Introduction Electrical instrumentation is important for every engineering discipline. Instrumentation is the design and use of an electrical system to collect and process physically meaningful data. Examples of instrumentation applications include non-destructive testing, monitoring of biological activity, and environmental monitoring. A non-destructive testing application illustrates principles that are typical of most instrumentation systems. In non-destructive testing applications, sound or electromagnetic energy is transmitted into an object or structure. Reflected or transmitted energy is detected by a sensor. The signal output by the sensor is conditioned, then converted into a digital signal for processing by computer. The computer processing creates a two or three dimensional image of the object under test. Almost all instrumentation systems include sensors, signal conditioning, conversion into digital signals, and computer processing.

Instrumentation Examples The following are examples of instrumentation systems. These examples are organized by engineering discipline. Note, however, that there is significant overlap between disciplines in terms of applications and types of sensors.

Biomedical Engineering Electrical sensors, coupled with advanced signal processing techniques implemented using high-speed digital computer processors, have dramatically improved the ability of physicians to diagnose and treat medical conditions. They have also had a significant impact on our understanding of biological processes at the cellular and chemical level. Electrical SensorsElectrical sensors are often used to measure biological electrical activity. Electrical sensors typically consist of electrodes that are attached to the skin; voltage differences between electrodes are related to electrical activity deeper within the body. For example, electroencephalograpy (EEG) measures voltages associated with brain activity. Electrocardiograms (ECG or EKG) measure voltages associated with cardiac function. Ultrasonic SensorsUltrasonic imaging provides a non-invasive method of imaging almost any body tissue using high-frequency sound pulses. A typical ultrasonic imaging system contains a transducer that both transmits the sound pulses into the body and receives the sound echoed from the body. The transducer converts the received sound pulses into voltages, which are then processed using a computer to create an image of the body tissue. X-ray SensorsX-ray imaging, and in particular Computed Axial Tomography (CAT), provides a three-dimensional image of tissue. In this process, X-rays are passed thorough the body and received by a very sensitive X-ray detector. The detector converts the received X-ray intensity into a voltage, which is then processed by a computer to create the three-dimensional image of the tissue. Magnetic Resonance ImagingMagnetic resonance and other related imaging techniques use strong magentic fields to orient the spin of atoms in the material to be imaged. As the fields are changed, the atoms emit small voltages that are sensed and converted into images.

Chemical Engineering We present sensors that are important for chemical process control, including mass flow sensors and temperature sensors. We also discuss sensors used in spectroscopy. Mass Flow SensorsMass flow sensors may output a voltage that is proportional to the rate of mass flow through the sensor. They may be used to monitor and control the flow of gasses within a chemical plant; they may also be used to monitor anesthesia delivery. Temperature SensorsTypes of temperature sensors include thermocouples, which produce a voltage that is proportional to the temperature difference between the two ends of the thermocouple; thermistors, whose resistance is proportional to temperature; and resistance temperature detectors (RTD), whose resistance is proportional to temperature. Chemical SensorsSensors are used to detect chemicals in many different applications. For example, chemical sensors are being installed in major airports to detect explosives in checked luggage. Chemical sensors are under development to detect the explosives in land mines, allowing mines to be detected and disarmed. SpectroscopyThere are many different approaches to spectroscopy. In infrared absorption spec- troscopy, chemical compounds are illuminated by broad-spectrum infrared light. The chemicals absorb light at particular wavelengths that are correlated with particular chemical structures (usually bonds). By examining the wavelengths that are absorbed, much about the structure of a chemical compound can be deduced. A spectrometer employs a sensor that converts light intensity (usually) into a voltage.

Civil Engineering Civil Engineering sensor applications include non-destructive testing, monitoring of structures, and environmental monitoring. Non-destructive Testing and MonitoringDepending on the structure to be tested, sound or electromagnetic energy may be used to image the structure. If sound is used, then ultrasonic transducers transmit sound energy into the structure and detect reflected energy. If electromagnetic energy is used, a source transmits energy into the structure and a detector measures the amount of energy that is transmitted through the structure. Monitoring of structures may be done using strain gauges. A strain gauge monitors expansion or compression of a component in a structure, and can be used to measure forces and loading. Typical strain gauges have a resistance that is proportional to strain. Environmental MonitoringSensors are used to detect and monitor many different environmental pollutants. Detected pollutants include metals and organic compounds.

