H13 Die Softening in Service
A die casting die does not crack because the thermal stress is high on its own. It cracks because the surface has softened until it can no longer carry that stress, so the deformation turns plastic and a crack opens. This page sets out the softening that controls thermal fatigue in premium grade H13, the hardness that a specimen loses against the temperature it reaches, and what happened when the hardness was brought back before the cracks started.
Softening is what starts the crack
In a die where there is no severe stress concentrator, the softening of the steel is the most important factor in crack initiation. That is the conclusion the source draws from its own test series. A higher yield strength means a greater resistance to plastic deformation, and the yield strength of a tool steel at temperature follows its hardness. As long as the surface can carry the compressive stress in each cycle without yielding, nothing starts. Once the surface has softened past a level set by the temperature and the stress, the compression goes plastic, a residual tension is left behind, and a crack opens at the first stress raiser it finds. The extension of the crack is governed the same way, by the strength of the steel ahead of the crack front rather than by the crack itself.
The hardness ahead of the crack tracks the temperature
The specimens in the source test were tempered to 44 to 46 HRC before cycling. The table below gives the temperature reached at the corner, the hardness close to the corner, the hardness measured ahead of the average maximum crack, and the resulting cracking, all at 15,000 cycles. The shortest immersion holds most of the hardness and does almost no damage. The longest immersion takes the corner down to 24.3 HRC, roughly twenty points below the tempered hardness, and the crack area rises by a factor of about eighty five against the shortest immersion.
| Item | 5 sec | 7 sec | 9 sec | 12 sec |
|---|---|---|---|---|
| Maximum temperature at the corner, F | 926 | 991 | 1087 | 1147 |
| Hardness at the corner, 0.01 in in, HRC | 36.9 | 33.8 | 24.5 | 24.3 |
| Hardness ahead of the average maximum crack, HRC | 36.9 | 33.8 | 31.6 | 29.2 |
| Total crack area, x 106 µm2, after 15,000 cycles | 1.97 | 5.9 | 108.56 | 167.72 |
The temperature, the hardness and the cracking for four immersion times, all in one premium grade H13 specimen series at 15,000 cycles. Hardness ahead of the crack is measured at a distance from the corner equal to the average maximum crack length.
Bringing the hardness back before the cracks start
The source also tested whether the resistance can be restored. One specimen was cycled for 2,500 cycles and then re-heat treated back to the original hardness, and the treatment was repeated every 2,500 cycles. Against a specimen held at 46 HRC and another held at 51 HRC, the specimen that was re-heat treated to 51 HRC after every 2,500 cycles gave the best resistance to heat checking. Restoring the strength while the surface was still uncracked held off crack initiation, and it also slowed the growth of the cracks that had already formed. The practical reading is that a die that is softened but not yet cracked is worth recovering, because the recovery acts on initiation and on propagation at the same time.
Softening that runs ahead of the tempering curve
The source checked whether exposure to temperature alone explains the loss. For the 12 second specimen the peak part of the cycle sits between 1100 F and 1150 F for about 5.5 seconds, which over 15,000 cycles adds up to about 23 hours at that temperature. On the tempering curve of the steel, a fall from 45 HRC to about 30 HRC is what 23 hours at 1150 F would produce, and the hardness measured 0.06 in in from the corner was 30.6 HRC. Further into the steel the temperature is lower and the time at temperature is shorter, yet the softened zone reaches deeper than the curve allows for. The source concludes that thermal exposure alone does not account for the softening, and that cyclic stress softening adds to it. Both act at once, so the softening of a die in service cannot be read off a tempering curve alone.
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. The softening resistance of the hot work family is compared on the hot work softening resistance page, and the preheat and working temperatures used to hold a die away from the brittle range are on the preheat and tempering temperature page. What the steel has to survive is set out on the thermal fatigue cracking page, and the cooling design that holds the peak temperature down is on the die cooling line design page. The general ranking of the grades is on the thermal fatigue resistant tool steels page, and the change in structure behind the loss of hardness on the H13 carbide coarsening page. The structure of the grade as quenched is described on the H13 microstructure page, the other failure modes on the die wear and failure page, and the way a grade is chosen for a press on the high pressure die casting selection page.
Before you use this as a die maintenance specification
This page is a reference summary of a published study and it is not an Aobo Steel specification. The temperature, hardness and cracking values are reproduced from the source, which tested one grade, one specimen shape and one set of casting conditions. The softening a production die reaches also depends on the die design, the alloy, the cycle and the cooling circuit. Any decision to re-heat treat a die is confirmed on the job.
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.