Добавил:
Опубликованный материал нарушает ваши авторские права? Сообщите нам.
Вуз: Предмет: Файл:
Displacement Measurements, Linear and Angular.pdf
Скачиваний:
178
Добавлен:
02.05.2014
Размер:
2.39 Mб
Скачать

FIGURE 6.32 A typical capacitive humidity sensor. The sensors have pore or cracked mosaic structure for the moisture in air or gas to reach the dielectric material. The characteristics of the dielectric material change with the amount of water absorbed, thus reducing the resistance and increasing the capacitance. The quantity measured can be either resistance, capacitance, or impedance.

There are many types of capacitive humidity sensors. Aluminum, tantalum, silicon, and polymer types are introduced here.

Aluminum Type Capacitive Humidity Sensors

The majority of capacitive humidity sensors are aluminum oxide type sensors. In these type of sensors, high-purity aluminum is chemically oxidized to produce a prefilled insulating layer of partially hydrated aluminum oxide, which acts as the dielectric. A water-permeable but conductive gold film is deposited onto the oxide layer, usually by vacuum deposition, which forms the second electrode of the capacitor.

In another type, the aluminum-aluminum oxide sensor has a pore structure as illustrated in Figure 6.32. The oxide, with its pore structure, forms the active sensing material. Moisture in the air reaching the pores reduces the resistance and increases the capacitance. The decreased resistance can be thought of as being due to an increase in the conduction through the oxide. An increase in capacitance can be viewed as due to an increase in the dielectric constant. The quantity measured can be either resistance, capacitance, or impedance. High humidities are best measured by capacitance because resistance changes are vanishingly small in this region.

In addition to the kind of transducer design illustrated here, there are many others available with a number of substantial modifications for particular properties, such as increased sensitivity or faster response. Although most of these modifications result in a change in physical dimensions or appearance, the sensing material of the transducer — the aluminum oxide — remains the same.

In some versions, the oxide layer is formed by parallel tubular pores that are hexagonally packed and perpendicular to the plane of the base layer. These pores stop just before the aluminum layer, forming a very thin pore base. Water absorbed in these tubules is directly related to the moisture content of the gas in contact with it. The porous nature of the oxide layer produces a large area for the absorption of water vapor. At low humidities, the capacitance is due entirely to the mixed dielectric formed between the

© 1999 by CRC Press LLC

oxide, water vapor, and air. However, at higher humidities, parallel conductance paths through the absorbed water are formed down the pore surfaces. Near saturation, this pore surface resistance becomes negligible, implying that the measured capacitance is virtually that between the very thin pore base and the aluminum core.

Tantalum Type Capacitive Humidity Sensors

In some versions of capacitive humidity sensors, one of the capacitor plates consists of a layer of tantalum deposited on a glass substrate. A layer of polymer dielectric is then added, followed by a second plate made from a thin layer of chromium. The chromium layer is under high tensile stress such that it cracks into a fine mosaic structure that allows water molecules to pass into the dielectric. The stress in the chromium also causes the polymer to crack into a mosaic structure. A sensor of this type has an input range of 0% to 100% relative humidity, RH. The capacitance is 375 pF at 0% RH and a linear sensitivity of 1.7 pF per % RH. The error is usually less than 2% due to nonlinearity and 1% due to hysteresis.

Silicon Type Capacitive Humidity Sensors

In other capacitive humidity sensors, silicon is used as the dielectric. The structure and operation of silicon humidity sensors are very similar to the aluminum oxide types. Some silicon-type humidity sensors also use the aluminum base and a thin-film gold layer as the two electrodes. The silicon dielectric has a very large surface area, which means that the sensitivity is still relatively large even if the sensing area is very small. This is an important feature with the increasing trend of miniaturization. Both sensor types are now typically found as extremely small wafer-shaped elements, placed on a mechanical mount with connecting lead wires. The formation of porous silicon is a very simple anodization process and, since no elaborate equipment is needed, devices can be made at relatively low cost. Also, by controlling the formation conditions, the structure of the porous silicon can easily be modified so devices can be tailored to suit particular applications.

In both the silicon and the aluminum oxide capacitive humidity sensors, the radii of the pores in the dielectric are such that they are specifically suited for water molecules. Most possible contaminants are too large in size to pollute the dielectric. However, contaminants can block the flow of water vapor into the sensor material, thus affecting the accuracy of the instrument. For example, in dust-contaminated streams, it may be possible to provide a simple physical barrier such as a sintered metal or plastic hoods for the sensor heads. Many sensors come with some form of casing to provide protection.

