Chapter 2

Errors in Measurement
Errors in Measurement
Errors in measurement systems can be divided into
- those that arise during the measurement
process
and
- those that arise during the transfer of signal
from a point of measurement to some other point
due to the corruption of the measurement signal by
induced noise
Errors arising during the measurement process can be
divided into two main groups, known as
systematic errors
random errors.
Definition of error
An error can be defined as;

The difference between the true value and the

measured value of quantity

Error (e) = difference between the measured

value and
the true value of a variable.

e = measured value − true value

Eg : displacement, pressure, temperature

Source of Errors
Source of errors

• Construction effects.
• Approximation of expressions.
• Calculation error.
• Environmental effects.
• Ageing effects.
• Strays and residuals.
• Insertion errors.
Errors in measurement
Various types of error in measurement

Absolute/relative error
gross error
systematic error
random error
limiting error
Absolute / Relative Error
 An error in measurement can be described as the
difference between the expected value and the
measured value of a variable, i.e
e = Yn – Xn

where:
e = absolute error
Yn = expected value
Xn = measured value

 When error is expressed in percentage of full

reading value, it is known as a relative error
e
(100)
ErrorY(%) =
n
Absolute Error
 If a resistor is known to have a resistance of 500
Ω with a possible error of ± 50Ω, then ± 50Ω is
an absolute error because 50 Ω is stated as an
absolute quantity.

Then, R = 500 ± 50Ω .

 Eg; The expected value of the voltage across a

resistor is 5.0 V. However, measurement yields a
value of 4.9 V. Calculate:
a) absolute error
b) relative error
Gross Error
 Largely due to human error.
Example:
i. Misreading of instruments.
ii. Incorrect adjustment.
iii. Reading Wrong Scale.
iv. Improper application of instruments.
v. Computational mistakes equipment faults.
 Gross errors can be improved by through these
3 methods.
i. Careful operator attention.
ii. Cross-checking of results.
iii. Frequent equipment calibration
Systematic Error
It describes errors in output readings of a
measurement system that are consistently on one
side of the correct reading - all errors are either
positive or negative.
It can also be described as type of error that
deviates by a fixed amount from the true value of
measurement. The error remains constant with
repeated measurements
Major sources of systematic errors in the output of
many instruments;
- Factors inherent in the manufacture of the
instrument
- Wear in instrument components over a period of
time
- Effect of environmental disturbances or
environmental inputs – zero drift and sensitive drift
- Disturbance of the measured system by the act of
measurement
Classification of Systematic Errors
Zero setting error.
Scale factor error.
Response Time.
Loading Error.
Zero setting error
Offset or zero setting error in which the
instrument does not read zero when the quantity
to be measured is zero
All readings are in error by the same amount.

Scale factor error or Multiplier

Scale factor error in which the instrument
consistently reads changes in the quantity to be
measured greater or less than the actual
changes.
Response Time Error
This is due to instruments/instrumentations
inability to follow the dynamic change in the
measured quantity.

Example:
The meter having a high inertia movement (or
damping to measure a signal which value
changes rapidly.
Loading Error

The instruments extract in sufficient energy from

the system under measurement.

Example:
By using low-impedance voltmeter to measure the
voltage of high-impedance source.
Examples
Examples of systematic errors of instruments are:

errors in measurements of temperature due to

poor thermal contact between the thermometer
and the substance whose temperature is to be
found - Disturbance of the measured system by
the act of measurement

