A tailor-welded blank (TWB) consists of two or more flat sheets that are welded in a planar configuration and subsequently stamped into a three-dimensional component. TWBs of steel are in prodcution use to improve automotive parts in terms of scrap, weight, cost, strength, stiffness, and quality. A tailor-welded blank (TWB) consists of two or more flat sheets that are welded in a planar automotive parts in terms of scrap, weight, cost, strength, stiffness, and quality. The same technology is more challenging for aluminum TWBs, because welding and forming are typically more difficult. Alloy 6111-T4 and 5754-0 blanks laser-welded at several rates and with several kinds of equipment were tested in tension and using the OSU Formability Test. The optimal weld speeds were identified for each welding process on the basis of tensile ductility and typical microstructures were examined. Press formability can be much greater than the inherent weld ductility would indicate, and the degradation of forming performance of aluminum TWBs is similar to that of steel TWBs. TWBs of both aluminum alloys have satisfactory longitudinal formability, but the heat-treatable 6111-T4 is severely limited under transverse loading because of a softer heat-affected zone.
1.0 Introduction
There is strong motivation to use TWB technology with aluminum automotive in order to produce aluminum TWBs which, by reducing engineered scrap and component mass, can favor the economic substitution of aluminum for this application (thus improving fuel economy and exhaust emissions). However, formability differences between steel and aluminum TWBs must be undertood for successful high-volume production using this technology.
Steel TWBs have shown advantages including reduction in cost, scrap, and weight of the body components, as well as part consolidation and improvements in structural integrity and dimensional consistency. Although steel TWBs can be made by various welding processes, laser welding is the most popular technique. C02 and Nd:YAG (Neodymium-Yttrium Aluminum Garnet) lasers are two common types of welding lasers, both of which may be continuous or pulsed. Laser-welded steel TWBs typically have a weld bead that is significantly higher in strength and hardness and lower in ductility than the surrounding base material.
Two fundamental modes of forming failures for steel TWBs are typically observed (see Figure). Type 1 corresponds to the parallel orientation of the weld relative to the principal stretch axis (longitudinal loading), and failure occurs across the weld because of the limited ductility of the weld relative to the surrounding base material. Type 2, which is more common in practice, corresponds to the perpendicular orientation of the weld relative to the principal stretch axis (transverse loading. Type 2 failure is related to the material properties in the weaker base material, which is constrained by the presence of the stronger to deform in plane-strain tension.
While the welding of steel has been thoroughly investigated and well-documented, aluminum alloys have proven more difficult to weld reliably. Difficulties include porosity and hot cracking in the fusion zone, poor coupling during laser welding because of the high reflectivity of the metal, and degradation of strength and loss of alloying elements in the fusion zone, which make welding aluminum alloys challenging from both welding process and formability standpoints.
2.0 Objective
* Establish the formability of CO2 and Nd:YAG laser-welded tailor blanks in Al alloys 5754-0 and 6111-T4, compare and contrast the results and measure formability in terms of weld process variables.
3.0 Experimental Procedure
Aluminum alloys 5754-0 and 6111-T4 were used for single-alloy welding (A-A, B-B). Alloy 5754-0 is a non-age hardenable alloy that is solid solution strengthened by magnesium (Mg). Alloy 6111-T4 is an Al-Si alloy which has been solution heat-treated and naturally aged, as indicated by the "T4" temper designation. During automotive manufacturing, this alloy will harden in the paint-bake cycle after forming. This feature is a result of a favorable distribution of precipitated particles, such as FeAI3, Mg2Si, and AI(Mn, Fe)Si.
Aluminum alloys 5754-0 and 6111-T4 were welded with several lasers at various weld travel speeds. Blanks were milled for precise edge straightness and then placed in a fixture that enforced accurate beam alignment during welding. Prior to welding, the edges of blanks were tack-welded to maintain a consistent gap during the welding process.
Uniaxial tensile tests were performed for base and welded materials with longitudinal and transverse weld orientation. Both orientations were used to estimate formability, while the longitudinal tests were also used to measure mechanical properties of the fusion zone. Three tensile specimen sizes were used, with smaller ones needed to isolate weld, In the smallest specimens, the weld zone represents approximately one-third of the cross-sectional area of the gauge section.
To investigate forming performance under typical conditions, the OSU Formability Test (see Figure) was employed, where the punch-height-to failure is taken as a measure of formability.
4.0 Results
Total elongation of transverse tensile specimensof 6111-T4 is nearly independent of weld speed and is very limited. In transverse loading, failure corresponds to complete specimen fracture in the weld zone because of the weaker material formed there after welding.
For 6111-T4 3 kW C02 welds, the trend of decreasing ductility with increasing weld speed most likely involves hot cracking, or weld solidification cracks. Examination of optical micrograph of an as-received 6111-T4 3 kW C02 weld (170 mm/s with low ductility) in a lengthwise view of the weld line and surrounding base material clearly showed large cracks in the weld zone, presumed to be weld solidification cracks, while examination of an 85 mn/s weld (the optimal weld speed) showed that large cracks are absent. (See Figure.) The explanation for the generally limited ductility of 6111-T4 welds is less clear. Small cracks or other defects may reduce ductility in 6111-T4 laser welds, but further experiments would be required to verify this.
