Numerical Modeling On Sand Casting Process of A Wind Turbine Part
Numerical Modeling On Sand Casting Process of A Wind Turbine Part
Keywords: Sand casting; ProCAST; Numerical simulation; Filling and solidification; Shrinkage and
porosity.
Abstract. In this research work, ProCAST software is employed to study the sand casting process of a
box-type part for wind turbine. The casting part is big in size, non-uniform in wall thickness, and
heavy in some local positions. Shrinkage cavity and porosity are very likely to from at these
locations. By numerical simulation, the influence of the parameters such as pouring temperature,
chills, riser on mold filling and solidification is analyzed. Simulation researches reveal that with
appropriate pouring temperature, correct number, size and location of chills and risers, a smooth mold
filling, reduced shrinkage and other defects are available and desired sound castings can be produced.
Introduction
Wind turbine manufacturers have to compete with other sources of energy such as natural gas and
coal. In order to produce cheaper wind electricity, larger and lighter wind turbines have to be
developed. Cast components make up much of the weight of the wind turbine. The majority of wind
turbine parts are made out of the challenging ductile cast iron because of its good castability and
mechanical properties [1]. However, due to the graphite expansion action and effect, the fluid feeding
channel may be blocked, and the feeding capability diminishes in the late period of casting
solidification. Thus, some defects such as porosity, shrinkage and crack may generate during casting
process [2]. On the other hand, the gating system and casting process parameters play some important
roles on mold filling and solidification [3, 4]. For example, turbulent mold filling may cause oxide
formation, entrapped air, mold erosion, sand inclusions, which would reduce the strength, toughness
and quality of the part[5, 6].
To prevent these casting defects, every new design of casting process is tested and refined
experimentally by using prototyping which would greatly increase the cost and time consumption of
trial production. Numerical simulation is widely used and accepted in casting design as a way to
reduce hardware prototyping and to improve the parts design and manufacturing processes, and
therefore could reduce the costs associate with prototyping and speed up process design by predicting
the casting defects that might be developed during the actual production.
The purpose of this paper is to demonstrate how the application of numerical optimization
techniques can be used to develop an effective optimization process for gating and riser system
design. Mold filling and solidification processes of the castings were simulated with the ProCAST.
The simulation results indicated that gating and riser system parameters significantly affect the
casting quality. This virtual approach and optimization technique can be applied to the foundry
industry, which is evidently superior to typical trial-and-error approaches.
All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans
Tech Publications Ltd, www.scientific.net. (#556902479, Tübitak Ulakbim, Çankaya - Ankara, Turkey-06/04/21,10:53:00)
1368 Frontiers of Manufacturing and Design Science II
Mathematical model
The mold filling process that closely related to molten metal flow, heat transfer and mass transfer
process is a flow process at changeable temperature accompanied with heat loss and solidification.
The governing equations include mass, momentum and energy conservation which were described in
previous published papers [7, 8] and will not be mentioned in this paper. However, the determination
of position and movement of free surface is a key for the calculation of unstable flow of free surface in
mold filling process.
the volume of fluid (VOF) method [9], which is especially helpful for complex 3D geometries, is
used to track the free surface. In VOF method, a new function F is introduced to stand for the
fractional volume of a mesh cell that is filled with fluid. Thus the value of F is unity at any point
occupied by fluid and zero otherwise. the transport equation for F is given by
∂F
+ ∇ ⋅ ( Fv ) = 0 (1)
∂t
Experimental verification
To verify the numerical reliability of ProCAST in casting process, a X-ray setup and a simple mold
with a bottom gate system are employed to record the mold filling process. The flow pattern of melt
metal filling the mold cavity at difference time is captured and shown in Figure 1. A injection to the
right cavity wall can be clearly observed in the mold cavity. This injection not only entraps air in the
molten metal, but also erodes the sand mold and may results in defects such as air pores and sand
inclusions. It is observed that the simulation results shown in Figure 1 provide a good prediction of
the flow pattern.
