A transducer is a device that converts one form of energy into an other form for some specific purpose. It might seem like such a device is quite specialized, but in fact transducers can be found everywhere. Their variety is practically unlimited. Transducers can be categorized as passive versus active, or as input versus output, or according to the type of energy with which they are supplied or the type that they produce at the output.
PASSIVE and ACTIVE TRANSDUCERSAny device that operates without the need for a source of external power is called a passive device. We might, e.g., build an audio low-pass filter using a coil and capacitor ( ill. 1-1A). No battery is necessary for this circuit to work; all we need to do is supply audio power at the input and connect a speaker or headset to the output. Thus, the coil-capacitor filter is a passive network. It’s also a transducer, being actuated by audio power of one kind (broadband) and supplying audio power of another kind (restricted-band). ill. 1-1. Passive (A) and active (B) audio low-pass filters. You can no doubt think of many examples of passive transducers right away: microphones, speakers, headsets, light bulbs, antennas. An active device must draw power from an external source in order to work. This power may be derived from an electrochemical battery or from a power supply that operates from the commercial mains. An active transducer thus has at least three ports: input, output, and power. We can design an audio filter, having a characteristic response almost identical to that of the passive le vice in ill. 1-1A, using an operational amplifier. Such a circuit is shown in ill. 1-1B. Other examples of active transducers include amplifiers, oscillators, tape recorders, radio receivers and transmitters, and electric eyes. INPUT and OUTPUT TRANSDUCERSA transducer may be used to provide the power needed by a device or circuit. In this application we call the transducer an input device. A common example of an input transducer is a microphone. We connect it to the input of an amplifier. When a transducer is used to convert some quantity or variable into something we can sense directly (see or hear), we call it an output transducer. A speaker is an output transducer. Some transducers can be used interchangeably as input or output devices. An example of this is an electric generator, which is normally used as an input transducer to provide electric current to a circuit. When certain kinds of electric generators are supplied with the power they normally produce from mechanical power, they operate as motors, which are output devices. EXAMPLES OF TRANSDUCERSThe following are commonly used transducers:
ill. 1.2. A simple mercury-activated temperature sensor. ill. 1-3. Block diagram of a simple radio receiver. ill. 1-4. Block diagram of a simple radio transmitter. ill. 1-5. Cross-sectional diagram of an X-ray tube. TRANSDUCER APPLICATIONSWe can use transducers to accomplish many different tasks. Some common applications are as follows. (This is by no means complete sampling.) Communication The most obvious examples of transducers employed in communications are telephone sets, radio transmitters, receivers, and antennas. But there are many different types of such transducers. We might want to communicate by using digital methods or analog methods. The antenna system (if electromagnetic communication is used) might employ linear or elliptical polarization. We might want to encode and decode the signals for some reason. Increasingly, modulated light, transmitted via optical fibers, is being used for communications. Lasers and photoelectric cells thus replace antennas and carrier-current converters. An example of a complete communications system is shown in ill. 1-6. ill. 1-6. A complete two-way electromagnetic communications system. Control Transducers are routinely used to regulate some variable, such as temperature, electric current or voltage, frequency, brightness, or loudness. Control may be accomplished manually, as with the volume control on a hi-fi set, or automatically, with a thermostat e.g.,. An automatic control system generally requires a sensor transducer that reacts to changes in the parameter under control. The sensor is connected to another transducer that operates according to the sensor output. The sensor in a thermostat is a thermal switch, usually a bimetallic strip, although a thermocouple or thermistor may also be employed. The sensor regulates the operation of a furnace or air conditioner, which is itself a form of transducer that generates thermal energy from electricity or from the potential energy in a fossil fuel. Electrochemical Potential energy in chemical form can be converted into electrical energy by means of a common cell or battery ( ill. 1-7A). The electrochemical cell consists of a solution or paste of some chemical compound, called the electrolyte, and two metal electrodes with which the electrolyte reacts. An excess of electrons appears at one electrode, and a deficiency of electrons exists at the other electrode. Thus, a potential difference is produced; when the two electrodes are connected to a circuit, current flows and power is generated. Some electrochemical transducers convert electrical energy into chemical form. An example of this is the electrolysis apparatus shown in ill. 1-7B. This device is very much like the electrochemical cell shown in ill. 1-7A. Some electrochemical transducers can work in both modes: electrical-to-chemical and chemical-to-electrical. Your automobile battery is an example of such a device, as is any rechargeable cell or battery. ill. 1-7. Electrolytic cells. At A, a common zinc-carbon dry cell; at B, electrolysis apparatus generating hydrogen and oxygen from an electric current. Electromechanical The motor and generator are the most common examples of electromechanical transducers. As we mentioned before, some motors can operate as generators, and some generators can operate as motors. Motors and generators can be designed to operate from either alternating current or direct current. Some electromechanical transducers are used to measure mechanical movement or