Purchasing Your First Robotic Welding System
by Chester L. Woodman
Lincoln Electric, Automation Division Posted 02/23/2009
If the welding industry has anything to learn from the past decade, it is that the stakes of doing business have permanently been raised. The 1990's were characterized by an enhanced emphasis on quality, reducing costs to compete in a global market, optimizing floor space for maximum output, training and maintaining a skilled workforce. As a result, the major question for the new millennium is not if any particular metal fabricator will adopt robotic welding technology, but rather when and how.
Automation will be a matter of survival, not only for large shops creating a high volume of pieces, but for the small fabricator, for whom the ability to raise productivity while reducing costs will determine the actual viability of his business.
More often than not, first-time buyers of robotic systems approach the purchase decision with trepidation. They are concerned about making a mistake - a big, expensive mistake. To address this fear, we must dispel the myths surrounding robotics in general, and specifically, robotic welding systems.
1) It takes a 'rocket scientist' to program a robot.
2) Only high-quantity production runs justify installing a robotic welding cell.
3) Installing a robot will solve all welding quality problems.
4) The robot operator must be a highly trained, skilled and compensated employee.
5) Robotic welding cells are very expensive and difficult to cost justify.
6) A robot can weld any part that can be welded manually or semi automatically.
7) Robots cannot be used to weld very large parts or assemblies.
8) The choice of a specific brand of robotic welding cell will be critical to success.
1) Programming a robot is very simple. Even workers who must overcome a language barrier can learn to program a robot in two days, thanks to the simple interactive screen on the pendant.
2) It is not necessary to dedicate a robot to a single task such as making only one-piece part. With the number of welding parts programs that can be stored in a robot control unit's memory, it is possible to go from one part to another part very quickly if the tooling nests are properly designed for quick change. Several different parts can be made in the same welding cell in a given day.
3) No robot can solve a welding quality problem all by itself. If the parts are not designed properly, the piece parts are not made properly, or the welding joints are not properly prepared or presented to the robot, there will be problems with quality.
4) Becoming a highly skilled welder takes years of experience, training and practice, whereas a robotic welding cell operator only has to load the part, press the appropriate buttons to activate the machine, and then unload the part. The training of a robot operator can literally take less than an hour.
5) Following the market trends of the hand calculator and the desktop computer, the actual dollar cost of a robotic welding cell has dropped dramatically in the past 10 years. During the same period software capabilities, programming ease, motion speed and accuracy have all been enhanced. The upshot of this is, that for a much lower cost, a robotic welding cell now offers far superior performance.
6) It is not true that a robot can weld any part that can be welded manually or semi automatically. Clamping requirements, access problems or specific positioning requirements may make the use of a robotic welding cell impossible or impractical.
7) Robots can be put on tracks or gantries, giving them the ability to weld parts that are 40 to 50 ft long and 8 or 10 ft wide.
8) Software support and expert part design assistance will help to provide success in robotic welding applications. In most cases, a specific piece of hardware will not be a key success factor. Application expertise, however, is critical.
Making the Decision
Once the myths about robotic welding cells have been debunked, a company will be ready to evaluate the benefits vs. actual costs. It cannot be stressed too strongly that this process will be most effective if it is done in conjunction with the proposed robotic cell vendor. Involving the vendor in the initial stages of piece part design will greatly enhance the fabricator's application of robotic technology to specific product line and production needs. Reputable vendors are happy to provide this assistance; they know that the successful implementation of a robotic welding system resides not in the hardware itself, but in understanding and meeting the fabricator's production needs.
The decision-making process begins with a detailed review of the following items: the parts to be welded; weld joint accessibility; repeatability of the parts; tooling nest (or fixturing) requirements; ways to compensate for distortion; and determination of the welding process to be used.
As the decision process evolves, the vendor and the fabricator will continue working together to determine the appropriate system accessories, including safety devices, the optimal layout for the robotic cell, manpower and training requirements, and service and maintenance requirements (internal vs. outside vendor support).
The robot cell layout must consider not just providing space for the work motion device, power source, robot controller and wire feed package, but how the piece part is delivered to the area, and how the finished part leaves the area. Workflow simplicity characterizes a good cell layout.
