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

Подождите немного. Документ загружается.

(A · d/2) = n

(Eq 4)

Writing in scalar form and solving for the component of d parallel to A, the following expression for in-plane

displacements is obtained:

(Eq 5)

That is, the component of d lateral to the line of sight is proportional to the number of fringes crossing point P and is

inversely proportional to the sine of the half-angle between the two views. The angle between A and d is called .

Fig. 9 Schematic of the observation of a double-

exposure holographic interferogram showing the vectors and

angles relating to in-plane displacement of a point on the surface of the object from position P to position P

See text for explanation of symbols.

It is possible to measure in-plane displacements from 5 up to 150 m (200 in. to 0.006 in.) using double-exposure

holographic interferometry (Ref 14). An error of ±3 m (±118 in.) can be expected on the large-displacement end of the

range because of the limiting size of the interferogram (the limiting magnitude of A).

For an object surface that undergoes a general displacement involving both in-plane and out-of-plane motion, the fringe

pattern can be viewed from several (at least three) directions to obtain the projection of the surface displacement of each

image point along at least three different sensitivity vectors, k

1,2,3

. The three sensitivity vectors can then be considered to

be a set of nonorthogonal coordinate axes that can be rotated and projected onto a conventional coordinate system to

obtain general object point displacements. Although the description of this process is a bit confusing, the mathematical

and computer algorithms used to perform these manipulations are quite straightforward.

References cited in this section

12.

J.Ch. Vienot, J. Bulabois, and J. Pasteur, in Applications of Holography, Pr

oceedings of the International

Symposium on Holography, Laboratory of General Physics and Optics, 1970

13.

J.E. Sollid and J.D. Corbin, Velocity Measurements Made Holographically of Diffusely Reflecting Objects,

in Proceedings of the Society of Photo-Optical Instrumentation Engineers, Vol 29, 1972, p a125

14.

J.E. Sollid, A Comparison of Out-of-Plane Deformation and In-

Plane Translation Measurements Made

With Holographic Interferometry, in Proceedings of the Society of Photo-

Optical Instrumentation

Engineers, Vol 25, 1971, p 171

Optical Holography

Revised by James W. Wagner, The Johns Hopkins University

Time-Average Interferograms

The characteristic function for time-average holographic interferograms differs from that of the double-exposure case. If

it is known that the object is undergoing strictly sinusoidal motion during the time of exposure of the holographic

interferograms, then the characteristic function is J

(K · d), where J

is the zero-order Bessel function of the first kind

and d is the vector displacement. This function behaves similarly to a cosine function with regard to its zero values;

however, it is not strictly periodic with zeroes existing at regular intervals. The first and second zeroes occur when the

argument is 2.4048 and 5.5201. After that, the zero values can be approximated by those given by the asymptotic limit for

large argument (large x); that is:

(x) cos [x - ( /4)]

(Eq 6)

For example, for the third fringe, the error is 15 parts in about 8600. None of the measurements to determine the values of

or (Fig. 8) is likely to be this accurate, so use of the zero values for the asymptotic limit is generally well justified.

Writing Eq 6 in scalar form and solving for the component of d parallel to K yields the following: For the first fringe:

|d| cos = 2.4048 /(4 sin )

(Eq 7)

For the second fringe:

|d| cos = 5.5201 /(4 sin )

(Eq 8)

For succeeding fringes, with n 3:

|d| cos = (n - ) [ /(4 sin )]

(Eq 9)

To a good approximation, the first fringe represents a displacement of about 3 /(16 sin ), with succeeding fringes

representing steps of /(4 sin ).

In addition to differences in the location of zeroes, the characteristic (Bessel) function for this case dramatically decreases

in amplitude with increasing fringe order. Because of the decreasing brightness of the fringes and the limited dynamic

range of the reconstruction film, it is difficult to record much more than seven fringes in photographically produced

reconstructions. Even when a superproportional reducer is used on the reconstruction negative to increase the visibility of

the higher-order fringes, it is difficult to work to much more than 30 fringes. In addition, it is difficult to work with a

slope on the object in excess of about 0.6% with either double-exposure or time-average holograms, because of the high

frequency of the fringes produced.

