HF butt welding a new process for tailored blanks

TWI Bulletin, September - October 1995

Steve Westgate is currently Technical Specialist - Resistance Welding in the Arc, Laser and Sheet Processes Department. He graduated with BSc Hons in Industrial Metallurgy at the University of Birmingham. Since joining TWI in 1972, Steve has had involvement in, and managed, a wide range of research projects in all resistance welding processes, including European and group funded projects. Topics have included: process and weldability studies, instrumentation, quality monitoring and NDT in resistance spot, seam and projection welding, plus flash, resistance butt and HF welding. Close contact is maintained with industry through an active consultancy and trouble-shooting role.

High frequency butt welding is a process recently developed for use in the automotive industry by Volvo Car Corporation. Evaluation of this process was conducted jointly by TWI, UK and Edison Welding Institute (EWI), USA, at Volvo's Olofström plant in Sweden. Steve Westgate summarises.

Tailored blanks have made an impact in the automotive industry for pressed sheet components. Larger pressings can be made than with standard-sized sheets, and dissimilar steel types or thicknesses can be joined. Resulting benefits are improved material use, potential weight saving and a reduction in subsequent assembly operations. Larger pressings could be used to produce complete body sides.

Currently, resistance mash seam welding is the major welding process used for tailored blanks. One machine manufacturer alone reports welding over 22M blanks per year. Typical welding speeds are 2-3 m/min. More recently, laser welding has been adopted for a number of applications and represents roughly 15% of the market. Potentially higher welding speeds are possible using lasers, but they require high accuracy cut edges or special procedures.

Now a prototype high frequency butt welding machine has been produced in an effort to increase productivity whilst minimising special requirements for edge preparation, see Fig.1. The evaluation study by TWI and EWI aimed to establish the influence of welding parameter and production variables, to identify the mechanism of weld formation and to assess the weldability of a range of steels including zinc coated and high strength grades.

Subsequently, the equipment was installed into production at Volvo, complete with automatic material handling equipment.

Equipment and welding sequence

The prototype welding equipment comprised a 390kW, 40kHz HF generator, a control and monitoring console, and the welding module with a manually loaded worktable with a rigid frame structure.

Parts to be welded are loaded on the worktable with edges to be joined positioned against stops. On initiation of the automatic sequence, the sheets are clamped close to the interface and edges brought into contact. The welding coil is lowered between the upper clamps to form a linear proximity conductor above the interface, see Fig.2. Current is delivered to the coil for a preset duration typically one second for one mm sheets. Eddy currents induced in the components heat the weld region. The sheet edges are then forged together with a controlled upset distance, allowing current to continue for a short period to complete the weld. Unclamping completes the sequence which is typically six seconds duration. Process control is confirmed by monitoring the main parameters during welding.

1 - fixed worktable,

2 - moving worktable,

3 - blanks,

4 - water-cooled induction coil,

5 - clamps.

Experimental study

The study comprised three main areas:

• Evaluation of the mode of operation and process mechanism.
• A study of the influence of the welding parameters and process variables using experimental design and statistical analysis.
• An assessment of the weldability of a range of materials representative of automotive sheet materials of interest in Europe and the USA.

Work was conducted on 500mm wide test samples, and weld quality was tested principally using the modified Erichsen cupping test and an angled bend test. In addition, a consistency test of 50 welds was conducted to test the machine under simulated production conditions.

Welding process mechanism

The results indicated that the welds were predominantly solid phase, but the material was raised almost to melting point to allow the small upset distance to be used. Although grain boundary segregation often occurred in the heat affected zone, there were normally no apparent defects revealed by bend testing in such welds. Parent material failure occurred in cupping tests. A typical weld section is shown in Fig.3.

A minor drawback of the process is that up to 10mm at the ends of the blank was not welded. This is because the nature of the eddy current distribution could not generate heat uniformly right up to the edges. However, this region would normally be within the blank holder and be trimmed from the final pressing.

Influence of process and production variables

Weld time was shown to be the most significant factor, and is used as the primary control parameter. Coil height and power level are the other main process parameters which control weld heat, and these are preset to suit the material combination being welded. Figure 4 illustrates results of an analysis of settings used for 1mm low carbon steel.

Quality of the sheared edge was examined and statistical analysis of weld quality under these conditions indicated no significant difference between good and bad sheared edges. However, consistent orientation of the burr is recommended. Initial contact of the sheet edges is preferred for the best results, but the presence of a 0.2mm edge gap (machined in one sheet edge) did not influence the Erichsen test values. In addition, the process was quite tolerant to 15% misalignment of the sheet surfaces.

Welds of good quality gave similar results in the Erichsen cupping test whether tested as-welded or with the upset removed. Thus, post-weld dressing may not be necessary, depending on the surface finish required.

Weldability of materials

A wide range of material combinations representative of tailored blank application has been welded successfully. These include coated and high strength steels in dissimilar thicknesses, as shown in the Table.

Material combinations studied

An example of the results is shown in Fig.5. This indicates the acceptable welding range for 0.8mm low carbon galvannealed steel (IZ) to 2.0mm hot dip zinc (HDZ) coated HSLA steel, together with a typical weld section made with the sheets aligned centrally. The form of the Erichsen cupping test failure and the extent of zinc removal is shown in Fig.6. It was also possible to weld with the sheet surfaces flush.

Presence of zinc on the coated materials did not appear to be a major problem in achieving weld quality and the galvannealed materials gave the best results. In the thickest material combinations (2.5 to 2.5mm low carbon and 2.0 to 2.0mm HSLA steels), acceptable weld quality could not be achieved because of inadequate heat balance through the sheet thickness.

The weld thermal cycle caused local hardening in the weld area with peak values between 160 and 210 HV depending on the material. These values fell gradually to the parent material value outside the heat affected zone and no softened zones were detected.

Process tolerance and consistency

The welding range was narrow in many of the material combinations studied, indicating that process control will need to be good to ensure repeatability during production. However, welding conditions were not fully optimised during these trials on the prototype equipment and further refinements in coil design are planned.

A consistency test of over 50 welds was conducted to simulate production conditions using 1mm low carbon steel and 1.5mm hot dip zinc coated rephosphorised steel. All the Erichsen cupping testpieces taken during the test failed in the parent material at a consistent penetration depth, despite progressive overheating of the welds due to the absence of water cooling in the prototype clamps. This led to the appearance of surface imperfections at the weld surface on bend testing as a result of liquation cracking.

Demonstration welds made in 1mm low carbon steel with a 980mm weld width illustrated the width capability of the prototype machine. Eight Erichsen test samples taken across the width fractured away from the weld at more than 80% parent metal penetration depth.

Conclusion

A detailed study of high frequency butt welding of tailored blanks suggests that the technique is an attractive alternative to existing techniques, with potential for higher productivity using standard sheared edge components.

The process may be applied to a range of steel types such as zinc coated and high strength grades of interest to the automotive industry, including sheet thicknesses in the range 0.8 to 2.5mm.

Consistency of weld quality has been demonstrated and the prototype equipment has been proved in a production application with automatic blank handling in Volvo's Olofström plant. During the startup phase some 75000 tailored blanks were welded and production is now at 3000 units per day.

References