BEKG1233

PRINCIPLES OF
INSTRUMENTATION
AND
MEASUREMENT

WEEK
CHAPTER
1 & 12 – INTRODUCTION (PART 2)
INSTRUMENT PERFORMANCE & STATISTICAL ANALYSIS
LECTURER: KYAIRUL AZMI BAHARIN
LEARNING OUTCOME
◦ By the end of the lesson the student will be able to:
a) explain about the static and dynamic characteristics of
measurement
b) describe about the errors that can occur in measurements
c) calculate error parameters.
d) apply basic statistical analysis principles to verify measurement
results.
Instrument Performance Characteristics
◦ Allow users to select the most suitable instrument for a specific
measuring task.
◦ Two basic characteristics:
1. Static – measuring a constant/steady state process condition
2. Dynamic – measuring a varying process condition
1. Static Characteristics
1.1 Accuracy 1.7 Threshold
1.2 Precision 1.8 Resolution
1.3 Tolerance 1.9 Sensitivity to disturbance/Drift
1.4 Range 1.10 Hysteresis
1.5 Linearity 1.11 Dead space
1.6 Sensitivity of measurement
1.1 Accuracy (and Inaccuracy)
◦ Instrument Accuracy = a measure of how close the output
reading is to the correct value.
Yn − X n
% error = 100%
Yn
Yn = expected value
X n = measured value Yn − X n
relative accuracy, A = 1 −
absolute error, e = Yn − X n Yn

% accuracy, a = A 100% = 100% − % error

1.1 Accuracy (and Inaccuracy)
Example 1.1:
Given that the expected value of the resistor is 80 . The
measurement recorded is 79 . Calculate the:
i. absolute error
ii. % (percentage) of error
iii. relative accuracy
iv. % (percentage) of accuracy

Ans: i. 1 , ii. 1.25%, iii. 0.9875, iv. 98.75%

1.1 Accuracy (and Inaccuracy)
Solution
i. e = Yn − X n = 80 − 79 = 1

80 − 79
ii. % error =
80
100% = 1.25%

1 79
iii. relative accuracy, A = 1 − = = 0.9875
80 80

iv. % accuracy, a = A 100% = 0.9875 100% = 98.75%

= 100% − %error = 100 − 1.25 = 98.75%
1.2 Precision
◦ Instrument Precision = a measure of consistency or repeatability
of measurement.

Xn − Xn X n = value of the n-th measurement

Precision, P = 1 − X n = average value of measurement
Xn

where, X =
 X n

n
 1.2 Precision Meas. No,
n
Meas. Value
Xn
1 98
Example 1.2: 2 101
3 102
From the value given in the table, calculate
4 97
the precision of the 6th measurement.
5 101

X=
X n
=
1005
= 100.5
6 100

n 10 7 103
8 98
X −X 100 − 100.5
P = 1− 6 = 1− = 0.995 9 106
X 100.5
10 99
 1005
Accuracy and Precision Relationship
1.3 Tolerance
◦ Closely related to accuracy.
◦ Defines the maximum error that is to be expected in some value.
◦ Usually written with a symbol 
Example: 100   5%
means that when you measure the resistance, the reading may have
a value between 95  and 105 .
◦ Sometimes the upper and lower tolerance may be different.
Example: 230 V (+10% −6%)
means the voltage reading may be between 216.2 V to 253 V.
1.4 Range
◦ Defines the minimum and maximum values of a quantity that the
instrument is designed to measure.
◦ The range must match the quantity we want to measure.
1.5 Linearity
◦ Output reading of an instrument should be linearly proportional
to the quantity being measured.
◦ Make it easy to interpret the value given by the instrument.
1.6 Sensitivity of measurement
◦ Sensitivity refers to how much the instrument output changes
when the quantity changes by a given amount.
scale deflection
Sensitivity, s =
value of measurand producing deflection
Example:
The following resistance values of a platinum resistance
thermometer were measured at a range of temperatures.
Determine the measurement sensitivity in ohms/C.

7
s= = 0.233  / C
30C
1.7 Threshold
◦ When an input to a measuring instrument is gradually increased
from zero, the input will have to reach a certain minimum level
before it can be detected by the instrument.
◦ This minimum level is known as threshold of the instrument.
◦ It can be in absolute values or as a percentage of full-scale
reading.
1.8 Resolution
◦ The smallest change that can be measured.

More uncertainty

Better resolution,
More precise.
1.9 Sensitivity to disturbance / Drift
◦ Environmental changes that affect measurement
◦ Two main ways, zero drift and sensitivity drift.