Optical Holography

Revised by James W. Wagner, The Johns Hopkins University

High-Resolution Interpretation Methods

As mentioned previously, consideration has been given only to the relationship between holographic interference fringes

and object surface motion. In fact, the appearance and apparent location of fringes in a reconstructed image depend not

only on displacement but also on object surface reflectivity and fringe brightness or contrast. Therefore, the intensity at

each point in a reconstructed holographic interferogram is a function of these three variables and not simply surface

displacement. Using high-resolution methods such as phase stepping (Ref 15, 16) and heterodyning (Ref 17), one can

compute directly all three variables at each point in the image to a degree of accuracy up to 1000 times better than can be

achieved by simple fringe counting. In this way, displacements as small as 0.25 nm (2.5 Å) can be detected in principle.

To perform either of these interpretation methods, independent control of the two interfering images must be available

during reconstruction. This is a natural consequence of real-time holographic interferometry because the reference and

object beams can be altered independently. In double-exposure methods, however, a dual-reference arrangement as

described previously must be used to permit independent control.

Phase-Stepping Methods. For phase stepping, several video images are recorded of the fringe pattern with a small

phase difference introduced between the reconstructed images prior to recording each video image. The phase shift can be

performed in several ways, but perhaps the most common method is to use a mirror mounted on an electromechanical

translation device such as a piezoelectric element. If the phrase shift imposed prior to each video recording is known, then

only three images need be recorded. Because the intensity at each point on the image is known to be a function of the

three variables described above, intensity information from the three images can be used to solve a series of three

equations in three unknowns. In addition to providing automated interpretation of fringe patterns, phase stepping affords

an increase in displacement sensitivity by as much as 100-fold ( of a fringe) relative to fringe-counting methods. In

practice, most investigators report a sensitivity boost of about 30.

Heterodyning Method. Still higher holographic sensitivity can be obtained with heterodyne holographic

interferometry. As with phase stepping, independent control of the interfering images must be provided either by real-

time analysis or dual-reference methods. Instead of introducing a phase shift between several recorded images, a fixed

frequency shift is introduced in one reconstructing beam relative to the other. Typically, acousto-optic phase shifters are

used to produce a net frequency shift of the order of 100 kHz. As a result, fringes once visible in the reconstructed

interferogram are now blurred because of their apparent translation across the image field at a 100 kHz rate. A single-

point optical detector placed in the image plane can detect this fringe motion and will produce a sinusoidal output signal

as fringes pass by the detection spot. By comparing the phase of this sinusoidal signal to that obtained from some other

point on the image, the difference in displacement or contour can be electronically measured. An entire displacement map

can be obtained by scanning the optical detector over the entire image. Because scanning is required, the speed of

heterodyne holographic interferometry is relatively slow. Sensitivities approaching of a fringe have been obtained,

however.

References cited in this section

15.

P. Hariharan, Quasi-Heterodyne Hologram Interferometry, Opt. Eng., Vol 24 (No. 4), 1985, p 632-638

16.

W. Juptner et al., Automatic Evaluation of Holographic Interferograms by Reference Beam Shifting,

Proc.

SPIE, Vol 398, p 22-29

17.

R. Dandliker and R. Thalmann, Heterodyne and Quasi-Heterodyne Holographic Interferometry, Opt. Eng.,

Vol 24 (No. 5), 1985, p 824-831

Optical Holography

Revised by James W. Wagner, The Johns Hopkins University

Holographic Components

The basic components of a holocamera are the:

• Light source (laser)

• Exposure controls

• Beam splitter

• Beam expanders (spatial filters)

• Mirrors

• Photographic plate or film holder

• Lenses

• Mounts for the equipment

• Tables to support the holographic system

Components and complete holographic systems are commercially available (Ref 18, 19).

Laser Sources

The characteristics of six types of lasers commonly used for holography are listed in Table 1. Helium-neon, argon, and

ruby lasers are the most common. Helium-cadmium and krypton lasers, although not used as frequently, can fulfill special

requirements for CW applications. Frequency-doubled Nd:YAG lasers are finding increasing popularity for pulsed

holographic applications.

Table 1 Wavelengths and temporal coherence lengths of the six types of laser beams in common use for

holography

Wavelengths

Temporal coherence length

Minimum

Typical

Type of laser beam

Electromagnetic

spectrum

in. mm

in.

