In Control

Chances are, when you think about robotic welding, you think about, well, robots. After all, the robot is the most-visible aspect of a welding unit, be it a stand-alone, cell or complete system. But when Tim Nacey, group manager, Robotics & Welding Equipment, Panasonic Factory Automation (Elgin, IL), talks about the subject, it is less about the robot arm—although Panasonic does produce them—and more about the control intelligence of the overall package. That, in his estimation, is what can make the biggest difference in automotive arc welding applications.

For example, Nacey talks about the controller. On the one hand, the controller ought to be incredibly fast and capable. On the other hand, it needs to be comparatively simple (after all, people need to be able to use the equipment on a daily basis, people who probably don’t have advanced degrees in electrical engineering). So they set about to develop a control that employs a 64-bit RISC processor, which he claims is 30 times faster from an internal calculation basis than what’s competitively available. Yet at the same time, they fitted the G2 control with a teach pendant that employs the Windows CE operating system. He explains, “Today, almost anyone coming out of high school is familiar with Windows.” He’s found that people take to the interface with little in the way of instruction, which is important, he suggests, when there’s employee turnover at the plant floor level.

Although you might consider a teach pendant to be one of the lesser parts of achieving high quality welds in, say, aluminum, it is emblematic of the approach that Panasonic is taking to the welding process. By concentrating on such things as easy user interfaces as well as on high processor speed, the robotic application can be greatly improved.

In addition to which, they’ve engineered the teach pendant such that it can be detached from the robot and plugged into a PC. This permits off-line programming to be performed. It also provides benefit from the point of view of teaching operators about the functionality of the equipment without the necessity of having a robot available for training.

One of the more significant developments that Nacey says they’ve come up with of late is an aluminum welding system that’s capable of welding 2-mm material at up to 200 ipm, which is more than twice as fast, he claims, than competitive MIG systems. This performance is a result of several things working in concert. For example, they’ve developed a servo torch with a planetary feeder. Apparently, in most welding applications, the wire feeder is at some distance away from the actual contact tip—perhaps three feet. It seems that what happens during a welding operation, as a robot maneuvers the torch around, there is a twisting effect on the wire, which causes the length of the wire to change, to move in and out in relation to the workpiece material. The result of that is arc “hunting.” That isn’t particularly conducive to attaining high weld quality, so weld speeds are kept on the slow side. With the system that Panasonic has developed, the wire is kept at a more consistent length because it is being pushed much closer to the weld gun: just six inches away. This eliminates the arc hunting and helps contribute to increasing weld speed. But they’ve also determined that there are other factors to take into account. For example, they’ve developed software for the power supply that monitors such things as the pulse width and pulse frequency and makes necessary adjustments to the waveforms within microseconds. Also, because of the thermal conductivity of aluminum, they’ve worked on the flow of the shielding gas so that quality welds are achieved even at these high speeds. Nacey says that particular interest in this system has been shown by manufacturers of aluminum engine cradles.

“In the past,” he notes, “the robot and the power supply were essentially two independent machines.” Each did what it was supposed to do, with the robot often being nothing more than a motion device that also turned things on and off. Now he describes a robot as being an element in a more “holistic” system, a system that is highly dependent on high-speed digital computation.

Rethinking Specials

ICA Cinetic Automation (Farmington Hills, MI), a supplier of special machines for assembly, is finding its business to be different nowadays, as it is providing vehicle manufacturers with systems, primarily for powertrain assembly. Mark Pehrson, vice president, Marketing & Business Development, suggests that whereas the company once generally worked in a mode where they’d build to order, now there is a greater acknowledgement that there is a competency that they bring to the party vis-à-vis system design and build, so they are given more responsibility in determining how to get the job done in the most effective manner.

Of course, these determinations can be made only in the context of what the prevailing requirements are in the industry. And among those requirements for systems—be they for, say, engine or transmission assembly—are greater standardization, reduced floor space, and functional flexibility. Toward that end, the company is offering what it is calling its “AgiLogix” approach to system design. Essentially, Pehrson explains, the systems that are created through this approach not only come in at a comparatively low capital investment (achieved, in part, through the utilization of more standard elements: after all, special, or custom, products always cost more), tend to have a considerably smaller footprint than traditional systems (achieved not only through much smaller components, including motors, drives and controls, but also as a result of reducing the amount of space between stations, predicated, in part, by greater reliability of the equipment), but which are also adjustable (through reprogramming, adjustment, or quick tool change). They’ve worked this through all of the elements of a system, including gauging, pressing, assembling, and washing. Among the assembly applications that Pehrson says AgiLogix is appropriate to are piston installation, head-to-block assembly, and transmission-to-case assembly.

