Home | Forum | DAQ Fundamentals | DAQ Hardware | DAQ Software
Input Devices | Data Loggers + Recorders | Books | Links + Resources
We’re now entering the second part of the guide, where we look in detail at the range of sensors available for measuring various physical quantities. As we study these sensors, we quickly come to realize that a wide range of different physical principles are involved in their operation. It also becomes apparent that the physical principles on which they operate is often an important factor in choosing a sensor for a given application, as a sensor using a particular principle may perform much better than one using a different principle in given operating conditions. It’s therefore prudent to devote this section to a study of the various physical principles exploited in measurement sensors before going on to the separate sections devoted to the measurement of various physical quantities. The physical principles to be examined are capacitance change, resistance change, magnetic phenomena (inductance, reluctance, and eddy currents), Hall effect, properties of piezoelectric materials, resistance change in stretched/ strained wires (strain gauges), properties of piezoresistive materials, light transmission (both along an air path and along a fiber-optic cable), properties of ultrasound, transmission of radiation, and properties of micromachined structures (microsensors). It should be noted that the chosen order of presentation of these is arbitrary and does not imply anything about the relative popularity of these various principles. It must also be pointed out that the list of technologies covered in this section is not a full list of all the technologies that are used in sensors, but rather a list of technologies common to several different sensors that measure different physical quantities. Many other technologies are used in the measurement of single physical quantities. Temperature measurement is a good example of this, as several of the sensors used are based on technologies not covered in this section.
Capacitive sensors consist of two parallel metal plates in which the dielectric between the plates is either air or some other medium. The capacitance C is given by C=E_a E_r A/d, where E_a is the absolute permittivity, E_r is the relative permittivity of the dielectric medium between the plates, A is the area of the plates, and d is the distance between them. Two forms of capacitive devices exist, which differ according to whether the distance between the plates is fixed or not.
Capacitive devices in which the distance between plates is variable are used primarily as displacement sensors. Motion of the moveable capacitive plate relative to a fixed one changes the capacitance. Such devices can be used directly as a displacement sensor by applying the motion to be measured to the moveable capacitor plate. Capacitive displacement sensors commonly form part of instruments measuring pressure, sound, or acceleration, as explained in later sections.
In the alternative form of capacitor, the distance between plates is fixed. Variation in capacitance is achieved by changing the dielectric constant of the material between the plates in some way. One application is where the dielectric medium is air and the device is used as a humidity sensor by measuring the moisture content of the air. Another common application is as a liquid level sensor, where the dielectric is part air and part liquid according to the level of the liquid that the device is inserted in. Both of these applications are discussed in greater detail in later sections. This principle is used in devices to measure moisture content, humidity values, and liquid level, as discussed in later sections.
++ Resistive Sensors
Resistive sensors rely on variation of the resistance of a material when the measured variable is applied to it. This principle is applied most commonly in temperature measurement using resistance thermometers or thermistors. It’s also used in displacement measurement using strain gauges or piezoresistive sensors. In addition, some moisture meters work on the resistance-variation principle. All of these applications are considered further in later sections.
Magnetic sensors utilize the magnetic phenomena of inductance, reluctance, and eddy currents to indicate the value of the measured quantity, which is usually some form of displacement.
Inductive sensors translate movement into a change in the mutual inductance between magnetically coupled parts. One example of this is the inductive displacement transducer. In this, the single winding on the central limb of an "E"-shaped ferromagnetic body is excited with an alternating voltage. The displacement to be measured is applied to a ferromagnetic plate in close proximity to the "E" piece. Movements of the plate alter the flux paths and hence cause a change in the current flowing in the winding. By Ohm's law, the current flowing in the winding is given by I=V/omega L. For fixed values of w and V, this equation becomes I=1/KL, where K is a constant. The relationship between L and the displacement, d, applied to the plate is a nonlinear one, and hence the output-current/ displacement characteristic has to be calibrated.
The inductance principle is also used in differential transformers to measure translational and rotational displacements.