Helium-neon 633 6330 Orange-red 152

6 457

422 4220 Deep blue 75 3 305

Helium-cadmium

325 3250 Ultraviolet 75 3 305

514 5140 Green 25 1 914

(a)

488 4880 Blue 25 1 914

(a)

Argon

(b)

25 1 914

(a)

647 6470 Red 25 1 914

Krypton

(c)

25 1 914

Ruby 694 6940 Deep red 0.8 0.030

914

1064

10640

Infrared

(d)

Nd:YAG

532

(e)

5320

(e)

Green

(e)

(d)

(a)

Typical temporal coherence length achieved when laser incorporates an etalon (selective filtering device) to make it suitable for holography.

(b)

Six other visible lines.

(c)

Nine other visible lines.

(d)

Varies with cavity design.

(e)

Frequency-doubling crystals

Helium-neon lasers are the most popular laser source when low powers are sufficient. Excitation of the gas is

achieved through glow discharge. These lasers have excellent stability and service life with relatively low cost. Another

type of laser is usually considered only when a helium-neon laser will not perform as required. A 20-mW helium-neon

laser in a stable system can conveniently record holograms of objects 0.9 m (3 ft) in diameter. (Within the limits of

coherence length and exposure time, as discussed previously, even larger objects could be recorded.) Such a laser

consumes 125 W of 110-V electrical power and operates in excess of 5000 h without maintenance. A 5-mW laser records

objects 460 mm (18 in.) in diameter and operates for more than 10,000 h without maintenance.

Helium-cadmium lasers are closely related to helium-neon lasers, with the following differences:

• Tube life is poor by comparison (approximately 1000 to 2000 h)

• The principal visible wavelength--422 nm (4220 Å) is 30% shorter (Table 1

), which provides increased

sensitivity and allows the use of recording mediums sensitive to blue light

• They have an output in the ultraviolet (325 nm, or 3250 Å), which is half the wavelength of helium-

neon lasers and produces doubly sensitive displacement measurements

• There is more danger to the eyes at the shorter wavelengths produced by helium-cadmium lasers

Argon and krypton ion lasers can be the least expensive holographic sources on the basis of light output per dollar.

Laser outputs of 1 W with 9 m (30 feet) of coherence length are available. Low-power argon lasers, however, are more

expensive than helium-neon lasers. The use of an argon laser should be considered over a helium-neon laser in the

following situations:

• When stability or dynamic conditions necessitate short exposures requiring high light power

• When the recording of large objects requires higher power to record good holograms

• When the recording medium requires high-power blue or green light

• When the holographic

system needs the higher sensitivity provided by the shorter wavelengths of the

argon laser

Argon lasers in the 1- to 4-W output range, equipped with an etalon (a selective filter required to achieve long

coherence length), are excellent holographic light sources; however, a heium-neon or a helium-cadmium laser may be

preferred for the following reasons:

• Argon lasers consume thousands of watts of electrical power and require water cooling

• Gas excitation is by electric arc, which generates high electrica

l and thermal loads on components and

makes reliability and stability lower than with a helium-neon laser

•

The output power is well above that which causes damage to the eyes, especially at the shorter

wavelengths produced by argon lasers. Most available data indicate that helium-

neon lasers are

incapable of causing the damage that could be caused by an argon laser. Therefore, safety requirements

for argon lasers must be more stringent

Krypton lasers are essentially the same as argon lasers except that the tubes are filled with krypton gas instead of argon

gas. The output wavelengths are longer (Table 1) and the power is lower than for an argon laser. A 2-W argon laser

produces 0.8 W in its most powerful line (514 nm, or 5140 Å), while the same laser device filled with krypton produces

0.5 W at 647 nm (6470 Å) and 1.3 W total. Argon and krypton gases can be combined in the tube to give custom outputs

over a wide range of wavelengths.

Ruby lasers use rods of ruby instead of a gas-filled glass tube as a lasing medium. Excitation of the medium is by

optical pumping using xenon flash lamps adjacent to the ruby rod. Ruby requires such high energy inputs to lase that the

waste heat cannot be removed fast enough to sustain continuous output. For this reason, ruby lasers are always operated in

a pulsed mode, and the output is usually measured in joules of energy per pulse (1 J of energy released per second is 1 W

of power). Peak output powers of ruby lasers exceed 10 MW, requiring extensive safety precautions.

