Module 6-1


Copyright 1987 by The Center for Occupational Research and Development

All rights reserved. No part of this book may be reproduced in any form or by any means without permission in writing from the publisher.

The Center for Occupational Research and Development
601 C Lake Air Drive
Waco,Texas 76710

Printed in the U.S.A.

ISBN 1-55502-024-0

See Idea Bank

(1) In electro-optical laboratory environments, few considerations are more critical than the way optical components and associated equipment are mounted.

(2) In Module 6-2, "Component Support," we’ll discuss the actual devices that are used to mount components such as mirrors, lenses, prisms, and diffraction gratings. In this module we will focus our attention on the surface on which the optical mounts are affixed. We will direct most of our discussion toward the requirement of nonpermanent optical assemblies. Most of the optical assemblies that you’ll construct in research laboratories are of the "experimental" nature. They are meant to be dismantled after serving a particular purpose. Permanent optical assemblies, such as the optics in a supermarket bar-code scanner, are custom-designed. They are not part of our discussion—although most of the same principles that we’ll discuss still apply.

(3) The main purpose of this module will be to help you to make a reasoned judgment of what type of device to select as the "foundation" of a particular optical system. Depending on the particular geometry of an optical setup, various options will be more favorable than others.

(4) For example, if you need to assemble an optical system that lies along a single straight axis, an optical rail or optical bench might be an appropriate choice on which to attach the various optical mounts. On the other hand, if the optical setup will by necessity occur in a plane, or even several planes, then an optical table or optical breadboard will usually be a more convenient choice.

(5) We’ll also discuss the major requirements of critical optical assemblies such as mechanical and thermal stability and vibration isolation. Most of our discussion will be about modern optical benches, tables and breadboards that have become de facto industry standards in laboratories that use lasers and related optical and electro-optical apparatus.

(6) Before you start this module, you should understand algebra, general laboratory and electrical safety, waveform analysis, and the use of transducers, oscilloscopes and low-power helium-neon lasers.


objectiv.jpg (6047 bytes)

(7) When you complete this module, you should be able to do the following.

1. Name four classes of optical benches. List one advantage and one disadvantage of each as given in the text.

2. Explain the following expressions in your own words:

• Vertical vibrations in a table

• Horizontal vibrations in a table

• Vibration isolation from table support

• Internal table damping from acoustically induced vibration

• Thermally induced movement

• Table rigidity

• Table flatness

3. List the advantages and disadvantages as given in the text (at least two each) of these optical tables:

• Lab workbench

• Lab workbench covered with thick metal plate with matrix of tapped mount holes

• Concrete slab on inner-tube support

• Granite slab with insert matrix, mounted on steel frame with rubber shock mounts

• Granite slab with insert matrix, mounted on pneumatic vibration isolators

• Steel table with magnetic mounts

• Metal honeycomb table, with stainless steel skin and matrix of tapped holes, vibration isolation

4. Align an optical bench parallel to an optical axis established by a laser beam.

5. With an accelerometer and oscilloscope, investigate the vertical and horizontal vibrations in an optical table, as follows:

a. Using a spherical object dropped to the table surface from a constant height, measure approximate frequencies of the vibrations thus excited. Compare such responses at various impact points.

b. Determine peak-to-peak amplitudes of the major resonances for the table, in terms of the patterns displayed on the oscilloscope.

c. Determine the time required for each major resonance to decay to half of its peak amplitude.

d. Compare the vibrational responses of the table in two horizontal directions, 90 apart, with those in the vertical direction.

e. By comparing peak amplitudes of vibrations on the table with those on the floor at the table base—both excited by identical impulses applied to the floor—appraise relative effectiveness of the table in damping floor-induced vibrations.



Optical Benches

General Comments

(8) An optical bench—sometimes called an optical rail—gives a stable, inline surface where optical system components and supports can be installed and adjusted to various relative positions and then held rigidly in place. The optical bench is most useful when the optical axis of the experimental arrangement or system lies along a straight line or in a vertical plane.

(9) Technicians use most optical benches for one or more of these applications.

a. Alignment of an optical system (such as a laser).

b. Evaluation of an optical system (such as measurement of the modulation-transfer function of a lens).

c. Support of a breadboard for an optical system.

(10) In each case the bench or rail is the stable base on which you mount the component supports. The mounting devices are called "carriages." The component support is mounted either directly onto the carriage or at an adjustable height above the carriage on a rod. These elements are all shown in Figure 1.


Fig. 1.
Elements of an optical-bench setup


(11) This description of optical rails or benches will be divided into four broad types:

Lightweight aluminum rails


Triangular rails

Flat-bed or "lathe-bed" benches

(12) Generally speaking, this list is shown in increasing order of quality and expense. You can get a good-quality one-meter-long aluminum rail for less than $200. But, certain complete flat-bed systems will cost more than $60,000.

