Virtual and Physical Testing of Third-Generation High Strength Steel
Evaluating a new high-strength steel’s ability to improve on an existing stamped-steel production part.
Developing lightweight, stiff and crash-resistant vehicle body structures requires a balance between part geometry and material properties. High-strength materials suitable for crash resistance impose geometry limitations on depth of draw, radii and wall angles that reduce geometric efficiency. The introduction of 3rd-generation Advanced High Strength Steels (AHSS) can potentially change the relationship between strength and geometry and enable simultaneous improvements in both.
This technical paper demonstrates the applicability of Third-Generation AHSS — that offers higher strength and ductility — to replace the 780 MPa Dual Phase steel in a sill reinforcement on the current Jeep Cherokee. The focus is on formability, beginning with virtual simulation and continuing through a demonstration run on the current production stamping tools and press.
One of the limitations in stamping body parts is that an efficient geometry typically requires material to be formed with small radii, high depth of draw and shallow draft angles. All of these characteristics are more difficult to achieve as the strength of the material increases. This difficulty limits the efficient use of First-Generation AHSS. This paper discusses material advances that improve the ability to use steel efficiently and provides a case study that illustrates the use of those advances – from concept, through design, simulation and validation on a production tool.
New Third-Generation AHSS is intended to supplement the existing portfolio of high strength steels for automotive applications. The strength/ductility balance of Third-Generation AHSS is superior to that of Dual Phase (DP) and martensitic steels and the relatively lean chemistry precludes the prohibitive costs and manufacturing issues associated with TWIP and austenitic stainless steels.
U.S. Steel developed a Third-Generation AHSS with a 980 MPa minimum tensile strength, calling it 980 XG AHSS. It is a commercially-produced steel with a microstructure comprised primarily of fine-grained ferrite/martensite and retained austenite; 980 XG3 steel exhibits an excellent combination of strength and ductility with UTS around 1000 MPa and total elongation around 25%. This material experiences >200 MPa strain-and-bake strength increase with no change in failure strain values.
Application of Third-Generation AHSS
In the context of automotive body stampings, increasing formability enables more complex geometry without excessive thinning or splits. Increased material strength enables less intrusion to the critical cargo — such as passengers, batteries and fuel systems — during a crash event. Before the commercial availability of 980 XG3 AHSS, the conventional method of simultaneously resolving splits and reducing intrusion was to select a steel grade with lower strength and increased thickness, thus increasing the total part weight.
Avoiding that weight increase with alternative high-strength materials such as press-hardenable steel or reinforced composites typically added prohibitive cost and joining concerns. Applying 980 XG3 AHSS enables a simultaneous increase in formability and strength compared to other AHSS, without an increase in component weight. When applied during early stages of body architecture development, an overall weight reduction can be achieved compared to vehicles designed with primarily first-generation AHSS.
Case studies have also shown that strategic application of 980 XG3 AHSS can replace press-hardenable steels with identical crash performance and component weight.
Identifying Candidate Part
Based on the properties of the 980 XG3 steel, there are several logical approaches to selecting a current production candidate part for a case study.
- Choose a DP590 or TRIP780 part to increase strength while maintaining current geometry
- Choose a DP980 part to improve geometry while maintaining strength
- Choose a DP780 part to improve strength and geometry simultaneously
Considering that the trial would be conducted in a current-production Jeep Cherokee tool, it was advantageous to choose a part that currently was 780 MPa or 980 MPa in order to minimize the risk to the die as well as limiting the expected increase in springback due to the higher yield strength. For this reason, parts that are currently DP590 were not considered for this case study and the vehicle’s Sill Outer Reinforcement was selected.
The reinforcement is currently in production in a DP780 material and contributes to the vehicle’s performance in the FMVSS 301 Fuel System Integrity high-speed rear-impact load case. The Rear Sill Reinforcement contains areas that are near to the thinning limit of the DP780 material, ensuring that the forming test would be non-trivial in a higher-strength material.
To evaluate a new material in a production stamping die designed for a different material, the following measures of success were established:
- The part forms without excessive thinning and without major wrinkles or splits, otherwise known as a “safe” stamping
- Finished part springback and twist either correlate to the simulation or fall within current production limits
Simulation software AutoForm was used to assess risk of thinning, wrinkles, splits, and springback before the trial could be scheduled. For new product designs, the final component CAD is imported to the software and then the stamping tooling and process are reverse-engineered to produce a safe stamping out of the specified material. In this case, the tooling and process already existed at the production stamping supplier.