The development of pulsed ruby lasers for holography has progressed with the need to record holograms of moving (or

highly unstable) objects. Ruby lasers have been extensively used to record the shock waves of aerospace models in wind

tunnels, for example. Most holographic interferometry done with a ruby laser uses a double-pulse technique. The tasks

that require a ruby laser are those that cannot be done with a helium-neon or an argon laser. Ruby lasers can routinely

generate 1 J, 30-ns pulses of holographic-quality and relatively long coherence length light, which is sufficient for

illuminating objects up to 1.5 m (5 ft) in diameter and 1.8 m (6 ft) deep.

The problem with ruby lasers lies in generating the two matching pulses required to record a suitable interferogram. Most

lasers can be either pulsed once during each of two consecutive flash lamp cycles or Q-switch pulsed twice in the same

flash lamp pulse to record differential-velocity interferograms. The pulse-separation time in the one flash lamp pulse

mode extends to 1 ms. Generating two matching pulses becomes more difficult as the pulse-separation time exceeds 200

ms because of the dynamic thermal conditions in the laser cavity. The result is images with contour fringes that modulate

displacement fringes, thus obscuring the information sought.

The operation of a ruby laser, when changing pulse-separation time or energy, requires the possible adjustment of flash

lamp voltages, flash lamp timing with respect to the Q-switch timing, Q-switch voltages, and system temperatures. These

conditions change as the laser system ages. Setting up the system requires many test firings to achieve stable

performance. In short, operation of the laser requires high operator skill. In addition, the high performance of these

systems requires care in keeping the optical components clean; buildup of dirt can burn the coatings on expensive optical

components. The periodic replacement of flash lamps and other highly stressed electrical and optical components is to be

expected. A helium-neon or a krypton laser is usually needed to reconstruct a ruby-recorded hologram for data retrieval.

Differences between recording and reconstruction wavelengths lead to aberrations and changes in magnification in the

reconstructed images.

Nd:YAG Lasers. Pulsed Nd:YAG lasers are constructed similarly to ruby lasers. Instead of a ruby rod, however, a

neodymium-doped yttrium aluminum garnet rod is substituted as the lasing medium. The Nd:YAG laser is more efficient

than the ruby system, but it operates in the near infrared at a wavelength of 1.064 μm (41.89 μin.). Frequency-doubling

crystals with efficiencies of approximately 50% are used to produce light at a more useful green wavelength of 532 nm

(5320 Å). All of the pulsed modes of operation available with the ruby system are also available with Nd:YAG system.

The reconstruction of pulsed holograms can be performed with an argon ion laser at 514 nm (5140 Å). Owing to

somewhat better thermal properties, continuous-wave Nd:YAG lasers are available with power capabilities well over 50

W (multimode), but their application in holography is still quite limited.

Exposure Controls

Most holographic systems control light by means of a mechanical or electrical shutter attached to the laser or separately

mounted next to the laser. More sophisticated systems have photodetectors in the optical system and associated

electronics that integrate the light intensity and close the shutter when the photographic plate or film has been properly

exposed. Holographic systems that require strobing capabilities use acousto-optic modulators that can modulate the laser

beam at rates up to at least 10 MHz and with 85% efficiency. It should be noted that a strobed system with a 5% duty

cycle will have an effective brightness of 5% of normal (a 20-mW laser is effectively a 1-mW laser).

Beam Splitters

A piece of flat glass is usually a sufficient beam splitter for a production holographic system designed for recording only.

If the system is to be used for recording and reconstruction or for real-time analysis, there are two approaches:

•

The less expensive system uses a beam splitter that splits 20 to 30% of the light into the reference beam;

a variable attenuator or a filter wheel is used to adjust the reference beam to the proper intensity

• The more expensive approach is to use a var

iable beam splitter, which consists of a wheel that varies the

split from 95-to-5% to 5-to-95% as the wheel is rotated

Beam Expanders and Spatial Filters

Beam Expanders. Expansion of the narrow laser beam is required to illuminate the test object as well as the

holographic film. A short focal length converging lens is often used for this purpose, ultimately causing the beam to

diverge for distances greater than the focal length of the lens. For high-power pulsed-laser sources, a diverging lens must

be used because the field strengths may become so intense at the focus of a converging lens that dielectric breakdown of

the air may occur.

