Recent advances in resistance welding technology,
alloys, and NDE techniques have helped to reduce manufacturing costs for aluminum sheet structures

BY Donald J. Spinella, John R. Brockenbrough, AND Joseph M. Fridy

The Starting Point: Uniform

Another factor often overlooked is how the current is distributed or flows across the electrodes and sheets. The majority of weld development occurs during the first half of the weld cycle in aluminum as opposed to the latter half for steel. Considering this observation with the shorter weld periods mentioned above, aluminum welds are developing within 25 to 33% of steel's total weld time. Assuming a steel weld occurs in about 200 milliseconds (ms), an aluminum weld primarily forms within 40 to 67 ms. This reinforces the need to have the proper weld pressure and electrode alignment available before the weld current is delivered.

Over time, electrodes will deteriorate or erode, causing changes in the uniformity of the current and pressure, which ultimately leads to undersized welds. Researchers have traditionally focused on reducing electrode erosion through a variety of surface coatings and treatments. Many refer to the tenacious aluminum oxide layer and how detrimental it is to resistance spot welding. Excessive oxide levels can cause erratic weld performance and expulsion, but solutions developed decades ago that remove surface oxides require much higher currents that substantially increase electrode sticking. In the automotive and commercial industries, it is common practice to resistance weld lubricated aluminum sheet with no special surface treatments. Issues regarding excessive oxidation have been addressed with improved mill controls, prelubricated aluminum materials, and proper storage techniques at end-manufacturing sites.

Identifying the key features that contribute to electrode wear is the first step toward establishing a robust production process. The issue for aluminum has never been producing 200 welds but rather strategies to reach 2000 welds before electrode replacement. Performance can only be optimized when the interaction between the material features and the process variables are understood. New techniques have been developed that help illustrate the behavior of the electrode's erosion on weld consistency. This has enabled process control and manufacturing strategies that can dramatically lower the overall manufacturing costs of aluminum resistance welding.

Understanding Electrode Erosion and Wear
Present research suggests the weld current flows through nonuniform fractures in the oxides of the sheet surfaces (Refs. 1, 2). The constriction in the current flow causes excessive heating, resulting in localized surface melting and alloying with the copper electrode. If the electrodes are not re-dressed, the process accelerates resulting in face diameter growth. The electrode geometry does not mushroom as commonly seen with steel but rather erodes away, retaining, in most part, the original sidewall (Ref. 3). Figure 1 illustrates several worn electrodes, as compared to the starting profile for electrode life testing of 1.0-mm 6111-T4 using AC pedestal welding equipment. As the electrodes erode, the net face diameter increases, reducing the average pressure and current densities, ultimately resulting in undersized welds. In addition to the diameter growth, the surface roughness of the electrode face substantially increases.

In order to develop the fundamental understanding of the mechanical contact along the electrode and faying surfaces, an approach was developed combining advanced surface topography measurement tools and three-dimensional finite element model (FEM) analysis (Ref. 4). The methodology is unique in that three-dimensional images of worn electrodes were used to create a model that predicts the mechanical contact distribution at the electrode to sheet and interfacial surfaces for various weld forces. The contact model can simulate a variety of effects such as alloy, temper, gauge, and weld force.

Results from the FEM analysis are shown in Figs. 2-4. The plots represent pressure maps along the electrode to sheet and faying surfaces for 1.0-mm 6022-T43. The baseline welding force for the particular electrode type and stackup is 2.67 kN (600 lb). Additional runs were also provided for 33 and 66% increases in weld force. The faying surface pressure map shown in Fig. 4 is a result of the contribution from both the top and bottom electrode contact regions.

Fig. 2 —Integrated weld pressure maps of top sheet for worn electrodes at 2.67- to 4.45-kN force. A — 2.67 kN; B — 3.56 kN; C — 4.45 kN.

Fig. 3 — Integrated weld pressure maps of bottom sheet for worn electrodes at 2.67- to 4.45-kN force. A — 2.67 kN; B — 3.56 kN; C — 4.45 kN.

Fig. 4 — Integrated weld pressure maps of faying surface for worn electrodes at 2.67- to 4.45-kN force. A — 2.67 kN; B — 3.56 kN; C — 4.45 kN.

The pressure maps show contact areas, displayed as blue, which are at or near the yield point of the sheet. Assuming that fractures in the surface oxides will predominately occur at these yield point regions, the pressure profile maps allow us to visualize where the current will flow across the welding interfaces. It is clearly seen that applying 2.67 kN of force with the worn electrode topographies produces little contact area where the faying surface is near the material yield point. The fraction of material contact area increases dramatically as the electrode force is increased to 4.45 kN (1000 lb).

