Indian Foundry Journal Indian Foundry Journal Indian Foundry Journal Indian Foundry Journal Indian Foundry Journal 21 Vol 60  No.8  August 2014 WAX RECYCLING CASTING SIMULATION MULTI-PHYSICS Multi-Physics Simulation and Casting Defect Prediction Using FLOW + S. Savithri 1 , Roschen Sasikumar 1 , Elizabeth Jacob 1 , Vivek Chourasia 2 , Y.V.M. Siva Kumar 2 and Baba Prasad Lanka 2 1 Computational Modelling and Simulation, CSIR-National Institute for Interdisciplinary Science & Technology, Thiruvananthapuram, 2 3D Foundry Tech Pvt. Ltd., Mulund, Mumbai, Corresponding Author: S. Savithri, E-Mail: sivakumarsavi@gmail.com In today’s global manufacturing environment, casting buyers want foundries to deliver the parts with the desired quality at short lead times. To achieve this, foundries are turning to computer simulation for analysing and optimising the design of the cast components as well as methoding (gating and feeding) design. Most of the casting simulation packages available in the international market are however, beyond the reach of small units, which form 80% of the Indian foundry industry. Keeping this in view, the computational modelling group of CSIR-NIIST has developed a highly efficient multi-physics-based software code for modelling mould filling and casting solidification, which is suitable for a wide range of metals and gravity casting processes. The software provides the filling sequence, temperature distribution, liquid fraction and cooling curves, and predicts major casting defects including air blow holes, cold shut, misrun, shrinkage porosity and hard zones. The code has been continuously improved during the last two decades and was named FLOW + , which was recently integrated with the casting design software AutoCAST. The combined programme enables foundry engineers to seamlessly carry out 3-D methods design, advanced coupled simulation, defects prediction, and casting optimisation within an extremely user-friendly environment. The mathematical model, functionality and results of the programme are briefly described in this paper, and illustrated with a few industrial case studies. Introduction In conventional metal casting process in foundries, the design of the methoding system for a given shape of the part depends very much on the knowledge and experience of the foundry engineers. The conventional trial and error method implies waste of material, melting energy and labour, as well as increased lead times to arrive at the optimal design that gives the desired quality with acceptable levels of yield. Foundry trials can be replaced by computer simulation of casting process. It is found that computer simulation significantly compresses the overall lead time to develop a new casting (right first time), and also reduces the level of defects by more than half (moving towards right every time) [1] . Casting process simulation requires a complete 3-D model of the mould including part cavities, feeding and gating system. The gating system takes liquid metal from the pouring basin to the casting cavities, and the feeders supply molten metal to compensate for metal shrinkage during solidification. Proper positioning and sizing of gating ensures complete, smooth and uniform filling of mould cavities, and freedom from related defects like blow holes and cold shuts. Similarly, proper location and sizing of feeders ensures directional solidification and prevention of shrinkage porosity inside the casting. The design of gating and feeding system is modified and verified by simulation until the desired level of quality and yield are obtained. At present, there are several casting simulation packages available in the international market. Popular ones include Magmasoft (Germany), Novaflow (Sweden), Procast (France) and Solidcast (USA) [2] . Most of these have established local offices or vendors in India to take advantage of the large number of foundry units (about 5,000). It is well-known that the average capacity of an Indian foundry is less than one fourth of that in Germany; indeed, more than 80% of Indian foundries are classified under small-scale industry. These foundries can benefit from computer simulation to improve their quality and yield, thereby move up the value chain,