Lecture 1 - Introduction To Measurement Systems

Measurements & Instrumentation
ELE-MEI-321; BECE3, BECE3, BETE3
A.O. Vweza, PhD
Department of Electrical Engineering
July 11, 2023
Dr. A.O. Vweza; Dpt. of Electrical Eng., MUBAS ELE-MEI-321, BECE3, BEEE3, BETE3
Lecture 1
Introduction to measurement systems: general
structure and performance terminologies
Outline
1 Lecture objectives
2 Introduction
3 General structure of a measurement system
4 Static calibration and performance terminologies
5 Overall instrument error and uncertainty analysis
Outline
2 Introduction
Learning objectives
By the end of the lecture you should be able to:

1 Describe the general structure of a measurement system.
2 Assess the static performance of commonly used sensors using

terms such as range, span, error, accuracy, sensitivity, etc.
3 Execute an appropriate uncertainty analysis corresponding to
the level and quantity of information available.
Outline
2 Introduction
Purpose of the measurement system
The purpose of the measurement system is to link the human

observer (via a display) or a control system (via feedback) to
a physical variable being measured/controlled in a process.
Calibration
Signal path
Sensor/transducer stage Signal conditioning stage Display stage
Feedback
Process
control signal
Control stage
Any commonly measured values that you know?
Purpose of the measurement system
The purpose of the measurement system is to link the human

observer (via a display) or a control system (via feedback) to
a physical variable being measured/controlled in a process.
Calibration
Signal path
Sensor/transducer stage Signal conditioning stage Display stage
Feedback
Process
control signal
Control stage
Any commonly measured values that you know?
Commonly measured values
Common measurands/information variables
Acceleration Density
Velocity Viscosity
Displacement Composition
Force/weight pH
Pressure Humidity
Torque Temperature
Volume Heat/light flux
Mass Current
Flow rate Voltage
Level Power
Outline
2 Introduction
Structure of a measurement system
General structure of a measurement system:
Quantity being Signal Signal Displayed

measured related to suitable value of
(measurand) Sensor/ measurand Signal for display measurand
Display
transducer conditioner
A sensor responds to the quantity being measured

(measurand) by giving an output signal which is related to the
quantity.
▶ E.g., a thermocouple is a temperature sensor. The input to the
sensor is a temperature and the output is an e.m.f., which is
related to the temperature value.
▶ Any more examples?

Display
A transducer converts the sensed information into a

detectable signal
WEBC09 11/06/2014
such as electrical or mechanical.
22:0:56 Page 406
▶ E.g., a potentiometer pressure transducer converts

translational displacement x (due to pressure) of bellows
sensor
406 into a 9measurable
Chapter electrical
Pressure and Velocity Measurements voltage E0 .
L
p1
p2 x
Rx
Eo
RL
Figure 9.11 Potentiometer pressure

Ei transducer.

Display
A signal conditioner takes the signal from the

sensor/transducer stage and modifies it into a form suitable
either for display or storage.
▶ E.g., the output from a thermocouple is a small e.m.f. and
might be fed through an amplifier to obtain a bigger signal.
The amplifier is the signal conditioner.
▶ Signal conditioning may also include: filtering, modulation,
level-shifting, analog-to-digital conversion, etc.

Display
A display system indicates or records the output from the

signal conditioner.
▶ E.g., a pointer moving across a scale or a digital readout, or a
recording device such as computer memory.
The following figure
Examples of shows an example of asystems
measurement measurement system.feedbackThe thermocouple is a
Detector and
transducer that converts temperature to a small voltage; the amplifier increases the magni-
electronics
tude of the voltage; the A/D (analog-to-digital) converter is a device that changes Laser the analog
signal to a coded digital signal; and the LEDs (light-emitting diodes) display the value of
Photodiodes
Digital thermometer
the temperature.
thermocouple LED display

amplifier
A/D Cantilever and tip

and
display Sample surface FIGUR
decoder with se
signal conditioner display/recorder

sensor/transducer
70

What about a mercury-in-glass thermometer?
60
Supplemental information important to measurement systems and analysis is
50
▶ Sensor/transducer:
provided in Appendix
▶ SignalA.conditioner:
Included are sections on systems of units, numerical preci-
40
30
sion, and statistics. You should review this material on an Stem
▶ Display: as-needed basis.
20
Bulb
1.3 THREADED DESIGN EXAMPLES FIGURE 1.3

