UDK 621.762:539.43
Original scientific article/lzvirni znanstveni članek
ISSN 1580-2949 MTAEC9, 48(6)837(2014)
FATlGUE PROPERTIES OF SINTERED DIN S1NT-D30 POWDER METAL BEFORE AND AFTER HEAT TREATMENT
LASTNOSTI SlNTRANEGA KOVINSKEGA PRAHU DIN S1NT-D30 PRl UTRUJANJU PRED TOPLOTNO OBDELAVO IN PO NJEJ
Marko Sori1, Borivoj Suštaršič2, Srečko Glodež1
1University of Maribor, FNM, Koroška cesta 160, 2000 Maribor, Slovenia 21nstitute of Metals and Technology, Lepi pot 11, 1000 Ljubljana, Slovenia marko.sori@um.si
Prejem rokopisa - received: 2013-09-30; sprejem za objavo - accepted for publication: 2014-01-31
The main focus of this study was to determine how heat treatment affects the dynamic properties of sintered steel. All the specimens were made of the DlN S1NT-D30 metal powder, but only half of them were additionally heat treated. Flat specimens were cold pressed and sintered. The second set was additionally heat treated to increase the strength. After the static mechanical properties were determined, the fatigue strength was investigated in a pulsating machine with a load ratio of R = 0. Wöhler curves were plotted and the parameters for determining the fatigue life (of and b) were calculated.
Keywords: powder metallurgy, fatigue, S - N curve
Glavni namen te študije je bil ugotoviti, kako toplotna obdelava vpliva na dinamične lastnosti sintranega jekla. Vsi vzorci so bili izdelani iz kovinskega prahu po DlN SlNT-D30, vendar je bila samo polovica vzorcev dodatno toplotno obdelana. Ploščati vzorci so bili hladno stiskani in sintrani. Druga skupina je bila dodatno toplotno obdelana, da bi se povečala trdnost. Po ugotovitvi statičnih mehanskih lastnosti je bila preiskovana trdnost pri utrujanju na napravi za pulziranje z razmerjem obremenitve R = 0. lzrisana je bila Wöhlerjeva krivulja in izračunani so bili parametri za določanje časovne dinamične trdnosti (of in b).
Ključne besede: prašna metalurgija, utrujanje, S - N-krivulja
1 INTRODUCTION
Sintering of metal powders is becoming an interest-
ing manufacturing process for large series, due to high performance, low costs, good accuracy and smooth
surfaces. The whole process consists of three main pha-
ses (powder mixing, automatic die compaction and sin-
tering) where different variables influence the mechani-
cal properties of a sintered component that mostly depend on the porosity1-7.
Like wrought-steel components, sintered-steel parts can also undergo additional heat treatment in order to improve mechanical properties. In7, different measures are taken to improve the mechanical properties of a sintered gear. lt is shown that sinter hardening is the most appropriate method for improving the wear resistance if the price and dimensional accuracy are considered as well.
ln many studies of the fatigue behavior, sintered specimens were pressed into rectangular shapes and afterwards machined into cylindrical specimens2 6 8. In our study, as-pressed standard P/M tensile-test specimens with rectangular cross-sections (Figure 1) were tested. Sharp edges that were a result of the pressing were grinded down, but not polished. ln order to avoid heating up the specimens due to the damping effects8, the testing was done at f = 10 Hz. Consequently, after 106
cycles it became too expensive to test the fatigue properties, which is why it was not possible to determine if the fatigue limit for this powder mix exists, like the one from5, or it is not to be determined even after 109 fatigue cycles, like in8.
The main goal of this study was to determine the fatigue life of the specimens between 104 and 105 cycles with minimum machining. It was found that hardening significantly improves static mechanical properties, but
Figure 1: Test-specimen dimensions Slika 1: Dimenzije preizkušancev
the difference gradually disappears when approaching 106 cycles of fatigue life.
