MEASUREMENT AND ERROR

CHARACTERISTICS OF MEASURING INSTRUMENTS

ANALYSIS OF MEASURED DATA
UNCERTAINTY ANALYSIS
CHARACTERISTICS OF MEASURING
Measurement is the process or act of determining the range, dimension,
extent, volume, or capacity of anything; it is the process of associating a
number with quantity by comparing it to a standard. The learning
outcomes of this lecture is to:
• Express the need for measurement and analysis of measured data
• Define technical terms related to a measurement such as accuracy,
precision, resolution, error, tolerance, etc.
• Describe the input/output relationship for a measuring equipment
(static calibration)
• Analyze the accuracy and precision of a measurement.
• Compare and contrast the accuracy and precision for a measurement.
• Use significant figures to express the precision of a measurement.
• Classify the measurement errors and list ways of reducing them.

2
Performance Attributes of Measurement

Response Connection Sensitivity

Resolution Range Linearity

Lag and
Precision Sampling Rate
Settling Time

3
Ideal Instrument

Physical World Information World

Operator

Device
Under Instrument
Test
Sensing Display
or and
Stimulus Control

Reference : Electronic Instrument Handbook (Figure 4.1)

4
A measuring system

Physical Quantity

Transmission User
Line Interface
Transducer

Signal Signal
Conditioning Processing

5
 More realistic measuring system

Physical quantity User

Interface
Feedback
Control
Power Indicator
Transducer In Record

Calibration Signal Signal

conditioning processing

6
Definitions in Measurement
• True value – standard or reference
of known value or a theoretical
value
• Accuracy: closeness to true value
• Error: deviation from the true value

High Accuracy

Low Accuracy

7
Ölçmedeki Terimlerin Tanımı
• True value – standard or reference of
known value or a theoretical value
• Accuracy: closeness to true value
• Error: deviation from the true value
• Precision: a measure of reproducibility or
agreement with each other for multiple
trials
• Sensitivity:
output(incremental)/input(incremental) High Precision
• Resolution: smallest change responded
• Linearity: departure from linear value
• Tolerance: maximum deviation allowed Low Precision
from the conventional true value.

8
–Exact measurement is
impossible
–All devices has
uncertainties
–All devices has a tolerance
value.

9
A measurement without an error analysis is
meaningless!
• A measurement is always limited by:
• Device capacity
• Human error
• Physics
• The purpose of an experiment is to deliver the best
possible measurement
• The deviation from the theory
• Should provide an error range

1
0
 Input Output Relation: Linear
Input A1 Measuring
Device Output
Input A2

𝜕(𝑂𝑢𝑡𝑝)
𝑆= = 𝑆𝑙𝑜𝑝𝑒𝑜𝑓𝐶𝑎𝑙𝑖𝑏𝑟𝑎𝑡𝑖𝑜𝑛𝐶𝑢𝑟𝑣𝑒
𝜕(𝐼𝑛𝑝)
𝑂𝑢𝑡𝑝𝑢𝑡 = 𝑆𝑥𝐼𝑛𝑝𝑢𝑡
Output
A2
A1
Sensitivity (S) is constant for
linear systems.

Input
1
1
 Input Output Relation: Non-Linear
Input B1 Measuring
Device Output
Input B2

𝜕(𝑂𝑢𝑡𝑝)
𝑆= = 𝑆𝑙𝑜𝑝𝑒𝑜𝑓𝐶𝑎𝑙𝑖𝑏𝑟𝑎𝑡𝑖𝑜𝑛𝐶𝑢𝑟𝑣𝑒
𝜕(𝐼𝑛𝑝)
𝑂𝑢𝑡𝑝𝑢𝑡 = 𝑆(𝐼𝑛𝑝)𝑥𝐼𝑛𝑝𝑢𝑡
Output B2

B1

Sensitivity (S) is a function

Input of the input.
1
2
Accuracy

CTV : Conventional True Value

1
3
 Example
A voltmeter is used for reading on a standard value of 50
volts, the following readings are obtained: 47, 52, 51, 48

1
4
Precision (Pr)
• Conformity:
• Ability to produce the same reading, also called
repeatability or reproducibility.
• Pr = max {(VAVG – VMIN ), (VMAX –VAVG )}
• # of significant digits
• [47, 52, 51, 48]

𝑉!"# = 49.5𝑉, 𝑉$%& = 52, 𝑉$'( = 47

𝑃𝑟 = 𝑚𝑎𝑥((49.5 − 47), (52 − 49.5))
𝑃𝑟 = 𝑚𝑎𝑥(2.5,2.5)
1

𝑃𝑟 = 2.5 5
Bias
• The difference between CTV and average value (VAV)
• Ideally, the bias should be zero.
• For a high quality digital voltmeter, the loading error
is negligible yielding bias very close to zero.