(a)t=0.23s (b)t=0.30s
(c)t=0.5s (d)t=0.76s
Figure 1. Comparison of experimental and simulation results
The calculation is carried out on the built mesh composed of 330000 elements for 200000 nodes. A
ferrite type nodular cast iron, QT400-18 (equivalent USA 60-40-18), is used to manufacture the wind
turbine part. With a good fluidity and casting behaviors, sound casting without defects is expected to
be produced.
Mold filling process is very important for producing sound castings. Many casting defects, such as
misrun, entrapped air, oxide inclusions, cold shut and etc. are closely related to this process. Fig 2
shows the mold filling process at different time. The free surface is visualized as an isosurface of F
with a constant value of 0.5. As shown in Fig. 2 (a), the melt enters the mold cavity from bottom
runner and ingates under the hydraulic pressure in sprue at the early filling stage. Because the pouring
flow rate is bigger than that flowing into the cavity through the bottom ingates, excrescent melt
establishes a pressure in the sprue rapidly and force the melt flow into the cavity from the top runner
and ingates (see Fig. 2(b)). This so-called mixed flow may make the melt splash, entrap air and
demolish the ideal temperature profile for subsequent feeding, as observed in Fig. 2 (b) (c).
As soon as the melt starts to fill the mold, the heat energy is extracted by the mold. However, as
shown in Fig. 2, the temperature variation can almost be ignored because of the fast pouring rate and
the heavy part. The obvious decrease of temperature can only be observed after some time later when
the mold is full filled. For example, the temperature filed in the casting at time 200s and 4000s are
given in Fig 3 respectively. As expected, the hot spots locate at some thick and heavy position in the
casting. This can also be observed in Fig. 4, which shows the solid fraction distribution in the casting
at transient moments and predicts the solidification procedure. The grey color indicates the
solidification starts from the surface of the casting, and the red color the melt liquid. As the modulus
of the riser is smaller than that of the hot spots in the casting, the riser freezes faster than the hot spots.
Thus, the riser can not provide sufficient melt to feed the shrinkage of the hot spots. Fig. 5 shows areas
that are predicted to be more likely to contain macroscopic shrinkage porosity.
1370 Frontiers of Manufacturing and Design Science II
Fig. 5 Shrinkage cavity and porosity (a) front view (b) back view
n the casting without chills Fig. 6 Setup of chills
Applied Mechanics and Materials Vols. 121-126 1371
Conclusions
Casting process of a wind turbine part at different conditions is simulated by ProCAST. The positions
of shrinkage cavity and porosity can be successfully predicted through analyzing mold filling and
solidification process. The shrinkage cavity and porosity can be reduced and even eliminated by
employing some appropriate chills and risers, and sound casting can be produced..
Acknowledgements
The authors acknowledge supports from the National Natural Science Foundation of China (No.
51074116), the National Science Foundation of Hubei Province (No. 2009CDB258) and Key Project
Supported by Education Department of Hubei Province.
References
[1] Q. Gui, G. Mei and L. Guo: Foundry Technology Vol. 25 (2004), p. 337
[2] L. Kumruoglu, A. Özer. J. Mater: Process Technology Vol. 197 (2008), p. 182
[3] A. Kermanpur, M. Eskandari, H. Purmohamad, M. A. Soltani, R. Shateri: Materials and Design
Vol. 31 (2010), p. 1096
[4] D. R. Gunasegaram, D. J. Farnsworth, T. T. Nguyen: J Mater Process Technology Vol. 209
(2009), p. 1209
[5] G. Lesoult: Mater Sci Eng A Vol. 413-414 (2005), p. 19
[6] S. Kumar, P. Kumar, H. Shan: J Mater Process Technology Vol. 204 (2008), p. 59
[7] F. Ilinca, J. F. Hetu: J. Num. Meth. Fluids Vol. 34 (2000), p. 729
[8] L. Baicheng, S. Houfa: J. Mater. Sci. Techol. Vol. 11 (5) (1995), p. 312
[9] C.W. Hirt, B.D. Nichols: J. Comput. Phys. Vol. 39(1981), p. 201