by Kennedy Smith

At a metrology lab’s disposal are myriad measuring tools, one of the most common being the coordinate measuring machine. Others, like hand-held gages and roundness measuring machines, ensure that your measurements are accurate and your parts meet specifications.

But what happens when you have to measure something so big that it doesn’t fit on the machine and using a hand-held gage would be futile? In these instances, many companies turn to noncontact inspection.

Large-part noncontact inspection is widely used in the automotive and aerospace industries, where it’s common to find extremely large parts in need of measurement, such as an aircraft wing or the body of an automobile. These industries have found that the most sensible solution is to use one of many noncontact devices capable of measuring large parts without wasting time or resources.

There are several devices to choose from when it comes to large-part noncontact measurement. As far as accuracy is concerned, many of these devices are comparable, but there are other factors to consider when choosing the right system, namely cost per measurement, speed and ease of use.

Photogrammetry

Photogrammetry is based on digital camera technology. The instrument takes a digital image of the target and was originally used with film cameras for mapmaking. “During the past 15 years or so, digital cameras have made photogrammetry more adaptable to large-part measurement,” says Gary Card, marketing manager for Brown & Sharpe.

Photogrammetry works by taking an image of the target and comparing it to something that would give it relative size; a yardstick is a rudimentary example. The user puts the yardstick somewhere in the image and captures the relative position between the target and that marker.

Modern photogrammetry systems use two or more digital cameras to capture an array of LEDs embedded on a probe. A computer uses the location of the LEDs to determine the position of a point in space.

Theodolites

Theodolites use optical sensors and encoders to map data points and record the position of a part. They typically contain an electronic system capable of triangulating targets in relation to the part, but this type of system doesn’t scan the part. Theodolites are a good choice when accuracy is more of a concern than speed.

The principle behind the theodolite is similar to that of the laser tracker in that it uses motorized devices to develop the horizontal and vertical readings, the angle readings, and the distance readings from the beam that it emits.

Walter Pettigrew of Carl Zeiss IMT Corp. explains how this technology works. “Let’s say you have three optical devices, and you point all three of them at a spot on the part,” he suggests. “You can triangulate and locate that feature. Then, if you focus it on another feature, you can relate the distance between them--much like surveyors on a construction site. You first have to establish three locations and get the distances between them. With basic geometry and trigonometry, you can calculate the part’s position.”

Laser trackers and scanners

Laser trackers utilize a laser beam to collect data from retroreflective targets located on the part. The beam senses movement of the mirrored target and follows its location. Data points are typically recorded at near 1,000 points per second. In order to calculate the information, a laser tracker uses a laser interferometer, two precision encoders and software. An advantage to using a laser tracker is that it provides consistent accuracy and repeatability.

For instance, FARO’s laser tracker has a range of 70 m. It sends out a laser beam to a spherically mounted retroreflector. The return beam from the SMR is followed by the tracker. The tracker reads the distance and horizontal and vertical angles to the target 1,000 times per second. Software reports the data in X, Y and Z coordinates. The laser tracker can be used with a single hand-held target or can “point and shoot” at an unlimited number of fixed targets.

The laser tracker and laser scanner are both highly precise, have large working ranges and utilize lasers. There are some distinct differences, however. The laser scanner is fully noncontact, whereas the tracker measures the location of a retroreflector that comes into contact with the part. Because the tracker relies on retroreflectors, there is some manual work involved in moving them to different locations on the part. The scanner, on the other hand, is automated.

MetricVision’s coherent laser radar scanner is an example. “Our scanner is very much like the laser tracker as a reader of azimuth, elevation and range, but unlike a tracker, we can automatically move it or program it to measure over a complete surface,” explains David Dozor, COO of MetricVision. “With a laser tracker, you need someone carrying a retroreflector. Our system is automated; it doesn’t require a cooperative target.”

MetricVision’s model V200 is distributed in certain markets as Leica’s LR200. It has an extremely large range as well. “We can take measurements down to a tenth of a millimeter in an area of up to 48 meters,” says Nicholas Bloch, vice president of global marketing and communication.

