Our Sites

Troubleshooting, traceability for tube, pipe producers

One system monitors welding, forming processes

Figure 1
A statistical distribution with little variation indicates the process is in control. A relatively tight distribution of weld bead height and width is evidence of a proactive strategy by the mill’s operator (blue data points, circled). A distribution with no clear pattern is associated with much less process control (red data points, dispersed).

Before the advent (and affordability) of the consumer GPS receiver, drivers used the car’s dashboard instrumentation and that was all. While useful for monitoring the condition of the engine and vehicle speed, the various gauges and meters don’t provide information about direction of travel, route, route options, and estimated time to destination. The GPS changed the way we drive. The latest ones provide up-to-date information on road construction and travel delays, turning the GPS into a feedback mechanism to make our driving more efficient. They help us avoid problems, take corrective actions, and make better decisions.

Like the GPS, weld and process monitoring accomplishes the same thing. The dials and gauges on the welding equipment are helpful, but a weld monitor provides much more information about weld integrity and allows the mill operator to make better, more informed decisions.

Many variables go into every weld. To a novice welder, it might seem that the right combination of gas, volts, amps, and speed produces a good weld, but the reality is far from this simple. Research in identifying weld defects and their causes has generated entire encyclopedias of information and thousands of journal articles.

Weld and process monitoring is a combination of measuring and monitoring weld conditions and the additional factors that make up a quality weld. Together, they provide a 360-degree view of the welding conditions. Monitoring both welding and the entire process—how the material edges match before the weld, weld bead size, weld bead shape, and so on—provides a proactive, rather than reactive, strategy.

Together they allow quality assurance personnel, supervisors, and managers to look for patterns, differences, and anomalies. This type of work is instrumental in tuning the process to make it more efficient. Many tube and pipe manufacturers work in a hypercompetitive field, so getting the setup right the first time to maximize the yield of salable product is critical to running a mill profitably.

Knowledge Is Power

After any sort of manufacturing anomaly occurs, such as a weld defect, a common refrain on the shop floor is, “I’ve seen this 1,000 times before.” If indeed the problem comes up every two weeks or so, and the person who said it is nearing retirement after a 40-year career, it’s not an exaggeration.

Dig into that statement a little deeper, and it’s troubling. First, why hasn’t the root cause been eliminated? Second, assuming it’s really a reference to a score of recurring problems with different causes and this mill operator knows a remedy for each one, what’s going to happen to all the information in his memory when he retires? It’s not likely that he’s going to transfer all of it to his replacement.

On the other hand, if you monitored the processes and recorded 1,000 anomalies, you’d have a good start in building a troubleshooting database. In statistics, 1,000 is a large enough sample to create a reliable distribution curve. In other words, you don’t need more samples. Recording 10,000 samples or 100,000 samples won’t create a more refined or comprehensive distribution curve. A sample size of 1,000 is sufficient. Better still, the 1,000 records can be shared among all of the operators and quality control personnel. Unlike the retiree’s memory, they aren’t prone to fading from memory as time goes on, and they can be analyzed more easily. They can be compared to other records and studied in the search for underlying causes.

A monitoring strategy becomes a company historian. Over time 1,000 occurrences converge to reveal statistical probabilities: why it happened, underlying conditions, and corrective actions. Welding without a process monitoring system is the same as driving without a GPS unit.

Taking Control

If you were to examine 1,000 tubes or pipes, a tedious task at best, it would take quite a while to determine how much control the operator had over the mill. On the other hand, looking at the distribution of just a dozen measurements of weld bead height and width as the tube or pipe moves down the mill tells an informed story (see Figure 1).

Figure 2
An asymmetrical weld bead indicates an alignment problem. In this case, the right-hand side shows some squeeze-out, but the left-hand size shows none. The rolls were applying too much pressure, and the material was riding too high on one side. Adjustments to the fin passes resolved this problem.

This monitoring system uses a laser mounted to the mill for accurate, real-time measurements. Use of the system is reflected in the blue data points in Figure 1. It shows that the mill operators have been trained in using the system’s feedback and samples with known good welds to fine-tune the mill’s operating parameters. The operators understand that as long as the weld bead dimensions are clustered in a small range—for this product, a height between 0.3 and 0.5 mm and width between 0.8 and 1.3 mm—the probability of a quality weld is very high. The monitoring system is both a historian and a GPS, helping them understand what has happened in the past so they can plot a course for the future.

The other data points, in red, are from a mill without any sort of monitoring system. The data was collected over a four-hour period and shows that the process is not in control. During this four-hour period the mill was stopped three times.

While statistics is the study of a collection of data, at least 1,000 data points, the welding monitor examines just one tube or pipe at a time as it comes down the mill. It sends an alert to the operator when it detects a change in the weld bead shape (see Figure 2).

Tying It Together

The ultimate monitor ties together both the forming processes and the welding parameters. By monitoring the line voltage and frequency, the weld current, and shielding gas parameters, this system calculates and monitors power (P = V x A) and thereby the weld heat (see Figure 3). This, in turn, affects the weld bead profile (see Figure 4).

The mill staff and quality control department now have a record of a cold weld condition. Nobody would have expected the mill operator to notice this, and indeed it did escape his attention, but the weld and process monitoring system did catch it.

A weld bead dimension monitor might have noticed the problem, and a weld monitor might have caught it, but a combination system provides a better chance of noticing the problem. It’s also more likely to notify the operator sooner rather than later, which helps to minimize scrap.

Traceability

With each passing day, metal fabricators and OEMs ask their vendors to be more compliant, retain more information, and make that information available when necessary. A robust and reliable monitoring system is one way a tube or pipe producer can be proactive in the battle against defects and have production reports available within minutes of a request.

Price will always be a factor, but depending on the item, it’s far from the dominant factor. Ultimately the OEM wants peace of mind, whether it’s making automobiles for consumers, heavy equipment for construction contractors, or spaceships for NASA. Peace of mind comes from having confidence in every component that goes into the final product, and that confidence stems from traceability.

About the Author

Pierre Huot

1016 Sutton Drive C5

Burlington, ON Canada Canada

905-331-6660