PARAMETER OPTIMIZATION OF CARBIDIC AUSTEMPERED DUCTILE IRON USING

P. Dhanapal et. al. / International Journal of Engineering Science and Technology Vol. 2(8), 2010, 3473-3482

linear graph is employed to identify suitable treatments and the interactions of the factors. The linear graph for the orthogonal array is shown in the fig.1.

Table 1. Factors and Levels - L32 (21 x 42)

Factor Levels Chromium Percentage 0.6 % 1.0%

Austempering Temperature 2500C 3000C 3500C 4000C Austempering Time 1 Hour 2 Hours 3 Hours 4 Hours

Fig.1.Linear Graph for L32 (21 x 42) OA.

2.2. Foundry Considerations for the Production of CADI The base composition of ductile cast irons usually is hypereutectic, where the carbon and silicon contents are typically 3.6 and 2.7 respectively [CE = 4.3]. High silicon content is to be retained to reduce the formation of the secondary carbides which will deposit in the grain boundaries and subsequent reduction of mechanical properties [3]. Thus, the first constituents to appear during solidification are graphite nodules, which nucleate and grow without any austenite, but eventually with austenite enclosing the graphite nodule.

Iron scraps raw material, turnings are melted in a medium frequency induction furnace, composition is adjusted by adding carbon, silicon and heated upto 1540oC. The melt is magnesium treated in a custom designed magnesium treatment ladle [Skaland]. The magnesium alloy consists of MgFeSi with 9wt % of Mg. The melt is transferred to pouring ladle and inoculated with FeSi (75wt% Si). Now, calculated amount of heated ferrochrome is added to increase the chromium level in the pouring ladle. Sand mold is prepared using ordinary silica and dried. The melt is poured into the prepared sand mold within the desired temperature and time. Long time interval may fad the magnesium.

Standard Y-block castings are prepared and the dimension of the casting is as per the ASTM standards. This casting production is done in a ductile iron foundry. The objective is any foundry producing ductile iron can make this CADI without altering their production line. The composition of the sample is analyzed using 40 element vacuum spectrometer and the results are shown in table.2. The material casted is called as Carbidic Ductile Iron [CDI] before heat treatment.

Table: 2. Chemical composition of the samples by wt %.

2.3. The Austempering Process

The Austempering process begins with austenitization followed by rapid cooling and maintaining that temperature for longer period. The austenitization is carried in a salt bath (Nitride & Nitrate mix) at 910oC, for two hours. Salt bath distributes the heat uniformly compared to the atmospheric heating [Dhanapl, et al (2009)]. In the Austempering process, the quenching media is held at a temperature above the martensite start temperature. Austempering temperature varies between 250oC to 400oC in nitrate salt mix.

Austempering time varies between one to four hours in a time interval of one hour. After austempering the specimens are cooled to room temperature in air quenching. This cooling rate will not affect the final microstructure as the carbon content of the austenite is high enough to lower the martensite start temperature to a temperature significantly below room temperature.

Melt Identification

C Si Mn P S Ni Cr Mo Cu Mg Zn Fe CE

CADI -0.6

3.593 2.705 0.500 0.038 0.009 0.008 0.630 0.013 0.406 0.046 0.015 91.92 4.507

CADI - 1

3.272 2.745 0.458 0.034 0.008 0.010 0.980 0.018 0.644 0.039 0.013 91.72 4.198

ISSN: 0975-5462 3475


2.4. Characterization

2.4.1. Microstructure examination Microstructure analysis is carried out on the specimens using metallurgical microscope. Specimens are polished by following standard metallographic procedures, etched in 5% nital [Collins and Watson, (1990)] and examined under optical microscope equipped with high resolution digital camera. The microphotograph is taken using the Nikon Epiphot-Dx microscope at various magnifications. The specimens are polished, etched in 10 % ammonium persulfate and the amount of carbide is measured using “Metal Plus Version-1.0” image analysis software.

