Dr Stephen Redmond School of Electrical Engineering & Telecommunications Ph: Rm: 458, ELECENG (G17)

Transcription

1 ELEC4623/ELEC9734: Semester 2, 2009 Dr Stephen Redmond School of Electrical Engineering & Telecommunications Ph: Rm: 458, ELECENG (G17) Medical Instrumentation (Webster): Chapter 2 Notes: on soon Session 2, 2009 ELEC4623/ELEC9734 1

2 Biomedical Instrumentation, Measurement and Design ELEC4623/ELEC9734 Lectures 5 & 6 Principles and Operation of Basic Transducers and Sensors Session 2, 2009 ELEC4623/ELEC9734 2

3 Outline Transducers Performance characteristics Resistive transducers Potentiometers Strain gauges Differential transformers (LVDT) Inductive transducers Capacitance transducers Piezoelectric transducers Bridge circuits DC Bridges Typical bridge configuration AC Bridges Photoelectric transducers Photoemissive tube Photovoltaic cell Junction cell Photoconductive cell Phototransistor Temperature sensors Resistance based Thermoelectric thermocouples Radiation thermometry Fiber optic sensor Optical biosensors Pulse oximetry Session 2, 2009 ELEC4623/ELEC9734 3

4 Transducers A transducer is a device which converts one form of energy to another A sensor is a type of transducer (with. an electrical output) Electrical/computing/biomedical engineers we are mainly interested in electrical transducers (sensors) for. physiological monitoring Transducer Sensors Actuators Physical parameter Electrical Input Electrical Output Physical Output e.g. Piezoelectric: Df Deformation -> >Voltage Voltage -> Deformation Session 2, 2009 ELEC4623/ELEC9734 4

5 Sensor performance characteristics Transfer Function (TF) Relationship between physical input signal and electrical output Usually, this relationship is represented as a graph showing the relationship between the input and output signal as a function of frequency (frequency response) For electrical systems frequency response is often dimensionless For transducers this is not true the TF might be in V/m, or V/ C etc. Alternatively the transfer function could be represented by a rational equation in s or jω (or even z for digital sensors) Y is usually the output and X the input For practical sensors, often n = 0,1 or 2 and m = 0 m m 1 bs m + bm 1 s bs 1 + b0 m m 1 m + m Y () s = X() s a s a s... as a Session 2, 2009 ELEC4623/ELEC9734 5

6 Sensor performance characteristics Sensitivity Defined in terms of the relationship between input physical signal and output electrical signal Generally the ratio of a small change in electrical output signal to a small change in physical input signal Thus it may be expressed as the derivative or slope (dy/dx) of the input-output characteristic with respect to physical signal (if relationship is nonlinear) or by TF if it a linear system Example: A Thermometer would have "high sensitivity" if a small temperature change resulted in a large voltage change. Typical units : Volts/Kelvin Session 2, 2009 ELEC4623/ELEC9734 6

7 Sensor performance characteristics Span or Dynamic Range The range of input physical signals which may be converted to electrical signals by the sensor Signals outside of this range are expected to cause unacceptably large inaccuracy (perhaps due to nonlinearity) This span or dynamic range is usually specified by the sensor supplier as the range over which other performance characteristics described in the data sheets are expected to apply Accuracy Generally defined as the largest expected error between actual and ideal output signals Sometimes quoted as a fraction of the full scale output Example: a thermometer might be guaranteed accurate to within 5% of FSO (Full Scale Output) Typical Units: Kelvin (for thermometer) Session 2, 2009 ELEC4623/ELEC9734 7

8 Sensor performance characteristics Hysteresis Some sensors do not return to the same output value when the input stimulus is cycled up or down States depend on their immediate history (i.e. memory) The width of the expected error in terms of the measured quantity is defined as the hysteresis Session 2, 2009 ELEC4623/ELEC9734 8

9 Sensor performance characteristics Nonlinearity (or linearity!): The maximum deviation from a linear response over the specified dynamic range Noise All sensors produce some output noise in addition to the output signal The noise of the sensor limits the performance of the system Noise is generally distributed across the frequency spectrum White noise is evenly distributed (uniform noise spectral density) Many common noise sources produce a white noise distribution Coloured noise contributes differently at different frequencies Session 2, 2009 ELEC4623/ELEC9734 9

