Homework chapter 3

3.3 Simple RTD. An RTD can be built relatively easily. Consider a copper RTD made of magnet wire (copper wire
insulated with a polymer). The wire is 0.1 mm thick and the nominal resistance required is 120 Ω at 20 oC. Neglect
the thickness of the insulating polymer.

(a) How long must the wire be?

(b) Assuming we wish to wind the copper wire into a single spiral winding 6 mm in diameter so that it can be enclosed
in a stainless steel tube, what is the minimum length of the RTD?

(c) Calculate the range of resistance of the RTD for use between -45 oC and 120 oC.

3.4 Self-heat in RTDs and errors in sensing. A platinum wire RTD enclosed in a ceramic body is designed to operate
between -200 C and +600 C. Its nominal resistance at 0 C is 100 Ω and its TCR is 0.00385/C. The sensor
has a self-heat of 0.07 0C/mW. The sensor is fed from a constant voltage source of 6 V through a fixed 100 Ω resistor
and the voltage across the sensor is measured directly. Calculate the error in temperature sensed in the range from 0
o
C to 100 oC. Plot the error as a function of temperature. What is the maximum error and at what temperature does it
occur? Explain.

3.5 Temperature sensing in a light bulb. Incandescent light bulbs use a tungsten wire as the light-radiating filament
by heating it to a temperature at which it is bright enough to produce light. The temperature of the wire can
be estimated directly from the power rating and the resistance of the wire when it is cold. Given a 120 V, 100 Ω light
bulb with a resistance of 22 Ω at room temperature (20 oC):

(a) Calculate the temperature of the filament when the light bulb is lit.

(b) What are the possible sources of error in this type of indirect sensing? Explain.

3.6 Accurate representation of resistance of RTDs. The Callendar–Van Dusen polynomials ((3.6) and (3.7)) can be
used either with published data from common RTD materials or the coefficients of the polynomial can be
determined from measurements. Suppose, one decides to introduce a new line of RTDs made of nichrome (a nickel–
chromium alloy) for the range between -200 oC and +900 oC. To evaluate the behavior of the new type of
sensors, one must determine the constants a, b, and c in (3.6) and (3.7). There are common calibration points that
guarantee exact known temperatures at which the resistance is measured. The common calibration points are the
oxygen point (182.962 oC, equilibrium between liquid oxygen and its vapor), the triple point of water (0.01 oC, the
point of equilibrium temperature between ice, liquid water, and water vapor), the steam point (100 oC, equilibrium
point between water and its vapor), the zinc point (419.58 oC, equilibrium point between solid and liquid zinc), the
silver point (961.93 oC), and the gold point (1,064.43 C), as well as others. By selecting the appropriate temperature
points and measuring the resistance at those points, one can determine the coefficients. The resistance measurements
for an RTD are as follows: R = 45.94 Ω at the oxygen point, R =50 Ω at 0 C, R = 51.6 Ω at the steam point, R = 58 Ω
at the zinc point, and R = 69.8 Ω at the silver point. The TCR for the nichrome alloy used here is 0.0004/ oC at
20 oC.

(a) Find the coefficients of the Calendar–Van Dusen polynomials using the oxygen, steam, and zinc points.
(b) Find the coefficients of the Calendar–Van Dusen polynomials using the oxygen, zinc, and silver points. Compare
with (a).

(c) Find the resistance at 150 oC and 800 oC. Compare the results obtained using the coefficients obtained in (a) and
(b) with those obtained using (3.5). What are the errors?

3.9 TCR and its dependence on temperature. The temperature coefficient of resistance (TCR) is not constant but
depends on the temperature at which it is given or evaluated. Nevertheless, the formula in (3.4) is correct at any
temperature, no matter what the temperature T0 is, as long as α is measured (or given) at T0. Suppose, α is measured
at 0 oC and is equal to α = 0.00385/oC (for platinum).
(a) Calculate the coefficient α at 50 oC.

(b) Generalize the result in (a) as follows: given α0 at T0, what is α1 at T1?

