Paper
Control the Distortion of the Large and Complex Shaped Parts by the Metal Injection Molding Process
2016 Volume 63 Issue 7 Pages 473-478
Details
2016 Volume 63 Issue 7 Pages 473-478
Metal Injection Molding (MIM) is an effective way to manufacture small components with low cost and high precision. However, in the case of large components, it becomes difficult to control the distortion and cracking because of the big shrinkage during debinding and sintering process. Therefore it is important to optimize the condition of each process to reduce the distortion of MIM compacts. Moreover, powder size is also one of the most important parameters. The small particle powder shows high shrinkage and high density as compared to large particle powder during sintering process. In this study, blending of both powders was conducted and the influence of powder size distribution on the distortion of complex shaped parts was evaluated. A coordinate measuring machine, which is a 3D device for measuring the physical geometrical characteristics of an object, was used to measure the distortion. Finally, through controlling the distribution of particle size, distortion of the comparatively large and complex shaped MIM compacts was successfully restrained.
Metal Injection Molding (MIM process) is the manufacture method composed of plastic injection molding and powder metallurgy, which has ability to produce the high degree of geometrical complexity of the component with high properties. MIM process is applicable for most of common engineering metals such as carbon steel, stainless steel, tungsten, nickel-based alloys, titanium alloys, and so on. MIM has been demonstrated to be cost effective manufacturing process of small components, complex shaped parts with high production volume1).
However, in the case of the large sized parts, it becomes difficult to control the distortion and cracking because of the remained binder in the thick part and the large shrinkage during debinding and sintering process2,3). These defects were affected by gravity, friction and inhomogeneity of green density4–6). Therefore it is necessary to optimize the condition of each process to reduce the distortion in the case of large sized MIM components.
Moreover, powder size is also one of the most important parameters. The small particle powder shows high shrinkage and high density as compared to large particle powder during sintering process.
Purpose of this study is to apply MIM process for large and complex shaped parts. In this study, blending of both powders was conducted and the influence of powder size distribution on the distortion of comparatively complex shaped parts was investigated.
Three kinds of 316L stainless steel powders with different particle size distributions were used. Each powder is called as Fine (AKT316L(2.5), Mitsubishi steel mfg co), Medium (AT316L-H, PF-5F, EPSON ATMIX CORPORATION), and Large (UltraFine®316L, CPP3126 (−325M), Carpenter Powder Products) in this paper. The particle size distribution is shown in Fig. 1. Fine powder has narrow particle size distribution ranging from 1 μm to 5 μm. Large powder has wide distribution. The average particle size of Fine powder is 2.67 μm, that of Medium powder is 4.46 μm, and that of Large is 13.57 μm. Table 1 shows the chemical composition of each powder. Fine and Medium powders were water atomized, Large powder was gas atomized. The particle shape is shown in Fig. 2. Fine powder shows irregular particles. Medium and Large powders are spherical particle.
Particle size distribution of raw powders.
| C | Si | Mn | P | S | Ni | Cr | Mo | O | Fe | |
|---|---|---|---|---|---|---|---|---|---|---|
| Fine | 0.04 | 0.9 | 0.1 | 0.002 | 0.007 | 12.1 | 16.0 | 2.0 | 0.55 | Bal |
| Medium | 0.03 | 0.4 | 0.1 | 0.016 | 0.005 | 12.3 | 17.9 | 2.1 | 0.39 | Bal |
| Large | 0.02 | 0.4 | 1.2 | n/d | n/d | 10.5 | 16.9 | 2.2 | 0.07 | Bal |
SEM images of SUS316L powder shape (a) Fine (b) Medium (c) Large.
In this study, each powder was blended to control the distribution of particle size. The Large powder was blended with smaller powder (Medium or Fine). Volume ratios of smaller powders were 5 to 50 vol%. These samples were named as F5, F10, F50, M5, M10 and M50 with smaller particle contents. All powders were blended in a V-shaped mixer for 90 min.
Fig. 3 shows the particle size distribution of each blended powder. The blended powders which contained 50 mass% of smaller particle (F50 and M50) show bimodal distributions. Fig. 4 shows the relationship between mean particle size of blended powders and blending ratio. Mean particle size of blended powder decreased with increasing Fine or Medium powder content. However, difference between both powders is less than 1 μm. Fig. 5 shows the relationship between smaller particle content and tap density. In this case, tap density decreased with increasing Fine or Medium powder but the difference is a quite large. The smaller particle content was affected on the tap density. In the case of low tap density, it is predicted the lack of fluidity during the injection molding process. Therefore, more binder was needed for their blending.
Particle size distribution of blended powder. (a) Large and small (b) Large and Medium.
Relationship between mean particle size and smaller particle content.
Relationship between tap density and smaller particle content.
The blended powder and polymer binders were mixed and kneaded for 1 hour at 150 °C. Binder material was consisted of 49 mass% of paraffin wax, 40 mass% of atactic polypropylene, 10 mass% of carnauba wax, and 1 mass% of stearic acid. The powder and binders were mixed as 69:31 in volume%.
