Our Sites

Advanced materials require advanced knowledge

Understanding resistance spot weld performance on AHSS

Mode 1 failure (left) shows that separation occurs through the opposing sheet around the base of the weld, resulting in a full- button pull. In Figures 1-8, nugget size is nominally 6 mm in diameter. Mode 2 failure (middle), also known as a partial-thickness fracture and partial-button pull, shows that at least 20 percent of the mating sheet thickness is removed during separation. Mode 3 failure (right) shows a partial-thickness fracture. At least 20 percent of the mating sheet thickness is removed during sheet separation without producing a button pull from the remaining sheet. Only a cavity is produced in the mating sheet (no hole typical of a button pull). These are three of eight failure modes that can occur during destructive testing of resistance spot welds made on AHSS.

Automotive structural designs that rely on high-strength steels to optimize weight, cost, and performance are becoming the standard for a new generation of lightweight body assemblies. Advanced high-strength steels (AHSS) offer additional opportunities to meet both rising safety and performance standards in vehicle design.

Additionally, many automakers prefer high-strength steels over other lightweight materials because of their relative cost and availability. To accommodate the expanded role these materials play in various body-in-white (BIW) components, steel manufacturers recently have introduced new AHSS. They include dual-phase (DP); transformation-induced plasticity (TRIP); complex-phase (CP); and hot-stamped, or martensitic (M), steel grades. These steels generally are designed to meet specific combinations of strength and ductility.

The first wave of AHSS (DP 600s) currently is being used in several models with varying degrees of difficulty. Of primary concern have been the failure modes of resistance spot welds during destructive testing. For conventional (mild) steel grades, peel testing typically results in a button mode of failure. However, during destructive testing of AHSS, spot welds also fail in several nontraditional modes.

During destructive testing, three factors generally drive failure modes:

  1. Stress state at the weld
  2. Fracture toughness of the weld metal and heat-affected zone (HAZ)
  3. Presence of solidification pores, cracks, or embrittled regions

The stress state refers to the inherent stiffness of the weld. Gauge, strength of the materials being welded, and the size of the weld determine stress state. Heavy gauges, high material strength, and small weld size all lead to high stress intensities at the weld and increase the susceptibility for non-button-type failures during destructive testing. Weld metal toughness generally relates to the hardness of the weld nugget itself.

For complex AHSS, higher weld metal hardness can result after welding, reducing nugget toughness. Reduced toughness increases the susceptibility for non-button-type failures. Finally, some AHSS grades contain other additions, such as phosphorus, that can lead to nugget porosity. Certain porosity distributions result in low-energy failure paths during destructive testing and also can lead to nontraditional failures modes.

Justifiably, the automotive industry has a vested interest in understanding failure modes for resistance spot welds during destructive testing. Given the range of steels (DP 780, 900, and TRIP) that will be introduced into automotive bodies over the next decade, automakers must recognize the implications of these nontraditional failure modes. Only by fully characterizing failure modes and the resulting weld performance can automakers introduce these new generations of steels with confidence.

Characterizing the Failure Modes

To fully understand the impact of various failure modes on a structure's performance, the industry first must be able to characterize those that can potentially occur. The American Welding Society (AWS) D8.1 committee on automotive resistance spot weld quality has been addressing this by looking at all of the potential failure modes that can occur during destructive testing of resistance spot welds made on AHSS.

Figure 4
Mode 4 failure shows that a partial-thickness fracture occurs with a partial-button pull and a partial-interfacial fracture in which the weld nugget partially separates through the faying surface plane of the weld.
Figure 5
Mode 5 failure shows a partial-interfacial fracture and a partial-button pull. The weld separates during fracture through the faying surface plane.
Figure 6
Mode 6 failure shows a combination partial-thickness and partial-interfacial fracture. The separation partially occurs through at least 20 percent of a part of the mating sheet thickness during fracture and partially through the weld nugget along the faying surface plane.

The committee came up with a fracture mode characterization system that uses a scale of 1 through 8 to semiquantify these failure modes. Figures 1-8 show welds that demonstrate the eight failure modes:

  • Figure 1is a mode 1 failure. This is a full-button failure in which the nugget pulls a full through-thickness volume of material from the opposing sheet without fracturing the weld nugget itself. This has been the conventionally accepted failure mode for mild steels.
  • Figure 2is a mode 2 failure, defined as a partial-thickness fracture and partial-button pull. In this mode, a button of at least 20 percent of the area is pulled from the opposing sheet. The weld nugget shows no evidence of failure.
  • Figure 3is a mode 3 failure, also known as a partial-thickness fracture. Failure occurs in the HAZ around the weld nugget, but not through the nugget itself.
  • Figure 4is a mode 4 failure, defined as a partial-thickness fracture, partial-button pull, and partial-interfacial fracture in which the weld nugget partially separates.
  • Figure 5is a mode 5 failure, a partial-interfacial fracture, partial-button pull, and a partial weld nugget separation.
  • Figure 6, is a mode 6 failure, a combined partial-thickness and partial-interfacial fracture. It is characterized by a fracture through at least 20 percent of part of the mating sheet thickness, as well as weld nugget separation.
  • Figure 7is a mode 7 failure, a full-interfacial fracture. This is a complete fracture of the weld nugget along the faying surface plane.
  • Figure 8is a mode 8 failure showing no fusion of the parent materials. This occurs when the sheets separate and show no evidence of nugget formation.
Figure 7
Mode 7 failure results in a full-interfacial fracture. The weld nugget experiences a complete fracture through the faying surface plane with less than 20 percent of the mating sheet thickness removed during sheet separation.
Figure 8
Mode 8, the "no fusion" failure mode, shows that the sheets separate during fracture. It reveals heating of the faying surface without the formation of a weld nugget (fusion) or a solid-state bond. With this mode, coatings that are soldered together undergo complete fracture.

