Welding in the '90s.
The keys to future economic survival will be maximizing product efficiency, rapidly assimilating new technologies, and increasing of professionalism throughout the industry. Technical literacy must be stimulated and coalitions formed among welding business and professional groups worldwide.
Towards this end, the American Welding Society (AWS) has been working to establish working relationships-the exchange of standards, educational texts, and commercial dialog-between AWS and peer societies in other countries. AWS has such agreements with Argentina, Australia, Brazil, Germany, Hungary, Japan, Korea, Mexico, China, Portugal, Singapore, Spain, the United Kingdom, and the USSR.
Yet, we still see too little cooperation among standard-setting bodies in the US. Reasons include our entrepreneurial spirit, a lack of coordination with the government, fear of violating FTC prohibitions on collusion, and in some cases, simply "turf protection."
The need for high tech
Increased emphasis will be placed on welding as a basic manufacturing technology, particularly the use of lightweight, high-strength materials, and higher levels of reliability and quality. To compete, US fabricators will need more automatic and semiautomatic welding for repetitive tasks to reduce costs, manufacturing defects, and environmental hazards.
The trend is to make welding safer and welding machines simpler through the use of microelectronics. Arcs are more stable and current pulses more precise for critical welding jobs. Workcells are becoming more integrated-manipulator, control, positioner, power source, wirefeeder-and cutting equipment more lightweight, portable, and modular. Welding software has become increasingly tailored to specific applications. Advanced materials
Slowly but surely, advanced materials will provide product differentiation and increased end-user value on a worldwide basis. New materials will fundamentally change all industry, and joining will become more broadly based than just the welding of metals. Everyone from designer to welder must become more knowledgeable about new materials and more accepting of related new fabrication technology.
While we don't expect a sudden shift to new materials, the trend towards them will persist-introduced first in relatively low-volume niche areas. Not all issues with these materials are resolved. Designers do not yet have all the necessary information to assure their reliability. Welding equipment manufacturers and fabricators must join to develop the technology to use them. Educators must train everyone from designers to manufacturing engineers to welders to use them also.
Design criteria must be made more realistic and practical. Over-design is too prevalent today because design engineers who do not have a good handle on what welding can do will typically overspecify welded joints. Inordinate concern about product liability makes the problem worse. When everyone in the design chain adds a 5% safety factor, we're soon at a 20% to 30%
The case for sensor control
Using sensors to help automate robotic arc-welding can result in considerable cost reduction, more precise positioning of the torch above the joint, improved weld quality, and reduced fixturing requirements and workpiece-preparation time. More importantly, it drastically reduces the tight-tolerance requirements for the workpiece, robot, and fixturing system.
Weld quality problems often occur with larger parts such as car bodies and frames. Repeatability and consistency are common concerns.
There are four choices in welding large structural components:
1. Manual welding which is subject to human error and high per-piece cost.
2. Dedicated welding machinery where joint location must be very accurate and edges need trimming by expensive die sets.
3. Robotic systems where the robot follows a programmed weld path and angle of attack, and workpieces must also be kept within tight tolerances to produce quality welds over long periods of time.
4. Sensor-controlled robotic arc welding where a noncontact measurement sensor on the robot arm guides the welding function. Weld parameters and robot position are adjusted based on these measured values, and seam-location searches and width variations can be easily accommodated.
Of the many types of seam sensors available, only a few can withstand the intense environmental conditions of a welding operation. The sensor must not lose accuracy or resolution in a high-speed environment of welding smoke, electromagnetic interference, and changing work surface.
One sensor manufacturer, Selective Electronic Inc (Selcom), Southfield, MI, has 400 seam-finding sensors in use worldwide. Their system includes a noncontact sensor, processing electronics, and controller that is easily integrated into the robotic-control system and programmed through the robot's digital outputs. It is particularly useful for thin-sheet resistance welding and plasma-arc welding requiring short welds and cycle times. The correct welding parameters are selected within a search time of 1.5 to 2.0 sec in three dimensions or 1 sec in two dimensions. Search accuracy is -0.16" ([+ or -] 0.4 mm), depending on the robot.
Compared to other gaging techniques, optoelectronic sensors are more adaptable to high-speed measurement that is accurate and repeatable. Newer fixed-point laser triangulation designs use solid-state laser diodes instead of gas-laser tubes and can be pulsed at high frequencies (16 or 32 kHz). This is particularly beneficial in background-light discrimination. Sensing of all types of surfaces is not affected by welding sparks.
A typical optical-triangulation sensor consists of an infrared light source, camera/detector, and electronics for signal amplification, light-source control, and measurement calculation. The Selcom Optocator is small in size, light in weight, and insensitive to welding spatter and fumes. As the distance from sensor to surface changes, the position of the focused scatter image on the detector shifts. That shift, through triangulation, directly translates into vertical displacement of the measured surface. Measuring range of the Selcom sensor is 1.25" (32 mm) with a resolution of 0.002" (0.06 mm).
During searches (without the arc actuated), start and stop points of the weld joint are defined in two or three dimensions. During welding, weld parameters can be adjusted for variations in the gap, and, depending on the sensor and robot used, accuracy is typically [+ or -]0.016" ([+ or -]0.4 mm).
A seat-welding installation at Saab Automobile, Kristinehamn, Sweden, chose the Selcom seam finder for these primary reasons:
1. Rapid, accurate, noncontact locating of thin-sheet joints.
2. Measurement unaffected by oil or paint on the part surface.
3. Wider tolerances on mechanical dimensions and workpiece positioning.
4. Easy integration of seam finder and the ABB robot.
Seam finding is used on the car seat's back rest. A slightly dished plate with holes and slots is welded to the seat side frames. Curve of the plate varies from plate to plate and requires a height search.
After the height is defined, lap joints are located and welded. One joint is a series of 0.12" slots that are filled and welded to an underlying part. The edge of each slot is first located, and the controller then offsets the weld program to start in the middle of the slot.