Sensors (part 3)

(<< cont. from part 2 )

6. Silicon Accelerometers

With the demand from automotive and other industries in the latter 1980s, several sensor manufacturers developed micromachined silicon ICs for sensing acceleration. Today, several component manufacturers, such as IC Sensors, Analog Devices, and Motorola, have families of silicon accelerometers. Basic sensor elements as well as signal conditioned versions are available.

6.1 Basic Principles of Sensing Acceleration

The principles of acceleration sensing were simulated using a weight and spring connected to a frame to develop silicon accelerometers using the piezoresistive properties of silicon and building capacitive structures with variation of effective capacitance between plates attached to a seismic mass of silicon. To simulate the basic mechanical analogy of accelerometers and minimize secondary effects that complicate the measuring process (IC Sensors, eoo; Quinnell, 1992) required several improvements in silicon-processing technology.

One such improvement was the advent of silicon fusion bonding (Quinnell, 1992). Fusion bonding, which bonds two wafers while preserving the crystalline structure of silicon, permits the creation of complex 3-D structures without introducing mechanical discontinuities or thermal-dependent stress. This structuring ability lets accelerometer manufacturers capture the seismic mass with a sealed cavity by bonding a cap and a base plate to the frame. By controlling the space between the mass and cavity, vendors can use the air sealed inside the cavity as a viscous damping fluid for the system's motion.

Silicon fusion bonding also provides an answer to another limitation: shock resistance. Simply falling off of a desk can produce a 200g shock when the sensor hits the floor. Despite silicon's toughness and flexibility, that kind of shock could break the springs in an accelerometer unless the seismic mass's motion is limited. Silicon fusion bonding allows the placement of bumpers and other mechanical stops to make the accelerometer much more shock resistant.

Devices now routinely handle shocks as great as 2000g.

In commercially available sensors, a single-or double-cantilevered or a membrane-supported mass is coupled with a piezoresistive or capacitive element.

6.1.1 Piezoresistive Sensors

FIG. 34(a) is a diagram of a single-cantilevered design of an accelerometer using piezo-resistive elements. Thin beams support one edge of a seismic mass, which is free to move within a cavity created by fusion bonding two additional wafers to the one containing the mass. Piezoelectric resistors fabricated at the beams measure the displacement by changing the resistance as the beams bend. The double-cantilevered approach shown in FIG. 34(b) supports the mass from two sides.

FIG. 34 Silicon accelerometer simulating the mass and spring: (a) single cantilevered, (b) double cantilevered

While single-cantilevered types are simplest and sensitive, they have drawbacks such as transverse sensitivity (Quinnell, 1992). Double-cantilevered types can be designed with self-compensating effects for transverse forces. IC Sensors and Lucas Nova Sensor manufacture piezoresistive type sensors.

FIG. 35 The basic arrangement of a capacitive sensor: (a) structure, (b) equivalent circuit

6.1.2 Capacitive Sensors

FIG. 35 shows a typical capacitive sensing device. In these devices, the mass is supported on all four sides and transverse sensitivity is very much reduced. Capacitive sensors use top and bottom plates to form a capacitor divider with a seismic mass that is temperature insensitive. Sensing the change in capacitance requires relatively complex circuitry, however.

In some capacitive accelerometers, such as the ADXL series from Analog Devices, the seismic mass is not a single block but a series of interdigitated fingers, as shown in FIG. 36. This allows the sensing of acceleration in the plane of the chip than in other types of sensors, where sensing is normal to the surface.

The drawback of complicated interface circuitry in a capacitive sensor is compensated for by the additional ability inherent in the capacitor structure. The presence of charge-carrying plates in the sensors provides a built-in means for applying an electrostatic force on the seismic mass. This capability lets the sensor be used in a closed-loop configuration.

Instead of letting the seismic mass move freely during acceleration, a closed-loop system applies a restoring force to the mass, keeping it relatively motionless.