Characterizing failure modes in this way allows them to be ranked in terms of severity and provides a tool for analyzing mechanical performance as a function of failure mode severity. Once it's correlated with mechanical performance, this failure mode ranking system can be used to assess spot weld quality in AHSS.

Relationships Between Mechanical Performance and Failure Mode

Concerns about variations in failure modes are increasingly related to the impact performance of spot welds. Specifically, it has been found that the relationships between failure mode and mechanical performance are particularly strong for impact loads that act to separate the sheets (peel modes). Some examples of the relationships among impact performance, steel strength level, and failure mode are presented in Figures 10-13. The testing for these tables was done using the specimen configuration inFigure 9. All steels are 1.5 millimeters thick.

Two types of data are presented: peak loads during testing and total energy to failure. The data is presented as a series of box plots. A box plot contains 50 percent of the data within its outer box dimensions and represents the total dispersion by the whiskers above and below the box. The median is shown within the box. Box plots offer a method for quick graphical data analysis for different categories of the dependent variable.

Figure 9
This specimen showing cross-tension impact is used to examine AHSS weld properties. The channels simulate the stiffness of actual components.
Figure 10
A plot shows the relationship between peak load during impact and sheet steel tensile strength. Interfacial fractures become increasingly more prevalent for steels stronger than 1,000 MPa.
Figure 11
This plot shows the relationship between energy absorbed during impact and sheet steel tensile strength. Interfacial fractures become more prevalent for steels with strength greater than 1,000 MPa.

The box plots for peak load and energy to failure as a function of steel type are shown in Figures 10and 11. The measured peak loads increase progressively with base metal strengths up to a level of about 1,000 megapascals (MPa). Above this level, performance progressively decreases.

It's important to note that the decrease in performance at the highest strength levels appears to be related to increasing degrees of interfacial failure. However, welds on steels over the entire range of strength levels show measured loads to failure roughly within a factor of 2, maintaining a base level of performance. The measured absorbed energy results almost parallel the strength results. Energy values are approximately the same for steels up to about 1,000 MPa and decrease (with increasing steel strength level) above this value. Again, the decrease in absorbed energies for steels with tensile strengths above 1,000 MPa appears to be correlated with increasing degrees of interfacial failure.

Figure 12
This plot illustrates the relationship between peak load during impact and weld fracture appearance code. RB represents no interfacial fracture; FIF represents full-interfacial fracture; and PIF represents partial-interfacial and partial-button fracture. Interfacial fractures become increasingly more severe with smaller numeric PIF rankings.
Figure 13
This plot shows the relationship between absorbed energy during impact and weld fracture appearance code. RB represents no interfacial fracture; FIF represents full-interfacial fracture; and PIF represents partial-interfacial and partial-button fracture. Interfacial fractures become increasingly more severe with smaller numeric PIF rankings.

Figures 12and 13show box plots comparing the energy and peak load to the actual weld button fracture appearance. The figures use the AWS D8.9-97 categories for degrees of interfacial fracture. On this scale, FIF is full-interfacial fracture, and PIF is partial-interfacial fracture. The suffixes indicate the degree of interfacial fracture, with 0 the lowest and 5 the highest. RB on this scale indicates a full-button failure.

This scale roughly correlates with failure modes 1, 5, and 7 described previously. As shown in Figures 12 and 13, performance decreases rapidly away from the full-button failure mode (RB). Similar peak loads and peak energies are shown for welds with failure modes of more than 10 percent interfacial failure.

When studying these plots, it's essential to notice that in no case does performance (tensile strengths and impact energies) decrease to zero. In all cases, some minimal level of performance can be used as a guideline for designing the material into a structure. The results presented here are a snapshot, indicating the dependence of performance levels on steel type and fracture mode for a given specimen geometry. Again, ultimate weld performance is defined by stress state and microstructure, as well as the harmful effect of discontinuities in the weld nugget.

Future work will define more quantitative relationships among these factors and the allowable design components for spot welds in new generations of AHSS.

Jerry E. Gould is chief engineer and Warren Peterson is senior engineer, both for the resistance and solid-state welding team at the Edison Welding Institute, 1250 Arthur E. Adams Drive, Columbus, OH 43221-3585, 614-688-5000, fax 614-688-5001, jerry_gould @ewi.org, wpeterso@ewi.org, www.ewi.org.

About the Authors

Jerry E. Gould

Chief Engineer, Resistance and Solid-State Welding Team

1250 Arthur E. Adams Drive

Columbus, OH 43221

614-688-5000

Warren Peterson

Senior engineer, Resistance and Solid-state Welding Team

1250 Arthur E. Adams Drive

Columbus, OH 43221

614-688-5000