May_June_AMP_Digital

A D V A N C E D M A T E R I A L S & P R O C E S S E S | M A Y / J U N E 2 0 1 9 2 4 One early modification to the AVT concept concerned the role of spot welds: For Alcan, the spot welds simply held the structure together until the ad- hesive had cured. For them, the vehicle was a bonded structure. Without any data supporting the long-term behav- ior of the adhesive in either crash or du- rability, Ford decided early on that the vehicle would be weld-bonded, with sufficient spot welds to support crash performance. The structural adhesive’s role would be limited to protecting the welds from fatigue loads and providing the continuous joint necessary for stiff- ness and noise, vibration, and harsh- ness (NVH). ALUMINUM PRESENTS CHALLENGES But even with that simplifica- tion, the structure design team oper- ated with three major blind spots. The first involved crash behavior of the front rails, essentially two flat U-chan- nels joined with spot welds. How would the weld-bonded front and rear rails behave in a crash? How would the re- quired folding patterns be achieved to maximize energy absorption? Would the notoriously weak aluminum spot welds be sufficient? What about the contribution of structural adhesives, if any? The first proposal was to build the required number of parts using pro- totype methods, but the cost estimate prompted a second look. The solution came from the Honda Service parts or- ganization, which unknowingly provid- ed the proper number of left-hand front rails from the recently released NSX through several discrete procurement disguises. Reverse engineering provid- ed the geometry and alloy properties, and soon the CAE analysts were able to start tuning their models using test data from actual parts. The second unknown was the fati- gue behavior of a weld-bonded struc- ture. Ford not only had zero experience with adhesively bonded structures, it had no baseline for the aluminum al- loys themselves nor their spot welds. Developing such a database was quick- ly moved to the top of the priority list and the task handed off to researchers in the materials department. Finally, the team had no official formability guidelines to guide the de- sign engineers towards parts that could be made with a reasonable degree of confidence. There was no computer modeling available beyond simple 2D section analysis. Worse yet, Ford’s CAD system at the time, PDGS, was a line- based system that was inadequate for visualizing 3D parts. The problem was solved by lining up all the stamped parts that made the body on long tables in a very large room. Armed with Alcan’s simple guidelines, stamping feasibility experts were soon poring over the parts and marking necessary modifications, which were transferred into CAD by the designers. Gauge-for-gauge trials were organized to further understand the new materials’ behavior in real tools. Further insights were gained by labora- tory testing and the new lubricant had to be evaluated for production use. The adhesive team was on a steep learning curve getting acquainted with Alcan’s testing protocols and durabil- ity targets. One of the issues was that Ford never single-sourced anything, not wanting to find itself held hostage to a supplier. Another minor issue was that Ciba-Geigy was not a Ford suppli- er. But the most difficult aspect was to set the performance targets that would characterize a successful product. All of Ford’s experience was based on acceler- ated tests that were known to correlate with steel, but they were suspected to be inadequate surrogates for alumi- num. Without prior experience with structural adhesives, the team needed to understand what degradation (if any) would be tolerable over the life of the vehicle and how to set the testing pro- tocols to verify it. Unforeseen issues cropped up repeatedly. For example, electrical grounding unexpectedly became a con- cern once it became clear that the stud welds that worked so well with steel could not readily be adapted to alu- minum. A whole new line of develop- ment had opened up, which required a speedy resolution. Other issues were anticipated, but required specialized resources like fasteners, as hundreds of them attach various components to the body structure. Engineers now had to identify the optimal coatings to pro- tect against galvanic corrosion. And then they had to procure them in time to support the upcoming builds. Meanwhile, Alcan was not sitting idle. The new builds required many dif- ferent gauges of various alloys, most to be pretreated and lubricated. Produc- tion schedules needed to be reserved and facilities readied—or even built and commissioned—like the prototype pro- duction pretreatment and lubricant line that was hastily built in Warren, Ohio, to support the upcoming material orders. On the Ford side, the project was quickly gathering momentum. By Janu- ary 1992, a little over five months after the kick-off, the design was advanced enough that the first material orders could be placed. By February, the pro- totype shop was kicked-off and the de- livery schedule set: The first parts and subassemblies would be delivered to Ford by July 10, 1992. A prototype body shop was installed in Ford’s pilot plant in Allen Park, Michigan. Detailed illus- trated build manuals had to printed. TABLE 1 — FORD TAURUS ALUMINUM PROJECT WEIGHT SAVINGS Automotive part Steel, kg Aluminum, kg Savings, kg % Savings Body structure (BIW) 270.4 145.2 125.2 46% Fenders 6.4 2.7 3.6 57% Hood 22.2 9.1 13.2 59% Trunk lid 12.0 5.4 6.6 55% Front doors in white 34.0 19.5 14.5 43% Rear doors in white 25.9 16.3 9.5 37% Total weight 371 198 173 47%