(13) The quality of an optical bench is measured in terms of how well it satisfies these two questions:

1. How rigidly can it hold component supports in place?

Rigidity is controlled by the material used in the bench and the size and shape of the bench cross section. For example, a granite bench is typically more rigid than a metal one. The larger and more massive benches usually are more rigid, and various-shaped cross sections (H-beam, triangular, and dovetail) are used to increase rigidity.

2. How accurately can optical components be translated along an optical axis when guided by a bench?

The answer to this question lies in the "straightness" of the bench. Metal and granite benches are machined flat and straight. They can be accurate (on large lathe-bed benches) to a thousandth of an inch.


Lightweight Aluminum Rails

(14) A typical example of a lightweight aluminum rail is shown in Figure 2. Stock lengths of rail are available from about 15 centimeters up to two meters, with either English or metric mounting holes. Attached to the rail are two typical carriages or carriers that accept rod-mounted optical components.

(15) Aluminum rails like these usually are designed with slotted mounting holes to allow secure attachment to optical tables, as shown in the figure. Many manufacturers also provide coupling plates so that two or more optical rails can be firmly attached at various angles to accommodate optical setups that require more than a single axis.

(16) A variety of carriages are also available. The carriages let you mount a wide variety of components and equipment such as lens holders, translators and rotators. A more complete discussion of these mounts is in Module 6-2.


Fig. 2
A typical aluminum rail with carriages shown attached to an optical table.
Photograph courtesy of Oriel Corporation.


(17) For added convenience, many optical rails have linear scales running along their length. The linear scale helps you to position equipment accurately. Vernier scales are also often attached to the carriages. These give an even more accurate determination of the position of components mounted on the rail.

H Beam

(18) An H-beam optical bench is used for rugged breadboards of heavy electro-optical or laser systems. One is shown in Figure 3a.

(19) The bench is made from a structural H-beam of either steel or aluminum. Channel often is used also. The top surface is milled flat. Components are either bolted down to the H-beam or held rigidly with C-clamps.

(20) The H-beam bench is relatively inexpensive and can be made in the shops. Its main disadvantage is that elements can’t be translated along the beam and remain in optical alignment. However, where you need this feature, you can modify an H-beam bench as shown in Figure 3b by milling edges and clamping the component to the side with a setscrew.


a. Laser breadboard on H-beam bench with C-clamps to hold component supports.


b. Modified H-beam bench to maintain component alignment during translation.

Fig. 3
H-beam benches


(21) In the early days of the laser industry, such "homemade" optical rails were quite common. In many cases, the optical components were held in mounts, affixed to the H-beam with double-sided tape! With the improved convenience and availability of commercial aluminum rails and ferro-magnetic optical tables and breadboards, such improvised mounting systems are an increasingly rare sight in industrial and university laboratories.


Triangular Rail

(22) A triangular optical rail (Figure 4) is the most widely used optical bench. The rail usually is made of aluminum, cast iron, or steel. It has an equilateral triangular cross section (about 2 inches per side) with grooves on the two top sides. The rails can be made straight to 0.005 inch. The grooves running down the length of both sides are used to lock the carriages in place with a screw. (See Figure 1.) Triangular benches are fixed directly to optical tables. Or, they’re mounted on cross-feet for precise leveling.


Fig. 4
Triangular optical rail


Flat-bed Benches

(23) Flat-bed optical benches (Figure 5) offer the ultimate in rigidity and straightness in an optical bench. They can be made from high-grade stainless steel or granite. The bed usually has a dovetail cross section for accurate, stable guiding and clamping of optical carriages. The benches are extremely heavy and expensive—they’re used for very special applications where high precision is required.


Fig. 5
Flat-bed optical bench


Thermal Effects on Bench Length

(24) Spacing between elements mounted on an optical bench varies if the temperature changes cause the length of the bench to change. This is due to thermally induced expansion and contraction of the bench material. (The following material has been moved here from pages 17 and 18.

When a material is heated, it will expand according to the relationship,

D L = aLo D T

Equation  2



Lo = Original length of the material

D L = Change in length of the material

D T = Temperature change, C

a = Coefficient of linear expansion for particular materials, C–1


The coefficient of linear expansion, a, varies with different materials. Typical values of a are given in Table 1.


Table 1. Typical Values of a


a (in C–1)

Aluminum (type 6061-T6)

23.4 10–6

Stainless steel (type 400)

10.4 10–6

Super Invar (31% Ni, 5% CO, 64% Fe)

–0.19 10–6 (a 20C)

Cast iron

10.8 10–6


0.55 10–6


3.2 10–6


7.2 10–6


The effect of thermal expansion on laser components attached to a stainless steel table are illustrated in Example A. wavelengths.