Articulated-arm and optical CMMs

Articulated-arm CMMs, such as the FaroArm, fall under the category of “large-part noncontact measurement systems,” not because they’re capable of scanning a large part but because they’re flexible and portable enough to work around a large part.

In the realm of optical CMMs, Zeiss offers the Eagle Eye. On a normal three-axis CMM, there’s another three-axis manipulator. On the end of that manipulator is a laser scanning device. In other words, it throws a laser line on the surface of the part and measures everything that the line comes into contact with.

“Part of the reason we have a sophisticated manipulator on the end of the machine is that it’s relatively important to keep the laser normal to the surface of the component,” explains Pettigrew. “Lasers on CMMs have been around for at least 20 years, if not longer. What’s different here is that the sampling rates and resolution are much higher.” Because there are times when contact measurement is integral, the Eagle Eye system is designed to switch back and forth from contact to noncontact.

Contact vs. noncontact

Pettigrew suggests that the best way to accurately measure something large is often to combine contact and noncontact measurement. “It’s sometimes difficult to read all the elements of a feature using noncontact measurement,” he asserts.

The main factor in choosing whether to use contact or noncontact is accuracy. “Certainly size is a consideration because some things simply exceed the size that a CMM can handle,” says Card.

Dave Genest, marketing director at Brown and Sharpe, notes another factor. “When a piece is large with many surfaces, it drives you toward noncontact inspection because it can take millions of points,” he adds. “You don’t want to spend the rest of your life measuring each part. You would tend to go the noncontact route if you’re dealing with multiple surfaces as opposed to geometric elements.”

Accuracy and cost

Experts agree that, on the whole, noncontact systems are not as accurate as contact systems. However, Brown & Sharpe’s research suggests that most large-part noncontact measurement systems have similar accuracy. “We looked at the laser tracker, theodolites, photogrammetry and articulated arms, and they all seem to have the same general range of accuracy,” reports Card. “For example, for a part that’s 30 feet long, you would get somewhere in the area of 0.002 to 0.005 in. accuracy regardless of what instrument you choose.”

This begs the question: How do you know which system is right for you? Card says it comes down to what you’re accustomed to and how much money you’re willing to spend. “You have to look at how often you’ll be using the instrument--what the payback period would be,” says Card. “If the piece of equipment is $200,000, you’d have to be able to justify that. A similar piece of equipment that could do roughly the same thing for about the same accuracy might go for $100,000.”

It’s important to keep in mind the cost per measurement when investing in a noncontact system. “When you talk about the cost of a system, especially in a resource-constrained environment, you have to think about the cost per measurement,” says Mark Shudt, vice president of marketing and sales at MetricVision. “In other words, figure out how often you’ll use the instrument before investing too heavily in one that might end up sitting in the corner of the metrology lab.”

On the horizon

Some experts suggest that indoor global positioning systems will become more prevalent as time passes. “It works similarly to our K-series machine,” says Card. “It uses infrared light and receivers, so it has a transmitter and receiver. There would be a number of transmitters set up around either the factory or the room, and receivers would be placed on either the part or a hand-held probe. A connection between the transmitter and receiver would pinpoint the location of the part.”

As for photogrammetry, emerging digital camera technology might make this method more user-friendly. “All you have to do is visit your local electronics store and look at all the digital cameras,” says Pettigrew. “The cheaper they get, the more accurate they get; and the bigger they get, the better the technology is going to be.”

“The core R&D that we’re pushing with as much of our recourses as available is algorithm development,” says David Dozor, COO of MetricVision. “Holes, gap and flush, and character lines are all things that our customers would really like to measure with the instrument. We’d like to control the scanner and laser radar to collect that data without a whole lot of aggravation.”

About the author

Kennedy Smith is Quality Digest’s associate editor. Letters to the editor regarding this article can be sent to letters@qualitydigest.com. Much of the content in this piece was contributed by Gary Card, marketing manager for Brown & Sharpe.