The tensile elongations for various longitudinal 5754-0 showed that failure initiated in the weld zone in most cases. Both longitudinal weld types (3 kW YAG and 5 kW C02) displayed a trend of decreasing ductility with increasing weld speed, although the change is less severe. This may be related to weld cracking behavior. Alloy 5754-0 welds generally have greater longitudinal ductility than the 6111-T4 welds.
5754-0 weld specimens of both welding procedures failed in the transverse weld at reduced elongations (10-15% strain) at low weld speeds; however, 5 kW C02 weld specimens produced at weld speeds greater than 170mm/s failed away from the weld and displayed impressive elongation-to-failure, approaching that of base 5754-0 (26% strain). The 5754-0 welds tested in the longitudinal orientation have similar ductility, but these welds have much different ductilities and failure modes in transverse loading. This suggests that longitudinal weld ductility is not critical for weld performance under transverse loading, which is consistent with observations made for 6111-T4 welds (and for TWB steel welds).
The best 6111-T4 TWBs are approximately 80% as ductile parent material under longitudinal loading, but only 25% as ductile transversely. The corresponding values for 5754-0 TWBs are 80% longitudinal and 95% transverse. It is interesting to note that these formability ratios are similar to ones measured by Saunders and Wagoner(9) for steel blanks: 60-85% in longitudinal loading and nearly 100% transverse. The origin of these differences is explored below.
The specimen-averaged hardening behavior of 6111-T4 for unwelded and longitudinally-welded miniature specimens was measured using miniature tensile specimens and an analytical technique to obtain stress and strain in the weld zone. Opposite to welded steel specimens, the aluminum weld zones have lower strength than the parent materials, and there is little difference between the weld types. The difference in apparent flow strength of YAG and C02 welds is attributable to the wider weld zone for C02 welds. (Typical stress-strain curves for optimally-welded specimens are shown in the Figure.)
The much lower strength in the weld zone (of high-quality, uncracked welds) in 6111-T4 is presumably related to typical solutionizing. To test this supposition, transverse and longitudinal tensile specimens were heat-treated to the T4 temper after welding; i.e., solutionizing at 560 - 580°C for about five minutes, followed by water quenching and aging at room temperature for two weeks. Nearly all of the parent material strength and much of the ductility was recovered by this treatment. The small remaining difference in ductility is most likely related to remaining small weld cracks or subtle differences in the effective heat treatment.
Knoop microhardness profiles for as-welded and post-weld heat-treated 6111-T4 specimens were evaluated. The parent material and heat-treated weld zone have Knoop hardnesses of approximately 80, whereas the weld zone before heat treatment has a Knoop hardness of 65. The grain structure for the specimen tested revealed large, columnar grains apparent in the weld zone, even though the microhardness values there cannot be distinguished from the adjacent parent material.
Engineering stress-strain behavior for unwelded and optimally-welded 5754-0 tensile specimens were compared. All curves displayed the saw-tooth-like pattern characteristic of serrated yielding. As with 6111-T4 welds, failure initiated in the weld zone in all cases (except some transverse 5 kW C02 190 and 210 mm/s ones); and, with this same exception, 5754-0 welds exhibited reduced longitudinal ductility relative to base material. Unlike 6111-T4 welds, 5754-0 welds consistently follow strain hardening curves similar to that of the parent material. Because 5754-0 has been annealed to its softest and most ductile condition, this non-heat treatable metal is expected to be relatively insensitive to the heat excursion of welding.
OSU formability tests were conducted on parent materials and longitudinal TWBs, and compared with results for aluminum-killed drawing-quality steel TWBs from the literature. The aluminum TWBs are 95% as formable as the parent material blanks, which is similar to the experience with steel blanks, which have formabilities of about 99% of the parent material. Other steel TWBs involving HSLA steel showed relative formabilities in the range of 90%.
One can conclude that the weld properties become a less important determinant of overall longitudinal formability as specimen width (normal to the weld) increases. For this reason, sheet stamping operations with TWBs are practical even though the inherent weld ductility can be quite limited.
5.0 Conclusions
The following conclusions may be drawn from the results:
1. Satisfactory forming performance for aluminum TWBs can be realized in spite of limited inherent weld ductility. The relative formability of aluminum TWBs to unwelded blanks can be similar to that of steel TWBs.
2. Alloy 5754-0 performs much better than 6111-T4 under transverse loading, because of softening of the HAZ in heat-treatable 6111-T4.
3. 6111-T4 TWBs can be heat-treated after welding to recover weld hardness and transverse formability.
4. Ductility changes with weld speed are in some cases correlated with weld cracking.
5. Optimal weld speeds were identified for all alloys and weld processes.
6.0 Deliverables
This project presents the mechanical properties of Aluminum Alloys
5754-0 and 6111-T4, as well as the causes for observed weld properties
and forming behavior. Welds with high inherent formability have been
identified, along with approaches to optimizing the formability of inferior
welds by weld orientation.
View Figures