In variable reluctance sensors, a coil is wound on a permanent magnet rather than on an iron core as in variable inductance sensors. Such devices are used commonly to measure rotational velocities. --- shows a typical instrument in which a ferromagnetic gearwheel is placed next to the sensor. As the tip of each tooth on the gearwheel moves toward and away from the pick-up unit, the changing magnetic flux in the pickup coil causes a voltage to be induced in the coil whose magnitude is proportional to the rate of change of flux. Thus, the output is a sequence of positive and negative pulses whose frequency is proportional to the rotational velocity of the gearwheel.
Eddy current sensors consist of a probe containing a coil that is excited at a high frequency, which is typically 1 MHz. This is used to measure the displacement of the probe relative to a moving metal target. Because of the high frequency of excitation, eddy currents are induced only in the surface of the target, and the current magnitude reduces to almost zero a short distance inside the target. This allows the sensor to work with very thin targets, such as the steel diaphragm of a pressure sensor. The eddy currents alter the inductance of the probe coil, and this change can be translated into a d.c. voltage output that is proportional to the distance between the probe and the target. Measurement resolution as high as 0.1 mm can be achieved. The sensor can also work with a nonconductive target if a piece of aluminum tape is fastened to it.
Basically, a Hall-effect sensor is a device used to measure the magnitude of a magnetic field.
It consists of a conductor carrying a current that is aligned orthogonally with the magnetic field. This produces a transverse voltage difference across the device that is directly proportional to the magnetic field strength. For an excitation current, I, and magnetic field strength, B, the output voltage is given by V = K ● I ● B, where K is known as the Hall constant.
The conductor in Hall-effect sensors is usually made from a semiconductor material as opposed to a metal because a larger voltage output is produced for a magnetic field of a given size. In one common use of the device as a proximity sensor, the magnetic field is provided by a permanent magnet built into the device. The magnitude of this field changes when the device comes close to any ferrous metal object or boundary. The Hall effect is also used commonly in computer keyboard push buttons. When a button is depressed, a magnet attached underneath the button moves past a Hall-effect sensor. This generates an induced voltage in the sensor, which is converted by a trigger circuit into a digital output. Such push-button switches can operate at high frequencies without contact bounce.
Piezoelectric transducers produce an output voltage when a force is applied to them. They can also operate in the reverse mode where an applied voltage produces an output force. They are used frequently as ultrasonic transmitters and receivers. They are also used as displacement transducers, particularly as part of devices measuring acceleration, force, and pressure. In ultrasonic receivers, sinusoidal amplitude variations in the ultrasound wave received are translated into sinusoidal changes in the amplitude of the force applied to the piezoelectric transducer. In a similar way, the translational movement in a displacement transducer is caused by mechanical means to apply a force to the piezoelectric transducer. Piezoelectric transducers are made from piezoelectric materials. These have an asymmetrical lattice of molecules that distorts when a mechanical force is applied to it. This distortion causes a reorientation of electric charges within the material, resulting in a relative displacement of positive and negative charges. The charge displacement induces surface charges on the material of opposite polarity between the two sides. By implanting electrodes into the surface of the material, these surface charges can be measured as an output voltage. For a rectangular block of material, the induced voltage is given by …
V = kFd/A ,
where F is the applied force in g, A is the area of the material in mm, d is the thickness of the material, and k is the piezoelectric constant. The polarity of the induced voltage depends on whether the material is compressed or stretched.
The input impedance of the instrument used to measure the induced voltage must be chosen carefully. Connection of the measuring instrument provides a path for the induced charge to leak away. Hence, the input impedance of the instrument must be very high, particularly where static or slowly varying displacements are being measured.
Materials exhibiting piezoelectric behavior include natural ones such as quartz, synthetic ones such as lithium sulphate, and ferroelectric ceramics such as barium titanate. The piezoelectric constant varies widely between different materials. Typical values of k are 2.3 for quartz and 140 for barium titanate. Applying Equation (1) foraforceof1 g applied to a crystal of area 100 mm2 and a thickness of 1 mm gives an output of 23 mV for quartz and 1.4 mV for barium titanate.
Certain polymeric films such as polyvinylidene also exhibit piezoelectric properties. These have a higher voltage output than most crystals and are very useful in many applications where displacement needs to be translated into voltage. However, they have very limited mechanical strength and are unsuitable for applications where resonance might be generated in the material.