Restricting the movement of the mass has two advantages. First, it improves sensor linearity by confining the motion to the linear region of the spring's restoring force. Second, it extends the range of a sensor beyond the limits imposed by its housing on the seismic mass's movement. In such force-feedback systems, the restoring force, not the actual movement, serves as the measure of acceleration.

FIG. 36 Arrangement in the ADXL50.

The ability to apply a force to the proof mass has an additional advantage: It gives the sensor a self-test capability. This capability is particularly important in systems such as automotive airbags, where you cannot test the system by actually accelerating it yet testing is necessary for safety or reliability.

6.2 Practical Components

6.2.1 Piezoresistive Example: Model 3255 from IC Sensors

The Model 3255 is a fully signal conditioned accelerometer containing two chips: the silicon sensing element and a custom integrated circuit (ASIC) for signal conditioning. The Model 3255 accelerometer is available in various measurement ranges. With a supply voltage of 5 V, the output voltage is 2.5 V at no applied acceleration and the output range is 0.5-4.5 V for the full acceleration range. The output voltage is ratiometric with the supply voltage and will track the supply voltage in the range 5.0 4-0.5 V. Only three connections need to be made to use the accelerometer: 5 V supply, ground, and signal output.

FIG. 37(a) is a photograph of the device showing the two chips and the sealed unit. FIG. 37(b) shows the arrangement of the sensor element.

FIG. 37 Model 3255 accelerometer: (a) photograph, (b) cross-section of accelerometer die, (c) functional block diagram, (d) simplified schematic diagram. (IC Sensors.)

6.2.1.1 Sensor Element

The silicon sensor element is shown in FIG. 37(b). A seismic mass and four flexures are formed using bulk micromachining processes. Each of the four beams contains two implanted resistors that are interconnected to form a Whetstone bridge. When the device undergoes acceleration, the mass moves up or down, causing four of the resistors to increase and the other four to decrease in value. This results in an output voltage change proportional to the applied acceleration. The eight resistors are interconnected such that the effect of any off-axis acceleration is canceled.

Silicon top and bottom caps are attached to the section containing the seismic mass and the beams. The silicon caps serve several purposes. Precision gaps are etched into the caps to provide air damping to suppress the resonant peak of the structure. Because the part is critically damped, the frequency response is fiat up to several kilohertz with little dependence on temperature. Small elevated stops on the top and bottom caps limit the motion of the mass to a fraction of the deflection at which fracture occurs. The caps also form a chamber around the seismic mass to provide protection during the later stages of manufacturing and its operating lifetime.

Last, the top cap allows testing the accelerometer in the absence of acceleration. When a voltage is applied to a metal electrode on the top cap, an electrostatic force moves the mass toward the top cap. This results in a change in output voltage proportional to the sensitivity and to the square of the applied voltage. It thus is possible to generate an "acceleration" using an external voltage and check the functionality of the mechanical structure as well as the electronics.

6.2.1.2 Signal Conditioning

The signal conditioning circuitry amplifies the output of the sensor element and corrects the sensitivity and offset changes that occur with overheating. As a result, the output signal is accurate and no trimming is required by the user. The data used to set the performance of the accelerometer is stored in fused registers within the signal conditioning IC. The signal conditioning IC converts the differential signal from the sensor element (nominally 4-5 mV) into a single-ended signal in the 0.5-4.5 V range while correcting for the temperature-related signal variations. Signals are processed by differential amplifiers throughout most of the circuit to minimize common mode effects and noise. Switched-capacitor circuitry is used to save space and because high-accuracy gain stages can be made easily. As a result, the compensated accelerometers are interchangeable with a very small total error.

The signal conditioning IC is made in 1.5 um CMOS technology and intended for 5 V operation.

The signal path is shown in the block diagram in FIG. 37(c). The output signal of the accelerometer die is processed by the following stages:

• The first stage provides a high impedance load for the sensor and amplifies the signal to maximize the dynamic range during subsequent processing.

• The offset of the sensor die is reduced to less than 0.5% of full scale at room temperature by adding a voltage generated by DAC 1. This DAC is controlled by a digital word representing the programmed offset value.

