Metalcutting Automation: Some Considerations

For the past several years there has been something of an on-going competition occurring in the metalcutting automation arena as high-volume systems are giving way, in some cases, to more flexible systems, and then the high-volume lines regain ground (or, more specifically, plant floor space). Because Lamb Technicon Machining Systems (Warren, MI), a UNOVA company, provides both types of systems, we decided to ask its Mark Tomlinson, vice president of Technology Integration, to provide some suggestions as to what people ought to take into account when assessing the right automation for their application.

Process Commonality

The first thing that Tomlinson recommends be kept in mind when Although this is a high-volume line for producing V8 cylinder heads—built for DaimlerChrysler's New Mack Engine Plant—its architecture is clearly more modular than high-volume transfer lines of the past. (Photo: Lamb Technicon). considering flexible and high-volume systems is that when it

comes to a given part, the processes will essentially be the same for both
types of systems. That is, if a hole needs to be drilled, then that's the case regardless of the volume of parts to be produced. In conjunction with this is the fact that there is likely to be commonality with regard to the tooling—though, as we shall see, there are some differences.

Hold It
Here's a flexible system for machining auto cylinder blocks. It consists of a series of CNC machiing modules and head changer machines. Note the blocks are not palletized. The robots are rail-mounted and used for load/unload operations. (Photo: Lamb Technicon).
Moving to the fixturing and transporting of the workpiece, we arrive at a point where there are clear system differences. For example, in a high-volume setup, where, generally, a single operation is performed at each station, the forces involved are often considerably different than those in a flexible system, where a multitude of cutting operations will be performed at a station (or, more accurately, within a CNC machine). Consider face milling. In a high-volume system, it is probably going to be done in one pass. In a flexible system, there are probably going to be multiple passes. Consequently, there is a difference in the clamping pressures for each type of system, which means there are fixturing differences. And because a flexible system is one in which there are multiple operations performed on a part in a setup, there are accessibility issues that have to be taken into account (e.g., there may be milling on one face and drilling on another, so there must be an open area where the drill can reach its intended target).

Move It

With regard to part transfer, Tomlinson points out that in the case of high-volume systems, there is a new architecture emerging: "The historic drive at the end of the machine is giving way to modular drives and transfer bars." This is providing the means by which there can be new functions added to an existing line by dropping in machine modules. (This modular approach also facilitates building and testing the line in the first place: since the individual pieces have their full mechanical functionality, there is the ability to assure that everything is as it should be far earlier than if it is necessary to wait for the entire line to be built and the drive at the end of the line engaged.)

In flexible systems, Tomlinson says that the trend is away from automatic guided vehicles (AGVs) and pallets fitted with fixtures. Tomlinson explains that the reason is the need to minimize the "streams of variation" that such devices can contribute to. That is, in the case of pallets, there are benefits realized from clamping once (e.g., in terms of consistent part location and management of cutting forces), but there can be difficulties related to maintaining and monitoring the traceability and repeatability of each palletized fixture, of which there are many in a flexible system running at production volumes. "Managing pallets is more difficult than managing parts on skids," Tomlinson notes. So a trend in flexible systems is to maintain the fixturing in the CNC machine tool and loading, unloading, and transferring parts in the free state.

Optimized Vs. Versatile

Another difference between the flexible and high-volume systems related to the physical characteristics of the tooling and fixturing is that in a high-volume system, a particular station is engineered to do one thing while the flexible system is fundamentally engineered to provide the opportunity to do plenty of things. So, for a high-volume system, the fixture, spindle and tooling are optimized at each station for a particular part. But consider just the spindle in a flexible system: it isn't sized for one specific part or operation; it must be capable of handling a variety. Consequently, there are likely to be tradeoffs, which can lead to processing and/or tooling issues.

Think about, Tomlinson suggests, valve seat and guide machining. In a high-volume system, there is a fixed tool that has the valve seat angles in place and a drawbar is used to put out a reamer to machine the guide. With a single-spindle machine this task can be exceedingly more complicated from the tooling point of view (e.g., a tool that has the seat configuration designed in and a reamer that pops out as a result of centrifugal force). The high-volume approach is simpler.

Controlled

In the area of controls for the equipment, there are big differences. Here the advantage is, in Tomlinson's estimation, with the flexible system because there are "standard" CNC controls. The consideration is wholly on the part programming logic. For the high-volume systems, on the other hand, the control architecture isn't as standardized—improved, yes, but standardized, no. Consequently, there is a need to work on both the machine logic and the part processing.

People Needs

There are workforce considerations, too, from the standpoints of required skills and number of people required to staff the systems. In the case of a high-volume system, the required personnel are generally limited to a loader, unloader and maintenance crew (that is either assigned to the line or central to plant operations). A flexible system needs more people. For one thing, the part transport is unlikely to be fully automated, so a number of part handling people may be needed to man the line (robotic handling is certainly a viable option). There must be a maintenance crew (who can be deployed as in the case of a high-volume system). But a big difference that Tomlinson points to is the need to have decision makers who will be able to decide what to do in the event that a cell within the system goes down. A high-volume system is either running or it is down. A flexible system based on work cells can continue to run, but the part flow must be determined. So there is a need for planners, schedulers and supervisors.

Downtime Data

Tomlinson explains that the flexible system introduces some uncertainty—at this point in time, anyway—when it comes to a part flow strategy. There is an abundance of data existing that describes the downtime for high-volume systems; after all, they've been used for decades. This data can be used to determine the number of parts that should be kept in a bank to handle the average downtime. But in flexible systems, this data isn't nearly as deep. What's more, there are many more problems that can arise.

For example, on flexible systems, one of the approaches to minimize downtime related to toolchange time is to have redundant tools in the toolchange mechanism, in addition to the variety of types of tools that are in place to perform the multiple operations. All of which means plenty of toolchanges per hour. But how many times can a toolchanger change before it fails? (And take into account the fact that in a given cell there are multiple machines, each of which has a toolchanger, each of which is changing tools on an on-going basis, so if one toolchanger goes, how long before the others follow?) There is a move toward using more high-speed machining spindles in flexible sys-tems. There are some concerns with the reliability of these spindles in production applications; downtime may be greater. And although CNC controls may be standard, there is still some question about control reliability. With time, these reliability issues will be worked out, just as they have been for the elements of high-volume machines; the high-volume machines simply have the advantage of having been around a lot longer.

Economic Advantage?

This time factor has also resulted in the development of a depreciation philosophy that tends to favor the high-volume approach. Tomlinson posits a situation where there need to be one million parts produced. A high-volume machine can handle it. A flexible system can be engineered to do it—but it may cost three times what the high-volume system does. When it comes to a five-year depreciation, the high-volume system provides what he describes as an "attractive return on capital." But the flexible system won't make the cut—under that metric.

Tomlinson points out that the flexible system can be readily retooled so that it may be able to do three or more programs. He cites a flexible DaimlerChrysler manifold system that has been retooled some six times since Lamb installed it in the mid-1980s. So when it comes to lifecycle costs, the flexible system can clearly be more economic. But unless more than just the sticker price is taken into account, flexible systems may be economically unviable.

http://www.autofieldguide.com/articles/010004.html