Many types of industrial processing equipment use sensors to provide information needed to monitor and control processes. This allows design parameters to be maintained or modified in order to manufacture products at the desired level of quality and throughput. Typical sensors measure variables such as temperature, pressure, flow, force, or position. The type of measurement as well as the sensor technology dictates the set of criteria that is important in specifying and sizing the appropriate sensor.
Linear position sensors measure absolute distance along a motion axis. They are available in several technologies, each having its own advantages and disadvantages. This article presents information on the application of magnetostrictive linear position sensors, which are gaining popularity due to their accuracy and reliability. A comparison of magnetostrictive sensors to other linear position sensors is also included.
Theory of operation
A magnetostrictive position sensor measures distance between a position magnet and the head end of the sensing rod. The position magnet does not touch the sensing rod, so there are no parts to wear out.
The sensing rod is mounted along the motion axis to be measured, and the position magnet is attached to the member that will be moving. The head includes an electronics module, which reports position information to a controller (or other receiving device) in the appropriate analog or digital format.
As shown in figure 1, a magnetostrictive position sensor includes five basic components — position magnet, waveguide, pickup, damp, and electronics module. A protective tube usually covers the waveguide.
The position magnet is a permanent magnet, often made in the shape of a ring, which travels along the sensing rod. The waveguide is housed within the sensing rod, and is a small diameter (approximately 0.30 to 0.80 mm) tubing or wire made from a magnetostrictive material.
Magnetostriction is a property of certain materials, including iron, nickel, cobalt, and some of their alloys, in which application of a magnetic field causes strain that results in a change in the size or shape of the material.
The waveguide is so named because a sonic wave travels in it during operation of the sensor. The sonic wave is generated by interaction between the magnetic field from the position magnet and a second magnetic field generated in the waveguide by the application of a current pulse (called the interrogation pulse) through the waveguide from the electronics module. The vector sum of the magnetostrictive strain from the two magnetic fields results in the generation of a torsional strain wave in the waveguide at the position magnet’s location, as shown in figure 1.
The strain wave travels in the waveguide, toward the head end, at about 2,850 m/s. At the head, a pickup device senses the arrival of the strain wave (called the return pulse). Another strain wave travels from the position magnet in the direction away from the head. This unused wave is eliminated by the damp in order to prevent interference from waves that would otherwise be reflected from the waveguide tip.
The electronics module applies the interrogation pulse to the waveguide and starts an electronic timer. After a time delay, which is proportional to the distance between the position magnet and the pickup, the electronics module receives the return pulse from the pickup and stops the timer. The magnitude of the time delay indicates the location of the position magnet. For example, at a measured distance of one meter with a waveguide velocity of 2,850 m/s, the time delay would be:
1 m ÷ 2,850 m/s = 0.35 ms
The electronics module then uses the time measurement to produce the desired output. Output can be a logic level pulse-width, an analog voltage or current, or a standard digital interface.
The interrogation rate can be controlled from an external controller, or can be internally generated at a rate anywhere from one to more than 4,000 times per second. This is the update rate, and is the frequency at which new position information becomes available at the sensor output. The maximum update rate depends on the waveguide length; a shorter waveguide allows a faster update rate to be used.
Selecting the appropriate type and size
Linear magnetostrictive position sensors are available in several housing configurations to enable mounting in a wide range of applications. Two hydraulic cylinder mount styles include the standard mounting of figure 1, and the two-piece version of figure 2. They both have rod and flange designs, which are capable of withstanding and sealing the high cylinder pressures. The high pressure mounting thread can be specified in English or metric units. Modular design allows replacement of the sensor cartridge without breaking the pressure seal to the cylinder.
The configuration of figure 2 is designed for installation in space-restricted clevis-type cylinders where the sensing element is separated from the electronics module by an interconnect cable.
Another popular way to mount a linear position sensor is by bolting its base to the machine frame, using a profile style housing. Examples of profile housings are shown in figure 3. Here, the sensing rod is enclosed within an aluminum extrusion that provides the mounting base for the sensor. The position magnet can be a bar magnet (“floating magnet”) passing along near the top of the extrusion (figure 3a), or it may be captured inside of a shuttle (“sliding magnet”) that rides in a rail as part of the extrusion (figure 3b).
