Method and Tool to 'AIDE' Arc Welding Operations
The goal of every organization is the continual improvement of its manufacturing processes. There is no exception for the arc welding industry. As all agree, information is at the heart of continuous improvement. Information is required to direct the very action necessary to improve processes or resolve problems. Without information organizations employ random acts with unknown benefits. This wastes valuable time and resources.
Information in the welding industry includes a myriad of data; material data, consumable data, fixture tolerance, part tolerance, weld process information, etc. Weld process information is a substantial and critical component to improving welding applications. A valuable Aide would provide the necessary tools to analyze and optimize the weld process. Moreover, upon such a tool a user could implement a paradigm as illustrated in figure 1.
Intentional collection, storage, and analysis of weld process data has largely been avoided. Welding for too long has been solely considered an art at the expense of good engineering practices. The art of welding many times is at the center of an organizations setup, maintenance, and launch of welding applications. As a result, organizations tend to lack documentation, setup methods, and historical information. Procedures derived from information produce common launch practices, universal “best practice” weld procedures, expected productivity and utilization metrics, and historical analysis of preventive maintenance practices to name a few. In a report released by the U.S. Department of Energy Office of Industrial Technologies titled Welding Technology Roadmap, the report stated “The transition of welding over the coming decade to a rigorous science based on physical data will contribute greatly to the industry’s success in achieving its vision.” Information is and will be increasingly more the cornerstone of improving and advancing weld practices.
A common opinion, which becomes a barrier, is information driven practices are for fortune 500 companies. That is sorely untrue. Large companies are highly dependent on medium and small suppliers to provide a myriad of components. It is obvious the overall assembly is only as good as the weakest link. Best practices must be employed throughout the chain. The benefits are huge; multiplied efficiencies, information sharing, common practices throughout the supply chain. Consider a mobile work force that can transition from industry to industry utilizing universal practices and tools similar to an electrical engineer employing data acquisition techniques to develop pc electronics and automotive electronics. The electrical engineer uses the same oscilloscope in each capacity. The electrical engineer requires the same information. This should be the same in welding. The welding professional will utilize a universal tool to capture the same information whether the application is a car frame or heavy industry tractor. Information will drive setup, maintenance and production activities.
A read of the industry suggests a new openness to embrace information and its effective utilization. In order to marshall in such an opportunity a welding Aide is proposed; a process independent method defining the steps for continuous improvement. Aide is a simple acronym and easy to remember. It can be the building block to a successful practice to continually improve arc weld operations.
Ø Step 1: Audit – gather real-time weld process information and store creating a historical database
Goal: Collect footprint of welding operations
Metrics: Machine utilization, productivity, Weld Signature® capture
Ø Step 2: Identify – analyze welding information creating knowledge base
Goal: Document weld process and measure process stability and repeatability
Metrics: weld process set point, repeatability of a given weld, number of welds per part, reasons for downtime.
Ø Step 3: Detect – establish control limits and alert on loss of process control
Goal: Bound weld process and alert when violated
Metrics: detect missing welds, detect incomplete welds, detect process anomalies
Ø Step 4: Engineer – use weld information coupled with welding knowledge to improve
Goal: Exploit weld process data to increase quality and velocity
Metrics: tighter process operating range, reduced downtime, reduced process faults
Aide is an iterative methodology which is continually implemented. Continuous improvement is by definition “continuous”. As such, the method requires a tool to automate the acquisition, analysis, and collection of weld process information. The Aide methodology is only as good as a tool that makes it possible. What is interesting is the implementation and study of the Aide methodology largely defines the necessary features of the tool. The tool must quickly and effectively source the information necessary to perform each Aide step. The goal is continuous improvement and the tool is an enabler to that end.
The Aide methodology will be described in detail along with the information type and format a tool needs to provide in support of each step.
A-ide: Audit
Literally the question needs to be answered as to what is occurring on the manufacturing floor? It is unfortunately a common unknown. A robust audit is at the heart of the continuous improvement process and appropriately the first step in the Aide methodology.