To emphasize further the lightweighting opportunity of 980 XG3 AHSS compared to the baseline DP780, a 6% thickness reduction was proposed. The effect of increasing strength and reducing thickness on crash results is elaborated in the subsequent Crash Safety Validation section.
Simulated formability assessment was considered overall safe on the 980 XG3 AHSS with reduced thickness, with two notable concerns. First is a slight risk of splits in the addendum, which is the area of the stamping used to control metal flow into the die. The authors agreed the risk was acceptable for a trial, as the addendum is cut off in subsequent stations and sent to scrap without affecting the final trimmed component.
Second, one section of blank edge was being drawn into the stake bead at the end of the press stroke. Stake beads are a critical component for controlling metal stretch and springback and require the material to be fully engaged in the bead to be effective. The developed blank shape was modified in CAD to ensure complete engagement in the stake bead for the duration of the press stroke.
Accurate springback simulation is an ongoing challenge. Great effort has been made for accurate predictions and software has improved. However, the result is still not consistent and prediction accuracy is case dependent. The magnitude of springback generally increases as the yield strength of the steel increases and the small increase in yield strength for this study is represented in the small increase in predicted springback.
Crash Safety Validation
Component test results showed that the average crush load (axial mode) is proportional to AVG x t3, where AVG is the average of the yield strength and the tensile strength (AVG = ½(YS+UTS)) and t is the nominal material thickness. Based on this relation at the component level, about 8.7% weight reduction can be achieved when replacing DP780 with 980 XG3 AHSS.
Thickness reductions between 4% and 9% showed relative crush intrusion in the study area comparable to the baseline without making any geometry changes, which is consistent with the 8.7% number estimated. To further refine crash simulation accuracy, a fracture material model (GISSMO) has been developed for 980 XG3 AHSS to predict fractures in crash simulation.
The right-hand (RH) die successfully produced finished 980 XG3 AHSS parts with no splits and with wrinkles comparable to current production. Additionally, the lower blank edge that was modified to stay in the stake bead during the press stroke was properly contained in the stake bead. When aligned and clamped to the existing inspection fixture in accordance with production quality control procedure, the trial parts passed inspection. The RH trial was suspended upon the completion of 20 samples.
Next, the left-hand (LH) die lineup was loaded into the press and prepared for trial in the same way as the RH die. Unlike the RH trial parts, 100% of the LH trial parts came out of the draw station with a split in the stake bead. The radius in the split area was measured on both parts and the LH radius was 15% smaller than RH.
Based on stretch bend fracture design guidelines developed by Shih, the LH radius violated the critical ratio of die radius to material thickness (R/t) for 980 XG3 AHSS. The stamper conceded that LH and RH dies were not always modified to match each other; they could be individually modified, provided they each make parts that meet the customer specification. Since the current production DP780 material was not splitting in this location, the decision was made to terminate the LH trial and return the die to regular production state without radius modification.
The next level of correlation assessment is to compare thinning percentage; 21 locations on the final trimmed part were selected for thinning comparison based on areas of high strain identified by the simulation. The resulting comparison indicates that the measured thinning was slightly less than the predicted amount in most areas. Areas with material thickening had good correlation.
The intent of the study was to compare the stamping performance of a new, higher strength material against a current production high-strength part. In accordance with the crash-safety validation simulations, a thickness reduction was implemented in the stamping trial to illustrate the potential for weight savings. By the measures defined in the Design section, the 980 XG3 AHSS trial was considered a success:
- Splits were not an issue on the RH part and the splits identified on the LH part could have been avoided by adhering to safe R/t guidelines established by the steel manufacturer
- The wrinkles correlated to the simulation, at the same locations and severity as the baseline part
- Finished-part springback and twist met or exceeded expectations
- The part thinning correlated to the simulation and was within safe limits
- The die was returned to baseline material production with minimal down time
This article is an edited and condensed version of SAE Technical Paper 2020-01-0757. The authors thank FCA U.S. LLC and United States Steel Corp. for permission to publish the paper, as well as “the many individuals at U.S. Steel for their support and thoughtful contributions.”