thermometer eq
Throughout the book, there are Examples, which show basic analysis calculations,
transducer, and
Outline
2 Introduction
Introduction
Sensor calibration is critical for maintaining the expected

operation of a measurement system.
A calibration applies a known input value to a measurement
system for the purpose of observing the system output value.
The known value used for the calibration is called the
standard. Rings a bell about role of MBS?
A calibration establishes the relationship between the input
information acquired by the sensor and the indicated output
signal.
Calibration can be static or dynamic.
Introduction

signal.
Introduction

signal.
Introduction

signal.
Introduction

signal.
Static & dynamic calibration

In a static calibration, 1.4 Calibrat
the applied input value to 10
the system under Measured values
Curve fit, y = f(x)
calibration does not vary 8
with time or space.
Output value, y (units)

y = f(x)
6
A range of known input

values is applied to the 4
K (x1) =
dy
dx
system and the system
x = x1
output values are

2
observed to develop a 0
x1
0 1 2 3 4 5
direct calibration curve for Input value, x (units)
the measurement system. FIGURE 1.9 Representative static calibration curve.
and space, is there lag, or is the output value input frequency dependent? Usually, such cal
tions involve applying either a sinusoidal signal of known amplitude and frequency or a

change as the input signal. The dynamic response of measurement systems is explored fu
Chapter 3.
Static & dynamic calibration
The calibration curve is the basis for fixing the output display
scale on a measurement system, or can be use to establish a
correlation y = f (x ) which can be used in later measurements
to ascertain an unknown input value based on the measured
output value.
A dynamic calibration determines the relationship between
an input of known dynamic (time-dependent and/or
space-dependent) content and the measurement system
output.
Usually, such calibrations involve applying either a sinusoidal
signal of known amplitude and frequency or a step change as
the input signal.
Definitions
Static sensitivity
The static sensitivity K is a measure relating the change in
the indicated output associated with a given change in a
static input.
It is the slope of a static calibration curve, i.e.,
K = K (x1 ) = dy is the static sensitiviy at a

dx x =x1
measurement input value x1 .
It is desirable in sensors to have a high and, if possible,
constant sensitivity.
▶ E.g., a sensor with response y = mx + c the sensitivity is
K = m for the entire range of values for x where it applies.
▶ For a sensor with response y = ax 2 + b the sensitivity is
K = 2ax , and it changes from one point to another over the
measurement range.
Definitions
Static sensitivity
static input.

dx x =x1
measurement range.
Definitions
Static sensitivity
static input.

dx x =x1
measurement range.
Definitions
Static sensitivity
static input.

dx x =x1
measurement range.
Definitions
Static sensitivity
static input.

dx x =x1
measurement range.
Definitions
Static sensitivity
static input.

dx x =x1
measurement range.
Definitions
Range and span

A calibration applies known inputs ranging from the minimum
(xmin ) to the maximum (xmax ) values for which the
measurement system is to be used.
These limits define the operating range of the system.
The input full scale operating range (FSO) is defined as
extending from xmin to xmax .
The span is the value difference of the range limits. E.g., a
transducer with a range of 0 to 100 N has a span of 100 N.
The input span may be expressed as ri = xmax − xmin .
The output full scale operating range extends from ymin to
ymax . Hence the output span for the FSO is expressed as
ro = ymax − ymin .
Definitions
Range and span

Definitions
Range and span

Definitions
Range and span

Definitions
Range and span

Definitions
Range and span

Definitions
Range and span

Definitions
Resolution
Resolution represents the smallest increment in the measured
value that can be discerned.
In terms of a measurement system, it is quantified by the
smallest scale increment or the least count (least significant
digit) of the output readout indicator.
Definitions
Resolution
Definitions
Resolution
Practice
L What is the span, range, and resolution for each of the following
measuring instruments.
Definitions
Accuracy and error

The exact value of a variable is called the true value.
The value of a variable as indicated by a measurement system
is called the measured value.
The accuracy of a measurement refers to the closeness of
agreement between the measured value and the true value.
Unfortunately, the true value is rarely known exactly in
engineering practice, and various influences, called errors, have
an effect on both of these values.
Definitions
Accuracy and error