• The temperature coefficient of offset (TCO) of the sensor is compensated for by adding a voltage generated by DAC 2 and DAC 3. This voltage is controlled by digital words representing the temperature and the programmed TCO value.

Both the offset and TCO voltages are derived from the supply to ensure that the signal remains ratiometric with the supply voltage.

• The signal gain is set by the value in DAC 4. The gain can be varied in a 5:1 range to allow for different full-scale specifications.

• The temperature coefficient of sensitivity (TCS) of the sensor is compensated in the next stage, built around a feed-forward loop using two DACs controlled by digital words representing the temperature and the programmed TCS value.

The sensitivity decrease over temperature is compensated for by increasing the signal gain linearly with temperature.

• The output bias voltage can be set to either 0.5 or 2.5 V by connecting an input pad on the chip to ground during assembly of the part. This allows signals to be processed with either a bipolar or unipolar range.

• A two-pole passive filter removes signals generated by the internal oscillator and switched capacitor networks. Switching noise is further minimized by having separate digital and analog internal supply lines and the differential signal processing.

• The final stage provides a low-impedance output for driving resistive and capacitive loads without influencing the signal. The output will go in a impedance "tristate" mode if the part is not addressed.

The temperature word that controls DACs 3 and 6 is generated by an ADC that digitizes the output of a temperature PTAT source driven by a bandgap reference. The temperature word therefore is linearly proportional to the temperature but does not depend on the supply voltage. For application and performance details, see IC Sensors (1995).

6.2.2 Capacitive Example: The ADXL50 from Analog Devices

The ADXL50 is a complete acceleration measurement system on a single monolithic IC. Three external capacitors and a +5 V power supply are all that is required to measure accelerations up to +50g. Device sensitivity is factory trimmed to 19 mV/g, resulting in a full-scale output swing of 4-0.95 V for a +50g applied acceleration. Its 0g output level is +1.8 V. A TTL compatible self-test function can electrostatically deflect the sensor beam at any time to verify device functionality. A functional block diagram of ADXL50 is shown in FIG. 38.

FIG. 38 Functional block diagram of the ADXL50. (Analog Devices, Inc.)

FIG. 39 ADXL50 operation: (a) sensor element at rest, (b) sensor momentarily responding to acceleration, (c) functional block diagram. (Analog Devices, Inc.)

The ADXL50 is a complete acceleration measurement system on a single monolithic IC. It contains a polysilicon surface-micromachined sensor and signal conditioning circuitry. The ADXL50 is capable of measuring both positive and negative acceleration to a maximum level of +50g.

FIG. 39(a) is a simplified view of the ADXL50's acceleration sensor at rest. The actual structure of the sensor consists of 42 unit cells and a common beam. The differential capacitor sensor consists of independent fixed plates and a movable "floating" central plate that deflects in response to changes in relative motion. The two capacitors are series connected, forming a capacitive divider with a common movable central plate. A force balance technique counters any impending deflection due to acceleration and drives the sensor back to its 0g position.

FIG. 39(b) shows the sensor responding to applied acceleration. When this occurs, the common central plate or "beam" moves closer to one of the fixed plates and farther from the other. The sensor's fixed-capacitor plates are driven deferentially by a 1 MHz square wave; the two square wave amplitudes are equal but 180 ° out of phase with one another. When at rest, the values of the two capacitors are the same, and therefore, the voltage output at their electrical center (i.e., at the center plate) is 0. When the sensor begins to move, a mismatch in the value of their capacitance is created, producing an output signal at the central plate. The output amplitude will increase with the amount of acceleration experienced by the sensor. Information concerning the direction of beam motion is contained in the phase of the signal with synchronous demodulation being used to extract this information.

Note that the sensor needs to be positioned so that the measured acceleration is along its sensitive axis.