some electrical quantity. A simple analog meter of the D’Arsonval type ( ill. 1-8A) is a form of electromechanical transducer that converts electric current into mechanical displacement. Two electromechanical transducers—a direct-current generator and a meter—can be connected to monitor linear or angular speeds ( ill. 1-8B). ill. 1-8. At A, a simple direct-current meter. At B, an electronic tachometer using a generator and meter. Measurement and Monitoring Any kind of transducer can, in some way, be used for measurement or monitoring purposes. We have just seen how electromechanical transducers are used to measure speed. Any measurement transducer is more commonly known as a meter, and the variety of metering devices in today’s technological world is overwhelming. The control panel of any sophisticated electronic device might have a dozen or more different indicating transducers. Testing Testing, for quality control, troubleshooting, or empirical design, involves transducers. Of course, measurement and monitoring devices play a central role in any test procedure. Peripheral transducers, such as radio-frequency sniffers or probes ( ill. 1-9) are employed with test apparatus in many instances. ill. 1-9. An RF sensor probe. EFFICIENCYThe operation of any machine, no matter what its function, is limited by its efficiency. Efficiency is defined, in percent, as the ratio of useful energy actually realized to the theoretical ideal. An efficiency of 100 percent is never obtained in the real world, al though in some cases performance is almost perfect. The efficiency of a passive transducer is easy to define. We can simply measure the input and output energies, in joules, over a given length of time and then compare them. If the input energy to a transducer is El and the output energy (in the desired form) is E2, then it will always be true that E1 > E2. We define the efficiency according to the formula Efficiency (percent) = 100(E2/E1) We may, in some instances, define efficiency according to instantaneous input and output power values (in the desired form) P1 and P2: Efficiency (percent) = 100(P2/P1) The instantaneous efficiency of a transducer may change, but it will always be less than 100 percent; that is, P1 > P2. Although in many cases the instantaneous power efficiency of a transducer is the same as the energy efficiency as measured over a period of time, it’s not necessarily true. Consider the example of an incandescent bulb whose power efficiency depends on the amount of voltage with which it’s supplied. The power efficiency increases with increasing voltage (up to a certain point). In this case the output power is that portion that falls within the visible-light range of the spectrum; infrared light and ultraviolet light are not considered part of the desired output power. The graph of ill. 1-10A illustrates a hypothetical graph of output power versus input power for a 100-W incandescent lamp. As long as we don’t vary the input voltage (and thus the input power) to the bulb, the instantaneous power efficiency and the energy efficiency will be the same because E1 = P1t and E2 = P2t where t is the length of time the bulb operates. But if we vary the input voltage over a period of time, the instantaneous power efficiency will change. But during the time span, the overall energy efficiency will be a constant—the average power efficiency during that time. The energy efficiency is actually the integral of the instantaneous power efficiency during the time specified. This is shown in ill. 1-10B. ill. 1-10. Output versus input for a hypothetical incandescent bulb (A), and Instantaneous power efficiency, integrated to obtain energy efficiency (B). The discrepancy between instantaneous power efficiency and energy efficiency is not unique to incandescent lamps. it's , in fact, a characteristic of almost all transducers. Transducer manufacturers specify the appropriate range of input power for best operation of a device. Efficiency is a little more complicated to define in the case of an active transducer, where the output energy/power is often larger than the input energy/power because of amplification. We must add another factor—the power-supply or battery energy/power—in order to obtain a meaningful measure of efficiency in an active transducer. If we let Eb represent the energy provided by the power supply or battery, and Pb represent the power provided by the supply at any given instant, then it will always be true that (E1 + Eb) > E2, and (P1 + Pb) > P2. The energy efficiency is then Efficiency (percent) = 100(E2/(E1 + Eb)) and the instantaneous power efficiency is Efficiency (percent) = 100(P2/(P1 + Pb)) As in the case of a passive transducer, the instantaneous power efficiency often changes with fluctuations in the input power. A class C amplifier is a good example of this kind of situation. If the signal input is very small, the output is practically zero, and the resulting efficiency is so poor that it can be considered zero. Yet, with adequate input power, the efficiency may approach 90 percent! RESOLUTION and ACCURACYFor certain kinds of transducers, efficiency is not of much concern. We’re not worried, e.g., about the efficiency of a digital ammeter; as long as the indication is readable, the device is working all right—provided the value is accurate. The resolution of an indicating transducer is the smallest incremental change that can be detected. An analog meter with a full-scale indication of 100 mA might have a resolution of 0.5 mA; This means that any change of less than 0.5 mA would result in no appreciable movement of the needle. The resolution can be expressed either in absolute terms or according to the proportion of the reading represented by this movement. Thus, a resolution of 0.5 mA (absolute) would be plus or minus 5 percent of 10 mA, plus or minus 0.5 percent of 100 mA, and so on. In the case of a digital display, the resolution is simply the smallest possible digital change. This can, as with an analog meter, be expressed either in absolute terms or as a percentage. A digital milliammeter might measure currents down to a tenth or a hundredth of a milliampere. Some