2 MATERIALS AND METHODS
The powder mixture used in this study can be classified as SINT-D30 according to the DIN standard.9 However, designations according to the other standards may be used: MPIF FD-020510 or JIS SMF 5040. The powder mixture used in this study was Höganäs Distaloy AB with an addition of the mass fractions w = 0.6 % of lubricant Kenolube P11 and w = 0.3 % of carbon in the form of graphite UF4. For a detailed chemical composition of the used powder mixture see Table 1, where it is compared to the standardized powders according to DIN and MPIF. Note that the MPIF standard suggests narrower limits for the alloying elements. In the last column of Table 1, the limits for the weight percentages of the alloying elements in powder mixture Höganäs Distaloy AB are also given.
Before the compaction of the specimens, the apparent density of the powder was 3.15 g/cm3 and the Hall flow rate was 29 s for 50 g. Flat specimens (Figure 1) were cold compacted with a pressure of 585 MPa and sintered for 30 min in a 10/90 hydrogen and nitrogen atmosphere at 1120 °C. After the sintering, half of the specimens were austenitized at 915 °C, oil-quenched and tempered for 1 h at 175 °C. Both sets of specimens had the final density of 7.07 g/cm3. Additional grinding of the specimens was done before the fatigue tests to remove the sharp edges, which were a result of the compaction process and could have significantly affected the results. However, the surfaces of the specimens were not additionally polished, thus, the average roughness at the thinned sections of the specimens was Ra = 0.76 ^m.
Table 1: Chemical composition of the specimens compared to the standardized SINT-D30, FD-0205 and commercially available powder mixture
Tabela 1: Primerjava kemijske sestave vzorcev s standardiziranima SINT-D30, FD-0205 in komercialno me{anico prahov
w/%
Fe
Cu
Ni
Mn
ICpnnliihp
Specimens
Bal
0.29
1 47
1.69
0 50
0 58
DIN SINT-D30
Bal
< 0.3
1 0-5 0
1.0-5.0
<06
MPIF FD-0205
Bal
0.3-0.6
1 3-1 7
1.55-1.95
0 4-0 6
Höganäs Distaloy AB
Bal
< 0.01
1 35-1 65
1.57-1.93
0 45-0 55
The static properties of the randomly chosen specimens from both sets were determined in a controlled environment at room temperature (22 °C) with a computer-controlled tensile-testing machine and a data-acquisition rate of 500 Hz. The displacement rate for all the quasi-static tensile tests was set to 0.50 mm/min. The stress-strain data was averaged and it is presented in the
results section along with the static properties for each set of the specimens.
Although dynamic tests are normally performed on rounded and polished specimens6, sintered components in practice usually do not undergo any machining before coming into use in service. Therefore, the specimens were only grinded to remove the sharp edges at the corners of the cross-sections.
Due to the rectangular section of the specimens, fatigue testing on a rotating-beam machine was not possible. Hence, it was performed on the same uniaxial machine where the static tests were done, but with a different configuration. To achieve the load ratio of R = 0, the load-control regime was induced in such a way that the maximum load was set. The loading frequency had to be set rather low, to f = 10 Hz, because the damping effects could have increased the temperature of the specimens and their cooling would not have been possible.
3 RESULTS AND DISCUSSION
The results of the static tensile tests show a good correlation to the values in the standards9-10. The Young's modulus of the sintered specimens is 130 GPa, which is the same as specified in the DIN standard and 10 % lower than in the MPIF standard for this material at a given density. For the hardened specimens a value of 142 GPa was recorded, which is 2 % lower than specified in the MPIF standard. The DIN standard does not give any value for this material after heat treatment, thus the values cannot be compared.