3.10 Semiconducting resistive sensor. A semiconducting resistive sensor is made as a simple rectangular bar 2 mm
0.1 mm in cross section and 4 mm long. The intrinsic carrier concentration at 20 C is 1.5 x1010/cm3 and the mobilities
of electrons and holes are 1,350 cm 2/(Vs) and 450 cm2/(Vs), respectively. The TCR for the particular device being
used here is 0.012/oC and is assumed to be unaffected by doping.

(a) If intrinsic material is used, calculate the resistance of the sensor at 75 oC.

(b) Now, the material is heavily doped with an n-type dopant at a concentration of 1015/cm3. Calculate the resistance
of the sensor at 75 oC.

(c) What is the resistance of the sensor at 75 oC if instead it is doped with a p-type dopant at the same concentration
as in (b)?

3.11 Silicon-resistive sensors and their transfer functions. A silicon-resistive sensor has a nominal resistance of
2,000 Ω at 25 oC. To calculate its transfer function, its resistance is measured at 0 oC and 90 oC and found to be 1,600
Ω and 3,200 Ω, respectively. Assuming the resistance is given by a second-order Callendar–Van Dusen equation,
calculate the coefficients of the equation and plot the transfer function between 0 oC and 100 oC.

3.12 Silicon-resistive sensor. An n-type silicon resistive temperature sensor is made as a thin film 2 mm wide, 0.1
mm thick, and 10 mm long. Mobilities of carriers go down with temperature whereas carrier densities are assumed
to remain constant in the range of interest. Electron concentration of the n-type-doped silicon used for the sensor is
1017/cm3 whereas the intrinsic concentration in silicon is 1.45x1010/cm3. To characterize the sensor, the mobilities of
electrons and holes are measured at 25 oC, 100 oC, and 150 oC as follows:

(a) Find the transfer function of the sensor as a second-order polynomial (resistance versus temperature).
(b) Calculate and plot the sensitivity of the sensor.
(c) What are the resistances and sensitivities at the three temperatures?

3.13 Thermistor transfer function. The transfer function of an NTC thermistor is best approximated using the
Steinhart–Hart model in (3.14) or (3.15). To evaluate the constants, a thermistor’s resistance is measured at three
temperatures, giving the following results: R = 1.625 kΩ at 0 oC, R = 938 Ω at 25 oC, and R = 154 Ω at 80 oC.
(a) Find the thermistor transfer function using the Steinhart–Hart model.

(b) Using the resistance at 25 oC as the reference temperature, find the transfer function using the simplified model in
(3.12).
(c) Plot the two transfer functions in the range 0–100 oC and discuss the differences between them.

3.14 Thermistor transfer function. A new type of thermistor rated at 100 kΩ at 20 oC is used to sense temperatures
between -80 oC and +100 oC. It is expected that the transfer function is a second-order polynomial. To evaluate
its transfer function, the resistance of the thermistor is measured at 60 oC as 320 kΩ and at +80 oC as 20 kΩ.
(a) Find and plot the transfer function for the required span using a second-order polynomial.
(b) Calculate the resistance expected at 0 oC.

3.15 Thermistor simplified transfer function. The transfer function of a thermistor over the range 0–120 oC is
required. The thermistor is rated at 10 kΩ at 20 oC. Three measurements are taken, at 0 oC, 60 oC, and 120 oC,
resulting in 24 kΩ, 2.2 kΩ, and 420 Ω, respectively. The simple exponential expression in (3.12) is used to derive the
model. However, since the model only has one variable, , one can choose any of the three temperatures to
derive the transfer function.

(a) Derive the transfer function using, in turn, the three measurements. Compare the values of β obtained.

(b) Calculate the errors at the three points for the three transfer functions.

(c) Plot the three transfer functions and compare them. Discuss the differences and the “proper” choice of a
temperature in deriving the simplified model.

3.16 Self-heat of a thermistor. A thermistor with a nominal resistance of 15 kΩ at 25 oC carries a current of 5 mA.
At an ambient temperature of 30 oC (measured with a thermocouple), the resistance of the thermistor is 12.5 kΩ. The
current is now removed and the resistance of the thermistor drops to 12.35 kΩ. Calculate the error due to self-heat of
the thermistor in oC per milliwatt (oC/mW).