ASM Metals HandBook Vol. 17 - Nondestructive Evaluation and Quality Control

Подождите немного. Документ загружается.

• Charge-coupled device sensors

• Holographic plates (see the article "Optical Holography" in this Volume)

Television cameras with vidicon tubes are useful at higher light levels (about 0.2 lm/m

, or 10

-2

ftc), while orthicons,

isocons, and SEC vidicons are useful at lower light levels. The section "Television Cameras" in the article "Radiographic

Inspection" in this Volume describes these cameras in more detail.

Charge-coupled devices are suitable for many different information-processing applications, including image sensing in

television-camera technology. Charge-coupled devices offer a clear advantage over vacuum-tube image sensors because

of the reliability of their solid-state technology, their operation at low voltage and low power dissipation, extensive

dynamic range, visible and near-infrared response, and geometric reproducibility of image location. Image enhancement

(or visual feedback into robotic systems) typically involve the use of CCDs as the optical sensor or the use of television

signals that are converted into digital form.

Optical sensors are also used in inspection applications that do not involve imaging. The articles "Laser Inspection" and

"Speckle Metrology" in this Volume describe the use of optical sensors when laser light is the probing tool. In some

applications, however, incoherent light sources are very effective in non-imaging inspection applications utilizing optical

sensors.

Example 1: Monitoring Surface Roughness on a Fast-Moving Cable.

A shadow projection configuration that can be used at high extrusion speeds is shown in Fig. 12. A linear-filament lamp

is imaged by two spherical lenses of focal length f

on a large-area single detector. Two cylindrical lenses are used to

project and recollimate a laminar light beam of uniform intensity, nearly 0.5 mm (0.02 in.) wide across the wire situated

near their common focal plane. The portion of the light beam that is not intercepted by the wire is collected on the

detector, which has an alternating current output that corresponds to the defect-related wire diameter fluctuations. The

wire speed is limited only by the detector response time. With a moderate detector bandwidth of 100 kHz, wire extrusion

speeds up to 50 m/s (160 ft/s) can be accepted. Moreover, the uniformity of the nearly collimated projected beam obtained

with such a configuration makes the detected signal relatively independent of the random wire excursions in the plane of

Fig. 12. It should be mentioned that the adoption of either a single He-Ne laser or an array of fiber-pigtailed diode lasers

proved to be inadequate in this case because of speckle noise, high-frequency laser amplitude or mode-to-mode

interference fluctuations, and line nonuniformity.

Fig. 12 Schematic of line projection method for monitoring the surface roughness on fast-moving cables

An industrial prototype of such a sensor was tested on the production line at extruding speeds reaching 30 m/s (100 ft/s).

Figure 13 shows the location of the sensor just after the extruder die. Random noise introduced by vapor turbulence could

be almost completely suppressed by high-pass filtering. Figure 14 shows two examples of signals obtained with a wire of

acceptable and unacceptable surface quality. As shown, a roughness amplitude resolution of a few micrometers can be

obtained with such a device. Subcritical surface roughness levels can thus be monitored for real time control of the

extrusion process.

Fig. 13 Setup used in the in-

plant trials of the line projection method for monitoring the surface roughness of

cables. Courtesy of P. Cielo, National Research Council of Canada

Fig. 14 Examples of signals obtained with the apparatus shown in Fig. 13.

(a) Acceptable surface roughness.

(b) Unacceptable surface roughness

Reference cited in this section

P. Cielo, Optical Techniques for Industrial Inspection, Academic Press, 1988, p 243

Note cited in this section:

Example 1 in this section was adapted with permission from P. Cielo, Optical Techniques for Industrial Inspection,

Academic Press, 1988.

Visual Inspection

Magnifying Systems

In addition to the use of microscopes in the metallographic examination of microstructures (see the article "Replication

Microscopy Techniques for NDE" in this Volume), magnifying systems are also used in visual reference gaging. When

tolerances are too tight to judge by eye alone, optical comparators or toolmakers' microscopes are used to achieve

magnifications ranging from 5 to 500×.

A toolmakers' microscope consists of a microscope mounted on a base that carries an adjustable stage, a stage

transport mechanism, and supplementary lighting. Micrometer barrels are often incorporated into the stage transport

mechanism to permit precisely controlled movements, and digital readouts of stage positioning are becoming increasingly

available. Various objective lenses provide magnifications ranging from 10 to 200×.