Spatial Filters. An unfiltered expanded laser beam usually displays diffraction rings and dark spots arising from

extraneous particles on the beam-handling optical components. These rings and spots detract from the visual quality of

the image and may even obscure the displacement-fringe pattern. For most CW holographic systems, laser powers are

sufficiently low that spatial filters can be used to clean up the laser beam. Spatial filters basically consist of a lens with a

short focal length and an appropriate pinhole filter. By placing a pinhole of the proper size at the focal point of the lens,

only the laser light unscattered by dust and imperfections on the surfaces of the optical components can pass through the

pinhole. The result is a uniform, diverging light field.

Pinhole Size and Alignment Specifications. A good spatial filter uses a high-quality microscope objective lens; a

round, uniform pinhole in a foil of stainless steel or nickel; and a mount that allows the quick and stable positioning of the

lens and pinhole. A complete analysis of the best pinhole size includes the factors of beam diameter, wavelength, and

objective power. If the pinhole is too small, light transmission will suffer, and alignment will be very sensitive. As the

pinhole size is increased, alignment is easier to achieve and maintain. As the pinhole size becomes too large, it begins to

allow off-center, scattered light to pass through, with the result that the diverged beam will contain diffraction rings and

other nonuniformities associated with dust and dirt. The pinhole will then begin to transmit information to construct the

diffraction field of the particle. This does not prevent the recording of holograms; it only generates unwanted variations in

light intensity. As a general rule, the magnification power of the objective multiplied by the pinhole diameter (in microns)

should equal 200 to 300.

The position of the pinhole should be adjusted at a laser power level below 50 mW. A misaligned pinhole at a high power

level can be burned by the intense point of light, rendering the pinhole useless. High-magnification spatial filters require

the most care. With proper alignment, standard pinholes will function without degradation when the laser output power in

watts multiplied by the objective magnification does not exceed 20.

Mirrors

Most holographic mirrors are front-surface coated. Second-surface-coated mirrors are generally unsatisfactory, because of

losses at the front surface and the fact that the small reflection that occurs at the front surface generates unwanted fringes

in the light field, which interfere with interpretation. Mirrors for holography do not need to be ultraflat. Inexpensive front-

surface-coated mirrors are readily available in sizes up to 610 mm (24 in.). Metal-coated mirrors are usually the least

expensive, but they can cause a 15% loss of reflected light. Dielectric-coated mirrors have greater than 99.5% reflectivity,

but are much more expensive than metal-coated mirrors and are sensitive to the reflection angle. All mirrors, like all other

optical components, require some care with regard to cleanliness. The cleaning of some mirrors is so critical that in many

applications it is best to use inexpensive metal-coated mirrors, which can be periodically replaced. The manufacturer

should be consulted in each instance as to the proper procedure for cleaning each particular type of mirror. Because

mirrors reflect light rather than transmit it, they are a particularly sensitive component in a holographic system. They must

be rigidly mounted and should be no larger than necessary.

Photographic Plate and Film Holders

Photographic plate and film holders perform the following two functions:

• They hold the plate (or film) stable during holographic recording

• They permit precise repositioning of the plate (or film) for real-time analysis

The first function is not difficult to achieve, but the second function is. If real-time analysis is not required, glass plates or

films will work in almost any holder or transport mechanism. Real-time work requires special considerations. The

problems inherent in real-time work can be handled by the use of replaceable plate holders, in-place liquid plate

processors, and nonliquid plate processing.

Replaceable Plate Holders. With replaceable plate holders, the photographic plate is placed in the holder, exposed,

and removed for processing. After processing, the plate is put back in the holder; the plate must be as close as possible to

its original position in the holder to permit real-time analysis. Some plate holders have micrometer adjustments to dial out

residual fringes. As a production method, the use of replaceable plate holders is very slow.