Computer Systems Engineering and Electrical Engineering Both of these disciplines use instrumentation to measure voltages, currents, or resistances associated with electrical circuitry. In addition, the development of sensor systems and the computer algorithms to evaluate the data collected from sensor systems falls within the range of electrical and computer engineering. Military SensorsThe military uses a wide range of sensors to monitor potentially hostile forces, track both friendly and hostile vehicles and troop movements, and to even to guide weapons to targets. These sensors collect electromagnetic, acoustic, seismic, and other signals from a wide variety of sources. Still images and video (both from optical and infrared sensors) must be interpreted to provide real-time information to battlefield commanders and analysts. Biometric IdentificationThere is an increasing need to be able to identify people based on biological information such as voice patterns, retinal patterns, facial features, finger prints, and electronically captured signatures. Sensors are used to collect this information; signal processing is used to process the information so that individuals can be identified or their identity can be verified.

Mechanical and Aerospace Engineering One application of sensors in Aerospace engineering is the development of fly-by-wire control systems for aircraft and other vehicles. Another mechanical engineering application of sensor technology is the control of modern internal combustion engines. Yet another is the use of strain gauges to measure loading of mechanical structures; this application is discussed above. Fly-by-wireA significant design advance for military and commercial aircraft is the concept of fly-by-wire. Unlike older aircraft designs, in which the controls are mechanically or hydraulically connected to the actuator surfaces, in a fly-by-wire system the actuator surfaces are controlled by electrical signals generated from sensors in the aircraft. The sensors include those that measure aerodynamic quantities that affect the aircraft performance and control, as well as those that measure the pilots’ control actions. Engine ControlTemperature, pressure, chemical, and mass-flow sensors are used in the control of modern internal combustion engines. The outputs of these sensors (typically voltages) are converted from analog to digital values and then input to a computer. The computer adjusts throttle settings and other controls within the engine to obtain the desired performance.

Components of a Typical Instrumentation System shows a block diagram of the components of a typical instrumentation system. These components include: Sensorconverts the measured value into an electrically useful value. Signal Conditioning"conditions" the signal from the sensor so that it can be sampled by the A/D converter. A/D Converterconverts the signal into a digital format. Computerprocesses, displays, and records the signal. We now describe each component in more detail.

Components of a typical instrumentation system.

Sensor The output of a sensor is a function of the physical quantity to be measured. Sensor outputs may be a voltage or current (from some temperature and pressure sensors, for example), a resistance (from a strain gauge, for example), or a frequency (from an accelerometer, for example).

Signal Conditioning The output of the signal conditioning is (usually) a voltage; the signal conditioning converts the sensor output to this voltage. The signal conditioning circuitry usually includes an amplifier. Amplifiers are characterized in terms of attributes such as gain, bandwidth and/or frequency response, linearity, harmonic distortion, and input and output impedance. In a typical instrumentation system, the gain of the amplifier is set so that the output voltage falls between lower and upper limits (for example, -10V to 10V) determined by the A/D converter. One commonly used type of amplifier is the Operational Amplifier (OpAmp). Op Amps have differential inputs: the output voltage is the amplified difference of two input voltages. Op Amps have very large gains (typically larger than 10 5 ). Most op amp circuits use negative feedback to achieve desired signal gains. Op amp circuits can be designed to provide voltage gain or attenuation, convert current to voltage, integrate or differentiate, and filter out noise or interference.

A/D Converter Analog-to-digital conversion consists of two operations: Samplingmeasuring the voltage signal at equally spaced points in time. Quantizationapproximating a voltage using a specified number of bits (usually 8 to 16)

Instrumentation Issues Issues in instrumentation systems include noise and signal bandwidth.

Noise Noise is undesirable variation in a signal. illustrates the effects of noise on a signal. Noise reduces the accuracy and repeatability of measurements and introduces distortion in signals. It creates errors in control systems.

(a) A signal without noise and (b) a signal with noise. There are several different types of noise: Thermal noise caused by the random thermal motion of charged particles in electrical components in the sensor and the amplifier. Electromagnetic noisefrom electrical wiring (e.g. 60Hz noise), electrical equipment (e.g. computers) or communication devices. Shot noise from quantum mechanical events. We reduce the undesirable effects of noise by grounding and shielding electrical connections, filtering (smoothing out variations in the signals), and averaging several measurements.

Signal Bandwidth Conceptually, bandwidth is related to the rate at which a signal changes. As illustrated in , a high bandwidth signal changes rapidly, while a low bandwidth signal changes more slowly. The rate of change of the signal in turn affects the rate at which the signal must be sampled; higher rates of change require more samples per unit time, as illustrated in .

(a) A high bandwidth signal; (b) a low bandwidth signal.

More samples are necessary to represent (a) a high bandwidth signal than (b) a low bandwidth signal. Every component in the instrumentation system has bandwidth limitations. Sensors do not respond immediately to changes in the environment. The amplifier output does not change immediately in response to changes in the input. The A/D converter sampling rate is limited.