Example A: Thermal Expansion of Laser Cavity

Given: An "open-cavity" HeNe laser is comprised of a Brewster-angle plasma tube and the HR and OC mirrors whose mounts are attached 1.00 meter apart to the top of a stainless steel optical table. Room temperature increased by 1.00C.
Find: How many longitudinal modes of the laser cavity the laser will "hop" through.
Solution: D L = aLOD L = (1.04 10–5)(1.00)(1.00)

D L = 1.04 10–5 meters
New longitudinal laser mode will form every time D L =

where: l = wavelength of HeNe laser = 6.33 10–7m
  = meters = 3.17 10–7 meters

Number of "mode hops" = =

@ 33


(32.3.4) Therefore, the longitudinal modes in the laser cavity will "hop"—in other words longitudinal modes will form, collapse, and reform—about 33 times as the optical table heats up. If we monitored the output of the laser with an optical power meter, we’d notice a periodic fall and then rise in the laser output.

(32.3.5) In laboratories where thermal stability is particularly important the optical tables often are made of Invar, a low-expansion alloy whose expansion coefficient is only –0.20 10–6 C–1. That’s about 55 times better than typical stainless steel! (End of material moved from pages 17 and 18.)


Care of Optical Benches

(25) Optical benches must be cared for to prevent warping, rusting, and nicking. A long bench should be supported from beneath at points approximately every two feet along its length. Such support will prevent the bench from bending under its own weight and the weight of the elements it supports. Use common sense to prevent loading a bench with elements that are too heavy.

(26) Since most benches are made of steel, they can rust. Use a coat of light oil—applied sparingly and then wiped off—on the bench and the carriages every several months. Do not use oil on the component supports.

(27) Be careful not to hit the bench surface with a carriage or other object. This can cause nicks and grooves that will prevent the carriages from moving smoothly along the surface and will frustrate attempts to maintain alignment.


Aligning a Bench to an Optical Axis

(28) Frequently a test will require that an optical bench be aligned parallel to an existing optical axis or light beam. If a laser or light beam defines the optical axis, you accomplish the alignment procedure by using an iris diaphragm or a set of cross hairss mounted on the bench with a carriage. You adjust the bench position until the laser beam passes through the center of the iris (or cross hairs) without perpendicular carriage adjustment at any position along the optical bench.

(29) Detailed steps for carrying out the alignment procedure, using a laser and a set of hairs, are outlined below:

1. SECURELY position the laser to establish an optical axis. Most benches are designed to be aligned to a horizontal axis, so try to establish the laser beam axis as close to horizontal as possible.

2. Obtain in index card and draw a set of perpendicular "cross hairs" in the center of the card.

3. Position the optical bench roughly so that it appears to be parallel below the laser beam. Move all adjustment screws to the center position of their turning range.

4. Center and secure the card to a mount such that the laser beam or the cross hairs is centered when the mount is placed on the far end of the bench. Do not secure the mount to the bench since it will be moved from one end of the bench to the other.

5. It is very important not to change the position of the card with respect to the mount any more after this! Whenever the card is to be moved, the bench mount containing the card must be moved with it.

6. Move the mounted  card to the near end of the bench. If the laser beam does not center on the cross hairs, turn the adjustment screws on the bench until it does.

7. Move the card to the other end of the bench. The laser beam should center on the cross hairs. If it does not, turn the adjustment screws on the bench.

8. Repeat steps 6 and 7 until you no longer have to touch the adjustment screws to line the laser beam up with the cross hair--on either end of the bench.


When Does an Optical System Require
a Table or Breadboard?

(30) A table (Figure 6) is desirable for breadboarding an optical system, test or experiment if any one or more of the following conditions exists or is required.

1. More than one optical path in the setup

2. Optical path runs in more than one horizontal direction

3. Vibration isolation is desired

4. Components are numerous or are relatively massive

(31) Typical examples of applications where you should use optical tables are listed below.

1. Multiple-path interferometers—Mach-Zehnder, Twyman Green, Michelson. Stability of the mirror separation is critical to a fraction of a micron. Vibration isolation is also desirable.

2. Hologaphy. Long-term vibration isolation is required. Rigidity and maintenance of specific separation between components also are required. See Figure 7. Tables for this use usually contain small inner tubes that provide both support and vibration isolation.

3. Research laser systems. The laser mirrors require rigid mounts to ensure stable output. Associated devices such as Q-switches, frequency doubling crystals, amplifiers, and optical power meters require a stable mounting surface. For convenience, components often are mounted at angles to the original laser beam path.

Fig. 6
Example of a high-quality optical table. This table has pneumatic isolation legs,
a lower accessory shelf (for mounting a laser) and beam-steering mirrors.
Photograph courtesy of Newport Corporation.


Fig. 7
A holographic setup mounted on a small multi-rail optical table.


4. Electro-optical systems development (examples: IR scanner, scanning imaging spectroradiometer, star trackers, laser radars). Requires support of massive components, stability, location of electronic, and thermal equipment immediately adjacent to the optical setup. See Figure 8.

Fig. 8
Electro-optical system test on an optical table


Desirable Characteristics of Optical Tables

(32) Generally, an optical table should provide a stable plane surface where optical components and supports can be rigidly placed or moved to any position or angle. The ability of a table to satisfy these requirements depends on the degree to which it has the following characteristics.