FIG. 39(c) shows a block diagram of the ADXL50. The voltage output from the central plate of the sensor is buffered and applied to a synchronous demodulator. The demodulator also is supplied with a (nominal)1 MHz signal from the same oscillator that drives the fixed plates of the sensor. The demodulator will rectify any voltage in sync with its clock signal. If the applied voltage is in sync and in phase with the clock, a positive output will result. If the applied voltage is in sync but 180 ° out of phase with the clock, the demodulator's output will be negative. All other signals will be rejected. An external capacitor, C1, sets the bandwidth of the demodulator.

The output of the synchronous demodulator drives the preamp, an instrumentation amplifier buffer that is referenced to -t-1.8 V. The output of the preamp is fed back to the sensor through a 3 Mr2 isolation resistor. The correction voltage required to hold the sensor's center plate in the 0g position is a direct measure of the applied acceleration and appears at the VpR pin. When the ADXL50 is subjected to acceleration, its capacitive sensor begins to move, creating a momentary output signal. This is signal conditioned and amplified by the demodulator and preamp circuits. The DC voltage appearing at the preamp output then is fed back to the sensor and electrostatically forces the center plate back to its original center position.

At 0g, the ADXL50 is calibrated to provide + 1.8 V at the VpR pin. With applied acceleration, the VpR voltage changes to the voltage required to hold the sensor stationary for the duration of the acceleration and provides an output that varies directly with the applied acceleration. The loop bandwidth corresponds to the time required to apply feedback to the sensor and is set by external capacitor C1. The loop response is fast enough to follow changes in gravitational level up to that exceeding 1 kHz. The ADXL50's ability to maintain a flat response over this bandwidth keeps the sensor virtually motionless. This eliminates any nonlinearity or aging effects due to the sensor beam's mechanical spring constant, as compared to an open-loop sensor.

An uncommitted buffer amplifier provides the capability to adjust the scale factor and 0g offset level over a wide range. An internal reference supplies the necessary regulated voltages for powering the chip and +3.4 V for external use.

Applications and further details on the ADXL series devices can be found in Analog Devices (Application Note G2112A).

7. Hall Effect Devices

The basic Hall sensor is simply a small sheet of semiconductor material. A constant voltage source forces a constant bias current to flow in the semiconductor sheet. The output, a voltage measured across the width of the sheet, reads near 0 if a magnetic field is not present. If the biased Hall sensor is placed in a magnetic field oriented at right angles to the Hall current, the voltage output is in direct proportion to the strength of the magnetic field. This is the Hall effect, discovered by E. H. Hall in 1879 ( FIG. 40). When a magnetic field, B, is applied to a specimen (metal or semiconductor) carrying a current, Ic, in the direction perpendicular to Ic, a potential difference, VM, proportional to the magnitude of the applied magnetic field B appears in the direction perpendicular to both Ic and B. The relationship is expressed in the form…

VH=KxlcxB

… where K represents a constant, the product sensitivity, which depends on the physical properties and dimensions of the material used for the Hall effect device.

The basic Hall sensor essentially is a transducer that will respond with an output voltage if the applied magnetic field changes in any manner. Differences in the response of devices generally are related to tolerances and specifications, such as operate (turn on) and release (turn off) thresholds, as well as the temperature ranges and temperature coefficients of these parameters. Also available are linear output sensors that differ in sensitivity or respond per gauss change.

FIG. 40 The Hall effect

7.1 Linear Output Devices

Linear output Hall effect devices are the simplest Hall sensor devices. Practical devices such as Allegro Microsystems' UGN-3605 give an output voltage response to applied magnetic field changes. Electrical connections for the UGN-3605 are given in FIG. 41(a). Applications of Hall devices are discussed in Swager (1989). The output voltage of the devices such as the UGN-3605 is quite small, which can present problems, especially in an electrically noisy environment.

Addition of a suitable DC amplifier and a voltage regulator to the circuit improves the transducer's output and allows it to operate over a wide range of supply voltages. Such combined devices are available and an example is the UGN-3501 from Allegro Microsystems.

