ADVANCING ASSEMBLY THROUGH TECHNOLOGY.

Although this is an industry characterized by pressure, this is a look at two aspects of what performance under pressure is all about.

Last April, General Motors had a snafu with the new sport utility vehicles (SUVs) that it had just launched, the Chevy TrailBlazer, GMC Envoy, and aids Bravada. The problem was sufficiently troublesome that the vehicle manufacturer instructed the -6,000 owners of the SUVs to leave them parked: they'd be picked up and towed to dealers for repair. The problem was with the lower control arm. The concern was with breakage of same.

Larry Stockline, president, Promess Inc. (Brighton, MI), says that shortly after the announcement, he received a call from GM about the Promess Electro-Mechanical Assembly Press (EMAP). The functionality of the system is largely predicated on two things: (1) the ability to dial in parameters including speed, force, distance, and time; (2) the use of sensors for closed-loop control of the process (this assures that what is dialed in is achieved). Promess doesn't provide machines; the EMAP package consists of the press head, motors, amplifiers, multi-axis controller, motion control, cabling, enclosure, PC, and Windows NT. The packages range in capacity from 1,000 lb. To 55,000 lb. and are used in a variety of auto applications, from fuel injector systems to seats.

Stockline recalls that during the telephone call he was given the opportunity to respond to 10 questions of both commercial and technical sorts; all had to be answered in the affirmative if Promess was to get this particular business with GM. Stockline says that he actually said "no" to one: He was asked if Promess could supply a system with a 35,000-lb. capacity. It's not that they can't; it's that Stockline told his interlocutor that if the application was to push bushings into control arms, then 35,000 lb. are way too much; the top end for this application, based on experience working with other vehicle manufacturers, is 11,000 lb. Usually, Stockline points out, it's less than 10,000 lb. "We were asked if we would bet our business on it," recalls Stockline. "I said, 'yes.'"

A second question that needed some additional input was whether a special machine builder could meet the requirements that GM had: Two machines in 10 days, installed and running in the supplier's plant. The machines would require four EMAP systems; they would normally take 12 weeks, but Stockline was confident that Promess could meet the timeframe. Stockline contacted Miller Tool & Die (Jackson, MI) to find out if they could build and integrate the equipment. Ordinarily, Stockline says, a machine of the type required would be built in about 20 weeks. Miller Tool & Die signed on for what was to become a highly accelerated, 24-hour-a-day job.

The machines met the deadlines. Six hours after the first machine was taken off the truck it was producing quality control arms. When the second machine arrived two days later, it was making parts within three hours.

Stockline acknowledges: "To fail could have brought these companies down."

He refers to all of the engineers and machine builders who were involved in this undertaking as "heroes in the shadows."

The SUVs were out and rolling to popular and critical acclaim.

The whole event seems to raise another issue, however, one of whether advanced automotive products, such as the three SUVs in question, can be produced with anything less than high-tech equipment, such as electromechanical, sensor-controlled press systems rather than tools like conventional hydraulic assembly presses. Stockline says that one of the objections that he hears with regard to his system is that process control of the steps prior to things like insertion should mean that things fit as specified (e.g., "I'll monitor the tolerances, so I'll know it is right.").

But as Mr. Murphy tells us, things rarely go as anticipated; parts can be dropped, machines can have transient errors. Parts won't be as expected. Stockline suggests that statistical sampling, the basis of many monitoring and control approaches, is insufficient. (He refers to it as a "1935 methodology.")

"Our system doesn't say if a part is 'good' or 'bad.' But it will tell you if it's 'different'"-as in requiring 35,000 lb. to be pressed into place when a fraction of this is sufficient-if the part is what it should be. Essentially, the system is programmed by simply inserting parameters for each of the process steps. The selected parameters are based on what would be a "good" assembly operation for the parts in question. You can then bracket those parameters plus and minus given amounts so as to provide a window for the process.

While acknowledging that his EMAP system has a higher price tag than conventional hydraulic units, he points out that it is best applied in situations where there are families of parts: as it is flexible, it can be adjusted to handle different insertion parameters. "With a hydraulic press," he says, "you get what you get."

It is also quieter than a hydraulic press and is greener in that it uses no hydraulic fluid or air.

Yes, it may be initially more expensive. But then there is always the price that's paid when things go wrong...