Thermal cycling and heat checking: where the cracks come from and how to slow them

The peer-reviewed review of heat checking in Journal of Materials Research and Technology (2023) is consistent with the older Engineering Failure Analysis literature on the underlying mechanism. A hot billet contacts the die surface and drives a fast surface temperature spike of several hundred degrees Celsius above the bulk. The surface wants to expand. The bulk underneath is colder and will not expand at the same rate, so the surface is held in compression by its own substrate. The billet releases. The surface cools faster than the bulk through radiative loss and lube-spray contact. Now the surface wants to contract against a bulk that has been warming up under it, and the surface is pulled into tension.

Every forging cycle puts the surface through one compression-then-tension swing. The peak strain on each side of the cycle is below the yield strength of tempered H13 on a clean cold start, which is why a fresh die does not crack on hit one. The plastic strain that does accumulate per cycle is small. It also accumulates without resetting. After a few thousand cycles the cumulative plastic strain on the surface exceeds the local ductility, microcracks initiate at grain boundaries and at carbide-matrix interfaces, and the crazing pattern starts to appear. The Calvo-García group at the University of Vigo (Materials, 2022) measured a high-cycle fatigue limit of roughly 980 MPa at 10⁷ cycles on H13 in dry conditions, with the finer-carbide B-grade outperforming the A-grade by a factor of two under corrosion-fatigue. The microstructure under the surface decides how fast the crazing starts. The thermal cycle decides how fast it grows once it starts.