Watch Out for Pitfalls
It should be clear by now that the vendor who views the order for a robotic welding system as a one-time-opportunity sale will not provide the total support net that is key to the customer's success. A vendor who understands the fabricator's business, and how to produce that fabricator's product with or without robotics, will provide vital input during the lengthy decision and design phase, as well as essential ongoing support after the system is on-line.
While it is true that most robotic welding systems are much simpler and easier to use than the average fabricator might think, it is also true that the move to automated welding systems presents a set of problems that did not exist (or at least were not significant) when a welder controlled the torch. The skilled welder is able to compensate for sloppy or ill-designed fixtures, varying trim lines and dimensions of piece parts. An experienced welder is also capable of overcoming problems with the welding equipment, wire delivery or the shielding gas supply.
Robotic welding systems demand closer attention to quality in the piece part manufacturing process. In addition, clamping and fixturing must be absolutely precise. These requirements simply point out the importance of considering tooling nest design issues early in the purchase decision process.
When a part is fixtured and the production tolerances become apparent for the first time, a human operator can make a choice between adjusting welding parameters and technique, or rejecting the parts. Now, robotic welding systems can incorporate vision capabilities that provide for adjusting electrode position and welding parameters to permit quality automated welding even on parts with variations.
When the proper decision-making process is followed, it is often possible to justify the cost of a robotic welding system by relying upon somewhat conventional measures. This has become truer in the past decade, as robotics have offered increasingly better performance at sharply lower prices, while labor and benefit costs have continued to rise.
The rigorous training necessary to produce skilled welders, the relative scarcity of trained welders in many regions of the country, and the potential effects of worker turnover are seldom-considered welding costs. The economic consequences of these factors may be subtle, but they are nonetheless significant.
The problem of finding enough highly skilled welders to fill a sudden, large order, has affected many companies in our industry at one time or another. Today, in fact, fewer new welders are being trained, and many experienced welders are approaching retirement. Any effort to cost-justify a robotic welding system should take into account the relative ease (and minor expense) of training an individual to load and unload a welding cell. This can take a matter of minutes or hours vs. the many years needed to develop a fully qualified welder operator. The scarcity of highly skilled labor means we must also factor into our cost-justification scenario the cost of poor quality, including rework, scrap and, most important, customer dissatisfaction.
Worker and environmental safety factors give robotic welding systems another edge in the cost-justification process. Failure to comply with OSHA and EPA standards can be expensive. Lastly, developments in electrode technology and welding torches, when combined with automation enhanced with vision systems, permit welding at speeds great enough to achieve overall cost reductions.
In California, a small manufacturing company that makes creepers for automotive mechanics suddenly experienced a large increase in the volume of orders for its product. With a permanent work force of about 20 employees, the company owner immediately tried to hire enough skilled welders to respond to the demand. He was unable to do so and resorted to ordering a robotic welding system that was shipped and installed within ten days. With some easy programming, simple tooling and straightforward operator training, the robot was productive very quickly. Using the system, one operator is able to weld as many creepers as four skilled welders previously did. For an investment of approximately $65,000, the creeper manufacturer was able to quadruple the productivity of a single worker and meet his customer's needs.
In a different type of application, a company that makes the steel crossbeams that support the bodies of tractor/trailer rigs ordered a robotic welding system. In this case, the company developed its own automated tooling to 'customize' the performance of a standard robotic welding cell to its specific needs. Employing several hundred people, it is a significantly larger company than the creeper manufacturer. The steel beam manufacturer invested about $135,000 in the robotic welding cell, and approximately another $50,000 to $60,000 to develop the automated tooling and fixturing. With this investment, productivity was improved more than 300 percent, and quality is excellent.
Quality, productivity and safety -- this is the triad of competitiveness. Robotic welding technology now delivers all three at a lower cost, with more flexibility and with greater ease of implementation than most of us could have imagined ten years ago. Remember, the stakes are higher than ever before. It is time for responsible fabricators, large and small, to discard the myths about robotics and investigate the realities for themselves.