The previous two blogs (#44 and #45) examined the monotonic tensile, axial fatigue and 4-point bending fatigue performance of SAE 8615 in a simulated through-carburized core and shallow-carburized conditions. This blog will expand the prior two studies to include the performance of SAE 8615 steel with a deeper case depth.
The previous blog examined the performance of SAE 8615 with a carburization of 2.5 mm depth (10 percent of gage radius) in monotonic tensile, axial fatigue and 4-point bending fatigue, including overload strain conditions. Significant increases both in monotonic tensile as well as cyclic properties of the material were observed in comparison with the simulated carburized core results. The equivalent strain versus reversals to failure graphs for axial and 4-point bending tests were similar. However, a significant difference was observed in overload conditions, with the overload axial test sample withstanding higher strain amplitudes when compared to the 4-point bending sample. Now, let us examine what happens if the carburization depth of 2.5mm is further increased to 5mm (20 percent gage radius). Again, we will be examining and making comparisons of the monotonic tensile, axial and 4-point bending fatigue cases including overload strain conditions.
As in the previous blog, the overload strain cycle applied to the specimen in this iteration consisted of repeated load blocks made up of one fully-reversed overload cycle followed by a group of smaller constant-amplitude cycles having the same maximum stress as the overload cycle. In overload testing, the overload cycles were applied at frequent intervals to maintain a low crack-opening stress resulting in the subsequent cycles being fully open. The overload cycle history is shown in Figure 1. NOL denotes the number of cycles in overload condition – one can see for this cycle, the strain amplitude is twice the subsequent small cycles. This high load simulates the real-life loading conditions of the material. εa,SC denotes the strain amplitude for small cycles, and εa,OL denotes the strain amplitude for the overload cycles.
Figure 1: Periodic Overload Fatigue Cycle
Monotonic Tensile Tests:
Tensile tests are conducted on two specimens, machined to standard dimensions. Various mechanical properties like the modulus of elasticity, yield strength, ultimate tensile strength, etc. are determined after plotting the engineering stress-strain curve, shown in Figures 2 and 3 for this steel grade. The properties are tabulated in Table 1.
Figure 2: The Engineering Stress – Strain Curve
Figure 3: True Stress-Strain Curve
Table 1: Monotonic Tensile Properties 
Comparing these values with those of the shallow carburized case (~2.5mm depth), we see a 10 percent increase in yield strength, 2.5 percent increase in the ultimate tensile strength and a 35 percent increase in the strength co-efficient. True fracture strength, percent elongation and percent reduction in area have decreased by 14 percent, 6 percent, and 6 percent respectively.
As is the case with other steels, when cyclic strains are applied to the SAE 8615 carburized steel, it first exhibits a transient response, and then stabilizes to a constant response (with time). This property is called cyclic stabilization. Figure 4 below shows superimposed monotonic and cyclic stress-strain curves. This curve is obtained after cyclic stabilization has occurred.
Figure 4: Composite Plot of Monotonic and Cyclic Stress-Strain Curves
The graph above shows the material is cyclically hardening type, with the demarcation between the two curves more clear than observed in shallow-case depth carburized steel.
Figures 5 and 6 show the true stress versus reversals to failure, and true strain versus reversals to failure. The parameters σf’, b, εf’ and c, which are calculated from these graphs, are used to characterize the cyclic properties of the material, which is tabulated in Table 2.
Figure 5: True Stress Amplitude vs. Reversals to Failure
Figure 6: True Plastic Strain Amplitude vs. Reversals to Failure
Table 2: Cyclic Fatigue Properties 
Comparing these values with SAE 8615 shallow-carburized case, we see carburization of 5mm case depth has significantly increased the fatigue strength co-efficient, cyclic strength co-efficient, cyclic yield strength and the fatigue limit values, by 25 percent, 37 percent, 23 percent and 13 percent respectively.
Effective strain-life curve, obtained when a material is subjected to periodic overload strain-controlled fatigue tests, is shown in Figure 7.
Figure 7: True Strain Amplitude vs. Reversals to Failure for Axial Fatigue Tests
The strain amplitude versus reversal to failure behavior is typical; with the fatigue life increasing as the applied strain amplitude (percent) is decreased. As expected, adding a periodic overload results in lower strain amplitude at the longer fatigue lives (overload data curve).
Bending fatigue with overload tests:
The true strain amplitude versus reversals to failure graph for the deep carburized SAE 8615 when tested in a 4-point bending configuration is shown in Figure 8 below:
Figure 8: True Strain Amplitude vs. Reversals to Failure in 4-point Bending Configuration
A comparison between the curves from equivalent axial and constant-amplitude 4-point bending tests is shown in Table 3 below:
Table 3: Equivalent Axial vs. CA 4-point Bending Values at Different Reversals to Failure
The strain amplitude levels sustained in equivalent axial and constant amplitude 4-point bending is significantly different, with the 4-point bending samples sustaining around 27 percent lesser strain amplitudes at each 1 x 103, 1 x 104, 1 x 105 and 1 x 106 reversals.
A comparison between the curves from overload axial and overload 4-point bending tests is shown in Table 4 below:
Table 4: Overload Axial vs. Overload 4-point Bending Values at Different Reversals to Failure
In the overload strain condition, remarkable differences can be observed between the axial and the 4-point bending test results. For a life of 30,000 (3 x 103) reversals, the axial specimen could sustain strain amplitude of 0.4 percent, while the 4-point bending specimen could do only 0.10 percent, a difference of nearly 75 percent. Same trend is observed in 200,000 and 600,000 reversals as well, with the differences being around 80 percent less in 4-point bending cases.