𝐵𝑖𝑎𝑠 = 49.5 − 50 = −0.5𝑉

1
6
 Example
A known voltage of 100 volts (CTV = 100 V) is read five times by a voltmeter and
following readings are obtained: 103, 107, 105, 102, 106

Pr = max {(VAVG – VMIN ), (VMAX –VAVG )}

= max{(104.6-102), (107-104.6)
=2.6V

Accuracy = max {(CTV – VMIN ), (VMAX –CTV )} The device is accurate by 7%

The device’s accuracy is 93%
= max{(100-102), (107-100)}
=7V

Bias = |AVG - CTV| = |104.6 - 100| = 4.6V

Corrected Readings : 98.6, 102.4, 100.4, 97.6, 101.4

1
7
Poor accuracy High accuracy
High precision High precision

Average accuracy Poor accuracy

Poor precision Poor precision 1
8
68W versus 68.0W

1
9
Error indication by significant figures
• Record a measurement with all digits of which we are
sure nearest to the true value.
• A voltage may be read as 117.1 V.
• indicates that the voltage, read by the observer to
best estimation, is closer to 117.1 V than to 117.0 V
or 117.2 V.
• Another way of expressing is that it indicates the
range of possible error. The voltage may be between
117.05 V and 117.14 V.

2
0
Example - Two resistors in series

• R1 = 18.7 W (3 significant figures)

• R2 = 3.624 W (4 significant figures)
• RT = R1 + R2 = 22.324 W (5 figures) = 22.3 W
• no value in retaining the 2 and the 4.
• If the (least significant) digit in the first place to
be discarded is less than five, it and the following
digits are dropped.
• 22.324 to 22.3; and 22.354 to 22.4.

2
1
 Sources of Error
sampling
Representative
sample preparation
homogeneous
vs. Loss
heterogeneous analysis
Contamination
Measurement
(unwanted addition)
of Analyte

Calibration of
Instrument or
Standard
solutions
Types of errors (uncertainties) and how to
deal with them

• Gross (human)

• Systematic
(determinate)

• Random
(indeterminate)

2
3
 Human error (Gross error)

• Misreading instruments
• Parallax effect
• Incorrect adjustment, or
forgetting to zero
• Erroneous calculations
• Improper choice of
instrument

Not possible to estimate their value mathematically

2
4
 Human error (Gross error)

• Misreading instruments
• Parallax effect
• Incorrect adjustment, or forgetting to
zero
• Erroneous calculations
• Improper choice of instrument
Methods of elimination or reduction
•Paying attention
•Awareness of instrument limitations.
•>2 observers to take critical data.
•>3 readings or reduce possible
occurrences of gross errors. 2
5
Determinate (or Systematic) Errors

• Sometimes called bias due to error in one

direction- high or low
• Known cause
• Result from mis-calibrated device
• Experimental technique that always gives a
measurement higher (or lower) than the true
value
• Operator
• Calibration of glassware, sensor, or instrument
2
6
 Systematic errors
Equipment errors Environmental errors
Examples Examples
• Bearing friction Changes in:
• Non-linearities • Temperature, humidity,
• Calibration errors stray electric and
• Faulty equipment magnetic fields.
 Systematic errors
Equipment errors Environmental errors
Examples Examples
• Bearing friction Changes in:
• Non-linearities • Temperature, humidity,
• Calibration errors stray electric and
• Faulty equipment magnetic fields.

How to estimate: How to estimate:

•Compare with more •Careful monitoring of
accurate standards changes in the
•Determine if error is variables.
constant or a proportional •Calculating expected 2
error changes. 8
 constant or proportional error…

Proportional
error influences
the slope.

Constant
error
influences
the
intercept.

2
9
 Systematic errors
Equipment errors Environmental errors

Reduction/Elimination: Methods of reduction

•Calibration •Hermetic sealing eqp.
•Ensure proper •Stabilize temperature and
operation humidity
•Correcting instrument •Shield equipment against
errors. stray magnetic fields.
•>1 method of
measuring a parameter.

3
0
Indeterminate (or Random) Errors

• Cannot be determined (no control over)

• Random nature causes both high and low values
which will average out
• Multiple trials help to minimize
• Deal with those using statistics

3
1
 Random errors

Examples
•Unknown events that cause small
variations in measurements.
•Quite random and unexplainable.

How to estimate
Take many readings and apply statistical analysis
to unexplained variations

3
2
 Random errors

Examples
•Unknown events that cause small
variations in measurements.
•Quite random and unexplainable.

Methods of reduction:
•Careful design of measurement to reduce
unwanted interference.
•Use of statistical evaluation to determine best true
estimate of measurement readings. 3
3
Upto here!