), the % graphite and the nodularity (in Number % or in % Area of Nodules) of the sample. This step is of crucial importance since, as it will become evident in this article, it will be the key in ensuring the results are meaningful and will be comparable to those obtained by a third party. The five main parameters which the user must specify to the IA system are: magnification, trap size, num- ber of fields observed, shape factor type and magnitude. Sec- tion 2.1 defines these parameters and describes their impact on the final result.
“features” are round graphite particles, malformed graphite particles, other inclusions, scratches (if any), and cavities. A sensitive gray scale filter is then applied on the images to distinguish graphite from the matrix and from other types of particles (and cavities). The IA system then applies the parameters selected by the user to calculate the nodule count (in Nodules/mm2
Image Analysis Parameters Magnification
The first parameter to fix is the magnification. With the cur- rent camera resolution available on the market (typically 0,6−0,8 µm/pixel), a magnification of 100X is generally thought of as an adequate compromise between acquiring a sufficient number of graphite particles to have decent sta- tistics and reproducibility, and the ability to distinguish the contour/shape of the graphite particles with an adequate number of pixels. However, when trying to characterize a thin-wall DI part, it is likely that the nodules will be small (short solidification/growth time available) and in that case, a 200X magnification might give better results when cou- pled with a smaller trap size.
Trap Size
The trap size is a feature offered by IA software which al- lows the user to eliminate particles smaller than a pre-de- fined size. This way, very fine surface imperfections (e.g. scratches) are not counted as graphite particles. For example, if a user selects a trap size of 10µm, the software will apply and rotate a 10µm x 10µm cage on each particle observed. If the particle fits entirely inside the cage, the particle will be eliminated from the count. On the other hand, if the particle does not fit inside the cage, then it will be kept and a shape factor algorithm will then be applied on it to determine if it is “round enough” to be called a nodule.
In general, the smaller the trap size selected, the higher the nodule count that will be obtained.6
Although this parameter
is only rarely reported when image analysis results are pub- lished, it is nevertheless one of the most important parame- ters to fix. There are currently no standards which determine a trap size to use. However, most laboratories tend to use a 10µm trap size (at 100X) for standard DI microstructural examinations and a smaller trap size (for example 5µm or less, as agreed upon with the buyer of the DI part) when the microstructure is composed of very fine nodules (1 - 15µm); as is the case for thin wall DI castings.
52
Number of Fields Analyzed
With each field of view being slightly (or sometimes signifi- cantly) different from each other, it is important to strike an adequate compromise between the rapidity of the analysis and the variance and statistical significance of the result ob- tained. Many labs use 25 as the adequate number of fields to use, which represents the analysis of several thousand graphite particles. Other labs, especially those without IA capabilities, examine only one or two fields of view to deter- mine if the cast part is in conformance.
Shape Factor
Once a feature is determined to be too big to be eliminated by the trap size, it is considered to be a graphite particle. However, the IA system has not yet determined if that par- ticle is a nodule or not. To be called a nodule, a graphite particle must be “round enough” to meet the mathematical criteria selected by the user. The image analysis software can use basic measurements such as area, perimeter and feret (measurement of an object’s size along a specific direction), and apply various mathematical relationships (i.e., shape factors) to determine if a particle is round. The most com- mon relationships are:
a) Sphericity Shape Factor: Ratio of area to particle perimeter.7
Sphericity Shape Factor = 4πAparticle / Perimeter2
b) Compactness Shape Factor: Ratio of area to con- vex perimeter.8
Compactness Shape Factor =4πAparticle rimeter2
/Convex Pe-
Figure 1. Schematic representation of the Sphericity shape factor.
Figure 2. Schematic representation of the compactness shape factor.9
International Journal of Metalcasting/Volume 8, Issue 2, 2014
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97