Fig.2. Microstructures of (A) DI, (B) ADI, (C) 1 % Cr-CDI and (D) CADI

2.4.2. Strength and hardness Brinell hardness tests are conducted using Model B-3000 Hardness tester with maximum load capacity of 3000 kgf. Impressions are formed using 10 mm diameter steel ball indenter and 3000 kgf load. Hardness is measured at four different places; average of the 4 values are taken for consideration. Impact toughness tests are conducted as per ASTM E 23 standard at room temperature using a Charpy Impact Testing Machine with 300 Joules capacity hammer and 4.5ms-1striking velocity. Tests are conducted on unnotched test samples of size 10x10x55mm. Average of the two values is projected in the table. The impact fractured surface is analyzed using Scanning Electron Microscope of JEOL MODEL JSM 6360 with magnification minimum of 25x and a maximum of 2 lakh.

ISSN: 0975-5462 3476


2.3.3. Wear Abrasion resistance of the material is measured using Pin-on-disk wear testing machine as per ASTM standard G99-05. Disc hardness of HRC – 65, Load of 98.1N is applied to the specimen, with a travel velocity of 1m/s and a distance of 10,000m is considered for the measurement of weight loss. The weight loss values are measured by means of a 0.01 mg precision scale. 3. Results and Analysis 3.1. Orthogonal Array The taguchi a method provides laying out the experimental conditions using specially designed tables called Orthogonal Array (OA). An appropriate choice of orthogonal array depends upon the degrees of freedom. The L32 (21 x 42) orthogonal array provides the required number of experiments (shown in the table.3). This array consists of 32 rows, each representing an experiment the columns are assigned to the factor levels. Linear graph shows interactions of factors are not necessary for the experiment. The plan of experiments is made of 32 tests in which the first column is assigned to the chromium content (Cr), second column to the austempering temperate (AT) and the third column for the austempering time (At). 3.2. Signal-to-Noise (S/N) Ratio Signal-to-noise ratio where signal represents the desirable value and the noise represents the undesirable value. Therefore, the S/N ratio consolidates several repetitions into one value, which reflects the amount of variation present. There are three S/N ratios available, higher is the better (HB) is used for the Impact toughness, Hardness and the lower is the better (LB) is applied to the wear loss. The equations associated with the S/N ratios are given below.

r

i iHB yr

NS1

2

11log10/

(1)

r

iiLB y

rNS

1

21log10/

(2)