An appropriate approach to stating this closeness of agreement
is to identify the measurement errors and to quantify them by
the value of their associated uncertainties, where an
uncertainty is the estimated range of the value of an error.
An error is the difference between the measured value and the
true value, i.e.,
Error = Measured value − True value
low end of the range is a major concern in selecting a measuring system. To minimize
Definitions uncertainty, the experimenter should select measuring systems such that important
readings will fall in the middle to upper portions of the range. For example, it would
adversely affect uncertainty if a 0-to-200°C thermometer were used to measure a room
temperature around 20°C. A 0 to 30°C thermometer would be far more appropriate.
There are other statements of accuracy, such as an accuracy stated as a percent of
reading. Manufacturers of high-quality measuring systems and components will nor-
Accuracy andmally
errorgive enough information about their products so that the experimenter can
determine the uncertainty in the measurement that is due to the measuring system
itself. The experimenter may have to enlarge the uncertainty interval to account for
E.g., a sensor can be
other error sources that result from the specific application.
specified as having an
Output of device (% of full scale)

120 Uncertainty
accuracy of ±5% of full 100
interval
range output. Thus if the 80 Ideal device
range of the sensor was, 60

40
say, 0 to 200°C, then the 20 Accuracy: 5% of full scale
reading given can be 0

20
expected toFIGURE
be 2.4
within 0 20 40 60 80 100
±10°C of the true

Accuracy as a percentage of full scale. Value of measurand (% of full scale)
reading.
Definitions
Accuracy and error

Errors can be random or systematic.
Random error causes a random variation in measured values
found during repeated measurements of the variable.
Systematic error causes an offset between the mean value of
the data set and its true value.
Both random and systematic errors affect a system’s accuracy.
Definitions
Accuracy and error

Random and systematic errors, accuracy, and precision.1.4 Calibration
(a) High repeatability gives (b) High accuracy means low (c) Systematic and random errors
high precision but no direct random and systematic errors lead to poor accuracy and
indication of accuracy. and high precision. precision.
FIGURE 1.10 Throws of a dart: illustration of random and systematic errors and accuracy.
value, and each throw can represent a measured value. The error in each throw can be calculated
as the distance between the dart and the bullseye. In Figure 1.10a, the thrower displays good
repeatability (i.e., small random error) in that each throw repeatedly hits nearly the same spot
Definitions
Hysteresis error
Hysteresis error refers to differences in the measured value
between an upscale sequential test and a downscale sequential
test.
A sequential test varies the input value sequentially over the
desired input range. This may be accomplished by increasing
the input value (upscale direction) or by decreasing the input
value (downscale direction) over the full input range.
The hysteresis error of the system is estimated by its
uncertainty, uh = yupscale − ydownscale , and
%uh max = (uh max /ro ) × 100, where ro is FSO.
Definitions
Hysteresis error
Hysteresis occurs when the output of a measurement system
is dependent on the previous value indicated by the system.
Such dependencies can be brought about through system
limitations such as
1friction or viscous damping in moving parts
2residual charge in electrical components.
? Any more examples?
Some hysteresis is normal for any system and affects the
repeatability of the system.
Definitions
Linearity, sensitivity, and zero shift errors

Many instruments are designed to achieve a linear relationship
between the applied static input and indicated output values.
The relationship between the predicted output value yL (x )
and measured value y (x ) is a measure of the nonlinear
behavior of a system: uL (x ) = y (x ) − yL (x ), where the
uncertainty uL (x ) is a measure of the linearity error that
arises in describing the actual system behavior by linear
relationship yL (x ) = a0 + a1 x .
Definitions
Linearity, sensitivity, and zero shift errors

The sensitivity error as described by its uncertainty, uK , is a
statistical measure of the random error in the estimate of the
slope of the calibration curve.
A zero shift (null) error results from a vertical shift of the
calibration curve, and quantified by uncertainty, uz .
Zero shift error can usually be reduced by periodically nulling
the output from the measurement system under a zero input
condition.
Definitions
Repeatability, reproducibility, and precision

The ability of a measurement system to indicate the same
value on repeated but independent application of the same
input provides a measure of the instrument repeatability.
Specific claims of repeatability are based on multiple
calibration tests (replication) performed within a given lab on
the particular unit.
Repeatability is an uncertainty based on a statistical measure
called the standard deviation, sx , a measure of the variation in
the output for a given input.
The value claimed is usually in terms of the maximum
expected error as a percentage of full-scale output:
%uR max = (2sx /ro ) × 100.
Definitions
Repeatability, reproducibility, and precision

