On the assessment of the quality of casting nonferrous small-size parts by numerical simulation.

Deac, Cristian ; Bibu, Marius ; Petrescu, Valentin 等

1. INTRODUCTION

The casting of small-size metallic parts, such as components of dental or bone prostheses, has become, in the last few years, a very important element in the activity of many specialized companies and even of many medical and technical laboratories.

This particular type of casting involves a high degree of precision and is very demanding in its execution, given the fact that parts often have wall thicknesses in the range of tenths of millimetres, the imposed tolerances being also very tight, sometimes in the range of [+ or -] 0,05 ... [+ or -] 0,1 mm (Deac V. et al., 1995).

This makes any quality problem in a part, even if it would seem insignificant in terms of size, to still become a major issue and the cause for an immediate rejection of that part.

Therefore, it is of critical importance for the manufacturers of such parts to be able to predict the outcome of a casting operation in terms of quality. This is however complicated by three factors: the variety and complexity of the shapes that need to be cast, the variety of available commercial casting materials and not least the variety of available technologies and equipments (Deac C., 2003).

It is thus no longer possible to rely on one's experience or on simple techniques used for determining a material's castability that in turn would give a clue on whether problems would or not appear in a cast part. The development of numerical simulation and modelling techniques and software can be of great help here. However, currently there are very few casting simulation software packages designed specifically for small parts, and they are much more expensive than software designed for the industrial casting of large parts.

Therefore, starting from the analysis of the defect types that might occur, and from their importance in castings, the authors of the present paper tried to determine whether an industrial casting simulation software can be customised and used for the accurate prediction of quality issues in small-size cast parts, i.e. whether a system designed primarily for the relatively large tolerances of industrial cast parts can be used also for small-size applications..

2. THE QUALITY OF CAST PARTS

Given the small-batch or often even unique part production style employed for the casting of small-size parts, very often the producers use the traditional "lost wax" technique for realising these elements. This technique implies realising a model of the part in a wax-like material, covering it with a so-called investment material, heating the assembly until the wax melts and then casting the metallic alloy in the resulting mold inside the investment material (Deac V, 1995).

The notion "quality of the cast parts", which refers to the presence or absence in the cast parts of defects such as pores, shrinkage holes etc., is strongly related to the notion of "castability" of the material of which the parts are made. The latter can refer to both the completeness of the casting, in the sense of the cast part proportion which could be correctly reproduced during the casting process, and to the casting precision, i.e. to the minimal dimensions of the details which could be reproduced. This explains also the different categories of casting defects that need to be taken into consideration in quality assessments. These categories are (Marxkors, 1999):

--Investment defects, related to the components of the casting system. The main causes for this type of problems are:

--secondary runners of inadequate length or diameter, or too many secondary runners connected to an under-dimensioned intermediate runner.

--an oversized or undersized casting cone, or a casting cone that is not exactly aligned with the main runner;

--strangled runners or strangled connections between runners;

--uncontrolled forming and hygroscopic expansion rate of the investment material, and differences to the metal's expansion rate etc.;

--Pores, determined by the existence of gas bubbles, initially dissolved in the molten metal, that remain trapped in the metallic mass during its cooling and solidification. The main causes for the occurrence of pores in a cast part are:

--inadequately dimensioned sprue and/or inadequately dimensioned gas removal runners;

--overheating of the molten metal before the actual casting;

--the presence of residual gases in the casting area;

--a too large casting pressure;

--a casting area that is too cold when casting starts.

The area with the highest probability to generate pores is the junction between the casting pattern (model) al sprue, although pores could be found also in the casting part itself. Actually, as proven by several casting experiments and also by the speciality literature, porosity is the main problem in small-size cast parts.

--Incomplete castings, related also to the problem of under-dimensioning. Among its main causes, we can cite:

--an insufficient amount of material used for melting and casting;

--the metallic material was not heated to the optimal casting temperature, or there was a too long delay between the moment of reaching of this temperature and the moment of the actual casting;

--the molten alloy starts flowing from the casting cone into the sprue before the actual casting force (e.g. centrifugal force) sets in;

--the casting mould contains gases but there are no gas evacuation runners etc.