7.2 Digital Output Devices

The addition of a Schmitt trigger threshold detector with built-in hysteresis, as shown in FIG. 42, gives the Hall effect circuit digital output capabilities.

When the applied magnetic flux density exceeds a certain limit, the trigger provides a clean transition from off to on with no contact bounce. Built-in hysteresis eliminates oscillation (spurious switching of the output) by introducing a magnetic dead zone in which switch action is disabled after the threshold value is passed.

FIG. 41 Linear output Hall effect device: (a) device connections for UGN-3605, (b) amplified version. (Allegro Microsystems.)

An open-collector NPN output transistor added to the circuit gives the switch digital logic compatibility. The transistor is a saturated switch that shorts the output terminal to ground wherever the applied flux density is higher than the on trip point of the device. The switch is compatible with all digital families. The output transistor can sink enough current to directly drive many loads, including relays, triacs, SCRs, LEDs, and lamps.

FIG. 42 Digital output Hall effect switch.

8. Humidity and Chemical Sensors

8.1 Humidity Sensors

Humidity, usually understood to refer to the water content of the air, also can be sensed using silicon-based sensor elements. Relative humidity (RH), which is the ratio of absolute humidity to saturation humidity, has a value between 0 and 1 (0% and 100%). Several techniques are used to measure the relative humidity using capacitance, resistance, conductivity, and temperature measurements. Thermoset polymer or thermoplastic polymer-based materials are used on silicon or ceramic-based substrates to measure the RH. A comparison of RH sensors is available in Microswitch (1997b). Capacitive RH sensors dominate both atmospheric and process measurements and are the only type of full-range RH measuring devices capable of operating accurately down to 0% RH. Because of their low temperature effect, they often are used over wide temperature ranges without active temperature compensation.

Thermoset polymer-based (as opposed to thermoplastic-based) capacitive sensors (see FIG. 43) allow higher operating temperatures and provide better resistivity against chemical liquids; vapors such as isopropyl, benzene, toluene, and formaldehyde; oils; common cleaning agents; and ammonia vapor in concentrations common to chicken coops and pig barns. In addition, thermoset polymer RH sensors provide the longest operating life in ethylene oxide-based sterilization processes.

An example of a thermoset polymer-based capacitance RH device family is the Hycal IH36XX series from Microswitch. These devices come with on-chip signal conditioning and provide a fairly linear ratiometric output based on the DC supply.

In operation, water vapor in the active capacitor's dielectric layer equilibrates with the surrounding gas. The porous platinum layer shields the dielectric response from external influences while the protective polymer overlayer provides mechanical protection to the platinum layer from contaminants such as dirt, dust, and oil.

A heavy contaminant layer of dirt, however, will slow down the sensor's response time, because it will take longer for water vapor to equilibrate in the sensor.

FIG. 43 Thermoset polymer-based RH sensors: (a) basic construction, (b) relative humidity IC, (c) output voltage of IH-3602 vs. relative humidity. (Microswitch.)

8.2 Temperature and Humidity Effects

The output of all absorption-based humidity sensors (capacitive, bulk resistive, conductive film, etc.) are affected by both temperature and %RH. Because of this, temperature compensation is used in applications that call for either higher accuracy or wider operating temperature ranges. When temperature compensating a humidity sensor, it’s best to make the temperature measurement as close as possible to the humidity sensor's active area; that is, within the same moisture microenvironment. This is especially true when combining RH and temperature as a method of measuring the dew point.

HyCal's industrial-grade humidity and dew point instruments incorporate a HyCal 1000 f2 platinum resistance temperature detector on the back of the ceramic sensor substrate for unmatched temperature compensation measurement integrity. No on-chip signal conditioning is provided in these high temperature sensors ( FIG. 43(b)).

