between the upper stage and lower stage compressors is the main obstacle for commercialization in the low heat rate range. The oil mass concentration is measured by the Fourier Transform Infrared Spectrospopy (FTIR) method with an attenuated total reflection cell (ATR). This method was calibrated on a separated test loop and compared to measurements of mass-sampling and of density by a Coriolis effect densitymeter. Infrared analysis gives a precision of < 0.1% of oil concentration and the detection limit was identified at about 0.3%. Density measurements cannot be used in the ordinary temperature range of the liquid line in the heat pump. Mass sampling and the FT-IR measurements give corresponding results. An air-water heat pump (10 kWth, R-407C) was build in the laboratory to perform comparison tests in the single-stage and two-stage heating modes. The main elements of the heat pump are: two scroll type compressors, a brazed plate condenser, a finned tube evaporator, an economizer heat exchanger and two electronic expansion valves. Defrosting is done by cycle inversion with a 4-way reversing valve. Measured improvements on the two-stage heating mode of 30% for the COP and 24% for the heat output comparing to the corresponding test point (A-7/W50) in the single stage configuration have been shown. Important reductions of the maximum system temperature at the outlet of the compressors enable an extended range of application (including the point A-12/W65). In the two-stage heating mode an oil transfer rate of 5 - 10 g/min from the high pressure stage to the low pressure stage compressor was detected in all tested conditions, which is strongly limiting the continuous service of the heat pump. For long term running tests an additional balancing system composed of an oil pump and an oil vessel has been installed. The circulating oil concentration is between 0.2% and 0.4% for the two-stage configuration as measured with the FT-IR method. In single-stage mode the concentration is lower than 0.2% and only the mass sampling technique provided reliable results. Sensibility analysis in changing the regulation parameters did not give concluding results in order to design a stable configuration of the two-stage heat pump. In absence of integrated two-stage scroll compressors intermediate solutions have to be considered. These consist in the implementation of a booster compressor or the implementation of an oil separator at the outlet of the high pressure stage compressor coupled with a oil return management or a periodic shut down of the heat pump with external balancing of the oil level in the crankcase. An adapted heat transfer model for refrigerant-oil mixtures was developed, in order to investigate the theoretical impact of oil in the evaporator.

2003

Analyse de la migration d’huile dans les pompes à chaleur mono- et biétagées

Daniel Favrat, Michele Zehnder

2003