Figure 4

Sometimes the transverse fracture planes of a ductile torsional crack are readily identifiable. The smear marks of the ductile crack in Figure 4 are an example.

Figure 5

The crack shown in Figure 5 is more challenging. Multiple people have tried to convince me that this is not a crack surface at all. It is, they say, a result of the shaft fracture surfaces rubbing on themselves after the crack event. So how can you distinguish these two scenarios? Figure 6 shows a side view of the fragment of Figure 4. Note that the longitudinal grinder marks have been twisted as a result of the torsional deformation, which preceded the crack event. Often there is some sort of longitudinal feature that can be used for confirmation of a suspected torsional crack. But for the crack of Figure 5, it was QUITE challenging to find any evidence of permanent twisting of the material adjacent to the shaft.

Figure 6

Note that UNLIKE tensile loading, there is no reduction of area due to plastic deformation from torsion. The fracture surface was cut off about 1 cm in from the end, and the remaining stub was heated in 50% hydrochloric acid for about an hour. This is much longer than the usual “deep etch” time used for revealing grain flow. The lab technician gave up on revealing any features, and gave the part back to me (Fig. 7). After lengthy “staring,” I found a faint trace of a longitudinal feature that may have been the result of a shallow discontinuity in the original hot-rolled bar stock used to make this shaft. This feature is highlighted with purple arrows, while a plain red line shows the axial direction. The shaft took on a permanent twist of about 10 degrees prior to separation. This is significant deformation and is consistent with our definition of a macro ductile crack.

Figure 7

Note that the common basic shape of a macro ductile torsional crack is the SAME as that of a macro BRITTLE axial tension crack!

YOU HAVE TO UNDERSTAND THE LOADING GEOMETRY to make a valid determination of whether your part cracked in a ductile or brittle manner.

Has your heat-treating operation ever made a crack that looked like any of these transverse cracks? Did someone tell you they were brittle cracks and therefore must have been heat treated wrong?

What do you think?