Optical comparators (Fig. 15) are magnifying devices that project the silhouette of small parts onto a large projection

screen. The magnified silhouette is then compared against an optical comparator chart, which is a magnified outline

drawing of the workpiece being gaged. Optical comparators are available with magnifications ranging from 5 to 500×.

Fig. 15 Schematic of an optical comparator

Parts with recessed contours can also be successfully gaged on optical comparators. This is done with the use of a

pantograph. One arm of the pantograph is a stylus that traces the recessed contour of the part, and the other arm carries a

follower that is visible in the light path. As the stylus moves, the follower projects a contour on the screen.

Visual Inspection

Reference

1. P. Cielo, Optical Techniques for Industrial Inspection, Academic Press, 1988, p 243

Visual Inspection

Selected References

• Robert C. Anderson, Inspection of Metals: Visual Examination, Vol 1, American Society for Metals, 1983

• Detecting Susceptibility to Intergranular Attack in Austenitic Stainless Steels, ASTM A 262,

Annual Book

of ASTM Standards, American Society for Testing and Materials

• Detecting Susceptibility to Intergranular Attack in Ferritic Stainless Steels, ASTM A 763,

Annual Book of

ASTM Standards, American Society for Testing and Materials

•

Detecting Susceptibility to Intergranular Corrosion in Severely Sensitized Austenitic Stainless Steel, ASTM

A 708, Annual Book of ASTM Standards, American Society for Testing and Materials

• W.R. DeVries and D.A. Dornfield, Inspection and Quality Control in Manufacturing Systems,

American

Society of Mechanical Engineers, 1982

• C.W. Kennedy and D.E. Andrews, Inspection and Gaging, Industrial Press, 1977

• Standard Practice for Evaluating and Specifying Textures and Discontinuities o

f Steel Castings by Visual

Examination, ASTM Standard A 802, American Society for Testing and Materials

• Surface Discontinuities on Bolts, Screws, and Studs, ASTM F 788, Annual Book of ASTM Standards,

American Society for Testing and Materials

• Visual Evaluation of Color Changes of Opaque Materials, ASTM D 1729,

Annual Book of ASTM

Standards, American Society for Testing and Materials

Laser Inspection

Carl Bixby, Zygo Corporation

Introduction

THE FIRST LASER was invented in 1960, and many useful applications of laser light have since been developed for

metrology and industrial inspection systems. Laser-based inspection systems have proved useful because they represent a

fast, accurate means of noncontact gaging, sorting, and classifying parts. Lasers have also made interferometers a more

convenient tool for the accurate measurement of length, displacement, and alignment.

Lasers are used in inspection and measuring systems because laser light provides a bright, undirectional, and collimated

beam of light with a high degree of temporal (frequency) and spatial coherence. These properties can be useful either

singly or together. For example, when lasers are used in interferometry, the brightness, coherence, and collimation of

laser light are all important. However, in the scanning, sorting, and triangulation applications described in this article,

lasers are used because of the brightness, unidirectionality, and collimated qualities of their light; temporal coherence is

not a factor.

The various types of laser-based measurement systems have applications in three main areas:

• Dimensional measurement

• Velocity measurement

• Surface inspection

The use of lasers may be desirable when these applications require high precision, accuracy, or the ability to provide

rapid, noncontact gaging of soft, delicate, hot, or moving parts. Photodetectors are generally needed in all the

applications, and the light variations or interruptions can be directly converted into electronic form.

Laser Inspection

Carl Bixby, Zygo Corporation

Dimensional Measurements

Lasers can be used in several different ways to measure the dimensions and the position of parts. Some of the techniques

include:

• Profile gaging of stationary and moving parts with laser scanning equipment

• Profile gaging of stationary parts by shadow projection on photodiode arrays

• Profile gaging of small gaps, and small-diameter parts from diffraction patterns

•

Gaging of surfaces that cannot be seen in profile (such as concave surfaces, gear teeth, or the inside

diameters of bores) with laser triangulation sensors

• Measuring length, alignments, and displacements with interferometers

• Sorting of parts

• Three-dimensional gaging of surfaces with holograms

• Measuring length from the velocity of moving, continuous parts (see the section

"Velocity

Measurements" in this article)

These techniques provide high degrees of precision and accuracy as well as the capability for rapid, noncontact

measurement.