The piezoelectric principle is invertible, and therefore distortion in a piezoelectric material can be caused by applying a voltage to it. This is used commonly in ultrasonic transmitters, where application of a sinusoidal voltage at a frequency in the ultrasound range causes sinusoidal variations in the thickness of the material and results in a sound wave being emitted at the chosen frequency. This is considered further in the section on ultrasonic transducers.
Strain gauges are devices that experience a change in resistance when they are stretched or strained. They are able to detect very small displacements, usually in the range of 0_50 mm, and are typically used as part of other transducers, e.g., diaphragm pressure sensors that convert pressure changes into small displacements of the diaphragm. Measurement inaccuracies as low as _0.15% of full-scale reading are achievable, and the quoted life expectancy is usually three million reversals. Strain gauges are manufactured to various nominal values of resistance, of which 120, 350, and 1000 Ohm are very common. The typical maximum change of resistance in a 120-Ohm device would be 5 Ohm at maximum deflection.
The traditional type of strain gauge consists of a length of metal resistance wire formed into a zigzag pattern and mounted onto a flexible backing sheet. The wire is nominally of circular cross section. As strain is applied to the gauge, the shape of the cross section of the resistance wire distorts, changing the cross-sectional area. As the resistance of the wire per unit length is inversely proportional to the cross-sectional area, there is a consequential change in resistance. The input-output relationship of a strain gauge is expressed by the gauge factor, which is defined as the change in resistance (R) for a given value of strain (S), that is, gauge factor = dR/dS.
In recent years, wire-type gauges have largely been replaced either by metal-foil types or by semiconductor types. Metal-foil types are very similar to metal wire types except that the active element consists of a piece of metal foil cut into a zigzag pattern. Cutting a foil into the required shape is much easier than forming a piece of resistance wire into the required shape, which makes the devices less expensive to manufacture. A popular material in metal strain gauge manufacture is a copper-nickel-manganese alloy, which is known by the trade name of "Advance." Semiconductor types have piezoresistive elements, which are considered in greater detail in the next section. Compared with metal gauges, semiconductor types have a much superior gauge factor (up to 100 times better) but are more expensive. Also, while metal gauges have an almost zero temperature coefficient, semiconductor types have a relatively high temperature coefficient.
In use, strain gauges are bonded to the object whose displacement is to be measured. The process of bonding presents a certain amount of difficulty, particularly for semiconductor types.
The resistance of the gauge is usually measured by a d.c. bridge circuit, and the displacement is inferred from the bridge output measured. The maximum current that can be allowed to flow in a strain gauge is in the region of 5 to 50 mA depending on the type. Thus, the maximum voltage that can be applied is limited and, consequently, as the resistance change in a strain gauge is typically small, the bridge output voltage is also small and amplification has to be carried out. This adds to the cost of using strain gauges.
A piezoresistive sensor is made from semiconductor material in which a p-type region has been diffused into an n-type base. The resistance of this varies greatly when the sensor is compressed or stretched. This is used frequently as a strain gauge, where it produces a significantly higher gauge factor than that given by metal wire or foil gauges. Also, measurement uncertainty can be reduced down to _0.1%. It’s also used in semiconductor-diaphragm pressure sensors and in semiconductor accelerometers.
It should also be mentioned that the term piezoresistive sensor is sometimes used to describe all types of strain gauges, including metal types. However, this is incorrect as only about 10% of the output from a metal strain gauge is generated by piezoresistive effects, with the remainder arising out of the dimensional cross-sectional change in the wire or foil.
Proper piezoelectric strain gauges, which are alternatively known as semiconductor strain gauges, produce most (about 90%) of their output through piezoresistive effects, and only a small proportion of the output is due to dimensional changes in the sensor.
Optical sensors are based on the transmission of light between a light source and a light detector. The transmitted light can travel along either an air path or a fiber-optic cable. Either form of transmission gives immunity to electromagnetically induced noise and also provides greater safety than electrical sensors when used in hazardous environments.
--- Light source Light detector Transmission path (air or fiber optic) Modulation of transmission characteristics by measured variable
+++ Optical Sensors (Air Path)
Air path optical sensors are used commonly to measure proximity, translational motion, rotational motion, and gas concentration. These uses are discussed in more detail in later sections. A number of different types of light sources and light detectors are used.
