July 13, 2004
The automotive industry worldwide has experienced dramatic changes in the last 10 years. Challenges facing the industry include increasingly stringent safety rules; requirements for dramatic improvement in fuel consumption; and the necessity to maintain or even reduce the vehicle price, even as warranties are being extended.
The welding industry has responded to these challenges with new joining processes using intelligent control.
New welding processes, improvements to existing processes, and the combinations of two welding processes (called hybrids) have enabled increased joining travel speeds, the welding of previously unweldable materials, and the implementation of new designs and materials.
Realizing the full potential of these processes requires new intelligent process control techniques. Controls and sensing systems that can adapt to the actual shop tolerances are required to bridge gaps, manage the joint location variance, and compensate for variations inherent in the welding process.
In addition, inspecting the weld beads for defects and conformity to applicable standards reduces scrap and ensures stable product quality levels. Optimally, this weld quality information is included in a closed-loop control and fed back to the actual welding operation in as close as possible to real time.
Some of the processes that have advanced the most include gas metal arc welding (GMAW) and laser beam welding (LBW), as well as the hybrid combination of these two processes, called hybrid laser/GMAW.
Laser welding is considered to be one of the most precise and efficient ways to weld automotive and aerospace mechanical components. However, because the spot of energy projected onto the joint is small (see Figure 1), it is not easy to attain a perfect fit. In addition, the expense of the equipment and support personnel can make laser welding cost-prohibitive for automotive assembly plants.
Welding formed parts requires increased part and tooling accuracy to render the joint preparation and fit-up compatible with the requirements of the process.
Despite these constraints, the competitive automobile sector remains interested in laser welding because of its speed and efficiency, as well as its compatibility with new car design criteria and trends.
With Nd:YAG laser welding, the laser light beam is delivered through an optical fiber to the welding torch, which is connected to the numerically controlled machine or robot wrist that follows the preprogrammed joint path. Nd:YAG's method of automating the delivery of energy to the joint offers possibilities not achievable with CO2lasers, because CO2laser beam delivery requires the part to be welded to be moved, rotated, or tilted to access the area to be welded. Because it is more cumbersome to move large and heavy parts, the size and complexity of the parts that can be welded efficiently by CO2are limited.
Laser welding is an autogenous process because, in principle, no filler wire is needed. Its dense, high-energy properties enable simple joint designs. However, when joining parts with imperfections such as irregular gaps or mismatches, the absence of filler metal means that no additional material is available to bridge those gaps and to compensate for excessive joint variation. Filler metal also is useful in compensating for the loss in the final weld chemistry and to reduce the solidification cracking especially common with aluminum and galvanized material. In addition, it helps maintain joint strength.
Consequently, the position of the focal point in autogenous welding is critical and is dependent on the joint conditions. The position of the focal spot relative to the joint requires accuracy greater than 0.05 millimeter, which is stringent compared to typical arc welding processes. In addition, process operational parameters need to be maintained within a small window (see Figure 2).
This is particularly true with square butt joints such as those used with tailor welded blanks, but it also is true with 3-D parts. In addition, gaps greater than 0.1 mm may produce defects because the interaction between the light beam of energy and the edge of the part to be welded is either nonexistent or unstable; consequently, a fusion zone either is completely absent or only partially present.
This is why it is important to measure the lateral position of the joint. Measuring the vertical position of the joint top surface can help control or adapt the laser spot size to the joint fit-up conditions or gap.
Adding a second energy source and filler metal may widen the application range and increase the flexibility of laser welding; however, it is even more important to use laser sensing to measure the joint geometry and position to adjust the resulting additional parameters.
For example, when welding tailor welded blanks in which joint gaps are created by the geometric configuration of the blank itself (see Figure 3), combining Nd:YAG with precision cold-wire filler helps achieve full penetration and reduce or eliminate defects. In this case, the filler wire speed and its position relative to the keyhole can be controlled to improve weld quality. In-process measurement of the gap joint can provide the data to control the amount of filler metal needed to ensure the proper joint weld nugget formation and to bridge the gap.
Using hybrid laser/GMAW to adjust the process output characteristics can control even more operating parameters than Nd:YAG with wire filler, including real-time positioning of the aiming point of wire and GMAW current, wire feed speed, voltage, and pulse parameters.
For every welding process, the actual operating phase can be maintained inside a process stability envelope. Within this envelope, stable process speed, joint penetration capability, heat input, and metallurgical characteristics can be obtained. As the process stability envelope increases, the operational weldability of the process also increases.