10 Sensor performance characteristics Resolution The minimum detectable signal fluctuation (dependent on noise) Bandwidth All sensors have finite response times to an instantaneous change in physical signal -> >low-pass characteristic Decay times represent the time after a step change in the input signal for the sensor output to return to its original value (only with AC coupled sensors) -> high-pass characteristic The reciprocal of these times correspond to the upper and lower cutoff frequencies, respectively. The bandwidth of a sensor is the frequency range between these two frequencies Session 2, 2009 ELEC4623/ELEC

11 Resistive sensors - potentiometers Potentiometers measure displacement either linear or rotational Can excite with AC or DC voltages Output is linear (± 0.01% of FSO) provided not electrically loaded Frictional and inertial components of such potentiometers should be low in comparison with activation force to minimise dynamic distortion Session 2, 2009 ELEC4623/ELEC

12 Resistive sensors - strain gauges Resistance R is related to length l and area of cross-section of the resistor A and resistivity β of the material as: R = βl A βdl 2 dβ dr = β A lda + l A A dr dl da d β = + R l A β Poisson s ratio µ, relates the change in diameter D to the change in length l as dd dl so: = μ dr dl dβ D l = 1+ 2μ + R ( ) l β π A= da= = 4 4 A D 2 2π DdD da 2 dd D Dimensional effect Piezoresistive effect Session 2, 2009 ELEC4623/ELEC

13 Resistive sensors - strain gauges The Gage factor is fractional change in resistance G = fractional change in strain Δ R / R Δ β / β G = = ( 1+ 2μ ) + ΔL / L ΔL/ L G is a measure of sensitivity i i G primarily due to dimensional effects for metals For semiconductors, the piezoresistive effect dominates But higher G for semiconductor devices is offset by their higher temperature coefficient of resistivity (see next slide) Applications of strain gauge: respiration detection and other plethysmography pet ogap y( (volume change detection) Session 2, 2009 ELEC4623/ELEC

14 Properties of strain gauge materials Material Composition Gage Factor Temp coefficient of resistivity ( C -1 *10-5 ) Constantan (advance) Ni45, Cu ±2 Isoelastic Ni36, Cr8, (Mn, Si, Mo)4 Fe to Karma Ni74, Cr20, Fe3 Cu Manganin Cu84, Mn12, Ni4 0.3 to 0.47 ±2 Alloy 479 Pt92, W8 36t 3.6 to ±24 Nickel Pure -12 to Nichrome V Ni80, Cr to Silicon (p type) 100 to to 700 Silicon (n type) -100 to to 700 Germanium (p type) 102 Germanium (n type) -150 Session 2, 2009 ELEC4623/ELEC

15 Resistive sensors - strain gauges Instruments that use strain gages of semiconductor materials often incorporate temperature compensation Strain gauges may be bonded or un-bonded Bonded - to surrounding (backing) material Unbonded free to stretch Unbonded d strain gauge can be made of 4 sets of strainsensitive wires connected to form a Wheatstone bridge Sketch equivalent circuit Session 2, 2009 ELEC4623/ELEC

16 Resistive sensors - strain gages A bonded strain-gauge element, consisting of a metallic wire, etched foil, vacuum deposited film, or semiconductor bar as shown below, is cemented to the strained surface Deviation from linearity of these bonded gauges is typically 1% and compensation for temperature variations is made by using a second identical strain-gauge, which is not under strain It is common to use bonded strain-gauges in four-arm bridges as this provides for some inherent temperature compensation as well as yielding up to four times larger output signal Session 2, 2009 ELEC4623/ELEC9734

17 Resistive sensors - strain gages Strain gages are generally mounted on cantilevers and diaphragms and measure the deflection of these materials Session 2, 2009 ELEC4623/ELEC

18 Resistive sensors - strain gages Elastic-resistance strain gauges are also used extensively in biomedical applications, especially in cardiovascular and respiratory dimensional or plethysmographic (volume change) measurement These simple systems normally consist of silicone rubber tubing filled with mercury or with an electrolyte or conducting paste. (NOTE: Mercury is being phased out! Google: As mad as a hatter ) The ends are sealed with metal electrodes and as the tube stretches, the diameter decreases and the length increases thus causing an increase in resistance The resistance/unit length of typical gauges is ohm/cm. The elastic strain gauge is linear within 1% for maximum extensions of 10% of their length but for larger extensions the linearity rapidly falls Session 2, 2009 ELEC4623/ELEC