Figure 2: Comparison of static response for sintered and hardened specimens
Slika 2: Primerjava statičnega odgovora sintranih in toplotno obdelanih v7nr/^pv
r
The average ultimate tensile strength for the sintered specimens was 532 MPa and the elongation at fracture was 2.16 %. The hardened specimens had a much higher ultimate tensile strength, averaging at 842 MPa, and their elongation at fracture was 0.86 %. Therefore, the heat treatment had a significant effect on the static tensile properties - increasing the tensile strength and reducing the ductility (Figure 2). A comparison of these values with the standards shows that the tensile strength is higher than specified in the DIN standard only for the sintered specimens and it is lower than specified in the MPIF standard for both sets of specimens. See Table 2 and Figure 2 for a detailed comparison of the static properties.
Table 2: Average static properties Tabela 2: Povprečne statične lastnosti
Sintered specimens
DIN D30
MPIF FD-0205-52.5*
Hardened specimens
MPIF FD-0205-130HT**
£/GPa
130
130
145
142
145
^m/MPa
532
460
575
842
965
A/%
2.16
1.75
0.86
< 1
*Material designation code FD-0205-52.5 is not found in the MPIF standard. The values are linearly interpolated according to density between FD-0205-50 with a density of 6.95 g/cm3 and FD-0205-55 with a density of 7.15 g/cm3.
**Material designation code FD-0205-130HT is not found in the MPIF standard. The values are linearly interpolated according to density between FD-0205-120HT with a density of 6.95 g/cm3 and FD-0205-140HT with a density of 7.15 g/cm3.
Several dynamic tests were done at different load levels. Data points were plotted in a lg ^a - lg N diagram and afterwards the method of least squares was used to find parameters A and b in the Basquin's equation (Eq.1), which suggests a straight-line relationship in a double logarithmic graph1112:
a a = A(N)b	(1)
where Oa is the applied alternating stress for N cycles. Parameter A represents the amplitude fatigue strength for 1 cycle and it is only a theoretical value. Parameter b indicates the slope of the Wöhler line on a logarithmic scale.
For the sintered specimens, the calculated parameters were As = 494 MPa and bs = -0.121. The calculated values for the parameters of the hardened specimens were Ah = 787 MPa and bh = -0.153. However, Equation 1 is often written in a slightly different form11:
o a = O f' A(2N)b	(2)
where parameter A is substituted with 2b • Of'. Fatigue-strength coefficient Of' represents the theoretical amplitude stress at N = 0.5 and it is roughly equal to the actual tensile strength Of for most wrought-metal materials. It can be easily calculated from Equation 1, if parameters A and b are known, by inserting the value of 0.5 for N. Parameter b is the same in both Equations, (1) and (2). Fatigue-strength coefficients of's and Of h were calculated for both sets of specimens and they are 537
Figure 3: Comparison of S - N lines and marked parameters Slika 3: Primerjava linij S - N in označeni parametri
MPa and 875 MPa for the sintered and hardened specimens, respectively.
The calculated S - N lines for the sintered and heat-treated specimens are compared in Figure 3, where the values of the fatigue-strength coefficients (Of's and Of'h) and slopes of both lines (bs and bh) are marked. When comparing the fatigue strengths at 104 cycles, the calculated value from the S - N line is 192 MPa for the hardened specimens and 162 MPa for the sintered specimens. From Figure 3 it is evident that the difference in the fatigue strength gradually dissipates. The fatigue strength at 105 cycles is almost the same for both sets -135 MPa for the hardened and 123 MPa for the sintered specimens. Figure 3 also suggests that the S - N lines would cross each other after 106 cycles, but this is inconclusive, because there are no data points after 106 cycles. Therefore, on the basis of the available data, the amplitude strength cannot be determined at 106 cycles for either of the specimens and additional testing should be performed to find if the S - N lines cross each other after 106 cycles.