Light sources suitable for transmission across an air path include tungsten-filament lamps, laser diodes, and light-emitting diodes (LEDs). However, as the light from tungsten lamps is usually in the visible part of the light frequency spectrum, it’s prone to interference from the sun and other sources. Hence, infrared LEDs or infrared laser diodes are usually preferred.
These emit light in a narrow frequency band in the infrared region and are not affected by sunlight.
The main forms of light detectors used with optical systems are photoconductors (photoresistors), photovoltaic devices (photocells), phototransistors, and photodiodes.
Photoconductive devices are sometimes known by the alternative name of photoresistors.
They convert changes in incident light into changes in resistance, with resistance reducing according to the intensity of light to which they are exposed. They are made from various materials, such as cadmium sulfide, lead sulfide, and indium antimonide.
Photovoltaic devices are often called photocells. They also are known commonly as solar cells when a number of them are used in an array as a means of generating energy from sunlight. They are made from various types of semiconductor material. Their basic mode of operation is to generate an output voltage whose magnitude is a function of the magnitude of the incident light that they are exposed to.
Photodiodes are devices where the output current is a function of the amount of incident light.
Again, they are made from various types of semiconductor material.
A phototransistor is effectively a standard bipolar transistor with a transparent case that allows light to reach its base-collector junction. It has an output in the form of an electrical current and could be regarded as a photodiode with an internal gain. This gain makes it more sensitive to light than a photodiode, particularly in the infrared region, but has a slower response time. It’s an ideal partner for infrared LED and laser diode light sources.
+++ Optical Sensors (Fiber Optic)
Instead of using air as the transmission medium, optical sensors can use fiber-optic cable to transmit light between a source and a detector. Fiber-optic cables can be made from plastic fibers, glass fibers, or a combination of the two, although it’s now rare to find cables made only from glass fibers as these are very fragile. Cables made entirely from plastic fibers have particular advantages for sensor applications because they are inexpensive and have a relatively large diameter of 0.5_1.0 mm, making connection to the transmitter and receiver easy.
However, plastic cables cannot be used in certain hostile environments where they would be severely damaged. The cost of fiber-optic cable itself is insignificant for sensing applications, as the total cost of the sensor is dominated by the cost of the transmitter and receiver.
Fiber-optic sensors characteristically enjoy long life. For example, the life expectancy of reflective fiber-optic switches is quoted at 10 million operations. Their accuracy is also good, with _1% of full-scale reading being quoted as a typical inaccuracy level for a fiber-optic pressure sensor. Further advantages are their simplicity, low cost, small size, high reliability, and capability of working in many kinds of hostile environments. The only significant difficulty in designing a fiber-optic sensor is in ensuring that the proportion of light entering the cable is maximized. This is the same difficulty described earlier when discussing the use of fiber-optic cables for signal transmission.
Light in; Shutter switch; Light out
Two major classes of fiber-optic sensors exist-intrinsic and extrinsic. In intrinsic sensors, the fiber-optic cable itself is the sensor, whereas in extrinsic sensors, the fiber-optic cable is only used to guide light to/from a conventional sensor.
In intrinsic sensors, the physical quantity being measured causes some measurable change in the characteristics of the light transmitted by the cable. The modulated light parameters are one or more of the following:
Sensors that modulate light intensity tend to use mainly multimode fibers, but only mono-mode cables are used to modulate other light parameters. A particularly useful feature of intrinsic fiber-optic sensors is that they can, if required, provide distributed sensing over distances of up to 1 meter.
Light intensity is the simplest parameter to manipulate in intrinsic sensors because only a simple source and detector are required. The various forms of switches are perhaps the simplest forms of these, as the light path is simply blocked and unblocked as the switch changes state. Modulation of the intensity of transmitted light also takes place in various simple forms of proximity, displacement, pressure, pH, and smoke sensors. In proximity and displacement sensors (the latter are sometimes given the special name fotonic sensors), the amount of reflected light varies with the distance between the fiber ends and a boundary. In pressure sensors, the refractive index of the fiber, and hence the intensity of light transmitted, varies according to the mechanical deformation of the fibers caused by pressure. In the pH probe, the amount of light reflected back into the fibers depends on the pH-dependent color of the chemical indicator in the solution around the probe tip. Finally, in a form of smoke detector, two fiber-optic cables placed either side of a space detect any reduction in the intensity of light transmission between them caused by the presence of smoke.