A clevis mounting system is also available, as shown in figure 4. This is similar to the profile housings of figure 3, but the position magnet is moved via a metal rod, with a clevis on the rod end and also on the opposite end of the housing. The sensor housing is supported through the clevis mounts.
When determining the proper size of a magnetostrictive position sensor to order for a particular application, it is important to consider the length and alignment criteria of the sensing rod and position magnet. There is a minimum distance allowable between the head end of the sensor rod and the position magnet. This is to prevent interaction of the position magnet with the pickup, and is called the null. The specified length of the null depends on the mounting configuration of the sensor. In figure 5, it is 50.8 mm; so, the motion system and sensor mounting alignment must be designed so that the front face of the position magnet will be no closer to the mounting flange of the sensor than 50.8 mm. The front face of the position magnet is the face closest to the sensor head.
At the sensor rod tip (the end opposite the head), there is an unusable area in which the damp is housed. This is called the dead zone. Like the null, the system must be designed so that the front face of the position magnet will come no closer to the tip than the specified dead zone distance. In figure 5, the dead zone is 63.5 mm.
For example, when ordering a model with the dimensions shown in figure 5: If the motion axis has a travel of 2 m, then a sensor with a stroke length of 2 m should be ordered. The total length of the rod, from the flange face (at the head) to the rod tip, will be:
2 m + 50.8 mm + 63.5 mm = 2.114 m
The standard power for industrial sensors is 24 Vdc, but some older systems use 15 Vdc. Typical industrial linear magnetostrictive position sensors operate over a power supply voltage range of 13.5 to 26.5 Vdc.
The signal from the transducer, as measured by the electronics module, is a time delay. This is shaped into a digital pulse when the sensor is specified with a start-stop interface. In operation, the user supplies a digital pulse to request a reading (starting a timer at the same time), and the sensor returns a stop pulse. The time between the two pulses indicates the position magnet’s location.
A second type of output signal interface is pulse-width modulation (PWM). Here, the sensor internally generates its own start pulses. The user reads a pulse width that varies with the reading.
Analog current or analog voltage outputs are common interfaces. The signal can be 0 to 20 mA, 4 to 20 mA, or 0 to 10 V. Automotive sensors with a 5 V supply have a 10 to 90% ratiometric output.
Several standard digital communication interfaces are also used including CANbus (several popular versions are supported), SSI (serial synchronous interface), Profibus, and Interbus-S.
Controlling of gap between rollers:
Figure 6 is a pictorial of a sensing and control system for maintaining a specified roller gap. Sensors are mounted along the roller adjustment axis, with position magnets mounted at each end of the movable rollers. The controller accepts the sensor signals and sends the control signal to the servo motors. When using the magnetostrictive model, the external controller can be eliminated, since this type of magnetostrictive linear position sensor includes the controller function within the sensor head.
An advantage of the magnetostrictive sensor, over other types of linear position sensors, is the ability to read the position magnet even when there is a barrier between position magnet and sensing rod. For example, the barrier can be the cylinder wall when the position magnet is part of a piston, or a transmission case when measuring gear position, etc. This is possible whenever the material directly between the position magnet and rod can be a non-magnetic material. Common materials for this duty include plastics, ceramics, aluminum and nonferrous metals, and some stainless steels.
Another advantage unique to magnetostrictive position sensors is the ability to measure multiple magnets while using one sensing rod. This allows the generation of more than one measurement by simply adding position magnets.
Some sensor models accept more than 20 position magnets. In an injection molding machine, for example, the injector motion, mold closing, and ejector can be measured using only one sensing rod. Or, a slitting machine can measure the positions of all of the knives using only one sensing rod and adding a position magnet for each knife.
Comparison of technologies
In addition to magnetostriction, several other technologies are used for measuring linear position.
When “designing in” a linear position sensor, many factors must be considered. Proper attention must be paid to matching the sensor to the application requirements regarding power input, signal output, housing style, mounting configuration, sensing stroke, and the ability of the sensing technology to make the measurement under the application conditions.
Questions or comments? Contact the editor, Frances Richards, at: [email protected]