1. (32.1.1) Flatness. Table flatness is desirable so that if a mirror support is moved from one position on the table to another, the mirror isn’t "tilted" in angle. Flatness also is needed to prevent "rocking" of the component support because its bottom plane and the tabletop are not matched.

(32.1.2) A surface is defined to be flat to within 0.00X inch if no point on the surface is more than "X" thousandths of an inch from an imaginary reference plane through the surface. Modern honeycomb optical tables are flat to about 0.005 inch over lengths of 8 or more feet.

2. (32.2.1) Rigidity. The static measure of rigidity is the "stiffness" of a table surface. Rigidity describes the degree to which a table bends under its own weight and the weight on the components it supports. Rigidity is determined by the type of support under the table, the composition and structure of the table material, and the table thickness.

(32.2.2) It’s obvious that a 0.5-inch-thick aluminum table surface five feet square, supported by four legs (located near the corners) will flex in the center due to its own weight and the weight of anything it supports. However, it may not be as obvious that four-foot by eight-foot granite tables will have noticeable differences in rigidity if they’re only eight inches thick instead of fourteen inches thick. Rigidity in bending improves as the square of the table thickness.

(32.2.3) Optical tables with metal honeycomb cores and ferromagnetic stainless steel top surfaces have become essentially a de facto industry standard for most applications in electro-optical research laboratories. Rigidity of such tables isn’t as good as a granite table of similar dimensions. But, it’s usually more than adequate for the vast majority of typical laboratory applications. The graph in Figure 9 shows the static deflection of the center of a 4-foot-wide honeycomb table as a function of both table thickness and table length. For five-foot-wide tables multiply all deflections by 0.80. This deflection is caused solely by the weight of the tabletop. It doesn’t take into account additional load applied by equipment mounted on the table surface.

Fig. 9
Plot of static deflection of a 4-ft wide steel honeycomb-core optical table.
Courtesy of Newport Corporation.

(32.2.4) Figure 10 shows how a table surface will deflect when it experiences a point load at its center. Such a load occurs when heavy equipment—such as lasers and spectrometers—is mounted on the table.

(32.2.5) Although equipment doesn’t apply load to the table at a single point, the point-load model is useful in estimating the actual deflection that can be expected. You can figure the additional deflection of the center of a honeycomb table due to a point load at the center from Equation 2.

Deflection of the center of a honeycomb optical table under a central point load.
Figure courtesy of Newport Corporation.


Equation 2


where: P = Force exerted by the point load (in lbs)

L = Length of the panel (in ft)

b = Width of the panel (in ft)

H = Thickness of the pane (in ft)

T = Thickness of the skins (in ft)

E = Young’s modulus for the skin material (in lb/ft2)

G = Shear modulus for the core (in lb/ft2)

With these units, the deflection Thing.gif (84 bytes) will come out in ft, as shown in Example B.

See Idea Bank


Example B: Deflection of an Optical Table

Given: A piece of equipment with a weight of 100 pounds (mass of 45.4 kg) is placed in the middle of a 4-ft 10-ft steel honeycomb table. The stainless steel "skins" of the table are 3/16 inch thick. Total panel thickness is 8.0 inches. Assume that the equipment produces a point load at the center of the table.
Find: Maximum deflection of the center of the table.


For a steel honeycomb

E @ 2.90 107 = 4.18 109

G @ 2.25 105 = 3.24 107

H = 8.0 in. = 0.667 ft

P = 100 lb

T = in. = 1.56 10–2 ft

Image618.gif (2773 bytes)


(32.2.6) Deflection in this realistic example is quite small. But, when aligning certain optical systems, this small movement could be critical. For example, in holography, motion on the order of a fraction of a wavelength of light can cause significant problems. Compare the table deflection to the wavelength of a HeNe laser to get a better feeling for the potential significance of the motion.


(32.2.7) Therefore our table deflection represents a motion of almost 19 wavelengths of light. This degree of motion could ruin the results of a sensitive experiment such as holographic image recognition.

3. Thermal Expansion. When a material is heated, it will expand according to the relationship , D L = aLo D T, as was discussed earlier.

As comparision of the values of a shown earlier in Table 1 indicates that metal tables are more susceptible to thermal expansion than granite. The significance of this is that temperature fluctuations of a centigrade degree in a laboratory could change a one-foot optical-path length in an interferometer on an aluminum table by about 11 wavelengths of HeNe laser light.

Image620.gif (2162 bytes)

4. (32.4.1) Vibration isolation. Vibrations induced into the surfaces can cause slight but significant movement in the table, and consequently the optical components. Vibrations can be induced in the table from the floor through the support structure by mechanical vibrations, or through the air by acoustic disturbances.

(32.4.2) Vibrations are vertical vibrations (where table ends flex up and down) and horizontal vibrations (where the table moves laterally). Generally, the vibrations of concern are relatively low-frequency (less than 100 hertz). In addition, the tabletop itself can vibrate in a number of torsional and bending modes. Several of the lower order modes are shown in Figure 11.

