The properties of magnetostriction have received much attention lately with their recent application in torque sensors. One of the earliest uses of magnetostriction, however, has been in linear position measurement. Sensors based on this technology are nearly immune to the effects of electrical noise, vibration, and temperature. In addition, because they are non-contact devices, they have a long operating life. They also offer the longest stroke ranges of linear position sensors.
How these sensors use magnetostrictive properties, though, is slightly different from the techniques used in torque sensors. In position sensors, the Villari and Wiedemann effects play important roles in determining position.
The effects of the moment
Like other linear position sensors, magnetostrictive versions have one part that moves with the machine component or device while the other part stays stationary. Instead of physical contact and the resultant wear, though, interaction between the parts is achieved through a magnetic field.
Two magnetic fields detect the position of a movable object. A third changes the result of that detected position into a signal usable by a controller.
While the position magnet tracks the motion of an object, such as a machine tool or cylinder, it supplies an axial magnetic field – one of the two fields used in position detection. The second field comes from a stationary wire made of magnetostrictive material, which is known as the waveguide. That field is generated when the sensor electronics send a current through the wire. The current creates a uniform magnetic field along the wire length.
The Wiedemann effect occurs at the point where the axial field intersects the wire's field. The wire undergoes torsional strain due to the interaction of the magnetic fields.
Because the current is applied as a pulse of approximately 1 or 2 μsec duration, the torsional-strain pulse travels in the wire like a sonic wave. It travels at the speed of sound in the waveguide material, approximately 3,000 mps, and is linear, repeatable, and insensitive to temperature, depending on the alloy.
As the sensor electronics apply the current, they also start a timer. When the torsional-strain pulse is created, it travels outward from its point of origin toward both ends of the sensor. At one end, a device called a pickup detects the wave and stops the timer. The elapsed time indicates the distance between the position magnet and the pickup. At the other end, a device called a damp absorbs the strain pulse to prevent any interference by reflections from the end of the waveguide.
The pick up
Applying a magnetic field to magnetostrictive material causes stress which changes its physical properties. Conversely, applying a stress to the material changes its magnetic properties, such as permeability. This "reverse" magnetostriction is the Villari effect, which the sensor's pickup device uses to help change the position data into an electronic signal.
A small piece of magnetostrictive material, the "tape," is welded to the waveguide near one end. The tape passes through a coil and is magnetized by a small permanent magnet called the bias magnet. When a sonic wave propagates down the waveguide and then the tape, the tape's magnetic flux density changes due to the change in permeability (the Villari effect) and produces a voltage output pulse from the coil (the Faraday effect). The electronic circuitry detects the voltage pulse and converts it into the desired output, which can be dc voltage, current, PWM, or start-stop digital pulses. The output can transmit across the CANbus, Profibus, Serial Synchronous Interface, HART, and other communication protocols.
Continue to next page
Other magnetic benefits
Use of a magnetic field eliminates the need for sensor components to touch. This non-contact eliminates surface wear and prevents problems with dead spots that can occur with other sensors. In potentiometers, for example, the wiper slides along the surface of a resistive element. In time, the positions it frequently tracks will become dead spots because of the wear.
Control applications often have a slight dither at 60 Hz. Such constant motion will eventually create dead spots on a potentiometer, sometimes in a matter of days if a sensor tracks only one position. For example, at the 60 Hz rate, potentiometers with a life rating of 100 million cycles and sensing one position point will develop a dead spot in about 20 days. Normally, however, sensors track several positions along their range, so it takes a few months to form dead spots.
Lastly, magnetostrictive sensors are absolute rather than incremental devices, so position is accurately known at power on. Incremental sensors, such as a linear optical encoder, require a set reference before they can indicate a position change. They must move to this reference position on power up and whenever the count in memory is corrupted because of noise or other causes.
The automated edge
The automotive industry was looking for a linear position sensor that could accurately find position, operate for many cycles, and that would be stable over time and temperature. The automakers needed many of these sensors and they also wanted them low in cost. Automation techniques are meeting these needs. Automatically assembled position sensors are not as accurate as the hand assembled versions, but they are lower in cost. As volume increases and cost continues to drop, these sensors will likely fit many applications that were previously out of reach.
Dave Nyce is Director of R&D at MTS Systems Corp., Cary, N.C.