Thermal Fatigue Cracking in Die Casting Dies

The four ways a die fails

The source lists four main failure modes for an aluminium die casting die. Physical erosion, also called washing, where a fast flow of melt scours the surface and gets worse when hard particles are carried in the melt. Chemical attack, or corrosion, where the melt dissolves the steel and forms interphase layers when the relative motion between the steel and the melt is small. Gross cracking, a cleavage fracture through the die that can be complete and sudden, and that starts from a combination of thermal and mechanical stress. And thermal fatigue cracking, also called heat checking, which is the slow crack network that this page is about.

How the stress is built and released in each cycle

Thermal stress appears when the free expansion or contraction of the steel is restrained. In a die the restraint is mostly internal, because a temperature gradient across the section means the hot surface wants to expand more than the cooler mass behind it. For a bar with fixed ends the stress from a temperature change is the expansion coefficient multiplied by the elastic modulus and by the change in temperature. In a biaxial condition the same product is divided by one minus the Poisson ratio. Both the expansion coefficient and the modulus change with temperature, so the full stress never develops, but the gradient is what drives it.

When the melt is injected, the surface is put in compression against the cooler interior. When the part is ejected and the lubricant is sprayed, the surface cools faster than the interior, the compression is relieved, and tension appears. Once the local deformation turns plastic, cycle after cycle, a crack starts. The crack then grows along a plane at right angles to the largest tensile stress, through the stages of initiation at the surface, linking of small cracks at the surface, growth of the crack network into the depth, and finally the growth of the largest crack until the die fails.

Longer contact with the melt raises the temperature and the damage

The source test dunked premium grade H13 specimens into molten aluminium held at 1350 F for 5, 7, 9 and 12 seconds, with the specimen withdrawn for 24 seconds between immersions and water run through the internal cooling line throughout. Immersion time was the only variable. Cracks were measured after 5,000, 10,000 and 15,000 cycles. The table gives the results at 15,000 cycles. The peak temperature rises by 221 F from the shortest to the longest immersion, and the total crack area rises from 1.97 to 167.72, about eighty five times larger.

The effect of immersion time on the temperature at the corner of a premium H13 specimen and on the cracking measured after 15,000 cycles. Immersion time is the only variable, and the cooling line was 1.5 in in every case.

Hardness falls from the corner inward

The cracking follows the softening, and the softening follows the temperature. The table below gives the hardness measured from the corner inward at 15,000 cycles for the same four immersion times. The 5 and 7 second specimens hold most of their hardness to within 0.1 in of the corner. The 9 and 12 second specimens lose twenty or more points at the corner and are still soft at 0.2 in into the steel.

Hardness in HRC measured from the corner inward on the four specimens after 15,000 cycles. The greater the immersion time, the deeper and softer the softened band.

Estimating the stress at the corner

With the expansion coefficient and modulus of H13, 6.9 micro inch per inch per F and 30,500 ksi, the axial stress range at the corner of the 12 second, 1.5 in cooling line specimen works out at about 145 ksi, using the temperature range measured at the corner. That figure agrees closely with the computer model the source ran for the same specimen. During immersion the corner is in high compression, and because the yield strength of the steel falls at temperature, that compression can go plastic. What is left behind is a residual tension, well below the yield strength but high enough to start a crack at any stress raiser such as a machining mark or a corrosion pit.

What this means on the job

The steel in these tests is premium grade H13, the chromium hot work steel covered on the H13 tool steel page and held as die stock on the 1.2344 ESR H13 supply page. The general resistance of the hot work grades is compared on the thermal fatigue resistant tool steels page, and why H13 is the default for a die is set out on the why H13 steel is used in die casting page. The cooling design that holds the peak temperature down is on the die cooling line design page, and the softening behind the crack on the H13 die softening page. The other failure modes are on the die wear and failure page and the tool and die failure analysis page. The loss of hardness at temperature is on the hot work softening resistance page, the properties that set the gradient on the thermal conductivity page, and the way a grade is chosen for a press on the high pressure die casting selection page.

Source, Effect of Design Factors on Thermal Fatigue Cracking of Die Casting Dies, D. Schwam, J. F. Wallace and S. Birceanu, Case Western Reserve University, final technical report to the U.S. Department of Energy, Award No. DE-FC07-00ID138486, October 2004.