19 Inductive displacement transducers Self-Inductance (L) is the property of a circuit element to produce a back-emf to resist a change in magetic flux For a coil with n turns, enclosing a total flux of Φ=BA (flux density x area of loop) the inductance is: L = nφ i Faraday s law of induction states: v = n di dt = dφ dt i.e. for constant rate of change we get a constant voltage Session 2, 2009 ELEC4623/ELEC v L

20 Inductive displacement transducers Mutual-Inductance is the property of a circuit element to produce a back-emf to resist a change in magetic flux in another circuit element Total flux enclosed by a loop is the sum of the fluxes from the loop and its neighbouring loops Consider two loops: L 12 = L 21, are the mutual inductances between the coils Any change in either current induces a voltage Session 2, 2009 ELEC4623/ELEC

21 Inductive displacement transducers (a) Self inductance (b) Mutual inductance (c) Differential transformer Any inductor L can be used to measure displacement by varying any of the coil parameters: Can be affected by external magnetic fields Differential transformer: Output Vcd = Vce Vde (since one coil is reversed) e Session 2, 2009 ELEC4623/ELEC

22 Linear variable differential transformers (LVDT) Probably the most widely used displacement transducers Very robust, can detect displacements in the nanometre (nm) range Formed from a transformer in which the coupling between the primary and a split secondary depends on the position of a ferromagnetic plunger as shown below The two halves of the secondary winding are connected in series opposition. Thus the output is zero when the plunger is in the centre position and increases on either side but with opposite phase Session 2, 2009 ELEC4623/ELEC9734

23 Linear variable differential transformers (LVDT) Oscillator frequency is usually between 60 Hz and 20 Khz. Commercial units have typical responsivity (sensitivity) in the range of mv/volt excitation/millimetre displacement, full scale displacement of mm and a linearity of ± 0.25%. As the core moves through the null position, the phase changes by 180 whilst the magnitude of the output voltage Vo is proportional to the displacement x An ordinary rectifier-demodulator cannot distinguish between directions and hence a phase sensitive demodulator is required Session 2, 2009 ELEC4623/ELEC

24 Capacitive transducers The capacitance C of a parallel plate capacitor of area A and separation d, containing a dielectric of relative permittivity ε r is given by C = ε rε 0 A d ε 0 is dielectric permittivity of free space The capacitance may thus be modified d by changing ε, A or d, giving rise to the three basic types of capacitive displacement transducers (see next slide) (a) Variable permittivity (b) Variable area (c) Variable separation In practice three plate capacitive transducers (d) are used as these relate displacement linearly to the differential change in capacitance Session 2, 2009 ELEC4623/ELEC9734

25 Capacitive transducers Q: For (d), given: Show that: ε0εr A ε0εr A C1 =, C2 = d x d + x x C 1 C = 2 d C 1 + C 2 Session 2, 2009 ELEC4623/ELEC9734

26 Piezoelectric sensors What is piezoelectricity? Strain/stress causes a redistribution of charges and results in a net electric dipole (a dipole is kind of a battery!) Biomedical applications: Accelerometer (motion, posture) Microphone (heart sound, BP) Ultrasound (imaging, blood flow) A piezoelectric material produces voltage by redistributing charge under mechanical strain/stress Session 2, 2009 ELEC4623/ELEC

27 Piezoelectric sensors Generate an electrical potential when we apply mechanical strain (sensor) An applied electrical potential results in a physical deformation of the material (actuator) Quartz is the best known naturally occurring piezoelectric material Flexible piezoelectric plastic polymer film (polyvinylidine fluoride - PVDF) is a recent development Session 2, 2009 ELEC4623/ELEC

28 Piezoelectric transducers For piezoelectric material, charge density q is directly proportional to applied force f : q = kf k = thepiezoelectric strain coefficient is typically 2x10-12 C/N for quartz Since the opposite surfaces are metallised, a capacitor is formed The voltage corresponding to the charge produced d is given as: kf v = = C kfx Aε r ε 0 A better equation describing the physical processes involved is simply (K = proportionality p constant in C/m, and x = deflection in m) q = Kx Session 2, 2009 ELEC4623/ELEC

29 Piezoelectric transducers (a) equivalent circuit of piezoelectric sensor where C S and R S = sensor capacitance and leakage resistance, R A and C A = amplifier input resistance and capacitance, C c = cable capacitance, q = charge generator R s C s C c C a R a (b) modified equivalent circuit with current generator replacing charge generator Note that for voltage amplifier as shown, amplifier and cable input characteristics are about same magnitudes as sensor so affect circuit C R Require a charge amplifier (or integrator) Session 2, 2009 ELEC4623/ELEC