Fig11.jpg (33534 bytes)

Fig. 11
Low-order bending and torsional modes of a uniform rectangular plate.
Figure courtesy of Newport Corporation.
(Fig. 11 has been changed, but the figure is not changed to red to indicte this)


(32.4.3) Bending and torsional modes of the table surface can be reduced greatly in amplitude by constructing the core of the table from an appropriate damping material. Most modern optical tables use various metal honeycombs for this purpose.

(32.4.4) Honeycomb construction was developed in great part for use in aircraft construction. Honeycombs provide excellent internal damping. And, they provide good mechanical strength with relatively little weight. Figure 12a shows the response of a 4-foot 10-foot 8-inch solid granite slab to an impulse. Note that after 0.30 second there’s still an appreciable amplitude to the "ringing" of the tabletop. By comparison, a similar impulse is applied to a same-size steel honeycomb tabletop. The resulting "ringing" is damped out in approximately 30 milliseconds (Figure 12b). That’s a dramatic difference in damping rates.

Fig. 12
Responses of solid granite plate (a) and metal honeycomb optical tabletop
(b) are shown in lower plots when they’re subjected to impulse forces
shown in upper traces. Figures courtesy of Newport Corporation.


(32.4.5) Vibrations that result from motion of the laboratory floor usually occur in approximately the 5-Hz to 30-Hz range. Modern isolation optical table systems usually are suspended on pneumatic legs that are designed to resonate mechanically at about 1 Hz when loaded with a tabletop. The mechanical resonance is far lower in frequency than typical floor vibrations. So, little energy is transferred from the floor to the table surface.

(32.4.6) The pneumatic isolation support system combined with the metal honeycomb damping materials make modern optical tables an excellent surface on which to conduct vibration-sensitive experiments. For these reasons—in addition to convenience and cost—steel honeycomb table systems with pneumatic supports have replaced solid metal or granite tabletops in most electro-optical laboratories.

5. (32.5.1) Ability to attach a component holder firmly in place. Optical components normally are held by mounts that have precision-flat bases. This allows the experimentalist to slide the base around on the table until the component is positioned correctly. When the mount is in the proper position, it’s desirable to firmly attach it to the table.

(32.5.2) As we’ll see in Module 6-2, several methods of attaching mounts to surfaces commonly are used in electro-optical laboratories. The most common types of attachment are: bolts into threaded holes drilled in the table or bench surface, inertial mounts (heavy bases) and magnetically held mounts. Modern honeycomb-construction optical tables and breadboards usually are supplied with a standard pattern of threaded bolt holes and a ferromagnetic upper surface, to facilitate the mounting of components.


Types of Tables—Methods of Fixture Attachment

General Comments

(33) Several types of optical tables are used. Some are very crude (and inexpensive) and some are very elaborate (and expensive). It’s important to know how much quality (flatness, rigidity, vibration isolation and low expansion) you need for a specific application.

(34) If an optical system doesn’t require high-precision alignment and stability, it’s economically wasteful to specify a high quality. Conversely, if high-precision alignment and/or stability are required, a crude setup will waste the experimentalist’s time and possibly cause erroneous results.

(35) The following list describes most optical tables that are commonly used, along with the method of attachment of fixtures to the table.

1. (35.1.1) Workbench or office table. These crude surfaces often are used for simple, noncritical optical alignment or tests in most laboratories. Typical tests that are conducted on these surfaces are: a rough check on the paper output of gas laser (see Figure 13a); frequency response of LED communication system; and measurement of the time description of the output of a pulsed laser.

(35.1.2) Optical fixtures and components usually are mounted in separate stands, or on lab jacks, books, or boxes. Components can be held somewhat rigidly by double-sided tape. Considerable vibration can be introduced into this setup because fixtures don’t rest flat on the table, table legs don’t rest uniformly on the floor and table support is usually unsteady. And, many elements may be cantilevered.

a. Laser power measurement on a workbench


b. Laser breadboard on workbench with metal surface plate

Fig. 13
Laser workbench setups

2. (35.2.1) Workbench with metal surface plate. An inexpensive way to improve the optical table quality of a workbench is to mount a precision-flat, heavy metal plate on the top of the bench. (See Figure 13b.) This technique will give a rigid, flat surface and allow the component fixtures to be firmly attached by either mechanical or magnetic mountings. (We’ll describe these mounting techniques under "Metal Optical Tables.") Significant vibration isolation at frequencies greater than 100 hertz can be achieved by mounting the plate on compressible rubber pads.

3. (35.3.1) Smooth granite table. The classical form of optical table is a smooth granite slab (Figure 14a). These tables can be obtained in surface areas greater than 6 feet by 12 feet, thicknesses of 14 inches, and flatness of 0.000025 inch. Component holders are attached to the smooth granite surface by suction holders. (See Figure 14b.)