The governing principles of the audit step are
Ø Simple connection of acquisition tool and sensors (figure 2)
Ø Real-time, high resolution collection of process parameters
Ø Organized weld data and user friendly accessibility
Ø
The audit activity focuses on getting a signature of the welding operations. The audit collects the primary weld process parameters; arc current, arc voltage, wire speed, and gas flow. These parameters are collected in real-time for each weld. Proper weld annotation and organization is paramount. Utilizing a method like Part Tracking®, welds are associated with a unique part. Effective annotation maintains a descriptive record of each weld and allows eventual comparison of like welds.
A-I-de: Identify
The identification step of the Aide methodology is the data analysis step; productivity or weld process problems are identified and cured. The efficiency of the identification process is dependent on the data presentation capability of the data acquisition tool.
The objectives of the identify step are
Ø Evaluate weld process stability
Ø Access productivity
Ø
Ø Cure identified problems
A single view can visually capture that information yielding an easy assessment.
A second example is the stability of the weld process. The audit step has organized the weld data uniquely by annotating the welds by part number and weld number. For example, each weld number 1 across numerous parts can be viewed and compared. Likewise long term trends can be produced providing additional information as to the stability of the welding process.
A third example is machine utilization or productivity. A number of good welds per hour is fine except that the number of good welds per hour is supposed to be 5 times greater. Good data presentation will indicate periods of downtime and catalogue downtime reasons. These reasons can direct preventative maintenance and possibly significantly increase profit margins.
Ai-D-e: Detect
The detect step of the Aide methodology builds naturally upon the previous steps; an audit was taken, process and production problems were identified and cured. The detection step deploys control limits to alert production staff if those problems re-occur.
The objectives of the detect step are
Ø Detect missing welds
Ø Alarm on process problems
Ø Maintain process control
Ø
Ø Maintain productivity
Once again the robustness of the tool used to implement the detect step is all too important. User friendly and robust setting of control limits is required. The control limits govern the detectability of anomalous events. The ability to set, maintain, and manipulate the control limits is paramount to sustained use.
Through the viewing of process capability charts or the overlay of Weld Signatures®, control limits can be developed to differentiate the acceptable from the unacceptable.
Aid-E: Engineer
The objectives of the detect step are
Ø Resolve problems
Ø Predict trends
Ø Schedule preventative maintenance
Process trending and prediction of maintenance is highly valuable in preventing downtime. The engineering step of Aide focuses on converging the welding operation to its sweet spot of operation. Operating in the sweet spot allows minor process variations but with little affect on process quality thus optimizing up time.
The information content is directed to coupling human knowledge with information yielding the highest degree of improvement. Most welding technicians and engineers know their trade. They simply lack the information which provides a record of what happened or is happening.
Summary
The Aide methodology can be the building block to an optimized welding operation. The cornerstone to the method is information. That is why an effective tool which supplies the required information is a necessity. Arc weld monitors, like the ARCAgent™ family of products, are such a solution. ARCAgents™ are robust, flexible, and provide information in a variety of user friendly formats. All charts and graphs provided were generated by ARCAgent™ technology.
An organization deploying the Aide methodology must be committed to the hard work and effort required to continually assess their welding conditions. This requires personnel especially a champion. The benefits will be measurable including improved process knowledge, increased productivity, documentation, and a establishment of best practices. In the end, it is always better to prepare than repair. And who doesn’t want an Aide, especially an unbiased one?
ARClient® is a registered trademark of IMPACT Engineering, Inc.
Weld Signature® is a registered trademark of IMPACT Engineering, Inc.
ARClient®
software provides powerful PC based tools to
make each arc weld monitoring system work efficiently and effectively
for our customers.
ARCharter™ is our SQL database solution for advanced data analysis, reporting and charting. Off-the-shelf and custom reports available.