In order to determine the rheological behavior of feedstocks, spiral flow test was performed. In this test, F50 which showed the highest relative density in a previous study7) and M50 and L100 which were comparison with F50 were used. Fig. 6 shows the spiral flow mold. This mold cavity is 1.5 mm in thickness and 3.9 mm in width. Injection molding was carried out by a vertical type of injection molding machine with plunger. Fig. 7 plots the flow distance of feedstocks F50, M50, and L100 in a spiral flow mold for various injection temperatures. At the 100 °C, the mold was filled up with L100. Feedstock contained small powder (F50, M50) was shorter value than L100. The flow distance of F50 was shown the shortest value in this test.
Spiral flow mold.
The flow distance of feedstocks.
Fig. 8 shows the geometry and dimension of the injection molded parts. The specimen(a) is flat bar specimen. The specimen(b) is an aircraft component. The dimension of this part is about 35 mm in width, 43 mm in height and 65 mm in depth (about 100 g in weight for 316L stainless steel) which is relatively larger size compare to typical MIM parts. The part has some challenging features such as cylindrical shape, overhanging area, and sharp corner as shown in Fig. 8 (b). The cylindrical shape is challenging because it has less strength to resist the deflection under influence of gravity. The overhanging area is also challenging because it require the special setters to prevent buckling or broken during the debinding or sintering process. In addition, the crack also tends to appear at the sharp corner area during both processes.
The geometry of the specimen. (a) Flat bar specimen. (b) Comparatively large complex shaped specimen.
After the injection molding, the paraffin wax was removed in vapored heptane atmosphere at 58 °C for 4 hours. After the solvent debinding process, polymer binders were removed in thermal debinding at 600 °C for 2 hours. The projected part expected distortion of large complex specimen was supported by alumina setter during debinding process. The sintering was performed in the furnace in argon atmosphere at 1350 °C for 2 hours. Considering the influence of friction, the large complex shaped specimen was supported at one point as shown in Fig. 92).
Supporting method during sintering.
In order to consider the distortion of gravity during sintering, deflection test was performed using flat bar specimen. Distance between the supporting jigs is 40 mm. Deflection was evaluated at the center of the specimens. Fig. 10 shows the deflection test result. The specimen contained smaller particle decreased the deflection. Finally, the value of deflection made from F50 and M50 was less than 50 % of L100.
Deflection of flat bar specimens.
To evaluate the deflection and geometrical feature of the complex shaped specimens, a coordinate measuring machine (Legex 776, Mitutoyo) is used. A CMM is 3D device for measuring the physical geometrical characteristics of an object. Measurements are defined by a probe attached to the third moving axis of this machine. In this work, the line laser probe is used to measure the distortion and dimension of the specimen. The level of the deflection can be obtained from the measurement result. The measurement data is compared with the 3D CAD data to evaluate the deviation from the reference 3D CAD data.
In this study, the distortions of large shaped specimens were evaluated by deflection and out of roundness. The schematic of the measurement place was shown in the Fig. 11. Eight different positions of the deflection value were evaluated from the measured result. The positive value represents deflection outward from the surface of reference 3D CAD data, and the negative value represents the distortion inward from the surface. The out of roundness of external diameter and internal diameter were evaluated at different level of height. Fig. 12 shows measurement place for out of roundness. Measurement rounds of a–b and d–e represent the distortion of the small holes, c and f represent the cylinder section of the specimen.
Measured points of deflection.
Measurement of the out of roundness (a) The measured round (b) The measured section.
Fig. 13 shows the deflection value of comparatively large sized specimens in each process. In solvent debinding, the deflection of each specimen was quite small and did not have much difference. In thermal debinding, the deflection value increased from the solvent debinding. In the same as the previous process, the deflection of each specimen had a slight difference. However, by sintering, the deflection increased from that of thermal debinding and the deflection of each specimen showed large difference.
Deflection value at each process (a)L100 (b)M50 (c)F50.
Fig. 14 shows the out of roundness distortion in each process. The out of roundness after solvent debinding of L100 shows large value, especially at positions c and f. During sintering, the out of roundness of each specimen mostly increased from those of thermal debinding. Finally, L100 showed the largest value, while F50 showed the smallest value.
Out of roundness in each process (a)L100 (b)M50 (c)F50.
Fig. 15 shows the deflection values of three kinds of as-sintered specimens. The value of deflection was remarkably decreased by containing smaller powder. In particular, the value of deflection of F50 specimens satisfied the tolerance specified by ISO standard. It was confirmed the out of roundness value was affected by particle size distribution. In total, F50 has the lowest out of roundness value. These deflection tendencies correspond with deflection test result by flat bar specimen.
Distortion results of as-sintered specimen (a) Deflection (b)Out of roundness.
In this study, MIM process was applied to comparatively large and complex shaped parts. Moreover, the deflection and out of roundness were evaluated. Different particle sizes of powders were blended to show various particle size distributions. It was effective to blend the smaller particle powder to large particle powder for improving the dimensional tolerance. Distortion of the as-sintered F50 specimens was satisfied ISO standard.