30 Piezoelectric transducers Typical frequency response curve of piezoelectric transducer Exhibits mechanical resonance at high frequency (modeled by R, L and C in series) useful for clock frequency in oscillators can be suppressed by low-pass (LP) filter Input impedance of amplifier and capacitance of transducer + cable govern high-pass (HP) cutoff Session 2, 2009 ELEC4623/ELEC

31 Charge amplifier This is basically an integrator Create a short circuit to virtual ground of opamp Use feedback capacitor to capture current This converts charge to voltage Session 2, 2009 ELEC4623/ELEC

32 Wheatstone bridge Ideal for measuring small changes in resistance Output V G is zero when the bridge is balanced and R1/R2 = R3/Rx Resistance type transducers may be connected to one or more arms of the bridge circuit and the variation detected either by null-balance or deflection-balance circuits The first method uses a variable resistance in another branch to balance out the output and the other uses the actual amount of bridge unbalance to determine the change in transducer resistance Session 2, 2009 ELEC4623/ELEC9734

33 Wheatstone bridge Linear! Assume that all resistance values in the bridge are initially equal to R 0 and that R 0 << R i (the input resistance of the meter). If R 1 and R 3 increase by ΔR, and R 2 and R 4 decrease by ΔR, then Δ V = 0 ΔRV R. i 0 If the meter s internal resistance R i is now included in the calculation This reduces to the previous expression if R 0 << R i Use Thevenin equivalent to calculate: ( ΔR R / Ro ) Vi Δ Vo = ( R / R)[1 + ( Δ R/ R ) ] o i o Session 2, 2009 ELEC4623/ELEC

34 Wheatstone bridge It is common practice to incorporate a balancing scheme in bridge circuit as shown by introducing resistor R Y and potentiometer R X as shown To minimize i i loading effects R X is approximately 10 times the resistance of the bridge leg and R Y is selected to limit the maximum adjustment Strain gauge applications normally use a value of R Y ~25times the resistance of the bridge leg. AC balancing circuits are more complicated because a reactive as well as resistive imbalance must be compensated Session 2, 2009 ELEC4623/ELEC9734

35 Equivalent circuits Wheatstone bridge Typical bridge configurations I) Double push-pull: all four arms are identical transducers, with R 2 = R 3 = R 1 = R 4 = Rand ΔR 2 = ΔR 4 = - ΔR 3 = - ΔR 1 = ΔR ii) Single push-pull: R 2 and R 3 are transducers with R 2 = R 3 = R and ΔR 2 = - ΔR 3 = ΔR R 1 and R 4 are fixed and equal usually to R. iii) Simple transducer: R 2 is the only transducer R 1 = R 2 = R 3 = R 4 = R will cause a non-linear output for large changes in R 2 All for small changes in ΔR Session 2, 2009 ELEC4623/ELEC

36 AC Bridges R/2 R AC bridges are usually formed, by replacing the resistances in R2 and R3 in the DC bridge by a transformer This gives better isolation between the oscillator and the bridge and permits the two sides of the bridge to be exactly equal and opposite in excitation Another advantage is that the bridge requires only two resistors, both of which are usually transducers A single ended output is easily obtained relative to the centretap of the transformer V S V S 2 V S 2 R V S δr 2 R AC resistive bridge and its equivalent circuit Session 2, 2009 ELEC4623/ELEC

37 Capacitive and inductive bridges and equivalent circuits L/2 Inductive and capacitive bridges can be similarly obtained V S L1 R 1 L 2 R 2 V S 2 δl L Push-pull types are preferred and result in the configurations and equivalent circuits shown right L1 = L 2 = L,R1 = R 2 δ L1 = δ L 2 = δ L 1 1 : R C 1 R/2 2C V S V S δc 2 C C 2 Session 2, 2009 ELEC4623/ELEC

38 Photoelectric Transducers Employed in two ways in the measurement of physiological events in living subjects Photosensor functions as a detector of the changes in intensity of light at a given wavelength. Photosensor detects changes in light intensity in which wavelength is relatively unimportant. These functions are achieved by five different types of photoelectric transducers: photoemissive tube photovoltaic cell photodiode (junction cell) photoresistor (photoconductive cell) phototransistor Session 2, 2009 ELEC4623/ELEC