The term “reproducibility,” when reported in instrument
specifications, refers to the closeness of agreement in results
obtained from duplicate tests carried out under similar
conditions of measurement.
The term “precision”, when reported in instrument
specifications, refers to a random uncertainty based on the
results of separate repeatability tests.
Manufacturer claims of instrument precision must be based on
multiple tests (replication) performed on different units of the
same model, either performed in the same lab (same-lab
precision) or, preferably, performed in different labs
(between-lab precision).
Summary of measurement system errors
BASIC CONCEPTS OF MEASUREMENT METHODS

Examples of some common measurement system error
Downscale Actual data trend

Output value
Output value
Hysteresis
Best linear curve fit
Upscale
Input value Input value
(a) Hysteresis error (b) Linearity error
Maximum for
typical device
Typical shift
Nominal curve
tput value
tput value
(high)
for typical device
Nominal
Hysteresis
Outp
Outp
Summary of measurement system errors Best linear curve fit
Upscale
(a) Hysteresis error (b) Linearity error

Maximum for
typical device
Typical shift
Nominal curve
Output value
Output value
(high)
for typical device
Nominal
Minimum for
typical device Typical shift
K
(low)
(c) Sensitivity error (d) Zero shift (null) error

ut value
Nominal
Outp
Outp
Summary of measurement
Minimum for
typical device
system errors
K Typical shift
(low)
(c) Sensitivity error (d) Zero shift (null) error

Output value
Probable (±2s x)
data scatter band on
successive measurements
Input value
(e) Repeatability error
RE 1.12 Examples of some common elements of instrument error. (a) Hysteresis error.
nearity error. (c) Sensitivity error. (d) Zero shift (null) error. (e) Repeatability error.
for the device. The relationship between yL (x) and measured value y(x) is a measure of the
near behavior of aVweza;
Dr. A.O. system:
Dpt. of Electrical Eng., MUBAS ELE-MEI-321, BECE3, BEEE3, BETE3
f some known calibration errors affecting a typical pressure transducer is given in Table 1
Example
The value assignedmanufacturer’s
to each stated error isspecifications
its uncertainty.
Manufacturer’s Speciﬁcations: Typical Pressure Transducer

Operation
Input range 0–1000 cm H2 O
Excitation ±15 V DC
Output range 0–5 V
Temperature range 0–50∘ C
Performance
Linearity error ±0.5% FSO
Hysteresis error Less than ± 0.15% FSO
Sensitivity error ±0.25% of reading
Thermal sensitivity error ±0.02%∕∘ C of reading
Thermal zero drift ±0.02%∕∘ C FSO
FSO, full-scale operating range.
Outline
2 Introduction
Introduction
Distribution of errors on repeated measurements 5.2 Measurement Errors
Measured data
True value
x'
Systematic error
Measured value, x
p(x)
–x
Random
error in xi
0 2 4 i M
Measurement number
GURE 5.1 Distribution of errors on repeated measurements.

Overall instrument error and instrument uncertainty
An estimate of the overall instrument error is made by

combining the uncertainty estimates of all identified
instrument errors into a term called the instrument
uncertainty.
For M known errors, the overall instrument uncertainty, uc , is
calculated using the root-sum-squares (RSS) method as:
v
q uM
uX
uc = u1 + u2 + · · · + uM = t ui2 .
2 2 2
i=1
It is common to report final uncertainties at a 95% probability

level.
Overall instrument error and instrument uncertainty
The zero-order uncertainty of the instrument, u0 , attempts

to estimate the expected variation in the measured values.
1
u0 = (resolution) = 1 LSD
2
Design-stage uncertainty ud is intended to provide a quick
estimate of the minimum uncertainty to be expected based on
available information.
Design-stage uncertainty
ud = u02 + u2c
ertainty procedure in zero-order uncertainty Instrument uncertainty

u0 uc
ion used, a design-stage uncertainty estimate is intended only as

nt and procedures before
Dr. A.O. Vweza; a Electrical
Dpt. of test, and Eng., is never used
MUBAS for reporting
ELE-MEI-321, BECE3, BEEE3, BETE3
Example
Calculating uncertainty
A force-measuring instrument comes with a certificate of
calibration that identifies two instrument errors and assigns
each an uncertainty at 95% confidence over its range. Provide
an estimate of the instrument design-stage uncertainty. What
is the instrument standard uncertainty?
Resolution 0.25 N
Range 0 to 200 N
Linearity error within 0.20 N over range
Hysteresis error within 0.30 N over range
Example
Calculating uncertainty
We have:
▶ Linearity error uncertainty, u1 = 0.20 N (95%)
▶ Hysteresis error uncertainty, u2 = 0.30 N (95%)
Hence instrument uncertainty is
q q
uc = u12 + u22 = (0.20)2 + (0.30)2 = 0.36 N
The instrument resolution is given as 0.25 N, hence

u0 = 12 (0.25) = 0.125 N. Therefore the design-stage
uncertainty is
q q
ud = u02 + uc2 = (0.125)2 + (0.36)2 = ±0.38 N
Multi-variable uncertainty propagation
Please study Sec 6.5 of Figliola

