Moving metrology to the process, rather than keeping it just in the quality lab, will increase overall quality and make automation of manufacturing itself more efficient,” said Robert Wasilesky of Carl Zeiss Industrial Metrology LLC (Brighton, MI). “Metrology Optimized Manufacturing [MOM] is definitely a trend we are seeing.” He was careful to note that metrology does not enable manufacturing—it works well now—but optimizes it to work even better.
Optical metrology in its various forms is often the best method to enhance automation, according to him, because of its speed compared to other methods. Other advantages of optical methods include their ability to measure without marking or deforming delicate surfaces, such as autobody sheet metal, composite parts, or delicate surgical implants. Another example is to use metrology devices to precisely move robots, increasing their location accuracy from a few millimeters to 100 μm or better.
To expand MOM, Wasilesky said, means expanding customer education. Manufacturing engineers today may not be aware of how much more optimized their shop could be.
“I think there is a potential to increase manufacturing efficiency, especially in automotive, by comparative leaps and bounds through Metrology Optimized Manufacturing,” he said. The concept must be reaching manufacturers, and Wasilesky noted that the automation side of the metrology market is growing at close to 20% per year, whereas the rest of the quality-related market is growing at about 6%. He said this is due in large part to greater customer acceptance of optical metrology methods over contact probes, a sentiment shared by others in a number of conversations. The technology has gotten better, there are more choices of optical methods to choose from, and processing software hosted on increasingly faster computing platforms has improved dramatically. “While the principles are the same, it has just become much more capable,” he said.
A good example of this convergence of faster optics, better algorithms and powerful computing is the Zeiss AIMax cloud optical 3D sensor. Classified as a structured light system, it relies on only one projector and one sensor. This makes for a more compact sensing head, but perhaps more significant is that the system is programmed to measure features quickly, typically in less than 0.5 sec—think of holes, slots, rivets, or T-pins—from a point cloud rather than providing the point cloud as the final result. This makes for a system that is ideal for in-line inspection, especially for automotive sheet metal and BIW (Body in White) components. Wasilesky stated the system, including sensor and robotic movement, could provide 300 μm or better of measuring accuracy to Six Sigma.
Zeiss also offers solutions for near-line measurements as well as faster, optical replacements for CMMs on the shop floor for off-line measurements that are also near the point of production. The AIBox is such a complete system aimed at getting CMM levels of accuracy without a trip to a quality room. The AIBox is capable of measuring less than 40 μm of spherical error to the VDI 2634 standard using both a digital fringe projection system with a standard photogrammetry device. It is for large parts up to 2000 mm in size and 1000 kg in weight standard.
One of the more useful optical metrology devices for embedding in an automated system are video systems on precision movement tables, such as the Nikon iNEXIV series. Since the measuring system includes a camera, it is easily adapted to vision tasks, opening new possibilities, such as reading bar codes or QR codes that are common with parts today, according to Nate Frost, product manager for Nikon Metrology (Brighton, MI). “We can also read text data, either a part number on a part or on a printed paper that might accompany the part.” This takes automation to the next level, because reading a bar code and identifying the part means the device can call up the correct measuring program and automatically record the results in a report for that individual part.
This could be especially useful as the Smart Factory and Industrial Internet of Things continue to develop. “These are not traditional automation tasks,” Frost said. “Automatic part loading on a measurement system should be something that any metrology supplier can accommodate. [Bar-code reading is] instead related to making decisions,” he said. It removes the human factor while enhancing quality.
An efficient means of creating a parts program is just as important. Following a common trend in the industry, creating programs from reading a CAD model offline is now available in Nikon’s CMM-Manager 3.6. This is the latest version of the support software for Nikon’s vision systems. “The previous software used only a joystick to teach-in a program,” Frost said. “This is fairly new to video measurement programs.” Why? Because, in contrast to off-line programming a contact probe or even a laser line sensor, a video program must consider lighting, edge effects, glints from reflective surfaces and other effects unique to video. “Simulations can be run and inspection results can be verified in real time. In fact, the more complete the CAD model, the more efficient the inspection process becomes,” said Frost. “With vision, autofocus laser, rotary indexer, and tactile input, we can even measure features and geometry you can’t see with a traditional vision system.”
Why does he think there is a push for automation? “The common denominator is that companies are leaning out their work-staff and operations,” Frost said. “They are expecting their employees to be more flexible. They want them to be able to run a CNC mill and a metrology system. We are seeing this across the board, but particularly in job shops where employees need to be a jack of all trades.” Automating metrology will make this that much easier.
“The precision and speed of optical measurement techniques is what makes it so good for automation,” agreed Andreas Blind, vice president of sales, marketing and services for Jenoptik (Rochester Hills, MI). Jenoptik provides a wide range of metrology sensors, primarily but not exclusively for in-process control. These range from air gauges and contact gauging as well as advanced, high-speed optical sensors. “Each has its place. Optical techniques tend not to work well in environments that are dirty or oily, however optical is far more flexible and is especially useful when the parts cannot be touched, or when you need speed,” he said.