The model results would suggest forces higher than the baseline promote more uniformly distributed yielding at the faying surface with worn electrodes. In support of these findings, Fig. 5 shows examples of undersized peel specimens that were produced at the end of a constant force (2.67 kN) electrode life trial on 0.9-mm 6022-T43. The welds only occurred within a small area of the electrode face contact. Additionally, very little if no heating was observed in the bulk of the faying surface contact region, suggesting poor contact in these areas. These figures are very similar to Fig. 4A, which shows only small, localized pressure points at the faying surface capable of oxide disruption and current transfer. The model suggests that changes in the weld force may help improve the uniformity of the pressure and current distribution at the electrode and faying surfaces.

Strategies for Improving Electrode Life

Electrode Dressing

While the type and frequency of electrode cleaning is open to debate, the overall effect is to disrupt the pitting and erosion phases, thereby eliminating the gross changes in electrode topography and diameter. The main motivation for dressing is the observation that extremely high-quality welds can consistently be achieved when the electrode has minimal wear, i.e., early in the electrode life process. Secondary advantages such as reducing primary demand through eliminating current stepping, longer cable life, and reduced inspections/teardowns may also be realized with electrode cleaning. In spite of the advantages of electrode cleaning, many applications cannot incorporate this type of equipment due to infrastructure and timing constraints.

Polarity and Dissimilar Gauge Stackups
The majority of production applications commonly weld various gauges within a stackup. The polarity effect observed in DC power supplies may provide an additional level of complexity. In general, DC systems may have lower electrode life and a greater tendency of electrode-to-sheet sticking than comparable AC systems due to the increased erosion of the positive (anode) electrode. A series of trials was designed to develop an understanding of the effects of both polarity and stackup orientation for Alcoa's welding guidelines.

Table 2 shows the overall test matrix for welding 0.9- and 2.0-mm 6022-T4 with MP404 lubricant. Four stackup combinations were evaluated (designated from top to bottom): 0.9 mm to itself, 0.9 mm to 2.0 mm, 2.0 mm to 0.9 mm, and 2.0 mm to itself. The polarity of the medium-frequency direct current (MFDC) power supply was specified such that the top material in the stackup contacted the positive (anode) electrode, which was also connected to the pressure stroke. Additionally, two types of electrode geometries were used in the testing: truncated (A-nose) electrode with a face diameter sized according to the guidelines in Table 1, and an F-nosed electrode with a 76-mm radius. All tests results shown in Table 2 employed no current stepping or electrode dressing.

The results showed that significant increases in electrode life performance were obtained when the positive electrode was in contact with the thicker gauge in the stackup. Additionally, the performance was enhanced when the positive electrode geometry was truncated as compared to a full radius. The general trend indicates electrode life improved as the positive electrode's face diameter decreased and its orientation was placed against the thicker material.

Current and Force Stepping
Current stepping is an effective process control strategy employed in steel resistance welding that compensates for electrode growth and wear. Very little information has been published on current stepping for aluminum applications. The obvious implication is that the equipment must be designed to handle currents that may reach between 40 and 45 kA at the end of the stepping sequence. Alcoa has recently completed a large testing program to evaluate the effectiveness of current stepping. In conjunction with the FEM analysis mentioned previously, we also explored the effect of increasing the weld force during electrode life. The current and force were adjusted after every 100 welds during the electrode life tests. The actual stepping factors were developed through empirical techniques. These data are intended to be a starting point and may change slightly depending upon the application and equipment type.

Fig. 7 — MFDC electrode life performance of 0.9-mm 6022-T43 with MP404 lubricant (Alcoa electrode face diameter, no current stepping, no force stepping).	Fig. 8 — MFDC electrode life performance of 0.9-mm 6022-T43 with MP404 lubricant (standard electrode face diameter, current stepping, no force stepping).
Fig. 9 — MFDC electrode life performance of 0.9-mm 6022-T43 with MP404 lubricant (standard electrode face diameter, same current and force stepping factor).

Assuming that electrode dressing is unavailable, Figs. 8 and 9 illustrate the impact of current and force stepping. Figure 8 shows the electrode life trial results produced with current stepping only. The addition of current stepping dramatically reduced the occurrence of undersized welds and increased the electrode life to 1600 welds, approximately a 2¥ improvement over the baseline of Fig. 7. While the vast majority of peel diameters were between 5√t and 6√t, the maximum diameters approached 7√t after 1500 welds. Figure 9 illustrates a trial with a force stepping profile identical to the current step. The minimum weld button size started to decrease at 1500 through 1800 welds but in general exceeded 4√t. After 1800 welds, the minimum size exceeded 5√t until the trial ended at 2300 welds, approximately 3¥ higher than baseline.