In the general case, consider the result, R, to be a function of
M measured variables x1 , x2 , . . . , xM , i.e.,
R = f (x1 , x2 , . . . , xM ). The propagation of uncertainty in the
measured variables is given by
v
uM
uX ∂R 2
uR = t uxi ,
i=1
∂xi
where uxi represents the uncertainty associated with the best

estimate of x1 and so forth through xM .
Example

Consider the power consumption in a resistive electric circuit
(P = VI), if the voltage and current have been measured and
found to be
V = 100 ± 2 V
I = 10 ± 0.2 A
Calculate the uncertainty in the computation of the power.
Example

We have ∂P/∂V = I = 10, and ∂P/∂I = V = 100. Hence
s
2 2
∂P ∂P

uP = uV + uI
∂V ∂I
q
= (10 × 2)2 + (100 × 0.2)2
= 28.3 W
i.e., 2.83% uncertainty.
Practice
Try the following...

Manometers are pressure-measuring devices that determine a
pressure by measuring the height of a column of fluid. We would
like to achieve an accuracy of 0.1% of the maximum reading, 10
kPa. This is to be done by using a type of manometer called a well
manometer, which has an uncertainty of 1/10 mm in reading the
scale. Estimate the uncertainty that can be tolerated in the density
of a gage fluid, which has a nominal value 2500 kg/m3 .
[Ans: 2.4 kg/m3 ]
Next lecture
Essential electronics for signal conditioning

Wheatstone bridges, op amps, and comprators.
Please study Bolton Chap. 3, 3.1-3.5, Figliola Chap. 6, Sec
6.4, 6.6.

Lecture 1 - Introduction To Measurement Systems

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lecture 1 - Introduction To Measurement Systems

Uploaded by

Copyright:

Available Formats

Measurements & Instrumentation

ELE-MEI-321; BECE3, BECE3, BETE3

A.O. Vweza, PhD

Department of Electrical Engineering

July 11, 2023

3 General structure of a measurement system

4 Static calibration and performance terminologies

5 Overall instrument error and uncertainty analysis

3 General structure of a measurement system

4 Static calibration and performance terminologies

5 Overall instrument error and uncertainty analysis

By the end of the lecture you should be able to:

2 Assess the static performance of commonly used sensors using

3 General structure of a measurement system

4 Static calibration and performance terminologies

5 Overall instrument error and uncertainty analysis

The purpose of the measurement system is to link the human

Any commonly measured values that you know?

The purpose of the measurement system is to link the human

Any commonly measured values that you know?

Common measurands/information variables

3 General structure of a measurement system

4 Static calibration and performance terminologies

5 Overall instrument error and uncertainty analysis

General structure of a measurement system:

Quantity being Signal Signal Displayed

A sensor responds to the quantity being measured

General structure of a measurement system:

Quantity being Signal Signal Displayed

A transducer converts the sensed information into a

▶ E.g., a potentiometer pressure transducer converts

Figure 9.11 Potentiometer pressure

General structure of a measurement system:

Quantity being Signal Signal Displayed

A signal conditioner takes the signal from the

General structure of a measurement system:

Quantity being Signal Signal Displayed

A display system indicates or records the output from the

thermocouple LED display

A/D Cantilever and tip

signal conditioner display/recorder

1.3 THREADED DESIGN EXAMPLES FIGURE 1.3

3 General structure of a measurement system

4 Static calibration and performance terminologies

5 Overall instrument error and uncertainty analysis

Sensor calibration is critical for maintaining the expected

Sensor calibration is critical for maintaining the expected

Sensor calibration is critical for maintaining the expected

Sensor calibration is critical for maintaining the expected

Sensor calibration is critical for maintaining the expected

In a static calibration, 1.4 Calibrat

the applied input value to 10

the system under Measured values

Curve fit, y = f(x)

calibration does not vary 8

with time or space.

Output value, y (units)

A range of known input

output values are

the measurement system. FIGURE 1.9 Representative static calibration curve.

Range and span

Range and span

Range and span