a voltmeter might show a reading of 1 volt even

when it is disconnected from any
electromagnetic influence. This means the
systematic error is 1 volt and all measurements
shown by this voltmeter will be a volt higher than
the true value – use of an instrument having zero
error
 How to minimize systematic errors
- Identify all sources of error, eg; bent meter needles, poor
cabling practice
- Recalibration of instruments to eliminate systematic error
develop over a period of time because of wear in
instrument components
- Careful instrument design. A proper design and choice of
elements for instruments are able to improve the
sensitivity of an instrument to environmental inputs - For
instance, in designing of a strain gauge , it should be
constructed from a material whose resistance has a very
low temperature coefficient
- Select the appropriate instrument for measurement
- Calibration - errors due to un-calibrated instruments can
be rectified by increasing the frequency of recalibration
- Other methods – incorporate in the system , high gain
feedback circuit or other methods of opposing inputs
Random Errors
Random errors in measurements are caused by
unpredictable variations in the measurement
system
They are usually observed as small perturbations of
the measurement either side of the correct value –
positive and negative errors which occur about
equal numbers for a series of measurements made
of the same constant quantity
Random errors are due to unknown causes and are
observed when the magnitude and polarity of a
measurement fluctuated in a unpredictable manner
The concept of random error is closely related to
the concept of precision. The higher the precision
of a measurement instrument, the smaller the
variability (standard deviation) of the fluctuations
in its readings.
Random error is also called as statistical error
because it can be eliminated from a measurement
by statistical means because of it is random in
nature.

 Sources of random errors

- Imperfection in instruments
- Human errors
- Faults in measuring techniques - random errors often
arise when measurements are taken by human observation of an
analogue meter, especially where this involves interpolation
between scale points
- Electrical noise
- environmental condition changes
Classification of Random Errors
1. Rounding Error.
2. Periodic Error.
3. Noise.
4. Backlash.
5. Ambient Influences.
Types of random errors
Rounding error
- When a reading is between scale graduations and the
reading is rounded up or down to the nearest graduation.
 Periodic error
- This occurs when analog meter reading swings or
fluctuates about the correct reading.
 Noise errors
- The reading is altered by outside interference
 Backlash
- The reading either lags or leads the correct value because
of mechanical movement, friction or damping.
 Ambient influences errors
- Due to condition external to the measuring system such as
variation in temperature, humidity or atmospheric
pressure.
Minimizing random errors
Random errors can be minimized a skilled observed
using a well maintained quality of instruments

Random errors can largely be eliminated by calculating

the average of a number of repeated measurements,
provided that the measured quantity remains constant
during the process of taking the repeated measurements

This averaging process of repeated measurements can be

done automatically by intelligent instruments
(measurement system with microcomputer or
microprocessor embedded within the system for data
computation/manipulation). Also refers as smart sensors
or smart transmitters
Because the error is random in nature, it possibly be
minimized by statistical means
Statistical analysis of measurement data
Important because it allows an analytical
determination of the uncertainty of the final
result

 A large number of measurements is usually

required
 can be divided into 4:
 Arithmetic mean / average
 deviation
 average deviation
 standard deviation
Arithmetic mean/average:
- the most probable value of measured
variable
x1  x 2  x 3    x n n
xi
x 
n i 1 n

Where,
n = total number of reading
xn = nth reading taken
xi = set of number
Deviation:
The difference between each piece of data and
arithmetic mean

d n xn  x

 Algebraic sum of deviation

d tot d1  d 2    d n 0
iii) Average deviation (D):
- precision of a measuring instrument
- high D low precision
- low D  high precision

d1  d 2    d n
D
n
iv) Standard deviation:
- also known as root mean square deviation
- the most important factor in statistical analysis
- reduction means improvement in measurement

d12  d 2 2  d32  ....  d n 2


n 1

dn2

n 1
Example ;
For the following data find,
(a) The arithmetic mean (49.9)
(b) The deviation of each value (0.2,-0.2,-0.3,0.3)
(c) The algebraic sum of the deviation (0)
(d) The average deviation (0.25)
(e) The standard deviation (0.294)
x1= 50.1 x2= 49.7
x3= 49.6 x4= 50.2
Limiting Errors
 Most manufacturers of instruments state that an
instrument is accurate within a certain percentage
of a full-scale reading

 Eg: a voltmeter is accurate within ±2% at full-

scale deflection (limiting errors)

 However, with reading less than full-scale, the

limiting error will increase, therefore, it is
important to obtain measurements as close as
possible to full scale
Example 1
A 300-V voltmeter is specified to be accurate
within ±2% at full scale. Calculate the limiting
error when the instrument is used to measure a
120-V source?

