The problem is compounded by the techniques engineers have traditionally resorted to when compensating for variation during commercial production. On the one hand, manufacturers have employed various post-assembly inspection regimes that are often of dubious value at best. On the other, they have put processes in place that have increased costs while providing only minimal value.
The classic example of the former is sampling a percentage of the parts assembled during a particular production run and then using that as the basis for judging the quality of the entire batch. If, say, 10 percent of a company’s riveted widgets look good, the operator in charge of quality assurance signs off on the whole bunch.
Another approach is to set your pressing parameters to accommodate a worst-case scenario and then use that same setting for the entire production run. Let’s say a manufacturer finds that it needs to apply between 500 pounds and 700 pounds of force in a particular press-fit application, because of variation in the parts’ inner and outer diameters. The solution, then, is to apply 700 pounds of force every time, whether it’s needed or not.
In either case, it is often a human operator making the decision as to whether an assembly is a good one, often judging it simply by it’s “looks”-hardly a recipe for Six Sigma stardom.
Finally, there’s the strategy of narrowing the tolerance of the parts that go into a pressed assembly. The problem with this approach is that precision doesn’t come cheap. Eventually you reach a point where machining out the variation isn’t worth the effort.
Brains vs. Brawn
To solve these kinds of problems, assemblers are increasingly turning to “smart” closed-loop assembly presses that don’t just mash a rivet or press fit to a predetermined setting, but respond dynamically to real-time force and position feedback.These pressing systems consist of a servo-driven press with an integrated load cell and position encoder; signal amplifiers; a system controller; dedicated system software; and when needed, one or more external sensors to monitor additional process variables.
Working in concert, these components do far more than just set pass-fail limitations for a particular pressing operation. For example, the systems can be configured to make multiple attempts at creating an assembly, depending on the readings that are generated as the ram begins applying its force. They also allow engineers to fine tune their pressing operations in response to the resistance generated as the parts first come in contact and then slide or deform into their final configuration.
Finally, they create a system that can respond dynamically to parts variation by pressing to a functional level-the point at which the assembly performs the way it’s supposed to-as opposed to simply pressing to a particular force.
For example, in the press-fit application above, the system would use only enough force to drive the two parts together-and no more. If the inner part was on the big side and the inner diameter of the outer part a little bit tight, the system could employ a full 700 pounds of force to get the job done. However, if the inner and outer parts were both in the middle of the range, the system would use only as much force as was required, thereby avoiding damaging or deforming the part by pressing too hard.
In either case, the system’s position encoder lets it know when the ram has travelled the correct distance, as opposed to simply coming up against a hard stop.
It’s the same thing with rivets. By pressing to a position, engineers no longer have to worry about applying too little force to a slightly harder rivet or over-pressing and possibly damaging a rivet that’s a little softer.
A Part's John Hancock
Central to this kind of feedback-dependent pressing is something called a force vs. position “signature”-a curve that is created by plotting ram position on the X axis and ram force on the Y axis of a line graph.This signature not only allows engineers to study the precise manner in which the various parts interact during assembly, but to accurately determine whether an assembly has been correctly processed.
For example, the signature indicates the position at which the assembly was first contacted. It also shows the amount of force applied as the piece was inserted, molded or otherwise changed; the maximum applied force; the rate of change for the force; and the final position of the ram.
By studying the force-distance curve for a particular assembly, engineers can quickly identify whether the process was correctly executed, and if not, what went wrong. If a rivet’s material is too hard, or a press fit too tight, for example, the curve will be higher than it should be. If a rivet is too soft or a press fit too loose, the curve will be too low. In the event a rivet is missing or a part is misaligned, this fact will also be readily evident.
The signature for an ideal prototype can also be used to establish a baseline by which all parts will be judged during production. For example, using a “box” approach, engineers can create a number of maximum and minimum force and distance regions at key points the curve. As long as the force-distance curve for a particular assembly falls within these prescribed areas during a pressing operation, the assembly is considered to be a good one.
Another approach is to “teach in” the entire force-distance curve of a good, nominal part and use that as the guide, inputting enough variation above and below this standard to allow for some part variation while still ensuring that you are producing good parts.
In either case, monitoring each assembly’s force-distance curve provides an ironclad means of certifying an assembly for traceability and control of warranty costs.
Getting It Right
Perhaps most impressive is a smart press’ ability to respond to the data it is generating and then adjust it’s behavior accordingly. The result is a system that can accommodate parts variation, creating assemblies that function correctly in the field even if the parts aren’t “perfect” to begin with.“In applications where products are assembled by pressing one or more components together, the heart of this adaptive control approach is a measure-and-press, press-and-measure philosophy,” says Larry Stockline, president of Promess Inc. (Brighton, MI), a pioneer of this kind of manufacturing.
“Measuring critical part parameters or functions during the assembly process and then using this information in a closed-loop system to control the assembly process is the key to manufacturing better products at lower cost,” Stockline says. “Instead of producing the component parts to increasingly tighter tolerances and then assembling to fixed dimensions by pressing to dead stops or bottoming out in the part, the assembly process compensates for…variations.”
To illustrate this point, Stockline cites the example of a customer that uses a Promess Electro-Mechanical Assembly Press (EMAP) to install a spring and valve seat in a hydraulic pressure-relief valve.
Prior to installing the EMAP unit, the company struggled to create valves with an opening pressure of 1,000 ±200 psi. In fact, when a customer demanded that it cut the variation in half, the company was at a loss as to how it would do so without having to scrap nearly a fifth of its product. One solution was to upgrade the quality of the relief springs so they would meet a tighter specification. However, doing so would have been prohibitively expensive.
