Printer Friendly

Photoelectric sensors for industrial systems.

To select the proper photoelectric system for heavy-duty industrial applications, a variety of factors - such as the setting and the nature of the target - have to be considered.

Industrial control systems rely heavily on sensors to supply information on a target's presence or absence, size, location, speed of motion, and other factors. Design engineers can choose from a wide variety of contact devices, such as limit switches and mechanical feelers, as well as noncontact ones, which may be inductive, capacitive, or photoelectric.

Photoelectric sensors do not require physical contact; thus they neither remove energy from nor add it to the system they support. Mechanical sensors, when used to monitor products, can bend, move, or scratch the target and create defective units. In addition, they are generally limited in the speed of motion that can be handled and the size of the target. Such obstacles are overcome by substituting noncontact sensors, such as photoelectric ones.

There are five basic configurations for photoelectric sensors, each offering advantages for certain applications and each having its limitations. To evaluate which sensor serves a particular need, identification of the specific application requirement is not the only need; a number of factors relating to the target involved, the restrictions imposed by the installation conditions, and the environment where the activity will perform must also be considered.


A photoelectric system's basic elements include a pulse-modulated light-beam transmitter and a receiver arranged to accept the light signal. When the transmitted light beam is interrupted, the receiver's output is switched to sound an alarm, turn on a lamp, or trigger a control system.

A tiny gallium arsenide - semiconductor (or other semiconductor combination) light-emitting diode is the key component of the transmitter. Small reflecting dishes are included in the assembly to direct the light beam; lenses may be added to increase light output.

The five general configurations are through-beam, retroreflective, passive-reflective or proximity, convergent-beam or mark sensor, and optical-triangulation area-reflective. In a through-beam arrangement, a pulse-modulated light beam is transmitted from an emitter and directed at a separately housed receiver. With no target present, the beam passes to the receiver. When the beam is interrupted by a target, the receiver's output changes and produces an output pulse. This mode of operation is known as Dark-On.

In the retroreflective configuration, the emitter and receiver are housed in the same enclosure. When no target is present, the emitted light beam is bounced back to the receiver by means of a reflector placed behind the region where the target would appear. When the target does appear, the beam is blocked (similarly to the through-beam Dark-On mode), and the receiver output changes. A corner reflector is employed rather than a mirror because a mirror does not reflect light back to the receiver but directs it away at an angle equal to the angle of incidence. However, a corner reflector with a prism-type structure reflects light energy from the emitter to its adjacent receiver.

A diffuse-reflective system houses the emitter and receiver in the same package, as does the retroreflective sensor. The transmitted light will be directed back to the receiver not by a separate reflector but by striking the target. Thus, unlike the previous two arrangements, the transmitted beam does not reach the receiver if no target is present; only when the target appears will light energy arrive at the receiver and switch its output state. This is the Light-On mode.

Distances of up to 600 feet between emitter and receiver can be handled with through-beam systems; spaces as small as millimeters can be measured with fiber-optic proximity or diffuse reflectors. Through-beam is well suited for opaque and reflective objects, but for reliable performance target size should be greater than the diameter of the light beam. Since emitter and receiver are separated in through-beam systems, the cost of installation and wiring is relatively high; however, they are less susceptible to dust, dirt, and pollution than retroreflective or diffuse-reflective systems.

Simple and less costly to install, retroreflective sensors will operate up to 30 feet from their targets. They are not recommended for small-object detection, since the reflector required would have to be even smaller than the target, and thus fragile and difficult to clean. When shiny targets require detection, reflective polarized sensors are available.

The light energy available for diffuse-reflective or proximity sensors is dependent on target size, texture, and color; in addition, signal levels are further attenuated by light scattering and absorption. For this reason, high-gain amplifiers, with adjustable sensitivity controls, are generally included in the package. Proximity sensors are well suited for detecting transparent or translucent objects, such as empty or filled bottles. Target background is important, since a highly reflective surface can be confused with the intended target.


Popular autofocus cameras and area-reflective sensors both take advantage of optical-triangulation techniques. As shown in the figure on this page, output from a low-power LED is sent through a condensing lens to the target; some light energy is reflected back to the receiver's lens to produce a spot on a position-detector diode. Output voltage from a conventional photodiode is a function of the light intensity striking the p-n junction. In a position-detector diode, output voltage is determined by the point at which the target's reflected light beam produces a spot. Since this output is based on the light beam's angle of arrival rather than its intensity, objects of different colors - even multicolored ones - can be detected. Area-reflective sensors are designed to see objects at a specific preset distance and ignore all others. These systems are replacing the older convergent-beam sensors that relied on fixed lensing to set the range.

Triple-beam detection makes use of two position-detector receivers placed symmetrically on either side of the projected light beam. By averaging the two position-detector output signals, range measurement can be made with resolution to 0.001 inches. Other advantages include insensitivity to changes in color, ability to detect a wide range of objects from dark to clear, and reliable operation under extreme dirt and dust environments.


When tiny targets, limited space, and environmental hazards are involved, fiber-optic devices are often selected. Although they are more expensive than conventional devices, they offer substantial benefits and are often the only solution.

Just as electrical wires conduct electrical current, fiber-optic threads conduct beams of light. Basically, the optical fiber - whether plastic or glass - consists of a central core and a concentric sheath that acts as a reflecting mirror surface. As the light beam travels along the fiber, it strikes the sheath, is reflected back to the core, continues its travel, again bounces off the sheath to the core, and so forth; for a 1-foot run of fiber-optic cable, there may be as many as 500 reflections taking place. The range of fiber-optic cable is about 2 feet in through-beam applications, which is usually sufficient; for longer runs, lenses may be added to the tips of the fibers.