 Solution
The magnitude of the limiting error is:
2/100 x 300 = 6V
Therefore, the limiting error at 120 V is:
6/120 x 100 = 5%
(reading < full scale, limiting error increased)
Example 2

A voltmeter and an ammeter are to be used to

determine the power dissipated in a resistor. Both
instruments are guaranteed to be accurate within
±1% at full scale. If the voltmeter reads 80V on
its 150-V range and the ammeter reads 70mA on
its 100-mA range, calculate the limiting error for
the power calculation.
Note: The limiting error for the power calculation
is the sum of individual limiting errors involved
Significant Figures
- An indication of the precision of the measurement is
obtained from the number.
- Significant figures convey actual information regarding
the magnitude and the measurement precision of the
quantity.
- More significant figures, the greater the precision of
measurement.
e.g
1. One resistor is specified a resistance of 68Ω, It
resistance should be closer to 67Ω or 69Ω.
2. One resistor is specified a resistance of 68.0Ω. It
resistance should be closer to 67.9Ω or 68.1Ω
Therefore:-
68Ω = [ Two significant figures]
68.0Ω = [ Three significant figures]
Note: So that more significant figures an express a measurement of
greater precision.
3. A set of independent voltage measurements taken
by four observers was recorded as :-
117.02, 117.11, 117.08, and 117.03V.

Calculate :-

a). The average of voltage at three and four

significant values.
b). The range of error.
Solution
1. Average 
 Sum
n
117.02  117.11  117.08  117.03

4
117.06V .

Average (three significant figures) = 117V

Average (four significant figures) = 117.1V

Solution

2.Range VMax  V Avg Equ.1

117.11  117
0.11
Range VMax  VMin Equ.2

117.11  117.02
0.09
Solution

0.11  0.09
 Average 
2
0.2

2
0.1
 Average 0.1V

 Average 117.11 0.1V

Significant Figures
1. Retaining Digits

- When recording values of known precision, the convention

is to retain the last digit that is not uncertain by more then
ten units.
e.g: 10.4
9.4
11.4

2. Rounding Off

- The last retained figure is increased by one unit if the

number is greater then 5.
e.g
Number of 28.651, rounded to three significant
figures
 28.651 === 28.7
Addition Figures

-e.g
1. Add 826 ± 5 to 28 ± 3
solution….
N1 826 5(0.605%)
5
 100%
 N 2 628 3(0.477%) 826

=1454 ±8 (±0.55%)

8
 100%
1454
Addition
-e.g
2. Add 15.04 ± 0.03 to 7.32 ± 0.01
Ans:- 22.36±0.04(±0.179%)

3. Add 40 ± 5% and 10 ± 10%

Ans:- 50±3(±6%)
Subtraction
-e.g
1. Subtract 7.32 ± 0.01from15.04v ± 0.03,
Solution….N1 15.04 0.03
 N 2 7.32 0.01
=7.72 ±0.04 (±0.518%)

2. Subtract 628 ± 3 from 826 ± 5

3. Subtract 437 ± 4 from 426 ± 4
Multiplication
Multiply 10.1 ± 0.1 by 5.3 ± 0.2,

Step1- Multiplication of the worst- case values using

positive
limits.
==> 10.2 x 5.5 = 56.1
Step2- Multiplication of the original value
==> 10.1 x 5.3 = 53.53
Step3- Error calculation
==> 56.1- 53.5 = 2.6
Step4- Multiplication value and limits
==> 53.53 ± 2.6 (±4.86%)
Division
Divide 10.1±0.1 by 5.3 ±0.2
Step1 – The worst case of division: number A plus it’s
positive tolerance divided with number B less it’s
negative tolerance.
N1 10.2
 2.0
N2 5.1
Step2 - Division using the original number
N1 10.1
 1.9
N2 5.3
Step3 - Error calculation: (2.0 -1.9) = 0.1
Step4 - Final Result => (1.9 ± 0.1)