Zero drift Sensitivity drift Combination

1.10 Hysteresis
◦ Measurement is affected by past history
◦ Commonly found in instruments with springs and electrical
windings.
1.11 Dead space
◦ Range of input values where there is no change in output value.
◦ Instruments that exhibit hysteresis also displays dead space.
Types of Static Errors
Types of Static Errors
a) Gross errors
◦ Mainly due to human mistakes in
◦ reading the value
◦ using the instrument
◦ recording observations
◦ Can be minimized by being careful when taking measurements.
◦ Conduct measurement at least three times.
Types of Static Errors
b) Systematic Errors
◦ Predictable error
◦ Always affect the reading by the same amount or proportion.
◦ Also referred to as bias.
◦ Influence all measurements of a quantity alike.
◦ Can be further divided into three types
i. Instrumental error
ii. Environmental error
iii. Observational error
Types of Static Errors
b) i. Instrumental error
◦ Error caused by the instrument limitations.
◦ E.g. friction in bearings, irregular spring tension, thermal effects on
components etc.
◦ Can be avoided by:
◦ properly selecting a suitable instrument for a particular measurement
task.
◦ applying correction factor based on the error.
◦ calibrate the instrument against a standard.
Types of Static Errors
b) ii. Environmental error
◦ External condition that affect the measuring device (e.g.
temperature, humidity, pressure or magnetic fields)
◦ Example: Temperature reading of a machine may be affected by
heat from other equipment nearby.
◦ These errors can be avoided by removing the effect of the external
condition or shielding the device.
Types of Static Errors
b) iii. Observational error
◦ Error introduced by the observer.
◦ Most common is the parallax error.

Mistake  Error
Mistakes are an accident and is less formal.
Error is due to lack of knowledge and is more formal.
Types of Static Errors
c) Random errors
◦ Errors due to unknown causes
◦ Normally small but unpredictable.
◦ Even after you control all the environment parameters and conduct
the measurement as careful as possible, there will still be some
variation in your reading.
Sources of Errors
1. Insufficient knowledge of process parameters and design
conditions.
2. Poor design.
3. Change in process parameters, irregularities, etc.
4. Poor maintenance
5. Errors caused by person operating the equipment
6. Design limitations.
 2. Dynamic Characteristics
◦ When measuring is done, there is usually a time delay between measuring
and obtaining the reading.
◦ Many quantities fluctuate with time or have transient conditions.

◦ The three most common variations in measured quantity are:

i) Step change – when the primary sensing element (sensor) is subjected to
an instantaneous and finite change in measured variable.
ii) Linear change – when the sensor measures a variable that changes
linearly with time.
iii) Sinusoidal change – when the sensor measures a variable that changes
sinusoidally with constant amplitude.
2. Dynamic Characteristics
Dynamic characteristics of an instrument are:
i) Speed of response – how fast an instrument responds to
changes in measured quantity
ii) Fidelity – the degree to which an instrument indicates the
changes in measured quantity without dynamic error.
iii) Lag – the delay in the response of an instrument to changed in
the measured quantity.
iv) Dynamic error – the difference between the true value and the
value indicated by the instrument if no static error is assumed.
Statistical Analysis
◦ Important because it can provide further insight into the data. It
can be used to find patterns or trends.
◦ Can be used to analyze the uncertainty of the measurement
results.
◦ Statistical analysis can be used to minimize random errors.
◦ It cannot be used to remove systematic errors.
Arithmetic Mean
◦ Average value from the number of readings.

Deviation from the Mean

◦ How far apart each reading is from the arithmetic mean
◦ Deviation can be positive or negative, but the algebraic sum of all the
deviations must be zero.
Example 1.3:
The length of an object in mm is measured five times and the result is as follows:

Calculate: i) Arithmetic mean ii) Deviation of each value ii) Algebraic sum of
the deviations
No Value Deviation, d
1 49.7 49.7 – 49.86 = − 0.16
2 50.1 50.1 – 49.86 = 0.24
3 50.2 50.2 – 49.86 = 0.34
4 49.6 49.6 – 49.86 = − 0.26
5 49.7 49.7 – 49.86 = − 0.16
 249.3 0
249.3
x= = 49.86
5
Average Deviation
◦ Indication of how precise is the instrument used in the
measurement.
◦ Highly precise instruments has a low average deviation between
readings.
Average Deviation
Example 1.4:
Calculate the average deviation for the data given in Example
1.3.
 Standard Deviation, 
◦ A measure of the amount of variation of dispersion.
◦ A low standard deviation indicate that the values tend to be
close to the mean
◦ A high standard deviation indicate that the values are spread
wider apart.
For N < 30 For N  30