Scanning Laser Gage. Noncontact sensors are used in a variety of inspection techniques, such as those involving

capacitive gages, eddy-current gages, and air gages. Optical gages, however, have advantages because the distance from

the sensor to the workpiece can be large and because many objects can be measured simultaneously. Moreover, the light

variations are directly converted into electronic signals, with the response time being limited only by the photodetector

and its electronics.

Optical sensors for dimensional gaging employ various techniques, such as shadow projection, diffraction phenomena,

and scanning light beams. If the workpiece is small or does not exhibit large or erratic movements, diffraction phenomena

or shadow projection on a diode array sensor can work well. However, diffraction techniques become impractical if the

object has a dimension of more than a few millimeters or if its movement is large. Shadow projection on a diode array

sensor may also be limited if the size or movement of the part is too large.

The scanning laser beam technique, on the other hand, is suited to a broad range of product sizes and movements. The

concept of using a scanning light beam for noncontact dimensional gaging predates the laser; the highly directional and

collimated nature of laser light, however, greatly improves the precision of this method over techniques that use ordinary

light. The sensing of outside diameters of cylindrical parts is probably the most common application of a laser scanning

gage.

A scanning laser beam gage consists of a transmitter, a receiver, and processor electronics (Fig. 1). A thin band of

scanning laser light is projected from the transmitter to the receiver. When an object is placed in a beam, it casts a time-

dependent shadow. Signals from the light entering the receiver are used by the microprocessor to extract the dimension

represented by the time difference between the shadow edges. The gages can exhibit accuracies as high as ±0.25 m (±10

μin.) for diameters of 10 to 50 mm (0.5 to 2 in.) For larger parts (diameters of 200 to 450 mm, or 8 to 18 in.), accuracies

are less.

Fig. 1 Schematic of a scanning laser gage

There are two general types of scanning laser gages: separable transmitters and receivers designed for in-process

applications, and self-contained bench gages designed for off-line applications. The in-process scanning gage consists of a

multitasking electronic processor that controls a number of scanners. The separable transmitters and receivers can be

configured for different scanning arrangements. Two or more scanners can also be stacked for large parts, or they can be

oriented along different axes for dual-axis inspection. High-speed scanners are also available for the detection of small

defects, such as lumps, in moving-part applications.

The bench gage is compact and can measure a variety of part sizes quickly and easily (Fig. 2). As soon as measurements

are taken, the digital readout displays the gaged dimension and statistical data. It indicates the total number of

measurements taken, the standard deviation, and the maximum, minimum, and mean readings of each batch tested.

Applications. A wide variety of scanning laser gages

are available to fit specific applications. Measurement

capabilities fall within a range of 0.05 to 450 mm

(0.002 to 18 in.), with a repeatability of 0.1 m (5 in.)

for the smaller diameters. Typical applications include

centerless grinding, precision machining, extrusions,

razor blades, turbine blades, computer disks, wire lines,

and plug gages.

Photodiode Array Imaging. Profile imaging

closely duplicates a shadowgraph or contour projector

where the ground-glass screen has been replaced by a

solid-state diode array image sensor. The measurement

system consists of a laser light source, imaging optics, a

photodiode array, and signal-processing electronics

(Fig. 3). The object casts a shadow, or profile image, of

the part on the photodiode array. A scan of the array

determines the edge image location and then the

location of the part edges from which the dimension of

the part can be determined.

Fig. 3

Schematic of profile imaging. The laser beam passing the edge of a cylindrical part is imaged by a lens

onto a pho

todiode array. A scan of the array determines the edge image location and then the location of the

part edges from which the dimension of the part can be determined.

Accuracies of ±0.05 m (±2 in.) have been achieved with photodiode arrays. For large-diameter parts, two arrays are

used--one for each edge. When large or erratic movement of the part is involved, photodiode arrays are not suitable.

However, when part rotation is involved, stroboscopic illumination can freeze the image of the part.