Fig. 14
Smooth granite table and suction-type component holder

(35.3.2) So-called "inertial mounts" sometimes are used on smooth surfaces such as granite or steel. An inertial mount is a heavy base—it usually weighs at least 6 pounds—that contacts the table surface at three points. Such mounts are inexpensive and can be repositioned easily. Inertial mounts (or "gravity bases") are useful in optical setups where the components are unlikely to be bumped or otherwise inadvertently moved.

(35.3.3) Desirable properties of granite are:

• Naturally noncorrosive

• Resistance to warping or distortion

• Dense structure with low porosity

• Low thermal distortion

• Uniform hardness, exceeding that of hardened steel

• No crater raised by the impact of a sharp point or edge

4. (35.4.1) Granite table with threaded insert holes. A more rigid tie-down of component supports to a granite table is achieved with a matrix of holes in the top surface. The holes are filled with steel inserts, tapped for machine screw threads. (See Figure 15a.) The screw size for the inserts is usually 1/4–20. You can attach component supports to the table by screwing them directly into the table inserts, or by clamping them with bars screwed into the inserts. (See Figure 15b.) The matrix of inserts may be located as close together as 2 inches or as far apart as 6 inches, depending on the application.

a. Granite table with threaded insert holes


b. Attachment for component support

Fig. 15
Granite table

5. (35.5.1) Metal tables with tapped mounting holes. These tables are used similarly to the granite tables with inserts. Solid metal tables are usually not as thick or expensive as comparable granite tables. In certain electro-optic applications, it may also be desirable to use a metal table so that it can act as an electrical grounding plane for the system electronics.

(35.5.2) As we said earlier, most metal (steel or aluminum) tables aren’t solid. They have a metal "honeycomb" core structure covered with a thin metal skin (1/8 inch to 3/16 inch thick). (See Figure 16.) This composition provides increased rigidity and good damping of lateral vibrations.

Fig. 16
Metal honeycomb table with thin metal surface skin


(35.5.3) Similar tables are also available that are composed of a phenolic-impregnated paper honeycomb core and solid surface with a matrix of tapped holes.

6. (35.6.1) Smooth metal tables with magnetic mounting. Either solid or honeycomb-core tables are available with smooth working surfaces of ferromagnetic metal. Component supports are attached to these tables by magnetic mounts. (See Figure 17.)

(35.6.2) These magnetic bases are switched "on" and "off " by rotating a strong permanent magnet within the base. In the "off" position the magnetic flux travels almost entirely through the metal of the base. In the "on" position, the permanent magnet is rotated so that virtually all of the magnetic flux travels through the table surface.

(35.6.3) Depending on their size and cost, such magnetic bases can have holding forces of from 24 to 300 pounds. These bases work best on unpainted, ferromagnetic surfaces (such as type-410 or -430 stainless steel) with a skin thickness of at least 5 mm (3/16 in).

Fig. 17
Magnetic mounts to attach component supports to metal optical tables


(35.6.4) This technique should not be implemented where the presence of strong magnetic fields might temporarily distort or destroy an electro-optic component (such as image-converter tube assemblies).

7. (35.7.1) In many applications, excellent vibration isolation isn’t needed, but the convenience of a ferromagnetic mounting surface and predrilled holes is desired. In such cases, optical "breadboards" often are used as a lower-cost alternative to honeycomb isolation tables. Figure 18 shows one example of an optical breadboard.

Fig. 18
A modern ferromagnetic optical breadboard


(35.7.2) Like honeycomb isolation tables, these breadboards come in a variety of stock and custom sizes and hole patterns. Typical hole patterns are 1/4-20 holes on 1-in centers. But, metric holes (usually M6 on 25-mm centers) are also available. The 2-ft 3-ft breadboard shown in Figure 18 also has a slot along each side to permit equipment to be "hung off " the side using bolts and captive nuts.

(35.7.3) Some manufacturers also will finish the breadboard surface epoxy-based semigloss black paint to reduce stray reflections. Such finishes, while attractive, can be fairly easily marred. And they somewhat reduce the holding power of magnetic bases.

(35.7.4) If you need some degree of vibration isolation, you can suspend breadboards on firm tabletops with small-diameter inner tubes or sheets of foam rubber. In some cases, these "budget" isolation surfaces make sense as a significantly lower-cost alternative to pneumatic isolation tables.


Vibration Isolation

(36) As we said earlier in this module, optical tables are susceptible to induced vertical and horizontal vibrations. These vibrations are undesirable in applications that involve very precise optical alignment systems because they result in relative movement between the optical components mounted to the table.

(37) You can understand vertical vibrations within the table surface by visualizing a very small, periodic bending motion, or flexing, in the table—similar to a diving board. (See Figure 11.) These vibrations originate as floor or building resonant vibrations. They’re induced through the table supports. The frequencies of these vibrations typically range from 5 to 30 Hz.

