The term "Plug-and-Play Sensor" has a reassuring resonance, implying an easy-to-install—and use—smart sensor component you simply plug into a special socket. No struggling with printed data sheets, instruction manuals, and keyboarding; the newly matched pair of high-tech objects does all the work. The sensor and your favorite data acquisition (DA) system are now successfully exchanging information.
 Figure 1. The basic functional components of a plug-and-play smart sensor are defined by the IEEE 1451 standards. |
The plug-and-play (P&P) concept has been popularized by the PC industry to refer to the interconnection of computers and peripherals (printer, scanner, fax, etc.) with little or no effort on the part of the user. Despite some initial glitches, the computer P&P features embodied in open standards have become accepted and, to a great extent, even required. Its accessibility and ease of use for both the experts and the technically challenged are also appealing to the sensor industry, where the benefits are lower costs for installation and maintenance, and a longer life cycle.
Just What Is a P&P Sensor?
In the sensor industry, P&P denotes an enhanced sensor component with machine-readable information that enables the automated setup of a sensor interface. The setup process typically entails connecting a sensor to a DA system, a control system, a data network, or some other form of higher level system that requires sensor data. Collecting and interpreting these signals requires knowledge of certain key sensor properties and application details.
 Are P&P Sensors Smart? |
For the simplest case, these details include the sensor calibration factor, which converts measured electrical signals into the desired measured quantity, and the physical units of the measured value. This simple case assumes an ideal linear sensor and may be a useful approximation for many applications. Others might require more sensor details such as operational dynamic range, offset signal magnitudes generated when no measurand is present, frequency response, corrections for nonlinearities, compensation for temperature effects, and correction for aging.
This collected set of information is often called the Transducer Electronic Data Sheet (TEDS). Although each protocol has its own definition of fields found in the TEDS, the IEEE 1451 standards identify several types of required and optional TEDS that can offer more. With minimal user involvement, P&P sensors communicate some or all of these factors automatically when the sensor is physically connected to the higher level system (see Figure 1).
 The IEEE 1451 Family of Smart Sensor Standards |
P&P sensors are responsible for transferring sensor-specific properties to higher level systems, and therefore need a set of standard rules or protocols that tell each side of the communication channel what to expect from and send to the other. These protocols have been the object of much discussion and investigation. One set of standards is the IEEE 1451 [1], established by the Institute of Electrical and Electronics Engineers and described in the sidebar, "The IEEE 1451 Family of Smart Sensor Standards". Other organizations and application-specific industry segments have developed their own protocols. There is no "best" solution; the requirements of each application will help determine the best P&P approach.
These standards specify the details of the content transferred during sensor setup, the data format, and the technical information about the communication process between the sensor and the higher level system. Some systems intended for P&P sensors have an automatic discovery process that continually scans its ports to determine if a new device has been attached. Others require the system to initiate a setup process after the sensor is electrically connected. Once the setup parameters are transferred, the sensor's analog and/or digital signals can be converted into a calibrated physical quantity with known physical units based on the transferred information. Depending on how smart the P&P sensor is, it can transmit other relevant information such as sensor location, serial number, status, and most recent calibration date (see the sidebar, "Are P&P Sensors Smart?"). If the sensor is connected to a network, connection issues must also be addressed in the feature set; if it is not, other signal interface issues may need attention.
Architecture
P&P sensors can be designed with different levels of electronic integration, as governed by market and technology constraints. At one end of the spectrum, the sensing element and its intelligence—communications electronics, memory registers, and sensor signal processing circuits—can be monolithically integrated onto a single silicon die. This approach can yield the least expensive sensors if manufactured in numbers great enough to amortize the nonrecurring engineering costs associated with designing and fabricating a custom IC. One caveat is that the sensor's operational environment must be compatible with semiconductor (typically silicon) materials technology. For instance, if the sensor is intended to operate at 600ºC, a monolithically integrated solution would be inapropriate. A hybrid or discrete P&P approach would be needed to separate the electronics from the sensing element if they are to survive and do their work.
