This second of two parts on die failure analysis deals with actual case studies of real applications. In each case, dies failed and had to be modified or redesigned. These cases provide the rudimentary analysis of each failure and how it was corrected.

In this second of two parts, we examine failures that occur in the tooling during forging. In Part 1, we presented the fundamental aspects of stress within a die that has been loaded during the forging process. In this article, we will describe a number of case studies, each of which provides an example of a die failure, the analysis that was done based on stresses in the die and determination of the root cause of the failure. It is our intent that, through these examples, the reader will be able to understand some of the reasons for die failures and methods that can be used to determine their root cause. This understanding, in turn, can be applied to help minimize future die failures.

There are a number of sound economic and engineering reasons why prevention of tool failures is important. Tooling costs, of course, are significant, but the total cost of die failure is much higher than the simple replacement cost. A tool’s failure is riddled with hidden costs above and beyond those of tool replacement. For example, there are costs associated with scrapped or rejected parts, rush charges on die replacements, lost energy inputs, late deliveries, lost customer goodwill, etc. Clearly, replacement of broken or worn dies is more expensive than just their replacement.

Operating a lean manufacturing operation can be very problematic if tool life is short or erratic. Delivery problems, for example, become routine when tooling problems occur unexpectedly. Additionally, the risk of safety problems is high in overstressed tooling, especially if the overloads lead to catastrophic die failures. When the ultimate strength of a die material is exceeded significantly, a catastrophic fracture may occur, resulting in injury.

Because of these economic and operating reasons, it is important to understand the root cause of any die failure and to institute corrective action to prevent it from recurring. Through the case studies examined here, we hope that you will be able to learn from the failures in other shops and benefit from their problems.

We looked at case studies that occurred during a period spanning two decades to illustrate theory and confirm that it “connects the dots” to real shop-floor tooling failures. These examples cover a range of die failures and solutions.

Figure 1. This die produced parts that were undersized on inside features, while the outside diameter was in tolerance.

DIE CASE YIELDING

In 1989, the late Alfred Lau presented a case at the North American Forging Technology conference in Orlando, Fla. A hot-forged turbine disk (Figure 1) was dimensionally out of tolerance after forging. The features on the inside were undersized, while the outside diameter was in tolerance.

An investigation showed that the punch (top die) was in tolerance. The die, which formed the outside diameter, was oversized in the forging contact region. A subsequent stress analysis revealed that the effective stress on the die insert was in excess of the yield strength at temperature, resulting in plastic deformation. The plastic deformation resulted in an oversized forging on the outside diameter, which was reduced (swaged) during ejection.

Figure 2. This simulation shows a high effective stress in the die insert. The red (circled) region is the failure region.; Figure 3. The redesigned die still shows high stress in the contact region, but the thicker die wall prevents the die from yielding.

The analysis (Figure 2) showed a high effective stress in the die insert. The red region (circled) was the failure region. The root cause was high forging pressure and a die insert with inadequate longitudinal tensile strength. While the pressure was radial, induced stress in the longitudinal direction resulted in yielding. On the left side of the plot, the principal stress is shown (red is tensile and green is compression).

During the problem-solving period, quite a few alternative designs were analyzed. The production solution was simply to use a thicker die insert. After solving this problem, other like jobs were found to have similar issues and subsequently redesigned. The analysis of the redesigned die is shown in Figure 3. While there is still a high stress in the contact region, the through-wall strength prevents the die from yielding. Die life in production was extended more than tenfold until die wear became the failure mode.

Figure 4. This die fractured into two parts with the fracture surfaces facing the camera.; Figure 5. A stress analysis of the die insert showed a significant tensile stress in the failure region (colored red).

DIE-INSERT FRACTURE

Fast forward to 1995, when we deployed die stress analysis on a die failure for a cold-formed automotive component. The die-insert fracture (Figure 4) during the first operation was the most significant failure. The monolithic die was shrink-fitted into a steel case. It fractured prematurely into two parts (sometimes called “wafering”). The low-cycle-fatigue fracture initiated on the inside, then propagated outward toward the outside diameter.

A stress analysis of the die insert showed a significant tensile stress in the failure region, perpendicular to the fracture. The die material was tungsten carbide. This material has excellent wear and stiffness characteristics, but it is prone to low-cycle-fatigue (LCF) failures when subjected to cyclic tensile stress. The maximum principal stress is shown in Figure 5. In this plot, the red represents tensile stress and the green is compression.

Figure 6. Further analysis indicated that the axial stress (right) was the culprit that needed to be resolved.

Although the root cause of failure was clear, it took some further analysis to determine a solution. The components were evaluated to better understand the direction of the tensile stress (Figure 6). The hoop component of stress (center) indicated compression throughout. This type of hoop stress indicates that the shrink ring is doing an adequate job to support the insert, even with forming load applied. The axial component of stress was the culprit that needed to be resolved.