(38) You can isolate some vibration in a rigid-frame table support by placing rubber pads between the table legs and the floor and between the support and the table surface. You’ll get much more effective isolation by using air pistons in the table legs. (See Figure 19.) These pistons actually "float" the tabletop and provide isolation from vertical floor vibrations. Vibration-isolation systems are similar to a high-frequency cutoff filter, insofar as they pass low-frequency vibrations and isolate all frequencies above some design value. Air pistons provide vertical isolation for frequencies above 5-10 Hz, and generally require a source of compressed air in the laboratory.


Fig. 19
Air piston for optical table vibration isolation


Selecting the Correct Table

(39) The design and construction of a vibration-isolated optical table takes considerably more theoretical insight and working experience than we can give you in this module. The important points to learn are:

A general understanding of the kind of performance you can expect of crude and sophisticated optical tables

An estimate of the optical table performance required by the experiment or alignment to be performed

Knowledge of how and what to specify when you need an optical table

(40) The lab tasks in this module should provide some experience for the first point. An understanding of the optical experiment or alignment should provide data for the second point. Address the third point by answering these questions:

How much working area do you need?

How accurately must you maintain component spacings?

What maximum temperature variations do you expect in the laboratory?

What degree of table surface flatness is required?

What vibration isolation (if any) is required? How much vibration can be tolerated, and at what frequencies?

Do you need to fasten components rigidly to the table surface?

Would the fields produced by magnetic support attachments distort the operation of the electro-optical system?

Would a conductive work surface (ground plane) be beneficial or detrimental to the electro-optical system operation?

What’s the "long-term" use of the table? How much versatility do you need?

How much money is in the budget for this equipment?

(41) Quite often, an early consideration of the last item forces compromises in system requirements for the optical table.


exercise.jpg (6215 bytes)

1. Name four classes of optical benches. List one advantage and one disadvantage of each.

2. Define the following terms:

Vertical vibrations in a table

Horizontal vibrations in a table

Vibration isolation from table support

Internal table damping from acoustically induced vibration

Thermally induced movement

Table rigidity

Table flatness

3. List the advantages and disadvantages (at least two of each) of the following optical tables:

Lab workbench

Lab workbench covered with thick metal plate with matrix of tapped mount holes

Concrete slab on inner-tube support

Granite slab with insert matrix, mounted on steel frame with rubber shock mounts

Granite slab with insert matrix, mounted on pneumatic vibration isolators

Steel table with magnetic mounts

Metal honeycomb table, with stainless steel skin and matrix of tapped holes, vibration isolation.

4. In your own words, outline the procedures used in the laboratory to align an optical bench parallel to an optical axis established by a laser beam.

5. Outline the specifications of two optical tables of comparable size. Use the criteria given in the section, Selecting the Correct Table, and catalogs from suppliers of optical tables. From your outline, evaluate the effectiveness of the table in isolating low-frequency vibrations.

6. How much will the center of a 4-ft 8-ft 12-in steel honeycomb table deflect when loaded with a 150-lb spectrometer at its center. Skin thickness is 1/8 inch. The spectrometer applies a point load to the center of the table. Get any more data that you need from the example worked earlier in this module. Express your answer in mm.

7. In an experiment that’s particularly sensitive, you can allow the center of your optical table surface to deflect downward by an additional distance equal to only one wavelength of light (633 nm). Assume that you use the table from Exercise 6 and that the only significant load on the table is a point load in the center. What’s the maximum point load allowed? Express your answer in pounds.

8. During a day of use, the air temperature in an optics laboratory increases by 2.50 C. Compute the thermal change in length of three 8.00-ft-long optical tables used in the lab. The tables are made respectively of aluminum, type-400 stainless and super invar. Express your answers in multiples of the red HeNe laser wavelength.

material.jpg (5811 bytes)


2-ft 3-ft 1-in shock-mounted metal plate to fit on workbench

4 rubber shock mounts

Pneumatic vibration-isolated optical table, 4-ft 8-ft working surface

Baseball or tennis ball

Sandbag, about 15-25 pounds of mixed sand and various-sized rocks

Aluminum cube, 3-in sides, tapped to accept accelerometer mounting stud

HeNe laser


Optical bench

Cross target





procedur.jpg (6345 bytes)

1. Align an optical bench parallel to a given optical axis defined by a laser beam. Use the technique described in the discussion.

2. Determine the response of a table in each of the three axis directions to an induced vibration.

a. Arrange accelerometer, cable and scope as shown in Figure 20. The accelerometer should be in the vertical position (circular end facing up), and directly over the approximate center of the tabletop. This will now indicate vibration in the Z-axis direction. The accelerometer connecting cable shouldn’t be kinked (damage can result) or looped (inductive pickup can occur). Some scopes are also sensitive to vibration. They should be supported on vibration-damping pads. Foam rubber is effective.