In-place plate processing is accomplished by using a liquid-gate plate holder (termed a real-time plate holder),

which has a built-in liquid tank with appropriate viewing windows. The plate is immersed in the liquid in the tank

(usually water) and allowed to soak for 15 to 30 s. Upon exposure, the tank is drained of the immersing liquid, and the

plate is developed in place by pumping in the proper sequence of developing chemicals. After the plate is developed, the

developing chemicals are replaced with the original immersing liquid, and the hologram is viewed through the gate of

liquid. This procedure not only permits processing of the plate without disturbing its position but also eliminates the

problem of emulsion swelling and shrinking, which causes residual fringes in many real-time setups. Plate development

can take less than 30 s; total processing time is 1 min or less. Commercial systems are available that cycle the appropriate

liquids through the cell as well as provide film advance for holographic films in a continuous-roll format.

Another holographic camera system permitting in-place development uses a thermoplastic recording medium that is

developed by the application of heat. Such systems are available from at least two commercial suppliers. One system

permits erasure and reexposure of the thermoplastic film plate with cycle times of just under 1 min. The plates can be

reexposed at least 300 times. These systems and the high-speed liquid-gate processing systems mentioned above

eliminate many of the inconveniences associated with holographic film handling and processing.

Nonliquid Plate Processing. Other in-place processing systems have been devised. Nonliquid plate processing using

gases for self-development holds much promise for holographic recording. Photopolymers are promising as production

recording media because they can generate a hologram quickly and inexpensively. For one photopolymer film, the

photopolymer is exposed at a much higher energy level than is a silver emulsion (2 to 5 mJ/cm

versus 20 J/cm

or less

for silver emulsions). After exposure, the hologram is ready to use. However, to prevent further photoreaction during

viewing, the hologram is fixed by a flash of ultraviolet light.

Lenses

Lenses are required in some holographic systems. If the function of the lens is to diverge or converge a light beam, almost

any quality of lens will suffice. However, if precise, repeatable control is desired, the lenses may need to be diffraction

limited. Analysis of a proposed holographic system is sometimes best done by trial and error or by use of the best possible

components, rather than by attempting a complicated mathematical computation.

Lenses can be antireflection coated if needed or desired. For example, lenses used in pulsed ruby systems for diverging a

raw beam should be fused-silica negative lenses with a high-power antireflection coating. Some low-power ruby laser

systems, however, have operated satisfactorily with uncoated lenses. Guidelines for using lenses with ruby lasers are

available from the laser manufacturer.

Mounts

Mounts for the holographic components should be carefully chosen. Mounts that require adjustments should be kept to as

few as possible. All fixed mounts should be bolted or welded in place. Some attention should be given to mount material;

aluminum, for example, is generally a good material, but because of its high coefficient of thermal expansion, an

alternative material might be more suitable in a given application. Mounts that are rugged and rigidly built should be

selected. Holographic components should be positioned as close to the supporting structure as is practical.

Holographic Tables

Holographic components must be mounted with sufficient rigidity and isolation from ambient vibration to maintain their

dimensional relationships within a few millionths of an inch during recording and real-time analysis. As discussed

previously, the use of pulsed lasers to generate double-exposure holograms requires very little vibration isolation so long

as the separation between exposures is short. When vibration isolation is required, the designer must exercise care in the

design of the structure used to support the holographic system in order to isolate the structure from outside excitation.

This design involves the three following considerations:

• Building the structure with sufficient rigidity to reduce the deflection of componen

ts to within

holographic limits

•

Building the structure with sufficient damping capacity to absorb excitation energy and to prevent

excessive resonant-vibration amplitudes

• Building the structure with sufficient mass to increase inertia and therefore decre

ase response from

outside driving forces

Small, low-cost holographic systems are usually supported by one of a variety of vibration-isolation tables, which float on

three or four rubber air bladders or small inner tubes. The holographic components are screwed, clamped, magnetically

held, or simply set in place on the table. As the size of the holographic system (and therefore the size of the table)

increases, more care is needed to maintain stability. There are three basic types of large holographic tables:

