November AMP_Digital

FEATURE A D V A N C E D M A T E R I A L S & P R O C E S S E S | N O V E M B E R / D E C E M B E R 2 0 1 9 6 4 15 no further increase in fatigue strength [2] . The change in frac- ture mode to brittle failure limits the load carrying capability of quenched and tempered components, leading ASPPRC to identify microstructural improvements that could overcome this limitation. The initial investigation looked for ways to reduce aus- tenite grain size andmodify element segregation at austenite grain boundaries where the brittle fracture occurs. Previous work shows that refining the austenite grain size through multiple heat treatments on carburized steel parts can lead to an increase in bending fatigue strength [3] . However, the multiple heat treatment process is not practical due to dis- tortion and high cost. Induction heating offers the potential to reduce austenite grain size and limit segregation of substi- tutional elements to austenite grain boundaries due to the very short heating times. During induction hardening, heating times are reduced to seconds rather than the hours required for conventional carburizing. An example of an induction hardening cycle using a two-turn scanning coil for a 2-mm case depth on a 38-mmdiameter shaft is shown as the solid line in Fig. 2. LABORATORY SIMULATIONS ASPPRC used Gleeble 3500 test equipment to dupli- cate thermal cycles shown by the dashed line in Fig. 2 on test specimens large enough to enable characterization of material properties. Induction hardening of a series of alloys (11-mm 2 specimens) was simulated using a range of ther- mal heating cycles involving heating at 50°C/s to austenitiz- ing temperatures between 850° and 1050°C for times of two to 1000 seconds. Specimens for one group of tests contained nominally 0.56%carbon to ensure achieving hardness levelswell above the normal range where brittle fracture is common. The steels contained additions of Mn, Mo, Ni, and W as shown in Table 1, andwere previously heat treated by quenching from 900°C and tempering at 200°C. The steels also contain 0.011- 0.015% P, 0.009-0.013% S, 0.23-0.25% Si, 0.002-0.009% Al, and 0.0063-0.0074%N. No grain refiners such as Nb or Vwere added to the steels. Fig. 2 — Example of induction hardening thermal cycle (solid line) for surface hardening [4] . After simulated induction hardening, a sharp notchwas cut in the center of the test specimen using electro-discharge machining (EDM), and notched specimens were loaded to failure using three-point bending on a servo-hydraulic test frame as shown in Fig. 3(a). Areas evaluated on the fractured test specimens are shown in Fig. 3(b). EXPERIMENTAL RESULTS AND DISCUSSION Figure 4 shows measured austenite grain sizes for test-steel Mo. Austenitizing times at two and 10 seconds are consistent with induction hardening cycles, whereas the 1000-second austenitizing time is consistent with conven- tional furnace heating. Note that austenite grain sizes for the simulated induction cycles are much smaller than those for simulated furnace heating cycles. It is also important to note that restricting the simulated induction austenitizing tem- perature to below 900°Cmakes it possible to achieve austen- ite grain sizes smaller than 10 μm (ASTM 10). Figure 5 shows the relationship of peak bending loads to the austenite grain size during the fracture tests for test-steel W. For grain sizes larger than about ASTM 7 (32 μm), peak breaking loads were less than 7000 N and TABLE 1 — CHEMICAL COMPOSITION OF TEST STEELS Test-steel ID Element(a), wt% C Mn Ni Mo W DI, mm Base 0.57 0.80 0.01 0.00 0.001 28 Mn 0.55 1.44 0.00 0.00 0.003 46 Ni 0.55 0.79 2.12 0.00 0.008 50 Mo 0.58 0.80 0.01 0.25 0.001 49 W 0.56 0.77 0.00 0.01 0.560 – a) Values in red indicate targeted element content to evaluate induction hardening response. 14