39 Photoemissive Tube An evacuated or gas filled bulb with two electrodes (the cathode and anode) On the cathode there is a coating of a specially prepared material that releases electrons when illuminated For electron emission to occur, there are certain restrictions on the type of surface and the wavelength of the impinging g light. Electron emission is only possible if the wavelength is shorter (energy greater) than a certain threshold value (energy required to release an electron from the metal - work function) Session 2, 2009 ELEC4623/ELEC9734

40 Photoemissive Tube A relatively high voltage (10-200V) must be applied between the two electrodes The electrons released by light quanta are attracted to the anode The electron flow constitutes a current that is linearly proportional to the intensity of the incident light Both gas filled and vacuum photoemissive detectors respond quickly to changes in light intensity Gas filled response times are approximately 10 msec and for a vacuum approximately 10 nsec Note also that both exhibit a small current flow with no light (dark current). Typically 10-7 to 10-9 A Session 2, 2009 ELEC4623/ELEC

41 Photovoltaic Cells Develops a voltage that can drive a substantial current through a galvanometer or other low impedance circuit. Consist of a thin coating of some non-metal like selenium or silicon on an iron or steel backing. Above this is a thin transparent film of metal. The film and selenium are insulated from each other, and this region constitutes the barrier layer. When the barrier layer is illuminated, light quanta are absorbed, electrons are released, and a potential difference appears across the barrier. The transparent metal film becomes negative and selenium positive. Completion of the circuit between the two electrodes causes a current to flow The resistance between the electrodes decreases with illumination Session 2, 2009 ELEC4623/ELEC9734

42 Spectral characteristics of photovoltaic cells As we can see from the graph, selenium (Se) cells have a spectrum that covers the visible spectrum These devices, with a filter, have a spectral curve closely resembling that of the human eye Session 2, 2009 ELEC4623/ELEC

43 Photovoltaic Cells The relationship between light intensity and the voltage developed by the photovoltaic sensor is not linear if the device is operated without a resistive load If a resistive load is placed across the device, the current flow becomes more linearly related to light intensity as the resistance is decreased Hence, it is necessary to employ a low resistance galvanometer or measuring circuit to indicate light intensity if a linear scale is to be obtained The most undesirable feature of the photovoltaic cell in biomedical applications is its sensitivity to temperature changes as the load resistance is varied Many cells have an optimum value for the resistance that can be connected across the device to minimise the sensitivity change with temperature, but not necessarily the one that yields the maximum power transfer or linearity of current with light intensity Session 2, 2009 ELEC4623/ELEC

44 Photodiodes (Junction Cells) Consist of a P-N junction and produce a potential across the junction due to the creation of hole-electron pairs, which are direct result of the absorption of light quanta Usually P-I-N used to spread depletion electric field across wider distance Photoconductive (reverse bias) mode give linear response to intensity Many different types of junction exist, each designed to have a desired spectral sensitivity The response of junction photocells is short enough (ms to ns) to be used as light detectors in which there is a rapid modulation of illumination (e.g. pulse oximetry) Because of their small size, high sensitivity, and short response time, junction photocells see a wide variety of uses as light detectors. One use, that t is very important t in biomedical engineering, i is in optically coupled isolators In these devices a photo-junction cell is used to detect the light produced by an LED, mounted a few mm away. Thus, the desired signal modulates the light produced by LED, and the light is detected by the junction photocell In this way, the signal is coupled from one circuit to another without any electrical interconnection, thereby providing excellent electrical isolation Session 2, 2009 ELEC4623/ELEC

45 Phototransistors A considerable increase in current sensitivity but also increase in response time The incident radiation is caused to strike the base region, which is intentionally left unconnected Holes generated in the base region by the radiation cause the base potential to rise, forward biasing the base-emitter junction Electrons flow into the base from the emitter to neutralise the extra holes. Because of the close proximity of the collector junction, the probability of an electron combining with a hole is small, and most of the electrons are immediately drawn into the collector region, which is at a high positive potential As a result, the total collector current is much larger than the photo-generated current (β times as large) Session 2, 2009 ELEC4623/ELEC

46 Photoconductive cells (photoresistors) Consist of a thin film of a material (selenium, germanium, silicon, or a metal halide or sulphide) 100K When exposed to certain types of radiant energy, exhibits the 10K photoconductive phenomenon (decrease in resistance with light) In most photoconductive cells the conductance is nearly linear with high intensity Such devices are extremely 100 sensitive photo-detectors and are often employed as light-controlled switches Resis stance, R Resistance R Conductance 10K 1K 1K Illumination (ft c) 10 Conductanc ce ( mho) Session 2, 2009 ELEC4623/ELEC

