by clicking the arrows at the side of the page, or by using the toolbar.
by clicking anywhere on the page.
by dragging the page around when zoomed in.
by clicking anywhere on the page when zoomed in.
web sites or send emails by clicking on hyperlinks.
Email this page to a friend
Search this issue
Index - jump to page or section
Archive - view past issues
button in toolbar for more information.
Metal Casting Technologies : September 2007
63 0.04 - 0.12%, while compacted graphite iron is produced in a much tighter range of 0.01 - 0.025% after treatment. In CGI magnesium and/or rare earths, Ca, Ti and Al are added to molten metal to consume both sulfur and oxygen; therefore, the higher the sulfur content the more alloying elements are required for melt treatment. Because magnesium is a strong sulphide former, magnesium sulphide (MgS) inclusions are formed preferentially to manganese sulfide (MnS), which has been reported to act as a lubricant during the machining process of gray iron. A few publications have addressed reduced life expectancy of cutting tools when machining CGI1,4 but as published by Skvarenina et al.5 laser-assisted machining seems to be a feasible alternative for CGI machining. This paper addresses the microstructure transition of CGI at the mold/ metal reaction interface. The potential of flake skin formation at the casting surface jeopardizes the properties obtained in the bulk, and this can become especially important in critical application castings. Measurement methods for evaluating the depth and severity of the skin formed are presented together with the different metal chemistries studied and respective observations. EXPERIMENTAL PROCEDURE The test piece selected for the experiments was the penetration casting, an adaptation of the design used by Murton and Gertsman6 for the evaluation of surface defects, penetration and veining. It is a bottom gated three-part mold of an 8" cylinder diameter, which accommodates four 2"x 2" cylindrical cores, and features a center located 8" riser. A sketch can be seen in Figure 2. The mold is made from silica sand bonded with a phenolic urethane no-bake binder. The advantage of this design is the capability to test up to four different combinations under the same conditions. For the production of CGI on laboratory scale, the SinterCast Mini-System 2000, technology developed by SinterCast AB of Sweden, leader in the know-how for reliable high-volume production of CGI, was used. This system utilizes high-resolution thermal analysis and software. The operational procedure consists of additions of FeSiMg, rare earth, and magnesium and inoculation wires to control and correct the melt to the desired chemistry followed by sample checks using a patented engineered sampling cup device that provides a representation of both modification and inoculation indexes. Additional information on method and apparatus can be reviewed in reference 7. Heats of 300 lb. of molten metal were prepared using a coreless induction furnace. The standard charge was approximately: 50% sorel (nodular pig iron), 40% CGI returns, 8.5% steel and 1.5% 75%FeSi. Small additions of carbon raiser and alloying elements were made to adjust the composition. The target base iron chemistry monitored by a QuiK Lab unit was: 4.45% CE, 3.8% Ctotal and 2.0% Si; sulfur level was kept at about 0.015%. Tapping from the furnace into the ladle occurred at about 1500(C (2730(F). FeSiMg and rare earth were used for base iron treatment, and magnesium and inoculant wires were used for desired corrections as proposed by the SinterCast method. Once the base iron was treated and corrections were made to guarantee the metal composition within the CGI window, the molten material was poured at about 1370(C (2500(F) into the molds. The initial test comprised of a total of 12 molds poured in three days, four molds were cast per day (heat). Each mold had four cores, two were coated with a sulfur containing coating and the other two were left uncoated. The purpose of the selection of a sulfur containing coating was to investigate graphite morphology at the casting surface. The molds were made with silica sand, GFN 55-60, and the cores made with zircon sand, GFN 108-112. The use of zircon cores was to minimize casting surface defects, such as veining, that can cause localized disturbances and interfere with the skin measurements. The castings were cross-sectioned and samples were cut in an L-shape around the opening left by the cores. Two samples were taken around each core, one was closer to the riser (I: inside) and the other was on the opposite side (O: outside). For details see Figure 3. Samples were prepared for optical microscopy and photographed at 50X along 11/2" length extension. About 30 frames were recorded per sample. Images were aligned and depth of skin formation determined. Results were graphed and skin formation trends determined. In order to investigate microstructure transition, three heats were conducted. Heat 1 was designed to obtain a good CGI microstructure, heat 2 can be characterized as CGI of higher nodularity, and heat 3 was treated to yield ductile iron composition with nodularity >90%. The differences in chemistries were controlled by magnesium wire addition during Figure 2. Sketch of a 2"x 2" penetration mold Figure 3. A cross-sectional overview of sample preparation WHO'S WHO OF METALS -- ANNUAL 2007/8