material options, Isomol, Zircoat and Magcoat are used as coatings. Two magnesium alloys, AZ91D and SC1 and one aluminium alloy A356 are the cast materials. As all these alloys have a similar melting and relatively similar ranges of pouring temperatures, temperature could be considered as yet another factor, with a narrow range of variation from 690 to 770C (1274-1418F). Each experiment was repeated four times and the L9 experimental design and average re- sponses are all shown in Table 1. Compositions of different alloys considered are given in Table 2. Normally, experi- mental designs such as these aim at establishing optimum process conditions for the best values of end responses, in which case, different alloy systems would not be combined in the same experimental design. But the overall aim of the current experimental investigation is limited to find if any of these alloys exhibit any loss of essential mechanical charac- teristics when cast in RP moulds. These are initial trials to get an overall impression of the performance of the moulds, and so, cast metals are included as one of the variable factors in the Taguchi L9 experimental design.
The Signal to Noise (S/N) ratio is the measure of the robust- ness of a process and can measure the amount of variation due to uncontrolled (noise) factors. The S/N ratio is actually the transformation of the Mean Squared Deviation (MSD) which measures both the average and the standard deviation. The transformation takes into account the application; the higher the better (strength) or the lower the better (surface roughness) types of situations. The ‘Higher is the better’ S/N ratio is used to maximise the response and ‘the lower is the better’ S/N ratio is used to minimise the response.
The moulds for the current experiment are designed based on certain geometric relationships and experiences from prior testing. The final mould design, as shown in Fig. 1, comprises of a top gated inlet, cylindrical specimens for the final cast- ings, and a feeder situated at the end of the cylindrical cavities to allow for metal shrinkage, during solidification. The mould was parted to form the upper and lower halves, located by means of aligning studs at all corners. Clamps are used at the centre to prevent any movement due to the buoyant forces of the molten metal. Vent holes are also added along the length of the cylindrical specimens to improve permeability and help release gas build-up if any, inside the cavity.
134a (R134 Refrigerant) as per the recommendations by ASM, American Society for Materials. Foseco MAGREX 36 flux was also used to stop any violent combustion once the metal was brought up to its melting point and as a means of controlling fire incidents if any. The molten metal was filtered with the use of a 10 PPI ceramic foam filter near the bottom of the ceramic pouring cup. The pouring cup was also coated with the mould dressing to avoid any reactions with the molten magnesium.
S/NLB yi
Where: S/NHB
The casting trials were conducted using the induction furnace available at AUT University, with a custom made 1040 car- bon steel crucible constructed for the purpose of melting the two magnesium alloys, as a refractory crucible might lead to unwanted reactions. The temperature of the crucible was con- trolled by the machine controls and a CKY 500 K type thermo- couple was used to check the temperature of the liquid metal before pouring. Each 3D printed mould was baked at the op- timum values of time and temperature obtained from previous experiments. The foundry sand used was silica sand bonded by an ester hardened alkaline phenol-formaldehyde polymer resin. In the case of magnesium castings, the molten Mg was cov- ered by an inert gas, in this case a mixture of CO2
and HFC
= Higher is Better (HB) Signal to Noise Ratio = Lower is Better (LB) Signal to Noise Ratio
= Experimental Response r = Each Response Repetition
Once each casting was completed, the risers and sprue were cut off using a hacksaw and each cylindrical specimen was turned on a lathe into a tensile test specimen. Before ma- chining however, the cylindrical cast specimen surface was tested for quality by Taylor Hobson Form Talysurf50. Ma- chining of the tensile specimens was then conducted with low speeds and feed rates and a coolant to avoid excessive
Figure 1. The final mould design for the current experiment is shown above. (Left) View of bottom half of mould (Right) 3-D view of the mould assembly
International Journal of Metalcasting/Summer 2011
Sectional view
29
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