47 Comparison of Photoelectric Transducers Most important characteristics when evaluating various photoelectric sensors: spectral sensitivity (wavelength) response time type of output (voltage or current) linearity with illumination intensityi Spectral response of the human eye for night and day vision, along with the spectral sensitivities of photocells, photoconductors, and photoemissive tubes The numbers in brackets identify the spectral peaks for the S curves: P-T denotes a phototransistor Session 2, 2009 ELEC4623/ELEC

48 Photoelectric sensor comparison Type Advantages Disadvantages Photoemissive Wide spectrum, high h sensitivity, fast response, linear response. Requires high voltage. Only produces a small current when illuminated. Photovoltaic Low voltage for appreciable Slow response affected by current. temperature changes Photodiode Small size, fast response, (Junction cell) high sensitivity. Photoconductive cell (Photoresistor) Phototransistor Large resistance change with illumination, good sensitivity. High sensitivity, high efficiency. Thermal noise voltage Problems in relating signal and light intensity linearly Slow response Session 2, 2009 ELEC4623/ELEC

49 Biomedical application of photosensors: Absorption spectroscopy Many molecules in the human body absorb/emit light at various wavelengths In some interesting compounds the absorption/emission changes based on an altered property of the molecule Haemoglobin is one interesting example (bottom right figure) Session 2, 2009 ELEC4623/ELEC

50 Pulse Oximetry Detected light reflects concentration of absorbing substance, based on Beer-Lambert Law: I 0 = Incident light intensity I = Transmitted light intensity c = Concentration of absorber (include arterial and venous blood + tissue) ε = Extinction coefficient of absorber (function of wavelength) d = Path length of absorber Finger I = I 0 e LED Photodetector εdc IR light Typical sites: finger, ear, forehead Signal consists of pulsatile (AC) and non-pulsatile (DC) components Session 2, 2009 ELEC4623/ELEC

51 Pulse Oximetry Often quoted as SpO 2 : refers to the pulse oximeters estimate of SaO 2 (arterial oxygen saturation) [ HbO2 ] [ Hb] + [ ] SaO (%) 2 = HbO This is usually empirically determined (linear fit) from the ratio of red to IR fractional variations With good reason, it is a good approximation to theory R = ACred / DC AC / DC ir ir 2 red Session 2, 2009 ELEC4623/ELEC

52 Future trends: Wearable pulse oximeters Made possible by miniature LED and photosensors! Session 2, 2009 ELEC4623/ELEC

53 Temperature Sensors 1. Resistance based a. Resistance temperature detectors (RTDs) b. Thermistors 2. Thermoelectric thermocouples 3. Fiber optic sensor Session 2, 2009 ELEC4623/ELEC

54 Resistance Temperature Detector (RTD) RTDs are made of materials whose resistance changes in accordance with temperature Metals such as platinum, nickel and copper are commonly used They exhibit a positive temperature coefficient Session 2, 2009 ELEC4623/ELEC

55 Thermistors Thermistors are made from polymer/ceramic material Generally, they have a negative temperature coefficient (NTC) Ro is the resistance at a reference point (in the limit, absolute 0) Over a small dynamic range a thermistor can be linearisedi Self heating is a problem For a low currents, voltage-current relationship is linear For high currents it heats itself, reduces its resistance and hence the voltage drops again Session 2, 2009 ELEC4623/ELEC

56 Electronic interface (for thermistors and RTDs) OR RTD Session 2, 2009 ELEC4623/ELEC

57 Thermocouples Seebeck Effect When a pair of dissimilar metals are joined at one end, and there is a temperature difference between the joined ends and the open ends, thermal emf is generated, which can be measured in the open ends This forms the basis s of thermocouples es 1 VOUT = at + bt 2 2 Session 2, 2009 ELEC4623/ELEC

58 Fiber optic temperature sensors Nortech's fiber-optic temperature sensor probe consists of a gallium arsenide crystal and a dielectric mirror on one end of an optical fiber and a stainless steel connector at the other end Light produced by LED is transmitted along the optical fiber via total internal reflection Amount of power absorbed by the semiconductor crystal increased with temperature Session 2, 2009 ELEC4623/ELEC

59 References Most of this material is directly taken from the following: Webster: chapters 1 and 2 Principles of Applied Biomedical Instrumentation by L.A. Geddes and L.E. Baker, published by John Wiley & Sons, 1975 Piezoelectric transducer: Session 2, 2009 ELEC4623/ELEC