Jenoptik’s Opticline series of devices is a good example of a technique—a shadowgraph—packaged into a high-speed automated cabinet for measuring shafts. Shadowgraph systems are ideal for measuring rotating parts, even ones that are not axisymmetric, such as camshafts and crankshafts. The company notes there are over 2500 Opticline systems installed worldwide, in a variety of sizes and capabilities for measuring large or small pieces. For example, there are a series of Opticline stations with statistical process control (SPC) interfaces for rapid monitoring using the SPC system of choice for that factory. Accuracies are reported as MPe and are as good as (1.5 +L/200) μm for measuring diameters with a measuring speed as fast as 80 mm/s.
Another important function of optical in-process control is surface flaw detection, according to Blind. “Our Optisense technology uses a combination of unique sensors, illumination, automation, and software that helps us detect even the smallest defects in surfaces and on parts,” he said. Detecting surface flaws in the bores of engine cylinders is especially important with today’s focus on fuel economy and emissions. The advent of plasma sprayed bores replacing thicker and heavier steel sleeves in aluminum blocks makes surface flaw detection especially important. “Our IPS line bore sensors are available in different bore sizes, from 4 to 150 mm, depending on the kind of sensor one would use,” Blind said. “They can detect flaws down to 30 μm, and are usually automated, though they are available in semi-automated versions as well,” he said. He noted that automation is especially important to the automotive sector. “Eighty percent of our equipment is currently going to automotive,” he said.
Another unpleasant reality for metrology providers is that quality checks can often be viewed as an expense, simply lost time in production. That is according to Stefan Scherer, president and CEO for Alicona (Graz, Austria). He said metrology can add value. “If we are able to show that they can adapt their manufacturing process using metrology to proactively improve manufacturing and make it more flexible, we think that will make a big difference.” He especially believes the results from metrology are vital in the growing Smart Manufacturing or Industry 4.0 movement. “We need to give machines a smart eye,” he stated.
Alicona is an innovative company, founded on a relatively new metrology principle it calls Focus Variation. This technique provides high-resolution surface data using limited depth of focus, similar to but more advanced than confocal imaging or optical sectioning, since it also provides color data. It measures both surface form and roughness, and is especially useful in production settings, according to the company. “Our technology is useful for measuring small features with high precision or relatively small features, such as micro or precision manufacturing, whenever tolerances are in the range of 20 μm or less,” Scherer said. To work optimally, Focus Variation requires an Ra surface roughness of 9 nanometers or more. Data speed is up to 1.7 million points per second.
After first achieving success in measuring edge preparations for cutting tools, the company expanded its offerings, automating its sensors.
Automation is crucial to Scherer’s vision of a smart eye, and the company turned to collaborative robots, or cobots, in doing so. Why cobots? “We see a shift towards higher precision and at the same time less volume. In previous times, a manufacturer might make a million parts. Today, they might make 50,000 of one kind of part and then switch to producing 75,000 of another,” he said. He believes cobots enable workers in these flexible manufacturing environments, making them useful and adding value. “We have a cobot mounted on a mobile platform to bring it more easily to a machining center,” he said. The user checks a few key parameters to set up the machine for making the next set of part numbers, and then moves on to the next task.
The newest offerings from Alicona are two new cobot products for defined applications, one for long round tools and the other for turbine disks. The tool solution expands Alicona’s expertise in edge prep measurement with nine-axis measurements on a wheeled portable platform.
With the Alicona disk solution, the robot arm with attached measuring sensor is manipulated by an operator to the desired surface position. Two handles with integrated joystick are mounted on the sensor. By means of an app, a smartphone displays the live view for either manual or automatic precise positioning and measurement.
Another common trend is developing a standard automation solution around existing metrology devices. Metrology companies are now offering standard solutions that integrate robots, enclosures, and fixturing. Usually aimed at a class of problems, they can be tailored to meet a manufacturer’s throughput and tolerance requirements.
In 2014, Hexagon Manufacturing Intelligence (North Kingston, RI), introduced the 360 Smart Inline Measurement System, or 360 SIMS, using its WLS (white light scanner) structured light devices. According to Hexagon, an in-line system built around a WLS is ideal for automotive body-in-white production because the WLS is relatively insensitive to nearby vibration—it collects data in milliseconds. Making it even faster, the 360 SIMS collects only select features, such as gap-and-flush or holes, according to Amir Grinboim, Hexagon product manager.
Hexagon has now added to that inline offering with the 360 Flexible Measurement Cell, or 360 FMC, first introduced in September 2016. “An off-the-shelf solution is powerful and cost effective when it comes to installation, integration, and incorporating safety protocols,” Grinboim said. “It reduces the overall cost of engineering and design to do that once.” The safety protocols on the 360 FMC meets the ANSI RIA 15.06-2012 standard. Part sizes can range up to 2.2 m in length and weigh up to 1500 kg.
The basic 360 FMC configuration is an enclosure equipped with a single FANUC M-710iC/50 robot mounting a Hexagon WLS400A sensor, driven by the CoreView software package. It measures parts fixtured to one, two, or three separate turntables, depending on the configuration chosen. “A modular, flexible solution will enable our customers to use the cell not just for metrology, but also to address throughput,” explained Grinboim. “We wanted it to adapt to changing conditions through the entire manufacturing lifecycle.” Customers can add or reduce the number of turntables without a major redesign or re-integration effort, according to Grinboim. “Adding—or removing—turntables is plug-and-play.”
The 360 FMC does not need reference targets on the part, relying on the robot motion with compensation programs. “It measures to an accuracy of 100–150 μm, based on feature type, which is suitable for typical automotive sheetmetal applications,” said Grinboim.
Connect With Us