Diffraction Pattern Technique. Diffraction pattern metrology systems can be used on-line and off-line to measure

small gaps and small diameters of thin wire, needles, and fiber optics. They can also be used to inspect such defects as

burrs of hypodermic needles and threads on bolts.

In a typical system, the parallel coherent light in a laser beam is diffracted by a small part, and the resultant pattern is

focused by a lens on a linear diode array. One significant characteristic of the diffraction pattern is that the smaller the

part is, the more accurate the measurement becomes. Diffraction is not suitable for diameters larger than a few

millimeters.

The distance between the alternating light and dark bands in the diffraction pattern bears a precise mathematical

relationship to the wire diameter, the wavelength of the laser beam, and the focal length of the lens. Because the laser

Fig. 2 Self-contained laser bench micrometer

beam wavelength and the lens focal length are known constants of the system, the diameter can be calculated directly

from the diffraction pattern measurement.

Laser triangulation sensors determine the standoff distance between a surface and a microprocessor-based sensor.

Laser triangulation sensors can perform automatic calculations on sheet metal stampings for gap and flushness, hole

diameters, and edge locations in a fraction of the time required in the past with manual or ring gage methods.

The principle of single-spot laser triangulation is illustrated in Fig. 4. In this technique, a finely focused laser spot of light

is directed at the part surface. As the light strikes the surface, a lens in the sensor images this bright spot onto a solid-state,

position-sensitive photodetector. As shown in Fig. 4, the location of the image spot is directly related to the standoff

distance form the sensor to the object surface; a change in the standoff distance results in a lateral shift of the spot along

the sensor array. The standoff distance is calculated by the sensor processor.

Fig. 4

Schematic of laser triangulation method of measurement. As light strikes the surface, a lens images the

point of illumination onto a photosensor. Variations in the surface cause the image

dot to move laterally along

the photosensor.

Laser triangulation sensors provide quick measurement of deviations due to changes in the surface. With two sensors, the

method can be used to measure part thickness or the inside diameters of bores. However, it may not be possible to probe

the entire length of the bore. Laser triangulation sensors can also be used as a replacement for tough-trigger probes on

coordinate measuring machines. In this application, the sensor determines surface features and surface locations by

utilizing an edge-finding device.

The accuracy of laser triangulation sensors varies, depending on such performance requirements as standoff and range.

Typically, as range requirements increase, accuracy tends to decrease; therefore, specialized multiple-sensor units are

designed to perform within various specific application tolerances.

Interferometers provide precise and accurate measurement of relative and absolute length by utilizing the wave

properties of light. They are employed in precision metal finishing, microlithography, and the precision alignment of

parts.

The basic operational principles of an interferometer are illustrated in Fig. 5. Monochromatic light is directed at a half-

silvered mirror acting as a beam splitter that transmits half the beam to a movable mirror and reflects the other half 90° to

a fixed mirror. The reflections from the movable and fixed mirrors are recombined at the beam splitter, where wave

interference occurs according to the different path lengths of the two beams.

Fig. 5 Schematic of a basic interferometer

When one of the mirrors is displaced very slowly in a direction parallel to the incident beam without changing its angular

alignment, the observer will see the intensity of the recombined beams increasing and decreasing as the light waves from

the two paths undergo constructive and destructive interference. The cycle of intensity change from one dark fringe to

another represents a half wavelength displacement of movable mirror travel, because the path of light corresponds to two

times the displacement of the movable mirror. If the wavelength of the light is known, the displacement of the movable

mirror can be determined by counting fringes.

Variations on the basic concepts described above produce interferometers in different forms suited to diverse applications.

The two-frequency laser interferometer, for example, provides accurate measurement of displacements (see the following

section in this article). Multiple-beam interferometers, such as the Fizeau interferometer, also have useful applications.

The Fizeau interferometer has its greatest application in microtopography and is used with a microscope to provide high

resolution in three dimensions (see the section "Interference Microscopes" in this article).

Displacement and Alignment Measurements. Laser interferometers provide a high-accuracy length standard

(better than 0.5 ppm) when measuring linear positioning, straightness in two planes, pitch, and yaw. The most accurate

systems consists of a two-frequency laser head, beam directing and splitting optics, measurement optics, receivers,

wavelength compensators, and electronics (Fig. 6).