Even though the amount of “specialization” that exists in a system built today is far less than it was just five years ago, Pehrson suggests that through the implementation of things like vision systems, better controls, and adaptive tooling, five to 10 years from now there will be the potential of having, in effect, a “general” assembly line for engines or transmissions, something that would be sufficiently flexible to handle a mix of models, yet would be, in effect, a “standard” product.

One of the areas in automotive assembly that hasn't been automated to any great extent is trim and final, particularly interior assembly work. So providing useful and ergonomically oriented hand tools is essential. This, for example, is a pneumatic spray applicator from Henkel Loctite that's designed for interior trim work.

Sticking In Place

Want to make sure that interior trim or HVAC components are reliably fastened in place? Perhaps better than might be achieved with such things as duct tape, staples, or bead of adhesive? Then you might want to check out two new hand-held pneumatic spray applicators from Henkel Loctite (Rocky Hill, CT): they dispense 1.75-in. diameter spots of hot melt adhesives. The Hysol 175-SPRAY dispenses EVA hot melts (it gets up to 180°C), while the 175-SPRAY-HT goes up to 195°C for dispensing polyamide. (The heater housings for the units are enclosed for safety.) The airflow control permits the spray pattern to be adjusted for specific application requirements.

Seconds to Dollars

“Assembly line work has special dynamics that are not found in many human activities. By doing a thorough job of balancing the work among stations, we have in fact made every operator a bottleneck. Let us assume we have a balanced final assembly line for cars, using 500 operators to make 500 cars per shift and we add one second to the job of one operator. Then all other 499 operators must wait one second for each of the 500 cars they are assembling in a shift. The added labor time is therefore:

1 sec x 500 operators x 500 cars = 4,167 min >10 operator shifts

Adding one second to the job of one operator cots more than ten operator-shifts per day.”

So writes Michel Baudin in Lean Assembly: The Nuts and Bolts of Making Assembly Operations Flow (Productivity Press; New York; 320 pp.; $50.00), a thoughtful book that Baudin describes as being “about what should be done rather than how to do it.” The book is prescriptive, yet it is not a step-by-step approach to improving assembly operations. Through text, diagrams, drawings, photographs, and graphs, Baudin lays out various aspects of lean production (visual management; one-piece flow) but all in the context of assembly operations. While not all of these examples are automotive-specific, regardless of the industry, the approach transcends particulars.

Of course, managers and executives in the auto industry probably spend more than their fair share of time obsessing on hours-per-vehicle ratings. Consequently, Lean Assembly ought to be a book that each of them studies with some high level of seriousness because Baudin examines assembly from a variety of perspectives, from line layout to final inspection, to the required data collection methodologies in between. Much of this is common sense. Yet even when something is sensible, it is sometimes hard to implement, because, to trot out still another cliché, old habits die hard. Yet, when you take that one-second example into account, and calculate the cost of more than 10 additional operators per day, understanding and implementing some of Baudin’s prescriptions will be time well spent. But remember: You have to do it.

If you were to see this as a video rather than a still picture, you'd see that not only is the vehicle being moved along on the transport, but that the sections of the floor where the operators are standing are in sync with the vehicle line speed. Siemens Dematic calls this people conveyance "Easy Step."

On the Line

One of the fundamental advantages of the assembly line can also be a fundamental disadvantage. That is, assembly lines move. Which is a good thing. But from the point of view of assemblers, it isn’t always a beneficial thing. That’s because as the line moves along they may have to (1) walk backwards and (2) contort their bodies in order to get their particular component(s) installed. A consequence of this balancing act can be that after not too long a time the assemblers lose concentration on the task at hand and there can be quality problems.

Russ Fedun, with the Rapistan Material Handling Automation Div. of Siemens Dematic (Grand Rapids, MI—although Fedun is out of the Mississauga, Ontario, office), says that a few years back, he and his colleagues began to work on what has become the “Easy Step Packed Work Platforms.” “It’s a parallel conveyor that’s synchronized to the work conveyor speed.” Instead of walking along to keep up with the vehicle being assembled, the operator simply stands on the belt, which is covered in a variety of thick, oil-resistant, non-slip materials. This permits the operator to concentrate on getting the job done.

The conveyor comes in various widths (from 18 to 102 in. are standard) and lengths (from 10 to 75 ft. are standard). The height ranges from 4.5 to 7.5 inches. When asked about the lengths, Fedun says that the average is from 28 to 36 ft. This is predicated on a few factors, including (1) the length of time that one needs to complete an assembly task on a line (e.g., from 42 to 48 seconds) and (2) the fact that the Easy Step is modular: it can be moved with a forklift. This latter characteristic facilitates line configurations.

What happens if the assembler gets to the end of the line and hasn’t completed the task? Fedun answers that there’s a safety switch at the end of the belt that can be activated with a shoe heel, which shuts the lines down.