digital meters have a fixed number of digits along with a floating decimal, and their absolute resolution therefore depends on the value of the parameter to be measured, whereas the percentage resolution fluctuates by only one order of magnitude (the greatest accuracy being at a level where all of the digits are nines, and the least being where all of the digits are zeros). Other digital meters have the decimal point fixed, and their resolution is always the same in absolute terms but varies, sometimes tremendously, in terms of percentage over several orders of magnitude. ( e.g., a reading of 0.0003 mA in a display with four digits to the right of the decimal point has a resolution of plus 50 percent and minus 25 percent; but a reading of 100.0003 mA has a resolution of plus or minus 0.0001 percent.) The accuracy of a metering transducer is the maximum possible difference, as a percentage, that can exist between the actual value (as determined by a standard) and the indicated value. This is usually based on the full-scale reading. The error increases (in general) as the value decreases from full scale. Accuracy depends on the resolution and also on the calibration of the metering device. SENSITIVITYThe sensitivity of a transducer can be defined in two ways: the minimum amount of input that generates a detectable or measurable output; or the smallest amount of input change that results in a detectable or measurable change in the output. This quantity can be expressed in absolute terms, as a percentage, or as a decibel ratio. These two sensitivity figures will often (in fact, usually) differ for a given transducer. Both factors may also change along with changes in some other parameter. We might consider the example of a speaker that requires 15 mW of audio power in order to produce a noticeable sound at a pure sine-wave frequency of 1000 Hz. At low power levels a 5-mW change might produce a detectable change in the volume; as the power level is increased, it might require a larger and larger absolute increment to produce a noticeable change in the volume. A hypothetical response of this kind is shown in ill. 1-11 for a frequency of 1000 Hz. But the figures may—and probably will—be different if the frequency is 500 Hz, 2500 Hz, or if the input waveform is not perfectly sinusoidal. When defining the sensitivity of a transducer, then, we must be sure that we specify the values of variables that might affect the result, or else we will end up with meaningless figures. ill. 1-11. A hypothetical speaker output-versus-input curve. RESPONSE TIMESome types of transducers don’t produce an output signal until a certain time after the application of the input signal. An example of such a device is an ordinary incandescent bulb. It does not light up right away when the input power is applied, but instead takes a few hundredths or a few tenths of a second to reach full brilliance. You have probably noticed this especially with high-wattage bulbs such as movie-camera lights. Measuring devices, such as analog or digital meters, also require a short length of time to produce a reading. Thermostatic devices fall into this same category. There are many other examples. The response time of a transducer is defined as the time t, usually measured in nanoseconds, microseconds, milliseconds, or seconds, between the time t1 at which input is applied and the time t2 at which the output stabilizes ( ill. 1-12A). Response time can also be defined as the time t’ between the power-removal time t3 and the point t4 at which the output again stabilizes ( ill. 1-12B). These two response times, t and t’, are not, in general, the same for a given transducer at a particular power level. ill. 1-12. Rise (A) and decay (B) curves. Rise time is given by t, decay time by t’. HYSTERESISHysteresis in a transducer is closely related to response time, in that it’s an indicator of the degree of sluggishness with which the device reacts to changes in the input. A good example of hysteresis is provided by the operation of a thermostat. Suppose we set the thermostat for a furnace to 65 degrees F. while the initial temperature in the room is well below that value. The thermostat will at first turn on the furnace immediately. The temperature will increase (assuming the furnace is working). The thermostat won’t shut off, however, until the temperature is a little more than 65 degrees, because of sluggishness in the response. By the time the furnace shuts down, it might be 67 degrees in the room. Then the room will start to cool, but, again, because of slowness in the response of the thermostat, the furnace won’t switch on again until the temperature cools to, say, 63 degrees. After the initial warm-up of the room, then, we can expect the temperature to fluctuate within a span of 4 degrees ( ill. 1-13). This phenomenon is known as hysteresis. ill. 1-13. Hysteresis in a hypothetical thermostat. The furnace is on during times indicated by the heavy line. Hysteresis may or may not be wanted in a transducer device. In the preceding example it’s desirable to a certain extent because otherwise the furnace would cycle on and off constantly at intervals of a few seconds. But we would not want the hysteresis to be excessive; it would be uncomfortable, e.g., if the temperature went up to 72 degrees and down to 58 degrees before the thermostat initiated corrections. Hysteresis can be defined in various ways, depending on the function of the transducer. Usually, hysteresis is defined in terms of the maximum fluctuation that takes place in the output of a control or metering device. USEFUL LIFEAn important consideration in the choice of a transducer for a given application is the service life of the device. This is usually indicated in hours, but it can also be given in terms of the number of switching or control operations that the transducer can perform. Useful life is especially important for control, monitoring, measurement, and testing transducers. Expense is also a significant consideration because we don’t want to have to constantly replace a component that costs a lot of money. |
Updated: Friday, November 14, 2008 20:00 PST
