OPTIMIZATION OF MULTIGATE RUNNER IN LONG CASTINGS: A SIMULATION APPROACH

Transcription

1 913 OPTIMIZATION OF MULTIGATE RUNNER IN LONG CASTINGS: A SIMULATION APPROACH IRFAN AHMAD ASARI (Mechanical engineering department Aligarh Muslim University, Aligarh U.P mechirfaan@gmail.com) The fluid flow phenomena during mould filling are very complex. It is very difficult to investigate these phenomena through experimentation, since both the metal and mould are opaque and the high temperatures involved. Complete, smooth and uniform filling of mould with clean metal is achieved by an appropriate system, which comprises of one or more pouring basins, sprues, s and gates. The number, location, shape and size of the elements determine the sequence and rate of flow of molten metal into the mould cavity. There are however, no mathematical models to estimate the discharge and volume of flow through each gate. This is needed to balance the flow through multiple gates (by adjusting their position and cross-sectional area), and to ensure that the mould cavity is filled in an optimal time.. In this study the effect of design on the flow behaviour of molten metal is studied and the defect which are associated with flow behaviour, further a simulation based study on ANSYS fluent 14.0 is carried out on four designs of system with multiple gate for filling of a 3000 mm 3 volume casting. Keywords: system, ANSYS INTRODUCTION In casting process, system plays an important role to produce a high quality casting. A poorly designed system results in casting defects. A system controls mould filling process. The main function of system is to lead clean molten metal from ladle to the casting cavity ensuring smooth, uniform and complete filling. Hence to design a good system one must know the behaviour of fluid flow during mould filling process. Mould filling is a complex phenomenon, influencing both internal and external quality. The flow of molten metal after being poured is a transient phenomenon accompanied by turbulence, separation of the flow from the boundaries, dividing and combined flow at the junction, simultaneous heat transfer during the flow and onset of solidification. Moreover melt properties like density, viscosity and surface tension are continuously changing during the flow. All this together makes the filling analysis quite complex. An optimized design satisfying this entire requirement is obtained by experimentation through trial and error methods for a given casting geometry. However, this method takes a long time to get the optimal dimensions of the channels and also adds cost to the company. Another and more suitable approach is numerical simulation in which a model of the actual system is designed and analysis is done on that, which represents the actual mould filling process, so that we can predict the results in advance before producing actual casting.after analysing the results and flow pattern necessary changes are made in the previous designed system, and the model is run again. In this manner iterative simulations are done until an optimize design is not reached. Hence physical experiment is replaced by numerical experiment by this method. This process saves time by applying an accurate and precise numerical optimization technique. Goal of this research is to study the flow behaviour of molten metal in horizontal multi-gate system and comparison of four design in order to optimize mass flow rate through each gate and exit velocity through each gate. The casting of thin plate from non ferrous metal such as aluminium is very difficult through green sand casting with a non pressurised ratio. But if we calculate the parameters based on the law of fluid mechanics and empirical rule given in various literature we could get defect free casting[1]. A case study of a component (Ginny weight) manufactured in an industry(m/s

2 914 jadhao steel alloy) done and it was found that because of improper design the company was facing 25%-30% average rejection of product because of poor quality and defect. After correcting the ratio and system parameter according to ISO standard the productivity got improved from 42%-74% reducing total percentage of rejection down to 7.5% and yield increases from 86%-92%[2].Further to obtain a defect free and sound casting the location of the ingate and its parameter design plays important role. A proper location of gate lead to lesser flow hesitation,less pressure and clamp forces and good flow balance also the shape and size of ingate highly affect the soundness of casting produced[3].formation of shrinkage cavity and its location inside the casting can be easily predicted by using a computer program based on Finite Element Method. Thus in this way without actually performing casting process one can optimize the process parameter, through simulation, in order to obtain shallow localization and regular shape[4].those casting having narrow passages such as rib of flywheel,the location of ingate and riser plays vital role. By Generating 3-D model on a CAD modelling software (Catia V.5) and iteratively simulating it for a number of time by changing the location of riser and ingate on ADSTEFAN Simulation Software, an optimum design can be obtained.simulation result for optimum design shows that there is 98% reduction in shrinkage and good heat distribution with directional solidification has achieved[5].in a study the author have compared four design of system proposed for goose neck machine clamp. The analysis have been carried out on ANSYS 14.0 FLUENT to determine the optimum system[6].in another research simulation based study is carried out to find the effect of shape of ingate. two shapes of ingate,one is with flat gate entry and other is spoon fed gate entry was compared through simulation analysis on ROTORK FlOW 3D.It was found that spoon fed gate leads to uniform flow rate, lesser fill time and lesser defects such as cold shut and misrun and gas porosity etc[7].a CFD based analysis was done on a high pressure die casting(hpdc),die containing six cavities producing various test specimen.it was found that a tapered with a cross section area reducing proportional to the cavities volume ratio, instead of parallel with constant cross section significantly improve the filling pattern,in which filling time and pressure distribution were well balanced among specimen cavities having different volume[8].on the basis of above literatures a simulation based comparative study on a multiple ingate has been carried out in this current research work and the findings have discussed. 2. METHODOLOGY This case study was designed for such large casting which requires multiple gates for filling. Here the author has focused on design and location of gates in the in sand mould casting of non ferrous metals and alloys. In such a case where the mould cavity is filled through more than one ingate then the flow rate through each gate is not necessarily same. In such situation the stream of molten metal coming from various gate does not get fused together properly. Which causes cold shut and misrun defect in casting also the mechanical property at these location where the molten stream from various gate meets get deteriorated.also when the molten metal stream comes out from in gate and hit the bottom of mould cavity it erodes the sand from mould surface causing sand inclusion and other defects. Based on the above analysis four design of system have been compared on the basis of flow rate from each gate and velocity of molten metal steam from each gate in a multi gate system. Description of four proposed designs of getting system is tabulated below 3. TABLE: Specification of Four Design Design1 Design2 Design3 Design4 Parallel Parallel tapered tapered at both end No. of 2 on 4 gates on 4 gates, 2 on gates -4 both side one end of both side of of sprue sprue sprue