Fig. 20
Experimental arrangement for measuring table vibrations


b. Adjust the vertical amplifier of the scope for maximum sensitivity (5-mV/cm desired). Set the sweep speed at 0.1 msec/cm. Adjust trigger controls to initiate a sweep when a fingertip is brushed lightly across the top of the accelerometer. Triggering level will be near zero, stability control is Preset, slope control on Internal +, and trigger mode on AC slow. If you can’t adjust the trigger level to activate on the weak signal generated by touching the accelerometer, you can set the trigger mode on automatic.

c. Hold the ball accurately 30 cm above the middle of the long edge of the table (point A). Release it (be prepared to catch it on the first bounce) while you watch the scope trace pattern. If possible, reduce room light or shade the scope screen so that you can see a persisting image.

d. If the pattern height is less than one or two cm, increase the height from which you drop the ball. Make subsequent drops from the same height. Record this height as (h).

e. Examine the scope pattern to estimate the general range of frequencies present. The number of full waves (distance between successive peaks) per centimeter, divided by the time per cm, will give an approximate frequency value. For example: the pattern shown in Figure 21 has about 2.5 full waves in a length of 1 cm. With the sweep rate at 0.1 see per cm, the vibration frequency would be 2.5/0.1–25 hertz. If higher frequencies are indicated by the presence of complex wave forms and fine structure, increase the sweep rate while you excite the table (ball drop) until a measurable and reasonably repeatable pattern is produced. Compute its approximate frequency. Record all data in Data Table 1.

Fig. 21
Analysis of vibration data


f. Repeat Steps c, d, and e, releasing the ball from the same height (h) over points B and C on the table. Determine and record frequencies, as well as any significant difference in the amplitudes or number of frequencies identified.

g. Set the sweep rate at a lower value, so that the signal amplitude decays to about half its peak height in one sweep. Measure the maximum peak-to-peak height of the pattern. Convert to mV, and record as peak amplitude for that frequency.

h. Using a slower sweep, determine and record the time required for the signal to decay to half its peak amplitude.

i. Turn the mounting block on its side, so that the accelerometer top is parallel to the long side of the table. The accelerometer now will measure vibration in the Y-axis direction. Record data in Data Table 2.

j. Repeat the above procedure after moving the accelerometer to measure vibration in the X-axis direction. Record data in Data Table 3.

3. Determining effectiveness of table in damping floor-induced vibrations

a. Position the accelerometer as in Part 2, Step 1.

b. Locate a point on the floor, a few (2-3) meters from the table legs or base, preferably on a line bisecting the floor contact points or area, and perpendicular to either width or length of table support.

c. Set the vertical sensitivity of the scope to maximum. While you observe the pattern, drop the sandbag from a height that will cause a scope pattern of at least 2-cm peak-to-peak height. Note location of the drop point (X) and value of the drop height. Use these for subsequent drops.

d. Determine major resonances in pattern, their peak amplitudes and decay times (half-amplitude decay). Record for each frequency.

e. Relocate the accelerometer and block (accelerometer vertical) on the floor near the base at point B or C in Figure 22 or near the legs (point A) of the table support.

f. Measure and record accelerometer location. Reduce vertical sensitivity by a factor of 100 (0.5 V/cm). Drop the sandbag on the selected point (X) from the selected height while you watch the scope pattern. Adjust sensitivity to produce a pattern with peak-to- peak amplitude of 4 to 10 cm.

g. Identify and measure major resonances of floor, as you did for the tabletop. Convert peak amplitudes to millivolts. Record all data in a table that you prepare.

Fig. 22
Experimental arrangement


h. Compare floor and tabletop vibrations by computing the ratio of peak-to-peak amplitudes at both floor and table accelerometer locations. This ratio will be a measure of the effectiveness of the table in reducing vibrations of the frequencies observed.

4. If you have time, compare the effectiveness of three different tables in damping floor-induced vibrations. Repeat procedures outlined in Section 3. What’s your conclusion?


data.jpg (6254 bytes)

Data Table 1. Accelerometer Adjusted for Z Axis

  Height of drop ______________________

Type of drop ________________________


Frequency or


Time for
Half Peak



Drop point A        
Drop point B        
Drop point C        


Data Table 2. Accelerometer Adjusted for Y Axis

  Height of drop ______________________

Type of drop ________________________


Frequency or


Time for
Half Peak



Drop point A        
Drop point B        
Drop point C        


referenc.jpg (6229 bytes)

Ealing Electro-Optics Product Guide. Ealing Electro-Optics, Inc., 22 Pleasant Street, South Natick, MA 01760,617/655-6029.

Melles Griot Opt Guide 3. Melles Griot, 1770 Kettering Street, Irvine, CA 92714,714/261-5600

The Newport Corporation Catalog of Precision Laser/Optics Products. No 100. Newport Corporation (formerly NRC), 18235 Mount Baldy Circle, Fountain Valley, CA 92728-8020,714/963-9811.

Oriel Corporation Catalog Vol. I of Tables, Benches. Micropositioners, Optical Mounts. Oriel Corporation, 250 Long Beach Boulevard, Stratford, CT 06497-0872,203/377-8282.

Specification sheet and application data on accelerometers.


---Course Contents---

---Next Module---