While YS and microhardness (and their respective ±σ) were not expected to be affected, as the T7 conditions and the heat treat furnace used were exactly the same, both plastic elongation and UTS dropped due to the elevated porosity and both exhibit large ±σ. The true spread on the results is best seen looking at the distribu- tion and Weibull plots of Figures 8a, 8b, 9a and 9b. The effect of porosity on plastic elongation and UTS is due to the reduction of load bearing cross-sectional area, and the elevated stress concentration effect of large shrink- age pores during loading.5, 6, 7
Reducing the porosity sig-
nificantly, with liquid metal degassing, but keeping the same tensile test sample orientation with respect to the chill, will significantly increase the values of the plastic elongation and UTS, and lower the ±σ. When the tensile test sample’s orientation changed slightly in iteration #3, the value of λ2
and porosity dropped further and the
plastic elongation and UTS increased, but ±σ appeared to be similar to what was determined from iteration #2. The Weibull plots seen in Figure 8a and 9b certainly re- flect that more than an increase in plastic elongation and UTS occurred, but rather that the reliability increased considerably, a trend difficult to assess with the value of ±σ alone.
When Chill #2 is used, and the consistency in metal con- dition, test sample placement with respect to the chill, and tensile test conditions are identical, with only the source for machining the tensile test sample and load frame used being different. The variation expected in this case would be due to the position of the test bar along the length of the extracted blank, and any acceptable differences in the load frame, grip end clamping pressures, etc., between the two testing labs having the same accreditation under A2LA. For the three iterations using the Chill #2, the mi- crostructure as reflected in Table 4 is reasonably simi- lar, all ±σ overlap. Figure 9b indicates that the Weibull modulus essentially yields similar results. However the Weibull modulus for iterations #5 and #6 which had one lab responsible for sample preparation, are similar (m = 82.6 and 84.6), and both different from the modulus cal- culated for iteration #4 (m = 63.4), Figure 9b. The au- thors believe that the reason for this is the placement of the tensile test sample machined from the section blank relative to the chill. One end of the sectioned blank is the end that was chucked, while the opposite end was used to fabricate the tensile sample. The length of the chuck determined how far along the section blank (and of course
along a gradient of microstructure feature sizes such as λ2 and porosity) the reduced gauge section of the finished tensile test bar was. Due to the chill effect in fact, the size of the test blank section sampled a gradient of 10 µm in λ2
tests will be done on tensile samples fabricated by one of the two labs to eliminate this variation. The alterna- tive choice is the test sample machining where a chuck is not required, and a test sample template used for con- sistent sample placement to encompass the whole tensile test sample. This alternative would presumably improve consistency for both labs when machining test samples separately.
Tensile testing, and fatigue testing for that matter, have a high sensitivity to discontinuities as reflected in distribu- tion and Weibull plots, whereas the metallography reflects not only the bulk of the local microstructure but also the 2-D measurement of microstructure features such as po- rosity. The SEM images of Figure 4b and 4c showed two adjacent ~ 500µm shrinkage pores, and are presumed to be connected, when metallography reported a 250.5 ± 97 µm (largest measured in one sample at 450µm). As a re- sult, Weibull plots for metallographic features may yield a “hit and miss” ability to correlate with the resulting tensile properties, particularly when the spread between iterations (e.g. 5, 6 and 7) is very small.
Conclusions
1. Dissolved hydrogen levels below reported concen- tration thresholds are required to not only improve UTS and plastic elongation, but to improve the reli- ability of predicted mechanical properties, as dem- onstrated by the Weibull plots.
2. Test sample extraction from a casting section that exhibits a gradient in microstructure requires consistency, as this will produce measurable ef- fects on averaging results, and on reliability as demonstrated by the Weibull plots. Standard deviations were determined to be similar for the two conditions.
3. Changing the chill power (Chill #1 to Chill #2) resulted in a microstructural refinement, driving up ductility and improving reliability of predicted mechanical properties, as demonstrated by the Weibull plots.
4. With complete control of the casting conditions, followed by tensile testing practices covered by ac- creditation, the authors were able to get reasonable repeats (3X) in the average and standard deviations of YS, UTS and plastic elongation. However, with the use of the Weibull plots, consistency in sample preparation was believed to be the factor needed to improve the inter-lab agreement.
Acknowledgements
, thus it would be expected that not only a template be used to section the test bar for sample consistency, but that the allowable size of the chuck-end used is a factor as well. For the two A2LA laboratories used in this study for inter-laboratory testing it was deemed that future inter-lab
International Journal of Metalcasting/Winter 10
The authors would like to thank A. Rivera, A. Vespa, N. Meloche, B. Lane, I. Veska for the technical support in sam- ple reparation and testing. Michelle Martin is thanked for the critical evaluation of the manuscript.
43
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