Many sensing elements cannot be fabricated with semiconductor IC materials or technology, and require a hybrid approach to integration using techniques ranging from special hybrid ICs to discrete circuit boards.
In the former, ancillary silicon dies for communications, memory, and logic functions are packaged with the sensing element so that all circuit elements are attached to a common substrate and then enclosed in a single circuit package.
Discrete circuit boards have separately packaged IC devices to provide the electronics. The sensing element and the support ICs are typically attached to a common board, although for some extreme application environments there may be a separation between the electronics and the sensing element.
At the opposite end of the integration spectrum is the virtual implementation of P&P sensors. Here, sensor property information is retrieved from a separate data source such as an Internet connection to a manufacturer's Web site or a CD-ROM disk. Instead of making hardware modifications to the sensor or supporting circuitry, a virtual P&P sensor shifts the burden of gathering sensor properties information to the higher level system. Depending on the sensor's features, the DA system automatically transfers an ID number from the sensor to the system, or the installer will manually enter the ID for the attached sensor. The higher level system then downloads the information from the user-identified external data resource (e.g., Web site, CD-ROM, or hard disk) and extracts calibration constants and other data for interpreting sensor signals.
One advantage of this approach is that implementation requires limited or no circuit changes to the sensor; one drawback is that less application-specific information is available about it. For example, calibration data obtained from the external source may not be for the specific sensors in the installation but rather a general calibration coefficient for all similar sensors of the same model. Furthermore, details of an application, installation, or other information specific to the sensor may not be available in the virtual P&P sensor data resource.
Technology Hurdles
Despite the progress in P&P sensors, some factors remain to be resolved. Certain information about a particular installation may be useful for remote viewers of the sensor data. For example, the precise functional location within a system such as, "First elbow connector on coolant flow at auxiliary input line #3," as well as the broader geographic location, "Process facility at 141 Main St., Oak City, IL," are two fields a user might include with other, more standard sensor properties information. This type of information enhances the value of the sensor data, but the inclusion of local information complicates the P&P implementation.
Another type of sensor-specific information deals with recalibrating a particular device after aging has made the original calibration factors questionable. Will there be a simple way to recalibrate a P&P sensor that's been sitting on a shelf for a year? What about an installed sensor that's been operating for a year? And will virtual P&P sensors accommodate aging? You need to be aware of how each vendor's implementation takes these questions into account. The answers are highly dependent on the specific details of the design and implementation of the attached higher level system.
Future Directions
Several forces in the marketplace and the industrial technology labs are encouraging the development and deployment of P&P sensors, smart sensors, and networkenabled sensors. As industry standards such as the IEEE 1451 series are completed, they will provide a recipe for building smart sensor interfaces that will be compatible with other hardware and software.
The spread of network and wireless technologies has also enhanced the demand for smarter sensors. Both need key smart sensor features for better network integration. For example, a remote user's simple command from across a network to read a sensor's current value could be trivial or exponentially difficult, depending on the number of sensors queried and whether sensor standards (with common command sets) are used. In addition, sensor properties information will soon be required as the number of sensors connected to networks grows. A lack of details about the identity of which sensor is providing data across the network diminishes the network's value.
Summary
If there are any lessons the PC world can teach us about the value of plug-and-play technology, we can be sure that this new feature will soon be a standard requirement on most of the sensors we design, make, or buy.
For an online list of manufacturers and vendors of plug-and-play sensing devices, please visit www.sensorsmag.com/pnplist.
Reference
1. IEEE 1451.2-1997, IEEE Standard for a Smart Transducer Interface for Sensors and Actuators–Transducer to Microprocessor Communication Protocols and Transducer Electronic Data Sheet (TEDS) format, IEEE Instrumentation and Measurement Society, TC-9, SH94566, Sept. 25, 1998.
James Wiczer, Ph.D., can be reached at Sensor Synergy, Buffalo Grove, IL; 847-353-8200, [email protected], www.sensorsynergy.com.