3. THE NUMERICAL SIMULATION OF CASTING SMALL-SIZE PARTS

All elements listed above have to be taken into account when manufacturing a small-size part by casting, but at the same time, if we want to be able to correctly predict the outcome of a casting operation from the point of view of quality, we need to use a proper simulation software which can take all these problems into account.

Although in the last few years there have been some attempts to use numerical modelling techniques for a better understanding or even solving of aspects related to the melting and casting of dental alloys, by using software packages like MAGMASOFT, FLUENT or JSCAST, there has been no systematic approach to the usage of computer-aided modelling and simulation software for the optimisation of the design of small-size casting moulds (Wu, 2002; Bernad et al., 2002).

The authors of this study have focused on the software package MAGMASOFT, developed by the company MAGMA GmbH from Aachen, Germany and meant for the use with relatively large parts from the machine-manufacturing industry.

The parts included in the simulation were simplified crowns, of cylindrical shape and with cylindrical cavities. From the software's materials database, the base metal alloy Fe-Cr-Ni type SA-6NM / ASTM A743 was chosen, with following characteristics:

--chemical composition: 0.07 % C; 1 % Si; 1.5 % Mn; 13 % Cr; 4.25 % Ni; 0.7 % Mo; remainder Fe;

--melting point: 1467[degrees]C;

--casting temperature 1630[degrees]C.

The discretisation process carried out with the simulation software has led to a mesh consisting of 998.944 elements.

In order to assess the influence of the various factors, several simulations were run, aiming at realising three identical parts at once, using different values for the casting pressure, protective atmosphere pressure and nature, mould temperature, as well as different design layouts for the sprue and patterns in the mould and various investment material types.

Several output types were obtained and analysed, including the part's porosity, the temperature distribution in the parts (which could indicate a problem in the dimensioning of the runners etc.).

Figures 1 and 2 present the assessment of porosity by simulations carried out for two different sprue designs.

It can be seen that, when using a sprue that allows the direct feeding with molten metal of each part, the porosity level is reduced and pores tend to form mostly in the sprue, rather than in the cast part itself. This is confirmed by experimental castings.

However, when comparing the simulation result with that of actual castings of the parts (figure 3), beneath an obvious similarity, several "fineness" differences were noticed, caused most likely by the lack of sufficient precision in the simulation software. This is understandable given the actual usage domain of the employed software, which is the casting of large parts, where a difference of a tenth of a millimetre, for example, does not play a significant role.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

Therefore, it can be concluded that the usage of an unspecialised, unmodified simulation software cannot lead to a correct assessment of the quality-related problems in small-size cast parts. After modifying some of the software parameters, it was nevertheless possible to obtain a satisfactory result.

5. CONCLUSIONS

The researches described in this paper have proven that it is possible to predict the occurrence of quality-related problems in small-size parts obtained by casting. However, the appropriate program parameters have to be carefully tuned to the actual conditions, in order to obtain the desired precision.

The future researches will target a more detailed part modelling (e.g. more closely resembling a dental crown) and the realising of broader researches, aimed at developing a mathematical model that could be used for example to predict the position and precise size of pores inside a future cast part.

6. REFERENCES

Bernad, S.I, Susan-Resiga, R., Muntean, S. (2002) Fluent Numerical Simulation of the Mold Filling Process for Titanium Dental Casting Applications, Southeastern Europe FLUENT Users Group Meeting, Thessaloniki, Greece, 2002

Deac, C. (2003) Contributii la dezvoltarea tehnologica a topirii si turnarii unor materiale metalice utilizate in tehnica dentara. (Contributions to the technological development of melting and casting some metallic materials used in dental technics). Ph.D. thesis, "Lucian Blaga" University of Sibiu, 2003.

Deac, V. et al. (1995) Turnarea titanului in protetica dentara" (Casting of Titanium in Dental Prosthetics). Editura Universitatii din Sibiu, Sibiu.

Marxkors, R. (1999) Ursachen, Auswirkungen und Behebung von Misserfolgen (Causes, effects and repairing of problems). Deutsche Zahnarztliche Zeitschrift, vol. 54, nr. 7, p. 600-610.

Wu, M. et al. (2002) Numerical simulation of the casting process of titanium removable partial denture frameworks. Journal of Materials Science: Materials in Medicine, vol. 13, p. 301-306.