Polymer Type Capacitive Humidity Sensors

In some sensors, the dielectric consists of a polymer material that has the ability to absorb water molecules. The absorption of water vapor of the material results in changes in the dielectric constant of the capacitor. By careful design, the capacitance can be made directly proportional to percentage relative humidity of the surrounding gas or atmosphere.

In general, an important key feature of capacitive humidity sensors is the chemical stability. Often, humidity sensing is required in an air sample that contains vapor contaminants (e.g., carbon monoxide) or the measurements are performed on a gas sample other than air (e.g., vaporized benzene). The performance of these sensors, and in particular the silicon types, is not affected by many of these gases. Hydrocarbons, carbon dioxide, carbon monoxide, and CFCs do not cause interference. However, the ionic nature of the aluminum oxide dielectric makes it susceptible to certain highly polar, corrosive gases such as ammonia, sulphur trioxide, and chlorine. Silicon is inert; its stable nature means that these polar gases affect the sensor element to a far lesser degree.

Capacitive Moisture Sensors

Capacitive moisture measurements are based on the changes in the permittivity of granular or powder type dielectric materials such as wheat and other grains containing water. Usually, the sensor consists of a large cylindrical chamber, (e.g., 150 mm deep and 100 mm in diameter), as shown in Figure 6.33. The chamber is filled with samples under test. The variations in capacitance with respect to water content are processed. The capacitor is incorporated into an oscillatory circuit operating at a suitable frequency.

© 1999 by CRC Press LLC

FIGURE 6.33 A capacitive moisture sensor. The permittivity of material between two cylindrical or parallel plates with fixed dimensions changes, depending on the moisture level of the materials in the chamber. The variations in capacitance values with respect to water content is processed. The capacitor is incorporated as a part of an oscillatory circuit operating at a suitable frequency, usually at radio frequencies.

Capacitive moisture sensors must be calibrated for samples made from different materials, as the materials themselves demonstrate different permittivities. Accurate temperature is necessary as the dielectric constant may be highly dependent on temperature. Most of these devices are built to operate at temperature ranges of 0°C to 50°C, supported by tight temperature compensation circuits. Once calibrated for a specific application, they are suitable for measuring moisture in the range of 0% to 40%.

Signal Processing

Generally, capacitive type pickups require relatively complex circuitry in comparison to many other sensor types, but they have the advantage of mechanical simplicity. They are also sensitive, having minimum mechanical loading effects. For signal processing, these sensors are usually incorporated either in ac deflection bridge circuits or oscillator circuits. In practice, capacitive sensors are not pure capacitances but have associated resistances representing losses in the dielectric. This can have an important influence in the design of circuits, particularly in oscillator circuits. Some of the signal processing circuits are discussed below.

Operational Amplifiers and Charge Amplifiers

One method of eliminating the nonlinearity of the relationship between the physical variable, (e.g., twoplate displacement sensors) and capacitance C is through the use of operational amplifiers, as illustrated in Figure 6.34. In this circuit, if the input impedance of the operational amplifier is high, the output is not saturated, and the input voltage is small, it is possible to write:

© 1999 by CRC Press LLC

FIGURE 6.34 An operational amplifier signal processor. This method is useful to eliminate the nonlinearity in the signals generated by capacitive sensors. By this type of arrangement, the output voltage can be made directly proportional to variations in the signal representing the nonlinear operation of the device.

1 Cf

= òif dt = eex eai

= eex

(6.45)

1 Cx = òix dt = e0 eai

= e0

(6.46)

if

+ ix iai = 0 = if + ix

(6.47)

Manipulation of these equations yields:

 

 

 

 

 

 

e0

= −Cf eex

Cx

 

(6.48)

Substituting the value of Cx yields:

 

 

 

 

 

 

e0

= −Cf x eex

εA

 

(6.49)

Equation 6.49 shows that the output voltage is directly proportional to the plate separation x, thus giving linearity for all variations in motion.

However, a practical circuit requires a resistance across Cf to limit output drift. The value of this resistance must be greater than the impedance of Cf at the lowest frequency of interest. Also, because the transducer impedance is assumed to be purely capacitive, the effective gain is independent of frequency.

A practical charge amplifier circuit is depicted in Figure 6.35. In this case, the effective feedback resistance Ref is given by:

Ref = R3 (R1 + R2 ) R2

(6.50)

It is possible to reduce the output drift substantially by selecting the resistors suitably. The accuracy of this circuit can be improved further by cascading two or more amplifiers. In this way, a substantial improvement in the signal-to-noise ratio can also be achieved. In the inverting input, the use of resistor R4 is necessary because of bias currents.

© 1999 by CRC Press LLC