3 D CAD Model of the Four Gating System same set of conditions. Input conditions given are listed below. Filling assumed to be completely gravity filling. Pressure was assumed to be atmospheric Input Fluid Flow Parameters Pouring 20 degree C 900 degree C temperature Melting _ 660 degree C temperature Viscosity _ ~ Results and Graphs Results shows that the mass flow rate for design concept 1 highly varies from in gate 1 to 4.whereas for design concept 2 and 4 it has similar nature. Design 3 give lowest variation in flow rate. Average velocity and maximum velocities from each gates for all design shows a similar nature. It is found from the simulation results there is trapping and circulation of molten metal at the sharp bends and mould erosion at the bottom of the sprue (figure). The graphs are attached below. GRAPHS: Fig.1. CAD geometry of design 1,2,3 and 4 Filling process through the and gate for each design concept were simulated using ANSYS 14 fluent to determine the best design concept. The simulation was based on a transient 3 D situation; material used for simulation was aluminium having properties listed below 5. TABLE: Properties of Materials PARAMETERS TYPE CONDITIONS OF MOULD Material Green Aluminium sand Density 1.5 gm/cm gm/cm 3 Fig.2. Graph of Flow Rate at Ingates Since here the author only wants to focus on the flow behaviour of the molten metal therefore only energy and viscous model were used to simulate the flow. Number of mesh used was simulation was run for 250 iterations for each design concept under the Fig.3. Graph of Maximum Velocity at Ingates

4 916 Fig.7(a). High Velocity at Sprue Exit (before modification) Fig.4. Graph of Average Velocity at Ingates Fig.7(b). Low Velocity at Sprue Exit (after modification) Fig.5(a). Modified CAD Model of Design 3 Top View Fig.5 (b). Front View 7. Conclusion Since the flow rate from 4 in gates in design concept 1 is highly varying whereas design concept 2,3 and 4 gives somewhat uniform flow rate hence on the basis of flow rate criteria design 1 comes out to be worst and 2,3 and 4 gives more satisfactory results. But in case of design concept 2 and 4 the maximum and average velocities from in gate first and last is quit high and reaching closer to 0.5 m/s which would cause mould surface erosion(for non ferrous metal in gate velocity must be less than 0.5 m/s to avoid splashing[9]. Hence design concept 3 fits best in both the deciding criteria i.e. flow rate and velocity at ingate.further some modification have been done in the design concept 3 to avoid the high erosion at the bottom of sprue and negative pressure at sharp bends.simulation of this modified version of design concept 3 gives better results as shown below. REFERENCES Fig.6(a) Velocity Vector (before modification) Fig.6(b). Velocity Vector (After modification) [1] V. Anjo and R. Khan, Gating System Design for Casting thin Aluminium Alloy ( Al Si ) Plates, pp. 1 10, [2] S. V Gulhane and A. D. Shirbhate, Effects of Improved Gating System by Using Standard Gating Ratio, vol. 22, no. 4. [3] B. Ravi, A Holistic Approach to Zero Defect Castings, no. February, pp. 1 9, [4] F. Engineering, Simulation of shrinkage cavity formation during solidification of binary alloy, vol. 10, no. 1, pp , [5] N. Hebsur and S. Mangshetty, Casting Simulation for Sand Casting of Flywheel, IOSR J. Mech. Civ. Eng., vol. 11, no. 4, pp , 2014.

5 917 [6] M. I. Ansari and D. K. Singh, On the Analysis of Molten Metal Flow through Sprue in Casting Process, Anal. Molten Met. Flow through Sprue Cast. Process, vol. 1, no. 6, pp. 1 6, [7] B. V. Ramnath, C. Elanchezhian, V. Chandrasekhar, a. A. Kumar, S. M. Asif, G. R. Mohamed, D. V. Raj, and C. S. Kumar, Analysis and Optimization of Gating System for Commutator End Bracket, Procedia Mater. Sci., vol. 6, no. Icmpc, pp , [8] Y. Wang, K. Kabiri-Bamoradian, and R. A. Miller, Runner design optimization based on CFD simulation for a die with multiple cavities, 2011 Die Cast. Congr. Tabletop, pp. 1 6, [9] B. Ravi, Computer-aided Casting Method Design, Simulation and Optimization, Foundry, no. March, pp. 1 5, 2008.