8.3 Chemical Sensors

Numerous technologies are utilized in the chemical sensing industry (see Waiters, 1996). Silicon as a basic structure for chemical sensors has been investigated in numerous laboratories over the last 20 years. Based on metal oxide semiconductor gas sensors, some sensor manufacturers such as FiS Sensors ( Japan) offer a wide range of products coveting many applications, including carbon monoxide sensing, flammable gas detection, toxic gas detection, indoor air quality controls, and combustion monitoring and control.

FIG. 44 The SB series gas sensors: (a) sensing element, (b) structure for standard housing, (c) SB-50 housing, (d) pin layout, (e) equivalent circuits, (f) standard circuit. (FiS Inc.)

The sensing element used in these devices is a mini-bead-type semiconductor, composed mainly of tin dioxide (SnOe). A heater coil and an electrode wire are embedded in the element ( FIG. 44(a)). The element is installed in a metal housing, which uses double stainless steel mesh in the path of gas flow and provides an antiexplosion feature ( FIG. 44(b)). The sensor has three pins for output signal and heater power supply. The SB-50 uses an active charcoal filter as shown in FIG. 44(c). The conductivity of tin dioxide-based metal oxide semiconductor material changes according to gas concentration changes.

This is caused by adsorption and desorption of oxygen and the reaction between surface oxygen and gases. These reactions cause a dynamic change of electric potential on the SnOe crystal and results in a decrease in sensor resistance under the presence of reducing gases such as carbon monoxide, methane, and hydrogen.

FIG. 44(d) and (e) show the pin layout and the equivalent circuit.

FIG. 44(f) shows the standard circuit of the SB series. The applied heater voltage regulates the sensing element temperature to obtain the specific performance of sensors. A change in the sensor resistance generally is obtained as a change in the output voltage across the fixed or variable load resistor (RL) in series with the sensor resistance (Rs). The sensitivity characteristics of semiconductor gas sensors are shown by the relationship between the sensors resistance (Rs) and concentration of gases.

The sensor resistance decreases with an increase of the gas concentration based on a logarithmic function. The standard test conditions of each model are calibrated to meet a typical target gas and concentration; For example, methane 1000 ppm for flammable gas detection, hydrogen 100 ppm for hydrogen detection, or ethanol 300 ppm for solvent detection. FIG. 45 shows the typical sensitivity characteristics of the SB series. In these diagrams, the sensor resistance change is normalized by the Rs at specific conditions. For further details, see FiS Inc. (1996).

FIG. 45 Typical sensitivity characteristics of the SB series gas sensors.

9. IEEE P1451 Standard for Smart Sensors and Actuators

Sensors are used in a wide range of applications from industrial automation to patient-condition monitoring in hospitals. With advancement of silicon and Micro Electro Mechanical Systems (MEMS) technologies, more "smarts" are integrated into sensors. The emergence of the control networks and smart devices in the marketplace may provide economical solutions for connecting transducers (hereafter specified as sensors or actuators) in distributed measurement and control applications; therefore, networking small transducers is seriously considered by transducer manufacturers and users.

Control networks provide many benefits for transducers:

• Significant reduction in installation costs by eliminating many and long analog wires.

• Acceleration of control loop design cycles, reduction of commissioning time, and reduction of downtime.

• Dynamic configuration of measurement and control loops via software.

• Addition of intelligence by leveraging the microprocessors used for digital communication.

For anyone attempting to choose a sensor-interface or networking standard, the range of choices is overwhelming. Some standards are open, and some are proprietary to a company's control products. To remedy the situation, the IEEE Sensor Technology Committee TC-9 is developing the IEEE P1451, Standard for Smart Transducer Interface for Sensors and Actuators. The sensor market comprises widely disparate sensor types. Designers consume relatively large amounts of all types of sensors. However, the lack of a universal interface standard impedes the incorporation of "smart" features, such as an onboard electronic data sheet, onboard A/D conversion, signal conditioning, device-type identification, and communications hand-shaking circuitry, into the sensors. In response to the industry's need for a communication interface for sensors, the IEEE with cooperation from the National Institute of Standards and Technology (NIST), decided to develop a hardware-independent communication standard for low-cost smart sensors that includes smart transducer object models for control networks.