Combining processes such as laser welding with GMAW and Nd:YAG with cold wire offers a larger process stability envelope and operational flexibility than each process alone.
A workpiece's weldability is determined by both the geometric and metallurgical attributes of the joint. High-speed laser cameras calculate the joint and part data, compute the updated parameters in real time, and send the control point to the welding equipment. Adaptive control allows the adjustments to be made as changes are detected.
Controlling the process by measuring the joint attributes and setting the process parameters accordingly can help obtain a weld within the acceptable limits. For example, if the gap is between 0.5 and 1 mm, a weave routine can be used.
2-D vision systems can sense most of these attributes; however, for a camera to be useful, the joint or weld bead contour must be sensed with an accuracy of 5 to 10 microns at a pixel rate of about 200 MHz, which requires a 3-D camera. This degree of speed and accuracy enables the joint gap and mismatch to be measured in front of the torch, and the weld bead profile characteristics and defects to be detected behind the torch. 3-D contour digitization detects minute weld defects and acquires enough information to track the joint at a speed of 1 to 20 meters per minute, which is compatible with the laser welding process speed.
The laser camera's primary function is tracking the laser joint. Because this function is performed during welding in front of the laser and keyhole, the laser camera must not be affected by spatter, heat, or light emitted around the torch.
Processing and feature extraction depends on the joint type. Because the tracking precision must be within ±0.05 mm, the torch motion generated by the laser tracking system must be carefully controlled, accounting for the position of the focal point, the speed and the position of the torch, and the camera orientation in the machine reference coordinates system.
Defining the laser's actual focal spot—the tool-center point of the welding robot—is more difficult to establish than the GMAW wire tip, but it is important to pay attention to it for optimum tracking accuracy.
Seam tracking can be affected by several mechanical issues. A two-axis precision actuator can be mounted on a robot wrist or on the linear axis of a tailor welded blank laser welding machine (see Figure 4). This type of actuator must be free of backlash and precise to the micron. An alternative is to use an oscillating mirror to deflect the power beam.
Because laser welding is used for assembling critical components in high production, the acceptable reject rate must remain smaller than 1 percent.
Under these conditions, welds produced at several meters per minute cannot be inspected efficiently by the human eye, which is why special laser cameras have been designed that verify the geometric profile of the bead and identify defects larger than the limits set by the standards (see Figure 5).
Inspecting tailor welded blanks for defects and correct geometric bead conformity is the most common task performed today by laser vision-based inspection systems. However, other joint types also are critical in the body-in-white structures on which lasers are used to make stitch-mode penetration welds or laser brazed joints. These types of joints must meet mechanical strength, cosmetic appearance, and water tightness requirements.
Car body inspection at speeds up to 10 m/min. for pinholes as small as 0.3 mm in diameter requires special laser cameras that contain twin imagers. Each imager located inside the same mechanical camera reference frame shares the work to be performed by the inspection process.
The lead imager has a wide field of view but is less accurate than the trail imager. Its role is to track the brazed joint position, which varies depending on the position and movement of the car body within ±3 mm. The trail imager is very accurate but has a smaller field of view. The field of view can be perfectly centered by the control information it receives from the lead imager (see Figure 6).
The principles of tracking arc joints and controlling arc welding processes are identical to those for laser and laser hybrid welding.
Intelligent controls such as these have enabled high-speed tandem arc welding of steel components and have contributed to the improvement of aluminum chassis part weld consistency. This intelligent control can be introduced immediately at the prototype stage to produce key information about the part, fixture, and weld process capability. This information can be used to make changes before the final production assembly line equipment and tooling are purchased, thus avoiding later costly changes and launch problems.
Jeffrey Noruk is president of Servo-Robot Corp., 11121 Riverland Court, Mequon, WI 53092, firstname.lastname@example.org, www.servorobot.com; Jean-Paul Boillot is president and CEO and Frederic Arsenault is product manager with Servo-Robot Inc., 1370 Rue Hocquart, Saint Bruno, Quebec J3V 6E1E, email@example.com, www.servorobot.com.
The FABRICATOR® is North America's leading magazine for the metal forming and fabricating industry. The magazine delivers the news, technical articles, and case histories that enable fabricators to do their jobs more efficiently. The FABRICATOR has served the industry since 1971. Print subscriptions are free to qualified persons in North America involved in metal forming and fabricating.