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. Here are some ways to kick that performance up a notch.

While the great majority of robotic arc welding systems installed in factories today are doing useful work, over 90 percent of these systems are not running up to their full capability. This capability includes achieving quality, productivity and uptime targets as well as achieving the return on investment required by the initial investment. This article reviews how to bring more of these robots closer to their abilities.

HOW DOES ONE KNOW SUCCESS
WHEN ONE SEES IT?
How does one actually know whether their installed robotic arc welding system is performing successfully? There are two prime methods used to determine this. The first is the traditional return on investment (ROI), while the other utilizes Overall Economic Effectiveness (OEE) as the metric. Both can be used to help justify the initial purchase as well as to determine the systems’ success.

RETURN ON INVESTMENT
There are a multitude of different formulas used to calculate ROI, but the key to any of them is to make sure they account for both total savings and the total cost associated with both the capital and expenses related to the installation and operation of the robot cell. For example, it is easy to calculate the savings when operators are replaced because that is a defined cost, but just as important is to reflect the costs of related training that is needed for the robot technicians to keep the system running.

One needs to account for all costs and savings to come up with a true ROI so a correct decision can be made. Today most companies look for a payback in the 18-24 months range, which is quite doable if one targets the right application, installs it correctly and does this in a timely fashion.

OVERALL EQUIPMENT EFFECTIVENESS
OEE is a very good way to measure the performance of an overall factory as well as individual pieces of equipment. It is calculated with the following formula:

OEE = P x U x Q
where:
P = Productivity (parts/minute)
U = Uptime (percent of time robot ready to run)
Q = Quality (first pass yield or parts per million defects)

World class is > 90 percent and an acceptable value is > 80 percent, but even this is difficult to reach. This is a good metric because it serves to balance the striving for just making more parts with making only good parts. The key to attaining both the desired ROI and OEE is to “own the process”. One knows when this is true if one can answer “yes” to the following conditions:
– Critical variables that drive the process are known
– Critical uncontrolled (noise) variables that affect the process output are known and the process is insensitive to them
– Process capability is known (CpK = ?, >1.00)
– Effective process control procedures and control plans are in place

INSTALLING SUCCESSFUL ROBOTIC WELDING SYSTEMS IS LIKE COOKING YOUR FAVORITE SOUP
What do making a great tasting soup and installing a successful robot welding system have in common? Two things: Both require not forgetting a single key ingredient and also making sure each ingredient’s quantity and quality are correct. Here is the recipe for the alphabet soup of robotic welding:

A is for Adapting
While there are many key ingredients that go into the successful installation of any robotic arc welding system, for the purpose of this article only the top six will be described in detail. With this in mind, where better to start looking at the key ingredients for your robotic arc welding cell than with the letter “A”?

“A” is for adapting the welding process to the actual situation present in the factory. Prior to purchasing the robot system one must establish the true capability of the overall system by assuming the expected variability in parts, fixtures, welding parameters and people. If the joint location and geometry variability is greater than that a “blind robot” can compensate for, one must consider some form of sensing. The most versatile is laser vision, which can quickly find the joint location and adapt for normal variation in fit-up. Figure 1 shows an example of laser vision seam finding where the pre-programmed path of the robot is adjusted for every part that enters the cell to optimize weld wire location to insure maximum travel speed and quality are achieved.