In order to solve this problem, the die insert was split longitudinally. The resulting analysis indicated a lower principal stress (Figure 7). In production, the LCF failure was no longer a factor, with die-insert life improvements of significantly more than tenfold!

Figure 7. The die insert was split longitudinally, with the resulting analysis indicating a lower principal stress.

M16 RECEIVER DIE FAILURE

In 2003, a PRO FAST project was conducted to study and resolve an issue with the dies used to hot forge an aluminum receiver. This premature die fracture limited production runs, delayed delivery and added costs to the process. Figure 8 shows the M16 rifle and the receiver forging assembly.

The initial symptom of the problem revealed itself as a rough area of raised material in the corner of the forging. Figure 9 shows the forging with the raised region of material. An investigation uncovered a crack on the die surface. Figure 10 shows the correlation between the raised material on the surface of the forging and the crack in the die.

Figure 8. Photographs of the M16 rifle and the receiver assembly. The dies for the receiver-assembly forging were experiencing premature failure.; Figure 9. Photo showing the raised region on the surface of the forging, indicating that there is a problem (longitudinally, left of center).

A forging and die stress analysis was conducted using DEFORM-3D. Figure 11 is a plot of the effective stress in the die under the fully loaded condition. While the effective stress is higher in the corner region where the die cracked, these predictions were very close to the yield strength. Figure 12 examines the maximum principal-stress component throughout the die when it is under load. Although the forging pressure was relatively low, a tensile-stress component (150 ksi) was observed at the sharp corner. Inadequate lateral support resulted in a bending stress, and the small radius resulted in a stress concentration. This high principal stress at the corner was the root cause of the LCF failure.

An alternative design was developed using a pocket type of die holder with an insert. An interference fit was designed to keep the insert in a predominately compressive stress state during forming. Figure 13 shows a plot of the maximum principal-stress component in the redesigned die, which is predominantly compressive. The die holder is stressed below the yield strength with a reasonable safety factor.

Figure 10. Photos of the defect observed on the surface of the forging and the fracture in the die that produced it.

TROUBLESHOOTING SUGGESTIONS

Based on these examples, there are several suggestions that can be made to minimize the occurrence of forging-die failures. To prevent a failure, consider:
  • Reducing the load on the die
  • Increasing the strength of the die material
  • Increasing the load-carrying ability of the die by redesign
  • Increasing or enhancing the die support
It is also important to evaluate the risk of miscellaneous problems, including poor process controls, mid-stroke loading, volume/stroke control in closed-die forgings, assembly and/or thermal problems. Troubleshooting die failures can utilize the experience of a seasoned designer and the technical tools of an engineering analyst. No analysis will pinpoint the root cause of a problem if the actual process conditions are not included in the model. The ideal redesign is not derived mathematically.

Figure 11. Plot of the effective stress for the forging die. Though the value of the effective stress is highest in the region of concern, the magnitude of the effective stress is not extremely large.; Figure 12. Plot of the maximum principal-stress component in the die. For the critical region the value of the maximum principal-stress component is tensile, which contributes to the root cause of the die fracture.; Figure 13. Maximum principal-stress components in the redesigned die. The magnitude of the stress components in the critical region are low and compressive, which will minimize the possibility of fracture in this region.

SUMMARY

In this second portion of our two-part paper, we have presented several case studies of die failures. These provide specific examples of forging die failures. The proper analysis of these failures allowed the determination of the root cause for the failure. Once the root cause is known, a proper approach can be devised to delay or to avoid the failure in the future.

Acknowledgements
The support for this work from the PRO-FAST Program is appreciated. The PRO-FAST Program is enabled by the dedicated team of professionals representing both the Department of Defense and industry. These teammates are determined to ensure that the nation’s forging industry is positioned to meet the challenges of the 21st century. Key team members include: R&D Enterprise Team (DLA J339), Logistics Research and Development Branch (DLS-DSCP) and the Forging Industry Association (FIA). This work was originally prepared for the Forging Fundamentals 101 course by Scientific Forming Technologies Corporation and the Forge-It Team.

Co-author Dr. Chet Van Tyne is FIERF Professor, Department of Metallurgical and Materials Engineering, Colorado School of Mines, Golden, Colo. He may be reached at (303) 273-3793 or cvantyne@mines.edu. Co-author John Walters is vice president of Scientific Forming Technologies Corporation, Columbus, Ohio. He may be reached at (614) 451-8330 or jwalters@defor.com.

Note: Alfred Lau, whose work in the Die Case Yielding section is cited here, unfortunately passed away in February 2008. He held a number of technical positions at Wyman Gordon in Houston, Texas, since the 1980s and was a pioneer in applying die stress analysis to forging-die failures.