The IEEE P1451 standards effort, currently under development, will provide many benefits to the industry. P1451, "Draft Standard for Smart Transducer Interface for Sensors and Actuators," consists of four parts, namely:

IEEE 1451.1--Network Capable Application Processor (NCAP) information model, IEEE 1451.2 m Transducer to Microprocessor Communications Protocols and Transducer Electronic Data Sheet (TEDS) formats, IEEE P1451.3--Digital Communication and Transducer Electronic Data Sheet (TEDS) formats for distributed multidrop systems, and IEEE 1451.4--Mixed-mode Communication Protocols and Transducer Electronic Data Sheet (TEDS) formats.

In the process of writing the draft document, the working group has defined the smart transducer interface module (STIM), transducer electronic data sheet (TEDS), transducer-independent interface (TII), and a set of communication protocols between the STIM and the network capable application processor (NCAP).

TABLE 5 The ten lines that make up the transducer-independent interface (Courtesy of NIST)

FIG. 46 System block diagram depicting the transducer interface (Courtesy of NIST)

A system block diagram depicting the interface is shown in FIG. 46. A STIM is specified to include up to 255 transducers, a signal converter or conditioning, a TEDS, and the necessary logic circuitry to support digital communication with NCAP. The TEDS is a small physical memory containing manufacturer's information and data for the transducer in a standardized data format. The TII, a 10-wire digital interface with provision for hot-swapping a sensor to a network, is used to access the TEDS, read sensors, and set actuators.

FIG. 47(a) depicts a STIM and the associated digital interface as described in the P1451.2-1997 hot swap. The STIM shown here is under the control of a network-node microprocessor. In addition to their use in control networks, STIMs can be used with microprocessors in a variety of applications, such as portable instruments and data acquisition cards, as shown in FIG. 47(b).

FIG. 47 (a) Hardware partition proposed by P1451.2 and (b) possible use for the interface.

The origin and function of each signal line of the ten-wire interface is listed in Table 5.

1451.2 was adopted by the IEEE as a full use standard, designated as IEEE Std. 1451.2-1997. The IEEE Std. 1451.2-1997 can be applied standalone, or it can be used with P1451.1. The two documents together will define a standard interface for networked smart sensors and actuators. Likewise, the P 1451.1 information can be implemented in a sensor control or field network without 1451.2.

The IEEE Std. 1451.2-1997 standard and IEEE P 1451.1 D2.2 draft can be ordered from the IEEE customer service department by calling 1-(800)-678-4333 (IEEE) in the United States and Canada, 1-(732)-981-0600 from outside the United States and Canada, or by faxing 1-(732)-981-9667.

10. P1451 and Practical Components

Existing microcontrollers fall short of fully implementing the standard in silicon, either because of functionality or prohibitive cost. For example, the standard transducer interface module (STIM) portion of the standard specifies the sensor interface electronics, signal conditioning, data conversion, calibration, linearization, basic communication capability, and a non-volatile 565-byte TEDS. Some microcontrollers with integrated 8-or 10-bit ADCs or comparator-based slope conversion can implement most of the STIM functionality, but are limited in conversion speed and accuracy. Moreover, few available controllers have economically integrated analog conversion together with high-density EEPROM because of the additional process complexity requirements of both functions.

These limitations are overcome by recently introduced components such as the AduC812 MicroConverter (Leonard, 1998) from Analog Devices, which integrates key STIM elements with 12-bit, 5 us data conversion on a single chip for high-accuracy, fast-conversion-time applications such as battery monitoring, pressure and temperature management, gas monitoring, and leak detection. In a typical application, the AduC812 conditions and converts signals from various types of sensors, sends signals to actuators and display devices, and communications with the host microprocessor over signal and control lines.

The AduC812 MicroConverter is supported by a development system that includes documentation, applications board, power supply, serial port cable, and software. Provided on a 3.5-inch floppy disk, the software consists of an assembler, simulator, debugger, serial downloader, and example code.