Current stepping can be run exclusively, without force stepping, as is traditionally practiced in production. Force stepping adds another degree of precision since it maintains increased uniformity of the pressure and ultimately current density distribution across the welding interfaces. Additionally, force stepping may ultimately reduce electrode erosion over current stepping alone since it lowers the contact resistance at the electrode and sheet interfaces.

Power Supplies: Conventional and Low Energy
The majority of power supplies for aluminum applications that have recently come online employ medium-frequency direct current. There are multiple advantages but MFDC offers the greatest flexibility to weld a variety of gauges and materials within the smallest and lightest package. Medium-frequency direct current systems now weld both aluminum and steel components in the same welding cell with little downtime during changeover. Thus, the technology allows the end user to make greater use of capital, lowering manufacturing expenses while maintaining the flexibility to rapidly change products as demand requires. Additionally, MFDC technology is now being paired with servo weld cylinders on several aluminum closure panel applications.

There are alternate power supplies such as single-phase AC and conventional DC that can also produce welds with the same quality as MFDC. These technologies offer a lower capital cost but may require larger robots and primary power feeds, potentially offsetting any initial savings. Typically, AC power supplies yield better electrode life performance when compared to DC systems but are not as flexible for welding multiple materials or gauges within the same cell. Medium- frequency versions that can alternate polarity are also being developed that may offer the electrode life advantages of AC but the lower, three-phase input of MFDC.

Further Cost-Reduction Options: Nondestructive Examination
Destructively testing panels is the most common form of quality assurance employed in resistance welded aluminum structures today, but does it have to be? While the individual costs vary across suppliers, roughly $0.50 of every aluminum closure panel produced can be attributed to teardowns. In an industry where several million closure panels are produced each year, the total savings opportunity is estimated at over $1 million. Other issues typically not recognized, such as lost productivity, ergonomics, and scrap management systems, also contribute to the cost of destructive teardowns.

A multiyear research program has recently been completed between Alcoa and ScanMaster-IRT that suggests ultrasonic testing (UT) can substantially reduce the need for destructive teardowns for aluminum resistance welds. In this program, upward of 3000 aluminum resistance spot welds on 0.9-mm 6022-T43 were analyzed by a variety of operators. The results indicated that overall accuracy between the nondestructive evaluation and actual peel performance was more than 90%. More importantly, the discrepant weld accuracy, i.e., the accuracy of nondestructively classifying an undersized weld, exceeded 95%. Examples of A-scans obtained on portable UT equipment for both an acceptable and a discrepant weld can be seen in Figs. 11 and 12.

Fig. 11 — Ultrasonic A-scan of an acceptable aluminum resistance weld.

Fig. 12 — Ultrasonic A-scan of a discrepant aluminum resistance weld.

Conclusions
Recent advancements in resistance welding technology coupled with new alloys, process developments, and nondestructive examination techniques have reduced the manufacturing cost for aluminum sheet structures. Resistance welding technology now can process both aluminum and steel components in the same manufacturing cell, dramatically reducing the capital costs while expanding flexibility to meet every changing market demand. Several strategies that improve electrode life for aluminum automotive sheet components based upon frequent electrode conditioning, optimized starting electrode face diameters, current stepping, or force stepping were illustrated. Production cost savings in terms of reduced line stoppages, maintenance, and inspection/teardowns may be realized with one or a combination of methods. Additionally new technologies such as capacitor discharge and ultrasonic NDE were presented as potential cost-reduction opportunities for current and future aluminum resistance spot welding applications.

2. Fukumoto, S., Lum, I., Biro, E., Boomer, D. R., and Zhou, Y. 2003. Effects of electrode degradation on electrode life in resistance spot welding of aluminum Alloy 5182. Welding Journal 82(11): 307-s to 312-s.

3. Boomer, D. R., Hunter, J. A., and Castle, D. R. 2003. A new approach for robust high-productivity resistance spot welding of aluminum. SAE Technical Paper 2003-01-0575, Detroit, Mich.

4. Spinella, D. J., Brockenbrough, J., and Fridy, J. M. 2004. Fundamental effects of electrode wear on aluminum resistance spot welding performance. Proceedings of the Sheet Metal Welding Conference, Detroit, Mich.

5. Arrington, S. E. 1995. Twisting electrodes improves tip life and weld quality on resistance spot welded aluminum sheet. SAE Technical Paper Series 950717, Detroit, Mich.

6. Patrick, E. P., and Auhl, J. R. 1986. Resistance Welding Method. U.S. Patent 4,633,054.

7. Spinella, D. J., and Patrick, E. P. 2002. Advancements in aluminum resistance spot welding to improve performance and reduce energy. Proceedings of the Sheet Metal Welding Conference, Detroit, Mich.

8. The Aluminum Association. 2002. Welding Aluminum: Theory and Practice, Fourth Edition.