 ( dn )  ( xi − x )  ( dn )  ( xi − x )
2 2 2 2

= = = =
N −1 N −1 N N

* N = total number of readings

Standard Deviation, 
Example:
Standard Deviation, 
Example 1.5:
Calculate the standard deviation for the data given in Example
1.3.
No Value Deviation, d d2
1 49.7 49.7 – 49.86 = − 0.16 (− 0.16)2 = 0.0256
2 50.1 50.1 – 49.86 = 0.24 ( 0.24)2 = 0.0576
3 50.2 50.2 – 49.86 = 0.34 ( 0.34)2 = 0.1156
4 49.6 49.6 – 49.86 = − 0.26 (− 0.26)2 = 0.0676
5 49.7 49.7 – 49.86 = − 0.16 (− 0.16)2 = 0.0256
 249.3 0 0.292

 ( dn )
2
249.3 0.292
x= = 49.86 = = = 0.073 = 0.27
5 N −1 5 −1
Standard Deviation, 
◦ Is also used to measure confidence in statistical conclusion.
Standard Deviation, 
Example 1.6:
The normal distribution below is obtained after doing 1000
measurements. If  = 10, what is the percentage of data that falls
between value 70 and 90?
Distribution of Measurement
Example 1.7:
Voltage across a resistor is measured 51 times and the results are given in the
following table. Analyze the data using basic statistical analysis.
Distribution of Measurement
Limiting Error
◦ Most measuring instrument manufacturer specify accuracy within
a certain % of full-scale reading. E.g.  2% of full-scale deflection.
◦ This specification is called the limiting error.
◦ It means a full-scale reading, the error is guaranteed to be within
 2%.
◦ However, for a reading less than the full scale, the limiting error
increases.
Limiting Error
Example 1.8:
A 600 V voltmeter is specified to be accurate within  2% at full scale.
Calculate the limiting error when the instrument is used to measure a
voltage of 250 V.

Example 1.9:
A 500 mA ammeter is specified to be accurate within  2% at full scale.
Calculate the limiting error when the instrument is used to measure 300
mA.
 Summation of Error
◦ Maximum absolute error in sum of the two quantities is equal to
the sum of the absolute errors in the individual quantities.
When, Any mathematic operation that used A
• A and B = the value of instrument A and B and B will produce a summation error as:
• ΔA and ΔB = the limited or absolute error
for A and B Math Operation Summation of error
• %ΔA and %ΔB = average error for A and B A+B
 error = (A+B)
A–B
So, the unit and error for A and B can be AxB
written as : % error = (%A+%B)
A/B
• A = A + ΔA = A + % ΔA
• B = B + ΔeB = B + %ΔB
Summation of Error
Example 1.10:
Given for certain measurement, a limiting error for voltmeter at 70V is 2.143%
and a limiting error for ammeter at 80mA is 2.813%. Determine the limiting error
of the power.

Solution:
P = V x I = 70 V x (0.08 A) = 5.6 W
P = 5.6 (%V+% I) = 5.6 W  (2.143% + 2.813%) = 5.6 W  4.956%
Summation of Error
Example 1.11:
A 150 V voltmeter and 100 mA ammeter are used to measure voltage and
current in a circuit. The dc source is in series with a resistor. Both of these
instruments has the accuracy of ±1% full scale. If the reading on the voltmeter is
80 V and the reading on the ammeter is 70 mA, calculate the power and its
limiting error.

1% 150V = 1.5V
Solution: 1.5V
100% = 1.875%
80V
1% 100mA = 1mA
1mA
100% = 1.429%
70mA
P =(80 V x 70 mA)  (1.875% + 1.429%)
= 5.6 W  (3.304%)
Summation of Error
Example 1.12:
A Current of 2.0 ± 0.01 A flows through a resistor of 100 ± 0.2 Ω.
Calculate the power dissipated and the limiting error.
Solution:
1.2
Power dissipated, P = I R
2 Limiting error, P =  400 W = 4.8 W
100
= (2A) 2 100  = 400 W
P = 400W  1.2% (percentage)
0.01 P = 400W  4.8W (absolute)
% I = 100% = 0.5%
2.0
0.2
% R = 100% = 0.2%
100
% P = % I + % I + % R = 1.2%