• Honeycomb tables

• Slabs

• Weldments

Honeycomb tables, made of honeycomb-core sandwiched panels, are extensively used for holography. They can

range in thickness from 50 mm to 0.9 m (2 in. to 3 ft). The outer skin is usually a ferromagnetic stainless steel, but for

increased temperature stability, an outer skin of Invar can be used. Honeycomb tabletops weigh less than 10% as much as

a steel table of equivalent rigidity. They can be fabricated from vibration-damping materials to make them acoustically

dead. The tables are usually floated on three or four air mounts. Air mounts (generally forming a leg for the table) are

large air cylinders with rolling-diaphragm pistons that contact the table. A servovalve inputs or exhausts air at the cylinder

to maintain constant height of the leg and keeps the table level as components are moved about. Air mounts provide

excellent isolation by virtue of their low resonant frequencies, typically 1 to 2 Hz. The holographic components can be set

on the honeycomb table, attached with magnetic clamps, or screwed down utilizing an array of drilled-and-tapped holes in

the upper-skin centers.

Most solid tabletops used in holography are flat within 0.025 mm (0.001 in.) or less, while honeycomb tables (1.2 × 2.4

m, or 4 × 8 ft) are flat within 0.10 to 0.25 mm (0.004 to 0.010 in.). This difference does not hamper the performance of

most holographic systems.

Slabs are the least costly type of support for a large holographic system. They are usually made from steel or granite and

floated on a vibration-isolation system. Many low-cost supports have been laboratory constructed by floating a surplus

granite or steel surface plate on an array of tire tubes. It is usually difficult to dampen vibrations that reach the surface of a

slab. For this reason, the performance of a slab degrades when the test objects are large and/or the ambient noise level is

high (particularly from air-conditioning systems that emit low-frequency noise). This problem can be minimized by

selecting a material (such as gray iron) that has naturally high damping capacity rather than a material that has a ring.

Most components that are attached to three-point mounts need not be rigidly attached to the slab, but other components

(and particularly the object being vibrated or otherwise stressed) need to be rigidly mounted to ensure stability. To

facilitate the mounting of components, the slab top may require tapped holes, T-slots, or a coating of tacky wax, or it may

need to be ferromagnetic.

Weldments are heavily braced frames or plates generally designed as part of a portable or otherwise special system

used to analyze very large or unusual test objects. Weldments are generally used where slabs or honeycomb tables are not

suitable, although a slab or honeycomb-core sandwich panel may be part of the structure for mounting the components.

References cited in this section

18.

J.D. Trolinger et al., Putting Holographic Inspection Techniques to Work, Lasers Applic., Oct 1982, p 51-

19.

The Optical Industry and Systems Purchasing Directory, 34th ed., Laurin Publishing Company, 1988

Optical Holography

Revised by James W. Wagner, The Johns Hopkins University

Types of Holographic Systems

There are basically two types of holographic systems: stationary and portable. Both will be discussed in this section.

Stationary Holographic Systems

A holographic system is considered stationary when it is of such size, weight, or design that it can be utilized only by

bringing the test object and required stressing fixtures to the system for analysis. Stationary systems are usually dependent

on building services, requiring compressed air for the vibration-isolation system, electric power for the laser and other

electronic components, and running water for processing the holograms and cooling the laser (for example, as required for

an argon laser). Most stationary systems operate in a room with light and air control to achieve high stability and the low

light levels required for recording, processing, and viewing holograms. As the size of the table increases, the stability

requirements become more difficult to satisfy. As a result, the cost of a stationary system generally increases

approximately exponentially with object size.

Stationary systems are used in the following cases:

• Production line inspection of small objects

• Where required flexibility in the type and size of test objects is needed for developmental work

• Inspection of a large or awkward structure that cannot be holographed by a portable system

Portable Holographic Systems

A portable holographic system can be moved to the test object and operated with minimal setup time. A portable system

built for the Apollo lunar exploration program was designed to record holograms of lunar soil. It was battery powered,

weighed 7.89 kg (17.4 lb), and occupied less than 0.017 m

(0.6 ft

) of area. A portable holographic system used for

developmental work, particularly in wind tunnels, has been transported throughout the United States by semitrailer truck.

The components of the system are mounted in cabinets or in frames on wheels, and upon arrival at the test site, the system

is unloaded and set up in several hours. A portable system used to inspect sandwich-structure helicopter-rotor blades is

shown schematically in Fig. 5; a portable system used to inspect sandwich panels is shown in Fig. 10.