A simple form of accelerometer can be made by placing a mass subject to acceleration on a multimode fiber. The force exerted by the mass on the fiber causes a change in the intensity of light transmitted, hence allowing the acceleration to be determined. The typical inaccuracy quoted for this device is _0.02 g in the measurement range of +/- 5 g and +/- 2% in the measurement range up to 100 g.
A similar principle is used in probes that measure the internal diameter of tubes. The probe consists of eight strain-gauged cantilever beams that track changes in diameter, giving a measurement resolution of 20 mm.
A slightly more complicated method of affecting light intensity modulation is the variable shutter sensor. This consists of two fixed fibers with two collimating lenses and a variable shutter between them. Movement of the shutter changes the intensity of light transmitted between the fibers. This is used to measure the displacement of various devices such as Bourdon tubes, diaphragms, and bimetallic thermometers.
Yet another type of intrinsic sensor uses cable where the core and cladding have similar refractive indices but different temperature coefficients. This is used as a temperature sensor.
Temperature rises cause the refractive indices to become even closer together and losses from the core to increase, thus reducing the quantity of light transmitted.
Refractive index variation is also used in a form of intrinsic sensor used for cryogenic leak detection. The fiber used for this has a cladding whose refractive index becomes greater than that of the core when it’s cooled to cryogenic temperatures. The fiber-optic cable is laid in the location where cryogenic leaks might occur. If any leaks do occur, light traveling in the core is transferred to the cladding, where it’s attenuated. Cryogenic leakage is thus indicated by monitoring the light transmission characteristics of the fiber.
A further use of refractive index variation is found in devices that detect oil in water. These use a special form of cable where the cladding used is sensitive to oil. Any oil present diffuses into the cladding and changes the refractive index, thus increasing light losses from the core. Unclad fibers are used in a similar way. In these, any oil present settles on the core and allows light to escape.
A cross-talk sensor measures several different variables by modulating the intensity of light transmitted. It consists of two parallel fibers that are close together and where one or more short lengths of adjacent cladding are removed from the fibers. When immersed in a transparent liquid, there are three different effects that each cause a variation in the intensity of light transmitted. Thus, the sensor can perform three separate functions. First, it can measure temperature according to the temperature-induced variation in the refractive index of the liquid.
Second, it can act as a level detector, as the transmission characteristics between the fibers change according to the depth of the liquid. Third, it can measure the refractive index of the liquid itself when used under controlled temperature conditions.
The refractive index of a liquid can be measured in an alternative way using an arrangement where light travels across the liquid between two cable ends that are fairly close together.
The angle of the cone of light emitted from the source cable, and hence the amount of light transmitted into the detector, is dependent on the refractive index of the liquid.
The use of materials where the fluorescence varies according to the value of the measurand can also be used as part of intensity-modulating intrinsic sensors. Fluorescence-modulating sensors can give very high sensitivity and are potentially very attractive in biomedical applications where requirements exist to measure very small quantities, such as low oxygen and carbon monoxide concentrations, and low blood pressure levels. Similarly, low concentrations of hormones, steroids, and so on may be measured.
Further examples of intrinsic fiber-optic sensors that modulate light intensity are described later (level measurement) and (measuring small displacements).
As mentioned previously, phase, polarization, wavelength, and transit time can be modulated as well as intensity in intrinsic sensors. Monomode cables are used almost exclusively in these types of intrinsic sensors.
Phase modulation normally requires a coherent (laser) light source. It can provide very high sensitivity in displacement measurement, but cross sensitivity to temperature and strain degrades its performance. Additional problems are maintaining the frequency stability of the light source and manufacturing difficulties in coupling the light source to the fiber. Various versions of this class of instrument exist to measure temperature, pressure, strain, magnetic fields, and electric fields. Field-generated quantities such as electric current and voltage can also be measured. In each case, the measurand causes a phase change between a measuring and a reference light beam that is detected by an interferometer.