4 CONCLUSION
The main purpose of our study was to compare the median fatigue strengths of sintered and additionally hardened specimens between 104 and 105 cycles with a load ratio of R = 0. Before the dynamic testing, monotonic tensile tests were done comparing our values with the standard values for sintered materials910. The results showed a good correlation with the standard values, with some deviations that may have been caused by many variables in powder metallurgy (density, size, distribution of pores, chemical composition, sintering temperature, cooling rate from sintering temperature, etc.). The ultimate tensile strength for the sintered and hardened specimens was found to be 534 MPa and 842 MPa, respectively. Heat treatment also decreased the elongation at breakage from 2.16 % to 0.86 %. Therefore, for both wrought materials and sintered metals, hardening increases the strength and decreases the ductility.
2
Thereafter, the fatigue strength was investigated in a pulsating, load-control machine with the load ratio set to R = 0 at a frequency f = 10 Hz. The acquired data was then used to calculate the parameters in the Basquin's equation12 with the least-squares method and S - N lines for both sets of specimens were plotted in log-log diagrams. It turned out that, even though heat treatment increases the static strength with the difference being noticeable at 104 cycles, the slope of the S - N line suggests that in the case of high-cycle fatigue, heat-treatment contributions to the fatigue strength of the chosen sintered material are negligible. However, more testing should be performed to investigate the fatigue behavior after 106 cycles because the data from this study alone is insufficient.
The fatigue limit was not determined in this study because the testing was interrupted when the cycle counter surpassed 106 cycles. Furthermore, the testing up to 107 cycles or beyond at the frequency f = 10 Hz would be very expensive. However, the MPIF standard does give a guiding value for FD-0205-HT regarding the axial fatigue limit for 90 % survival after 107 cycles, which is 310 MPa at a load ratio R = -1, determined at f = 100 Hz. A comparison of this value with the data from our study is not possible because the tension mean stress (R > -1) reduces fatigue life.11
5 REFERENCES
1 N. Candela, F. Velasco, M. A. Martinez, J. M. Torralba, Influence of microstructure on mechanical properties of molybdenum alloyed P/M steels, Journal of Materials Processing Technology, 168 (2005), 505-510
2N. Chawla, X. Deng, Microstructure and mechanical behavior of porous sintered steels, Materials Science and Engineering A, Structural Materials: Properties Microstructure and Processing, 390 (2005), 98-112
3 W. Khraisat, L. Nyborg, Effect of carbon and phosphorus addition on sintered density and effect of carbon removal on mechanical properties of high density sintered steel, Materials Science and Technology, 20 (2004), 705-710
4H. Danninger, D. Spoljaric, B. Weiss, Microstructural features limiting the performance of P/M steels, International Journal of Powder Metallurgy, 33 (1997), 43-53
5	M. Kabatova, E. Dudrova, A. S. Wronski, Microcrack nucleation, growth, coalescence and propagation in the fatigue failure of a powder metallurgy steel, Fatigue & Fracture of Engineering Materials & Structures, 32 (2009), 214-222
6	S. J. Polasik, J. J. Williams, N. Chawla, Fatigue crack initiation and propagation of binder-treated powder metallurgy steels, Metallurgical and Materials Transactions A: Physical Metallurgy and Materials Science, 33 (2002), 73-81
7	W. Predki, A. Miltenovic, Influence of Hardening on the Microstruc-ture and the Wear Capacity of Gears Made of Fe1.5Cr0.2Mo Sintered Steel, Science of Sintering, 42 (2010), 183-191
8	M. Dlapka, H. Danninger, C. Gierl, E. Klammer, B. Weiss, G. Kha-tibi et al., Fatigue behavior and wear resistance of sinter-hardening steels, International Journal of Powder Metallurgy, 48 (2012), 49-60
9	DIN, Sintered metal materials, Part 4: Materials for structural parts, Berlin, 2010
10	MPIF, Standard 35, in Material Standards for PM Structural Steel, Princeton, 2007
11	R. I. Stephens, A. Fatemi, R. R. Stephens, H. O. Fuchs, Metal fatigue in engineering, 2nd ed., John Wiley & Sons Inc., New York 2001
12	O. H. Basquin, The Exponential Law of Endurance Tests, Proceedings of American Society for Testing Materials, 10 (1910), 625-630