Guest Blog,

FABTECH Q&A with Josh Pawley, Wolf Robotics

By Kelly Clark on on September 15, 2015

Josh Pawley, Applications Engineer, CWI with Wolf Robotics was nice enough to participate in a special Q&A for our blog this week to give us a preview of his session at FABTECH: Automation Application Strategies for Product Mix. Automation systems integrator with a strategic focus on heavy metal fabrication Wolf Robotics, A Lincoln Electric Company, has been integrating robotic systems since 1978. With over 8,400 robotic installations, Wolf deploys the world’s foremost talent in order to address the fabrication needs of today’s manufacturers. At FABTECH this year, Wolf will be presenting strategies and technologies that increase robotic efficiency and relevancy in high mix/low volume product environments. Learn more about Josh’s FABTECH session here and check out the Q&A below: How would you gauge the acceptance of automated welding by the job shop community? Just beginning? Growing aggressively? We continue to see more and more job shop installations, largely driven by the adoption of offline programming, modular workholding designs, belief in the benefits of automation, and further spurred by skilled labor shortages. These same drivers are relevant to product-oriented shops as well, as they apply to any High Mix/Low Volume operation. We see product manufacturing facilities adopting the same technologies and tactics that provide for robotic relevance in job shops because they enable both types of shops to stay competitive in today’s global manufacturing market. What might be holding back acceptance of automated welding in job shops? I think a lot of the hesitation has to do with outdated perceptions. Common impressions of robots revolve around them taking jobs and being too costly to program for new parts or irrelevant for high mix/low volume applications. However, these perceptions fail to take into account that robots still need human operators and programmers; and manual welders can then shift focus to the welding tasks that require higher levels of skill. In addition, a robotic welder will pull the operator away from welding fumes and other occupational safety hazards. Another unfortunate reality is that if manufacturers can’t improve processes to be more cost- and quality- competitive, jobs and products have a higher likelihood of being outsourced. A final note I’ll offer here is that oftentimes a decision-maker will hold a bad reputation for robots because they heard of an unsuccessful robot implementation. This failed installation is less a reflection of a robot’s capabilities and rather a reflection of either a non-ideal application or unrealistically set expectations. Both are prime cases of the importance of learning about robotic integration from a trusted source, analyzing a part portfolio to hone in on the best candidates for robotic automaton, and integrator transparency on expectations at the front-end of a project. How do modular workholding solutions contribute to quicker fixture creation? What type of time-savings are we talking about? The primary value add of modular workholding is in part changeover time. Where conventional bolt-to-platter tools for a specific part number may take 30minutes to multiple hours to changeover and calibrate, new quick-change or modular solutions can be switched out in a few minutes. This reduces robot downtime while allowing for a large part portfolio. What type of CAD files are needed to create the robotic tool path? Are there certain steps that should be taken to guarantee the best path? Generally, a CAD-independent file type such as SAT, STEP, or IGS can be imported into an offline robot programming software in order to create the robotic tool path. Most CAD design suites offer the ability to export their native file type into one of the interchangeable file formats. The best offline path will be the one that requires the least amount of point touch-up on the physical robot. That transferability of the offline program is generally not a function of the CAD export (the CAD file either comes in or it doesn’t), rather the accuracy of the tooling build relative to the design as well as the calibration of the robot system – those two factors have the largest impact on offline programming precision. Both part models and tooling models can be brought into the robot programming suite. This allows for robot reach confirmation, welding torch access checks, and to work out all tooling design kinks before actually fabricating the tooling. How does an automated welding setup cope with the variations in gaps that are common in many fabrications? Do upstream processes such as press brake forming have to deliver much more consistent tolerances than might otherwise be required for typical jobs? A major ingredient for robotic welding success is a consistent upstream process. While there are some advanced welding processes that can help us bridge gaps, gaps remain a large challenge for robots. For large groove welds, the root pass is usually welded manually and the robot is tasked to fill the groove, which takes inconsistent gaps out of the equation. Then, a robot can use a vision system to identify variances in the size of the groove along its length and adapt welding speed and weave to maintain a constant volume fill. Often, we can identify upstream processes that could be easily refined in order to present a more consistent part to the robotic cell. As these processes are identified and improved, the customer often realizes additional gains in weld quality, throughput, and less rework. How do shops maintain quality welds in these automated applications? When a part is first programmed, the weld parameters are “dialed in” to produce a quality and sound weld in a given weld joint. From then on, the robot simply reuses these dialed-in weld parameters. As such, there are three keys to maintaining quality welds:

Consistent upstream process. Presenting the weld joint to the robot as close as possible to the original joint where the “perfect weld” was developed will deliver the highest-quality results.

Repeatable tooling. Similar to the benefit of consistent upstream processes, the workholding fixture must be designed to hold the part in a repeatable manner.

Having a project champion on the customer side. If weld quality begins to slip, the project champion will be able to identify, troubleshoot, and interface with the integrator to get it resolved quickly.

Are job shops more likely to use an automated welding cell for just one welding process, such as gas metal arc welding, or has technology evolved to accommodate quick and easy changeovers to other processes, encouraging shops to consider using multiple processes in one cell? Multi-process robotic installations are becoming more common and economical – in job shops and product-oriented shops. Automatic and Manual changeover options are available, with automatic being the most common. A sample of the tools available include welding guns (with interchangeable neck lengths and styles), plasma cutting torches, oxyfuel cutting torches, grinders (with interchangeable grinding media), preheat torches, laser scanning cameras, and descalers (for silica islands on large solid wire welds or for slag on flux core welds). For high-throughput applications that use different types of wire, it is possible and plausible to change between two separate welding torches with separate wires – for instance, one weld torch setup with stainless steel and the other with a mild carbon steel. Can you share an example of where this type of welding activity is making a difference in a high-mix, low-volume fabricating environment? What were the motivations for the shop to head down this automated path? A recent job shop installation was a pre-engineered cell at a job shop that focuses on a large variety of off-road suspension components. Due to their customer base’s preferences, weld quality/consistency is paramount – something a robot excels at providing. An additional motivator was throughput and demand – the robot system cuts production cycle times in half (parts are being completed 53%-65% faster). The right strategy is to start with the highest-run parts that can start providing a noticeable ROI on the investment, and then add to the robot’s part list. This particular manufacturer plans to run over 100 different parts on their system, and will add additional duplicate systems in the future that will be able to run the same parts as their current cell with no additional programming. Can you describe how robotic welders were programmed 10 years ago? Robots used to be programmed exclusively “online” – where the programmer used the physical system to teach the robot its tool paths. The major downfall of online programming is that the robot cannot be in production while a new part is being programmed. This is solved by offline programming, where the programmer can be developing tool paths for new parts at his/her desk while the robot is downstairs still throwing sparks. The tool paths, weld sequencing, and kinks in the program can be worked out on the computer and then quickly downloaded to the robot. With proper calibration and fixture tolerancing, online “touch-up” of the tool paths is minimal or, in some cases, non-existent – significant reducing or erasing robot downtime for programming while increasing robot part capacity.

Join Us in Orlando

Expo Hours

Venue

About the Expo

North America's Largest Metal Forming, Fabricating, Welding and Finishing Event

From the Blog

FABTECH Q&A with Josh Pawley, Wolf Robotics

About the Expo

2024 Platinum Sponsors

2024 Sponsors

Tuesday Oct 15	9 AM — 5 PM
Wednesday Oct 16	9 AM — 5 PM
Thursday Oct 17	9 AM — 4 PM