GIVE YOUR ROBOT THE EYESIGHT AND INTELLIGENCE NEEDED TO BECOME A TOP LEVEL WELDER
Over 90 percent of installed robotic arc welding systems are not running up to their full capability. Jeff Noruk of Servo Robot shows some ways to kick that performance up a notch.
Posted: October 13, 2010
C is for Champion
This is one of the most important ingredients because a champion is needed to fight the negativity which is for sure going to come up during some phase of the project. The Champion is the one person who is truly committed to the success of the project and will keep the team focused and fight through the naysayers when problems occur. Ideally the Champion is also the project manager, economic buyer or the technical expert.
D is for Design for Manufacturing/Automation/Automated Inspection
First-time robot users must realize that every operation upstream before the robot welding cell must be reviewed to make sure they are compatible for automation. What this means is that one needs to understand the capability of operations like stamping, forming, casting, etc., as well as the variability introduced during the assembly of the components that are presented to the robot for welding. Depending on if the parts are pre-tacked or put into a fixture, further variability of the joint location and fit-up may be present. If this total joint variability does not fit within the capability of the welding process, one must again consider using a sensor to gather the right information to provide the required compensation to the robot path and welding parameters. See Figure 2 for an example of a real-time laser vision seam tracker.
I is for Inspection
Inspection is used for go/no-go decisions on whether a part meets minimum requirements but, just as important, automated inspection can provide trending information that can be used to prevent a defective part from being made. Right now companies are doing a hodge-podge of human visual inspection, including other NDT methods such as PT, MT, etc., as well as destructive testing such as cross-sectioning and bending. Instead of this haphazard approach, one needs to work upfront to develop an overall quality plan that will provide the ability to achieve Six Sigma quality and continuous improvement. Figure 3 shows an automated weld inspection system and the output screen indicating clearly what the weld measurements are, what defects are present, and how this information is categorized so that actions can be taken to improve quality. As the old saying goes, “what gets measured gets improved.”
T is for Team
A true team is made up of everyone having a vested interest in the success of the project. This team includes not just the company engineers, technicians and welders but also the purchasing agent, integrator, suppliers of peripheral equipment and even the companies supplying the steel, weld wire and gas. It is better to over communicate as to what is required to make the project a success than to keep anyone in the dark that could offer a useful solution to a problem. While it is important to have a great team, one must also put in charge of the team a leader who is the same Champion that was discussed earlier.
T is also for Technology
No one should install technology just for technologies’ sake. Before looking at automation, one needs to simplify and optimize your existing production. The automation project must have an acceptable ROI and, of course, one must prepare the people in the plant who must make it work so that they do not resist it and try to make it fail. One must look at cutting edge technology without going into the “bleeding edge” zone. See Figure 4 for an example of a new portable welding robot that can offer unique solutions where a large articulated robot may not be the right solution.
CONCLUSIONS
(1) Start with the end in mind. The end must include a suitable ROI with a world class OEE.
(2) For a project to be successful there must be a champion and a team in place.
(3) Use best practices that have resulted in thousands of successful robotic arc welding installations. Don’t reinvent the wheel.
(4) Use technology appropriately. Automating an out-of-control process will simply produce bad parts faster.
(5) Own the Process. This means the process capability is known and is acceptable. If not, consider using the appropriate sensor to compensate for excessive variability.