To solve the problem, the company equipped an EMAP press with an additional pressure sensor that allows it to press the valve seat to a specific spring-relief pressure as fluid flows through the valve at a fixed rate. By using this approach, the company accommodates part variation without compromising functionality. Instead of pressing to a single distance or force, it now presses to a performance standard-which is, of course, the whole point of a product like a pressure relief valve in the first place.
“Fluid pressure is measured and fed back to the controller in real time,” Stockline says, describing the system. “When the specified pressure is reached, the press ram is stopped and the assembly operation is finished. The resulting valves consistently provide a ±2-psi variance in operation. The closed-loop assembly process produces virtually no scrap, and allowed the manufacturer to eliminate a gauging station and produce valves at a lower overall cost.”
In another example of this kind of functionality, Bruno Maczynski, engineering manager at TOX-Pressotechnik LLC (Warrenville, IL) says one of his customers, a solar panel manufacturer, is using a TOX EPMK spindle to provide the pressing force in one of its laminating operations.
According to Maczynski, the laminate is heated and held together for anywhere from five to 30 minutes. During this time the press needs to adjust its position to accommodate for changes in part size.
“”The simple X-Y positioning system previously used would try and hold a constant position. When the part heated up and expanded in size, the holding force would ramp up significantly and fall out of specifications. When the part cooled down and shrank, the holding force would decay, and the part would also fall out of spec,” he says. “By using a [closed-loop system], the amount of scrap was decreased by almost 25 percent.”
In yet another application of this kind, Rick Wiltsie, vice president of sales and manufacturing for Kistler Stamotech North America (Shelby Township, MI), says his company recently implemented a system that both presses a bearing onto a shaft and then pre-loads that same bearing through the application of 5,000 pounds or force.
During the press-fit portion of this multi-stage operation, the system employs both force and distance control to ensure the bearing is positioned correctly. During the pre-load phase, however, it employs force control alone, while simply monitoring the distance. This positional data is then used to confirm the assembly is within spec and indicate the correct shim width needed to complete the operation.
According to Wiltsie, in performing these two tasks, the system is doing work that previously required two separate processing stations and an array of different measuring and gauging devices.
David Zabrosky, North American marketing manager for Schmidt Technology Corp. (Cranberry Township, PA), notes that, in addition to part variation, the precision and adaptability of a quality closed-loop servo press can be used to accommodate variation within the assembly system itself.
Even the sturdiest press, for example, will exhibit some kind of give-same thing with tooling. By monitoring the entire assembly cycle, a closed-loop system can accommodate any kind of flexing, compression or relaxing, whether it’s in an assembly’s constituent parts or the system itself, to ensure a quality product, Zabrosky says.
Barry Wright, applications engineer at Cincinnati Test Systems Inc. (Village of Cleves, OH), adds that these capabilities come in handy with a wide range of assembly types. According to Wright, low-cost products manufactured in high volumes can benefit from the precision and flexibility of a closed-loop servo press, the same as expensive, mission-critical components.
Cost Savings
If there is a downside to this kind of technology, it’s cost. Conventional press systems have been around for generations, and lower-tech models have the low prices to prove it. However, as is so often the case in manufacturing today, spending a little more upfront can mean big savings in the long run.Zabrosky says, for example, that by verifying the quality of each assembly, closed-loop servo presses ensure that bad assemblies are neither further processed nor integrated into a larger product, where they can incur serious rework or warranty costs. The presses also reduce scrap rates thanks to their ability to compensate for parts that might not otherwise go together correctly.
Upstream, a closed-loop servo press can allow an assembler to spec less expensive parts, because the process no longer relies solely on close tolerances for its success. Servo presses are also cheaper to operate than their hydraulic counterparts, because they require minimal maintenance and consume less power.
“What many assemblers have to do right now is machine parts to a high degree of precision. As a result the parts become more expensive,” says Zabrosky. “The servo can let you machine less expensive parts and still get a precision assembly…. A less engineered part coming in and a higher-end product coming out-a servo can do that.”
Keith Lowery, product engineer for the servo-press line at FEC Inc. (Macomb, MI), adds that by closely monitoring each assembly as it is being performed, closed-loop systems can prevent damage to both parts and fixtures.
“If anything gets caught in a fixture, a traditional hydraulic press will rip it apart. With a servo press we can stop the operation,” Lowery says. “The system can also tell the customer whether he has the right parts or not, identifying it instantly without ruining the part.”
For assemblers looking to save a little money up front, there is also the option of a system like those found in the Accu-Flex line manufactured by Beckwood Corp. (St. Louis). Featuring a hybrid system in which a motor moves the press ram via a hydraulic actuator, as opposed a mechanical one in a strict servo configuration, this kind of press offers precision at slightly less cost, especially at the higher end of the force range.
“Because of the compressibility of the hydraulic fluid, position and force control are not as tight as a ball screw…. But we can provide ±0.001-inch accuracy and one percent force control, which is well within required tolerances for many applications,” says Beckwood application and sales engineer Darrell Harrelson.
In terms of the types of industries and applications that can benefit, the automotive industry has been using this kind of closed-loop pressing technology for some time. However, Lowery says any industry can benefit wherever it is creating precision assemblies with a press, including consumer electronics and alternative energy products.
Zabrosky agrees, noting that while there are still plenty of applications for which a less precise pressing technology is more than adequate, there’s nothing like a closed-loop system for those assemblies requiring both flexibility and precision. “It’s a matter of control. With [these systems] you are controlling the process even while you are actually doing it,” he says.
Lowery notes that his company is currently developing a new, smaller 100-pound press to address the needs of customers performing lighter assembly operations, like those found in the electronics industry.
Along these same lines, Promess’ Stockline says his company has developed presses down to 45 pounds for use in light assembly operations, as well as presses with a 225,000-pound capacity for use in heavy-duty applications like fuel cell manufacturing.