The principle of phase modulation has also been used in the fiber-optic accelerometer (where a mass subject to acceleration rests on a fiber) and in fiber strain gauges (where two fibers are fixed on the upper and lower surfaces of a bar under strain). The fiber-optic gyroscope is a further example of a phase-modulating device.
Devices using polarization modulation require special forms of fibers that maintain polarization.
Polarization changes can be affected by electrical fields, magnetic fields, temperature changes, and mechanical strain. Each of these parameters can therefore be measured by polarization modulation.
Various devices that modulate the wavelength of light are used for special purposes. However, the only common wavelength-modulating fiber-optic device is the form of laser Doppler flowmeter that uses fiber-optic cables.
Fiber-optic devices using modulation of the transit time of light are uncommon because of the speed of light. Measurement of the transit time for light to travel from a source, be reflected off an object, and travel back to a detector is only viable for extremely large distances. However, a few special arrangements have evolved that use transit time modulation. These include instruments such as the optical resonator, which can measure both mechanical strain and temperature. Temperature-dependent wavelength variation also occurs in semiconductor crystal beads (e.g., aluminum gallium arsenide). This is bonded to the end of a fiber-optic cable and excited from an LED at the other end of the cable. Light from the LED is reflected back along the cable by the bead at a different wavelength. Measurement of the wavelength change allows temperatures in the range up to 200_ C to be measured accurately. A particular advantage of this sensor is its small size, typically 0.5 mm diameter at the sensing tip. Finally, to complete the catalogue of transit time devices, the frequency modulation in a piezoelectric quartz crystal used for gas sensing can also be regarded as a form of time domain modulation.
Extrinsic fiber-optic sensors use a fiber-optic cable, normally a multimode one, to transmit modulated light from a conventional sensor such as a resistance thermometer. A major feature of extrinsic sensors, which makes them so useful in such a large number of applications, is their ability to reach places that are otherwise inaccessible. One example of this is the insertion of fiber-optic cables into the jet engines of aircraft to measure temperature by transmitting radiation into a radiation pyrometer located remotely from the engine. Fiber-optic cable can be used in the same way to measure the internal temperature of electrical transformers, where the extreme electromagnetic fields present make other measurement techniques impossible.
An important advantage of extrinsic fiber-optic sensors is the excellent protection against noise corruption that they give to measurement signals. Unfortunately, the output of many sensors is not in a form that can be transmitted by a fiber-optic cable, and conversion into a suitable form must therefore take place prior to transmission. For example, a platinum resistance thermometer (PRT) translates temperature changes into resistance changes. The PRT therefore needs electronic circuitry to convert the resistance changes into voltage signals and then into a modulated light format, which in turn means that the device needs a power supply.
This complicates the measurement process and means that low-voltage power cables must be routed with the fiber-optic cable to the transducer. One particular adverse effect of this is that the advantage of intrinsic safety is lost. One solution to this problem is to use a power source in the form of electronically generated pulses driven by a lithium battery. Alternatively, power can be generated by transmitting light down the fiber-optic cable to a photocell. Both of these solutions provide intrinsically safe operation.
Piezoelectric sensors lend themselves particularly to use in extrinsic sensors because the modulated frequency of a quartz crystal can be transmitted readily into a fiber-optic cable by fitting electrodes to the crystal that are connected to a low-power LED. Resonance of the crystal can be created either by electrical means or by optical means using the photo-thermal effect. The photo-thermal effect describes the principle where, if light is pulsed at the required oscillation frequency and directed at a quartz crystal, the localized heating and thermal stress caused in the crystal results in it oscillating at the pulse frequency. Piezoelectric extrinsic sensors can be used as part of various pressure, force, and displacement sensors. At the other end of the cable, a phase locked loop is typically used to measure the transmitted frequency.
Fiber-optic cables are also now commonly included in digital encoders, where the use of fibers to transmit light to and from the discs allows the light source and detectors to be located remotely. This allows the devices to be smaller, which is a great advantage in many applications where space is at a premium.
A number of discrete sensors can be distributed along a fiber-optic cable to measure different physical variables along its length. Alternatively, sensors of the same type, which are located at various points along a cable, provide distributed sensing of a single measured variable.
NEXT: Sensor Technologies (part 2)
Updated: Sunday, December 22, 2019 19:03 PST