Glass fibers offer a number of advantages over plastic types: they are tougher, have less loss for a given length, and can operate at temperatures as high as 800 [degrees] F. Plastic fiber-optic sensors are cheaper, and can be cut and spliced more easily.


Hundreds of photoelectric sensors are offered to the design engineer. There are plenty to choose from, but what factors are important to consider in making a choice? Do you want to sense the presence or absence of objects, count them, check their specific location, or measure thickness or height? If you want measurement, to what resolution? It is important to identify limitations related to sensor size, mounting space, environmental restrictions, and available power sources.

The nature of the target is a critical element. What is it? Is there one target or a series of objects moving along that have to be identified and separated? If it's a single target, how big or small is it? A large shipping container or a hair-thin magnet wire? Is it opaque, clear, or translucent? Does it have any colors, and which ones? Is the object's body coarse or shiny? Will it reflect or absorb light energy, and to what extent? How will the target be seen by the sensor? Head-on for maximum exposure or at an oblique angle to permit only a brief exposure? In short, how difficult will it be for the sensor's light beam to detect the target, and how long will the target remain in view? When dealing with targets mounted on a rotating wheel or drum, don't be misled by small numbers. A small label rotating at a speed of 10 rpm mounted on a drum that is 12 inches in diameter will be traveling at more than 360 inches per minute.

Can the sensor be placed anywhere - above, below, or on either side of the target? If not, what are the restrictions? Against what background will the sensor be viewing the target? Are there dull or highly-reflective objects behind the target area? Will background colors blend with the target, making detection difficult?

The industrial environment where the system will be employed is also critical. Hostile gases, toxic fumes, oil and grime, and extreme temperatures can damage and destroy sensors and the monitoring processes involved. It is necessary to note all such factors, including provisions for fast replacement and sensor/reflector cleaning when dictated.

How slow or how fast is the target moving as it passes through the sensor's field of view? Speed of target motion relates to the length of time the target will be exposed to the sensor's beam. The faster the target motion, the more briefly the target will be seen by the sensor; thus, the possible need for a sensor with a fast response time. Response time is the time required for a sensor's output to appear after a light-to-dark or dark-to-light transition; these two parameters may not have the same values. For example, a sensor may have a 5-millisecond response for a light-to-dark change and a 6-millisecond time for dark-to-light response, for an 11-millisecond minimum response. Under ideal conditions, such a sensor could handle a little less than 100 targets per second (11 milliseconds is slightly slower than 1/100 seconds).

A common mistake that can raise system cost is simply to select the sensor with the fastest response time. But sensor response time is just one factor in the total system response. Solid-state direct-current output circuits contribute a small additional delay, while optically coupled field-effect-transistor output stages may add several milliseconds. Alternating-current circuits could add as much as 8 milliseconds, while electromechanical relays may add as much as 30 milliseconds or more.

Labor costs for mounting the sensors, aligning them for optimum operation, and running the necessary wiring add up. Costs for through-beam sensors can be high, since separate mounting is required for the emitter and receiver, with wiring running between both units. How much will it cost to pull wiring through conduits? If splices are involved, check whether junction boxes exist. Then there is the threat of potentially damaging electromagnetic radiation. It's cheap and fast to route sensor wiring within the bundle of cables in existing conduits. But unexpected bursts of high-current pulses from heavy-duty motor wiring adjacent to sensor wires can trigger false control system operation.


Kids enjoy sweet drinks and find straws useful to consume the product (and squirt their peers), so soft-drink manufacturers now include straws attached to the container. To thwart temper tantrums from tiny customers, it is essential to ensure that a straw comes with each drink. Machine-vision systems and laser optics can do the job, but a less expensive and more reliable alternative involves a lineup of fiber-optic sensors that informs a programmable-logic controller whether the straw is present or absent, and whether it is properly aligned on the container. Proper position can be monitored to within 0.025 inches. As each container passes through the sensor area, edge-detector outputs are produced and must appear in the proper sequence, or the item is rejected.

How about a 0.00175-inch-diameter magnet wire being pulled through a coil-winding machine? To detect wire breakage, a convergent visible-beam mark sensor can be used; such sensors are available in a wide selection of specific focal distances, making setup a simple task. The arrangement uses the Light-On mode, with no change in output from the receiver as long as the magnet wire reflects light back from the emitter to the receiver. The fiber-optic sensor is only 12 millimeters by 5.5 millimeters, and is coupled to a high-gain amplifier for reliable operation even under low light return conditions.

Many small components are supplied in tape-and-reel format. Parts count can be made using an area-reflective triple-beam arrangement. The sensor's sensitivity is adjusted so that the appearance of a relatively dark component on the reel activates the sensor to produce an output for the counter. When a component is missing from the reel, the white tape reflects light back to the receiver and a count is not added.

Imagine the disappointment of biting into a delicious sandwich cookie, only to discover the cream filling is missing. To avoid such a disaster, two triple-beam sensors are positioned over the assembly station where 700 cookies per minute are moving along. One of the two precisely focused beams monitors the height of the bare cookie while the other checks for the proper level of filling. If the filling is absent, the cookie is rejected.

These examples are representative of thousands of applications where photoelectric sensors are used in industry. There will always be special applications in which the expertise of the vendor may be helpful. Most applications accomplished with vendor assistance end up costing the customer a lot less than solutions evolved without the aid of vendor expertise.

Raymond Butow is a sensor product manager at Aromat Corp. in New Providence, N.J.
COPYRIGHT 1996 American Society of Mechanical Engineers
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1996 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Butow, Raymond
Publication:Mechanical Engineering-CIME
Date:Mar 1, 1996
Previous Article:Choosing the best mechanical motion components.
Next Article:Dropping to safety.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters