May 1, 2000, Contributing Author, Electronic Design

When an application calls for detecting a metallic target that falls within an inch of the sensing surface, inductive proximity sensors are apt for the task. First introduced in the early 1960s, these durable components have proven their mettle in the sensing arena. In fact, they're the best-selling sensing technology in the world. Their immunity to dust and dirt buildup suits them well for harsh industrial environments. Additionally, the standardized physical and electrical characteristics of the general-purpose, cylindrical types of these sensors simplify their use.

Naturally, designers make some common mistakes when applying these devices. Knowledge in several key areas, though, can help careful users avoid these pitfalls. Successful object detection requires an understanding of the fundamentals of sensor design. The criteria for choosing between the various styles of inductive proximity sensors also must be kept in mind. And, the significance of key sensor specifications and the effect of mounting restrictions on sensor implementation should be recognized.

An inductive proximity sensor has four components: the coil, oscillator, detection circuit, and output circuit. The target material, environment, and mounting restrictions all have an influence on these items and on the senor's operation, magnetic nature, and shielding. The oscillator generates a fluctuating, doughnut-shaped magnetic field around the winding of the coil, which is located in the device's sensing face.

When a metal object moves into the sensor's field of detection, Eddy currents build up in the object, magnetically push back, and finally dampen the sensor's own oscillation field. The sensor's detection circuit monitors the amplitude of the oscillation and, when it becomes sufficiently damped, triggers the output circuitry (Fig. 1).

Inductive proximity sensors come in shielded and unshielded versions. Without any shielding, the doughnut-shaped magnetic field generated by the sensor's coil is unrestricted. As a result, the sensor will be triggered when any metal object comes from behind, along side, or in front of the device. In a shielded sensor, a ferrite core directs the coil's magnetic field to radiate only from the sensor's detection face. Even unshielded inductive proximity sensors have peeled-back ferrite-core shielding, which gives them a longer sensing distance than the shielded versions. At the same time, this feature prevents false readings caused by objects behind the detection face.

There are five categories of inductive proximity sensors: cylindrical, rectangular, miniature, harsh environment, and special purpose. Cylindrical threaded-barrel sensors account for 70% of all inductive proximity sensor purchases. Years ago, this style's behavior was standardized by the CENELEC organization, which determined characteristics such as body size, sensing distances, and output levels. It's easy to understand why a designer would automatically select this general-purpose sensor, since it would be the right choice 70% of the time.

Yet experience has shown that there are many proximity-sensing applications where one of the other, specialized sensors can provide a better solution. Designers who automatically specify a general-purpose sensor may encounter problems that would vanish if another style were selected. Target material, environment, and mounting restrictions should guide the choice of sensor style.

In the world of inductive proximity sensors, not all metals are created equally. The familiar specification in technical data sheets refers to a "standard detectable object" made of an iron (ferrous) material. Other metallic materials, such as stainless steel, brass, aluminum, and copper, have different influences over the inductive effect. They're usually less detectable than iron, too.

Designers should determine two things. First, is the target material made out of iron or another metal? Second, is it possible for the target material to change in the application's future runs? To calculate the sensing distance of nonferrous metals, multiply the standard sensing distance by a reduction factor. Typically, this value is 0.8 for stainless steel, 0.5 for brass, 0.4 for aluminum, and 0.3 for copper.

A full-line sensor supplier will have a sensor solution for the detection of troublesome metallic materials. These special inductive proximity sensors are known as "nonferrous sensing" or "all-metal sensing." Nonferrous sensors will detect metals such as aluminum better than they sense iron, while all-metal sensors will pick up on all kinds of metal at the same sensing distance.

What separates the nonferrous and all-metal sensors from general-purpose inductive proximity sensors is the number of separate inductive coils included in the proximity-sensor head. The nonferrous and all-metal types contain two or three separate coils in the sensor head, while the general-purpose sensor has only one. Consequently, the nonferrous and all-metal sensing styles tend to be larger and more expensive than their general-purpose counterparts.

Environmental conditions can significantly affect the sensor. Extreme temperatures will reduce its operating life, causing premature failure. Hot temperatures will make it more sensitive, while cold temperatures will lower its resistance to shock. Nevertheless, a full-line sensor supplier can offer solutions to specific environmental conditions.

In certain applications, metallic "chips" or filings accumulate on the sensor's side or face. To account for this, some modern inductive proximity sensors contain embedded microprocessors that detect the slow buildup of these chips over time and teach the sensor to ignore their effects. These sensors are "chip immune." The flat-pack proximity sensor also resists the effects of chip buildup. With its slim profile, it's virtually unaffected by chip buildup when its sensing face is vertically exposed.

Sensors may be exposed to cutting fluids or chemicals for prolonged periods of time as well. This can cause traditional inductive proximity sensors to become brittle and crack, shortening their lifetimes. In such cases, designers must again turn to a specialized model. Proximity sensors dipped, coated, or shot from Teflon suffer no ill effects from the material in terms of performance or reliability. Teflon's added cost can be justified by the material's stability in the presence of cutting oils and corrosive chemicals. It also prevents weld slag buildup.

High-temperature environments pose another challenge. Inductive proximity sensors generally are self-contained devices that include their silicon amplifiers and detection circuitry inside the sensor-head housing. Self-contained proximity sensors are practical for most applications until environmental conditions begin to exceed the standard operating parameters for a silicon-based circuit. Normally, silicon-based circuitry operates between ?25°C and 70°C.
Separate Amplifiers May Be Needed

Under any temperature conditions beyond this range, the circuitry becomes prone to operating failure. Designers should then look for inductive proximity sensors that use separate amplifiers. Their sensor head contains the inductive coil and little else. The amplifier and detection circuitry can be located safely away in a remote, environmentally controlled area. Such sensors can resist temperatures as high as 200°C.

Inductive proximity sensors are strong representatives of the last decade's microelectronics revolution. Today, it's possible to manufacture a rectangular proximity sensor as small as 5.5 by 5.5 by 19 mm with an extended sensing range of 1.6 mm (Fig. 2). Advances in sensor miniaturization also result from the development of the separate in-line amplifier types. These sensors come with sensing heads as small as 3 mm in diameter and robotic cabling that enables the sensor head to move if necessary.

In some instances, space constraints prohibit the use of an inductive proximity sensor with a traditional cylindrical body. Fortunately, a wide variety of packages are available. Rectangular-shaped versions range from the subminiature (5.5 by 5.5 by 19 mm) to the flat-pack style (25 by 10 by 50 mm), all the way up to the limit-switch housing size (40 by 40 by 115 mm). A sensor in a limit-switch housing will vastly outlive a typical limit switch, which has mechanical contacts. A limit switch has a life expectancy of about 300,000 cycles, while a similarly shaped sensor in limit-switch housing can last up to 100,000 hours.

If a specialized sensor isn't required, designers can reliably fall back on the proven success of the traditional cylindrical type. But before a particular device is specified, it's important to investigate several areas to ensure a long-lasting and well-manufactured sensor.

A strong enclosure is crucial. The thicker the barrel housing in a cylindrical proximity sensor, the less likely it is to break because of overzealous installation techniques or incidental object collision. Keep in mind, however, that housing thickness varies from manufacturer to manufacturer.

Also, check the sensor to see if it's vacuum potted. Most proximity sensors are potted, but poor potting is almost worse than no potting at all. Air bubbles can be trapped inside poorly potted sensors. These bubbles cause undue stressing, which may lead to pc-board cracking and failure.

The cable must have proper strain relief, too. An inductive proximity sensor with a cable that protrudes directly out of the potting material is susceptible to breakage at the junction between the potting material and the cable. A proximity-sensor cable with this design also has a much weaker pull force. Strong, flexible strain relief can provide a sensor with a long life.

Even though they're used outside the sensing industry, certain terms have unique definitions when they're applied to inductive proximity sensors. Designers should understand what these terms mean before specifying a particular device.

When an inductive proximity sensor's data sheet refers to a standard detectable object, it describes the specified shape, size, and material that's used as the standard for examining the sensor's performance. This definition is important because the sensor's detection distance differs according to the shape and material of the object being detected. Generally, the standard detectable object will be an iron plate with a thickness of 1 mm and a height and width equal to the inductive sensor's diameter.

Detection distance is the position at which the inductive proximity sensor is triggered when a standard detectable object is moved in front of it in a defined manner. To determine this distance for a sensor with an end (or "front") detection surface, the sensor's center line is aligned with the standard detectable object's center line. Then, the object is moved toward the sensor's face until the sensor changes output states.

Detection distance is influenced both by the conductivity and the thickness of the target material. Highly conductive materials make poor targets for traditional inductive proximity sensors. Thick materials are harder to detect than thin ones. Both factors relate to the generation of Eddy currents in the target. A conductive material disperses Eddy currents, so the target becomes harder to detect. But a thin material, with its reduced ability to move current, causes a buildup of Eddy currents. This makes it detectable at greater distances.

The reset distance is the distance at which the inductive proximity sensor releases its output when the standard detectable object is removed from its field of detection. The difference between the detection distance and the reset distance is called the distance differential. Typically 3% to 10% of the overall detection distance, the distance differential is incorporated into the sensor's design to prevent its output from "chattering" (switching on and off erratically) due to noisy environments or detectable object vibrations (Fig. 3).

Today's quality inductive proximity sensors can have trigger points that are repeatable to 0.0001 in. To obtain such precision, though, the detectable object must be moved the reset distance away from the sensor after each time the sensor is triggered.

The setting distance describes the distance at which the inductive proximity sensor will trigger an output with the standard detection object, even if the detection distance has decreased due to temperature or voltage fluctuations. Of course, not every design will have the luxury of sensing the standard detection object described in the sensor data sheet.

The detection distance for an irregular object cannot be estimated from the manufacturer's data. Instead, it must be measured with a sample object. To do so, take the object in question and move it toward the sensor until the output changes state. The result is the detection distance for that particular combination of target object and inductive proximity sensor.

The setting distance for the target object can then be calculated by the following formula: new setting distance = (detection distance obtained by test with target object) × (setting distance of the standard detectable object)/(standard detection distance of the standard detectable object).

Sensor mounting requirements must be considered when the inductive proximity sensor is implemented into the design. Otherwise, there may be a reduced sensing distance, false triggering, or target non-detection. It's important to consider the effects of the mounting hardware itself as well as other metallic objects located near the sensor.

The device may be embedded into a metallic mounting fixture up to the point where the shielded sensor's face is flush with the mounting surface. This embedded mount protects the sensor from mechanical damage due to incidental contact with the target object. Even so, shielded sensors shouldn't be recessed into a metal mounting surface. Objects, materials, or opposing surfaces that aren't supposed to be detection objects should remain clear of the inductive sensor's face by a factor of three times the sensor's standard detection distance.

Unshielded sensors cannot be completely embedded into a metallic-mounting fixture. Because of their extended sensing distance, they're susceptible to the influences of surrounding metals. Designers, then, have to obey the factor-of-three rule for shielded types. The sensor must be surrounded by a metal-free area. This area must be equal to the sensor's size (or diameter, in the case of a cylindrical proximity sensor). It also must stretch in every direction, with a depth clearance of two times the sensor's standard detection distance (Fig. 4). Failure to meet clearance requirements can lead to false detection or reduced sensing distances.

When multiple inductive proximity sensors are mounted in close proximity to one another, either side by side or in opposing directions, the sensors can be subject to an effect called mutual interference. If one proximity sensor's field couples with the detection coil field of another, an inductance may generate a beat frequency in one or both of the sensors. This, in turn, causes the output of the proximity sensor to chatter.

Mutual interference problems can be insidious, due to their erratic nature. When inductive sensors are mounted side by side at distances closer than the sensor manufacturer's mutual-interference specifications, they may perform seamlessly. Then, they may suddenly display signs of chattering and false detection.

Separation distance specifications for sensors mounted side by side can vary according to sensor body type and from manufacturer to manufacturer. Always examine and adhere to the manufacturer's specification distances for mounting inductive proximity sensors to avoid potential mutual-interference.

Several options must be considered if the application and sensing require the inductive proximity sensors to be mounted close together. Shielded types allow for closer mounting. So do miniature inductive sensors, whose smaller size means decreased sensing distances and a smaller likelihood of mutual interference.

Some sensor manufacturers offer alternative frequency types. These sensors oscillate their magnetic coils at different cycle rates than corresponding standard inductive proximity sensors, preventing the inductive coupling that leads to output chattering.

Finally, if close sensor mounting cannot be avoided, the sensors can be multiplexed. Switching alternate sensors on and off and taking alternate reads can be a quick solution to a mutual-interference problem, provided that the application accounts for the corresponding reduction in sensor response time.

Equipped with an understanding of sensor operation, available sensor options, and the application's environmental conditions, designers can select the inductive proximity sensor that best fits their needs while delivering optimum performance.