ISSN: 0975-5462 3477


Table 3. Measured responses and computed values of S/N ratio

Ex. No

Process Parameters Responses S/N Ratio Factor

A Cr

Factor B AT

Factor C At

Hardness yh

Impact Strength

yi

Wear Loss yw

Hardness Impact

Strength Wear Loss

1 0.6 250 1 414 40 30 52.34001 32.0412 -29.54243

2 0.6 250 2 387 38 37 51.75422 31.59567 -31.36403

3 0.6 250 3 444 28 28 52.94766 28.94316 -28.94316

4 0.6 250 4 444 38 34 52.94766 31.59567 -30.62958

5 0.6 300 1 401 84 30 52.06289 37.14665 -29.54243

6 0.6 300 2 422 80 71 52.50625 38.0618 -37.02517

7 0.6 300 3 380 60 41 51.59567 35.56303 -32.25568

8 0.6 300 4 406 66 26.5 52.17052 36.39088 -28.46492

9 0.6 350 1 326 72 66 50.26435 38.48559 -36.39088

10 0.6 350 2 347 84 30 50.80659 38.48559 -29.54243

11 0.6 350 3 337 66 78 50.5526 36.39088 -37.84189

12 0.6 350 4 352 70 52 50.93085 36.90196 -34.32007

13 0.6 400 1 302 58 68 49.60014 35.26856 -36.65018

14 0.6 400 2 307 54 70 49.74277 34.64788 -36.90196

15 0.6 400 3 311 68 78.1 49.85521 36.65018 -37.85302

16 0.6 400 4 311 64 32.3 49.85521 36.1236 -30.18405

17 1.0 250 1 467 26 10.9 53.38634 28.29947 -20.74853

18 1.0 250 2 495 20 11.4 53.8921 26.0206 -21.13810

19 1.0 250 3 474 19 27 53.8921 25.57507 -28.62728

20 1.0 250 4 495 18 21.8 53.51557 25.10545 -26.76913

21 1.0 300 1 444 24 14.9 52.94766 27.60422 -23.46373

22 1.0 300 2 414 22 24 52.34001 26.84845 -27.60422

23 1.0 300 3 444 18 12 52.94766 25.10545 -21.58362

24 1.0 300 4 444 20 31 52.94766 26.0206 -29.82723

25 1.0 350 1 388 20 30 51.77663 26.0206 -29.54243

26 1.0 350 2 444 12 30 52.94766 21.58362 -29.54243

27 1.0 350 3 388 18 24 51.77663 25.10545 -27.60422

28 1.0 350 4 429 14 17.2 52.64915 22.92256 -24.71057

29 1.0 400 1 435 12 16 52.76979 21.58362 -24.08240

30 1.0 400 2 420 14 23 52.46499 22.92256 -27.23456

31 1.0 400 3 495 16 26 53.8921 24.0824 -28.29947

32 1.0 400 4 454 14 27 53.14112 22.92256 -28.62728 3.3. Data Analysis The average effects of other factors are computed and shown in table.4 (a-c). This table includes the ranks based on the delta statistics, which compare the relative value of the effects. It is the difference between the highest and the lowest averages for the factor chosen. Chromium content is the first controlling factor for all responses. For higher hardness and low wear loss chromium content of level 2 is the best and for the impact toughness level 1 performed

ISSN: 0975-5462 3478


higher value. The optimum level setting for the hardness should be A2B1C4. Table.3 (a). Experiment No. 20 shows these conditions and produces the optimal hardness result.

Table.4 (a). S/N Hardness Response table

Symbol Parameters S/N Ratio

Max.-Min Rank Level-1 Level-2 Level-3 Level -4

Cr Chromium

Content 51.246 52.955 _ _ 1.710 1

AT Austempering Temperature

53.084 52.440 51.463 51.415 1.669 2

At Austempering

Time 51.893 52.057 52.182 52.270 0.376 3

Total Mean value of S/N ratio = 52.1060dB The table 4(b) shows the average of selected characteristics of each level of factor for the impact toughness. The optimum level setting for the impact toughness is A1B2C1. Experiment No. 5 shows these levels of factors and this is the optimum impact toughness among the other values.

Table.4 (b). S/N Impact Toughness Response table

Symbol

Parameters

S/N Ratio Max.-Min Rank

Level-1 Level-2 Level-3 Level -4

Cr Chromium

Content 35.261 24.857 - - 10.404 1


28.647 31.592 30.724 29.275 2.945 2

At Austempering

Time 30.793 30.020 29.781 29.747 1.0453 3

Total Mean value of S/N ratio = 30.0702dB The table 4(c) shows the average of selected characteristics of each level of factor for the wear loss. The optimum level setting for the wear loss is A2B1C1. Experiment No.17 shows this level of factors and this is the optimum wear resistant of CADI.

Table.4 (c). S/N Wear Loss Response table.

Symbol Parameters S/N Ratio

Max.-Min Rank Level-1 Level-2 Level-3 Level -4

Cr Chromium

Content -32.965 -26.212 _ _ 6.7529 1


-27.220 -28.720 -31.186 -31.229 4.0088 2

At Austempering

Time -28.745 -30.044 -30.376 -29.191 1.6307 3

Total Mean value of S/N ratio = -26.977dB The Figure 3 shows the effect of factors on the responses. Figure 3(a) shows the effect of factors on hardness. Increasing the chromium content increases the hardness. Chromium is carbide stabilizer, increases the carbides in the material and the hardness. Due to the same reason the impact toughness of the material decreases, and wear resistance increases drastically. These are shown in the figure.3 (b) & (c). Higher levels of austempering temperature reduce the hardness. At the initial levels, it is steeper and attains a constant level. Increase of

ISSN: 0975-5462 3479


austempering time increase the hardness. This line is not steep when compared to other factors; this shows a less change in the hardness. Comparing the three factors the chromium content line is steeper; which means this affects the hardness drastically.

HARDNESS

50

51

52

53

54


S/N

Ra

tio

Cr AT At

A

IMPACT TOUGHNESS

24

26

28

30

32

34

36


S/N

Rat

io

Cr AT At

B

Figure.3. Response graph of S/N ratio (a) Hardness, (b) Impact toughness, (c) Wear Loss.

Among the plots in the figure 3(a), gradient of the chromium content plot is more. The same trend is followed in all the graphs 3(b) and 3(c). Effect of chromium content is higher when compared to other factors. Gradient of austempering temperature line is less when compared with the chromium content. So, the next controllable parameter is austempering temperature and the least is the austempering time. 4. Analysis of Variance The statistical procedure used most often to analyze data is known as the Analysis of Variance [ANOVA]. This technique determines the effects of treatments, as reflected by their means, through an analysis of their variability. The ANOVA establishes the significance of factors in terms of percentage contribution to the response and is also needed for estimating the variance of error for the effects and confidence interval of the prediction error. The F-ratio, is the ratio between variance due to the effect of the factor and variance due to error term. This is used to measure the significance of factor at the desired level.

ISSN: 0975-5462 3480


Table.5 (a). ANOVA for Hardness

Symbol Source of Variation

DOF Sum of Squares

Mean Square

F-Statistics F,1% P(%)

Cr Chromium Percentage

1 123740 123740 95.77091 6.84 23.93


3 234016 78005.3 60.3737 3.94 45.25

At Austempering

Time 3 4364 1454.67 1.125867 2.13* 0.84

Error 120 155045 1292.04 29.98

Total 127 517165 100.00

Table.5 (b). ANOVA for Impact Toughness.


DOF Sum of Squares

Mean Square



1 14535 14535 287.0250 7.12 76.04


3 1610 537 10.5959 4.16 8.42

At Austempering

Time 3 135.25 45 0.8903 2.19* 0.71

Error 56 2836 51 14.84

Total 63 19116 100.00

The F-statistics of factors chromium content and austempering temperature are significant upto 99% confidence in all the responses, but the austempering time is not significant. The percentage contribution of chromium content is higher in the impact toughness and the wear loss response. A maximum of 76% contribution is attained in the impact toughness. But the percentage of contribution of chromium content is not that much for the hardness. Austempering temperature contributes more to the hardness.

Table.5 (c). ANOVA for Wear


DOF Sum of Squares

Mean Square



1 5663.14 5663.14 67.427 7.12 45.69


3 1641.91 547.30 6.516 4.16 13.25

At Austempering

Time 3 386.47 128.82 1.534 2.19* 3.12

Error 56 4703.38 83.99 37.95

Total 63 12394.90 100.00

* Not significant upto 90% confidence. 5. Conclusion The factor that affecting the performance of the material has been determined using this analysis.

ISSN: 0975-5462 3481


Chromium content in the metal is the most significant factor for all the responses. (1% Cr) Level 2 is the optimal for more wear resistance and (0.6%Cr) Level 1 is the optimum for the hardness and the Impact toughness.

Austempering Time is the least significant for all the responses. An optimal parameter level calculated by the Taguchi method coincides with the experimental results.

Acknowledgments The author likes to thank the staff members in the Department of Metallurgical Engineering, PSG College of Technology, Coimbatore. The supports of Mr.N.Visvanathan, Managing Director and Mr.V.Venkateswaran, General Manager, AMMARUN FOUNDRIES Coimbatore for their cooperation and the facilities provided to do this study are also acknowledged with gratitude. References: [1] Arron Rimmer. (2006): Furnace is the key to CADI Solutions, Foundry Trade Journal. March. PP. 58-59. [2] Batra, U.; Ray, S.; Prabhakar, SR. (2004): The influence of Nickel and Copper on the Austempering of Ductile iron. Journal of materials

Engineering and Performance. Volume 13(1) February, PP. 64-68. [3] Brezina, R.; Filipek, J.; Senberger, J. (2004): Application of ductile iron in the manufacturing of ploughshares. Research Agricultural Engg.

50, (2) PP. 75-80. [4] Collins,W.K.; Watson,J.C. (1990).: Metallographic Etching for Carbide Volume Fraction of High-Chromium White Cast Irons. Materials

Characterization 24: PP.379-386. [5] Dhanapal. P.; Premkumar,R.; Mohamed Nazirudeen,S.S. (2009).: A study on the influence of heat treatment on the wear resistance of

alloyed Austempered Ductile Iron, Indian foundry journal, Vol: 55. No: 3 March. PP. 32-38. [6] Hayrynen, K.L. (1995): ADI Another avenue for ductile iron foundries, Modern Casting, 8 PP.35-38. [7] Hayrynen,KL. (2002): The Production of Austempered Ductile iron. World Conference on ADI. [8] Hayrynen, K.L.; Brandenberg, K.R.: (2003) Carbidic Ductile iron (CADI) – The New wear material, AFS Transsactions. PP. 03-08. [9] Jadoun, R.S.and etal. (2009): Taguchi’s optimization of process parameters for production accuracy in ultrasonic drilling of engineering

ceramics. Prod. Engg. Res. Devel. 3: PP. 243-253. [10] Keough John, R.,PE.; Hayrynen,K.L. (2000): Carbidic Ausempered Ductile iron (CADI) Presented on November [11] Laino, S.; Sikora, JA.; Dommarco, RC.(2008): Development of Wear Resistant carbidic austempered ductile iron (CADI), wear. 265. PP. 1-

7. [12] Olivera Eri, and etl. (2005): An austempering study of ductile iron alloyed with copper. J. Serb. Chem. Soc. 70 (7) PP.1015–1022. [13] Osama Muhammed and Etal. (2009): Using of Taguchi method to optimize the casting of Al-Si / Al2O3 Composites. Engg &Tech Journal.

Vol.27, No.6, PP. 1143-1150. [14] Prasad Rao, P.; Putatunda,Susil K. (2003): Investigations on the fracture toughness of austempered ductile irons austenitized at different

temperatures, Materials Science and Engineering. A349 PP. 136 – 149. [15] Rossella Surace and etal (2010): Application of Taguchi method for the multi-objective optimization of aluminium foam manufacturing

parameters. Int. J. Mater Form 3. PP. 1-5. [16] Seshan,S. (1998): Austempered Ductile Iron – The under exploited wonder cast iron, Indian foundry journal October, PP. 84 – 92. [17] Torbjorn Skaland. Developments in cast iron Metallurgical Treatments. Elkem ASA Research, Norway. [18] Trudel, A.; Gagne, M. (1997): Effect of composition and heat treatment parameters on the characteristics of austempered ductile irons.

Canadian Metallurgial Quarterly, Vol.36, No.5, PP 289-298. [19] Zhizhong SUN. etal (2008).: Gating system design for a magnesium alloy casting. Jou.Mater. Sci. Tech. Vol-24, No.1.PP. 93-95. [20] Zimba,J.; Simbi,DJ.; Navara,E. (2003): Austempered ductile iron: an alternative material for earth moving components. Cement & concrete

composites. Volume 25. PP. 643-649.

ISSN: 0975-5462 3482

PARAMETER OPTIMIZATION OF CARBIDIC AUSTEMPERED DUCTILE IRON USING

Documents