2. Experimental Procedure
In this paper, a solid solution strengthened ferritic ductile iron with 3.25 wt % Si was investigated. Melt was prepared in medium frequency induction furnace from pig iron, steel scrap, and ductile cast iron returns. The spheroidizing treatments and the preconditioning were performed in a dedicated ladle by adding 1.2% FeSiMg commercial alloy, containing 1 wt % RE, and 0.3% inoculant (75 wt % Si, 1 wt % Ca) using the sandwich method. After the spheroidizing and inoculation process, and just before pouring the iron into the molds, a metal sample was analyzed by optical emission spectrometry to determine the chemical composition. At the same time, in order to complete the alloy characterization, a thermal analysis was carried out. A standard cup for the thermal analysis, containing the same weight percentage of inoculant of the castings was filled and the cooling curve was then measured by using ITACA MeltDeckTM (ProService Tech, Borgoricco (PD), Italy).
The final chemical composition of the material is shown in
Table 1.
The carbon content was chosen to be 3.3 wt %, in order to maintain a near eutectic composition, with the Equivalent Carbon calculated using the equation:
[
25].
With the aim to evaluate the effect of increasing solidification times on microstructural and mechanical properties, three different geometries, with increasing section thickness were produced in a green sand automatic molding line. 15 molds were produced, each of them containing cast samples with geometries taken from the UNI EN 1563:2012 standard [
1]. In particular, the round bar-shaped (type b), the Y-shaped type III and the Y-shaped type IV were used, the relevant wall thicknesses of which were 25, 50, and 75 mm respectively.
From the round bar samples, only tensile test specimens were machined, while from the Y-shaped samples fatigue specimens were also obtained. In particular, the tensile test specimens were characterized by a net diameter of 14 mm, while the fatigue specimens had a rectangular net cross section of 10 × 15 mm2.
In order to evaluate the influence of increasing solidification time on the mechanical properties, specimens were taken from the three different cast samples. In particular, in the case of round bar shaped samples with a diameter of 25 mm, the tensile specimens were directly machined. Differently, in the case of Y-shaped samples, a block of about 25 × 25 × 175 mm
3 was cut before the final machining. The positions where the specimens were taken from are shown in
Figure 1; it can be noted that from each type IV sample, two specimens were obtained.
In order to compute the solidification time within each cast sample, numerical analyses were carried out by using the code Novaflow & Solid® (Version 4, Novacast, Ronneby, Sweden). The temperature dependent material properties for ductile iron and green sand have been taken from the database of the numerical code. A size element of 2.5 mm was used for the mesh. The temperature history measured by a virtual thermocouple positioned at the centre of each zone (
Figure 1) was used to calculate the solidification time of the whole zone. Tensile tests have been conducted at room temperature according to ISO 6892-1:2016 [
26] by using the INSTRON 5500R (Instron, Norwood, MA, USA) tensile test machine under strain rate control. Fatigue tests have been performed according to ASTM E468-18 Standard. A resonant testing machine (RUMUL Testronic 150kN, Russenberger Prüfmaschinen AG, Neuhausen am Rheinfall, Schweiz) was used with a sinusoidal tensile pulsating load at the frequency of about 130 Hz and nominal load ratio R = 0. Tests have been stopped at the total separation of the two parts of the specimens, or after reaching 10
7 cycles. The staircase method was carried out with an applied stress increment of 10 MPa in order to evaluate the fatigue strength corresponding to a fatigue life of 10
7 cycles.
Fatigue tests have been conducted on plain specimens taken from Y-shaped type III and IV cast samples. In particular for each sample, specimens were taken from three levels, numbered consecutively from 1 to 3 going towards the thermal center of the casting, as shown in
Figure 2.
In the case of type IV sample, six specimens have been obtained, while, due to the smaller dimension, it was possible to take only three specimens from each type III sample. A total of 60 and 30 specimens have been tested for type IV and type III sample, respectively.
Microstructural parameters, such as nodule count, nodule size, nodularity and matrix structure have been evaluated according to the ASTM E2567-16a standard, using an optical microscope and an image analysis software, on polished samples in the unetched and etched (Nital 5%) conditions.
Finally, the fracture surfaces of some broken specimens under fatigue loading have been examined using a scanning electron microscope (Quanta 2580 FEG, FEI, Boston, MA, USA).