Measurement and

Instrumentation
Lect 4
BSME 19-23 5th semester
Pakistan Institute of Engineering and Applied Sciences Nilore Islamabad
Lecture outline
• Quantification of systematic errors
• Quantification of individual systematic error components
• Calculation of overall systematic error
• Sources and treatment of random errors
• Statistical analysis of measurements subject to random errors
• Mean and median values
• Standard deviation and variance
• Frequency distributions
• Gaussian distribution
• Standard Gaussian tables
• Standard error of the mean
• Estimation of random error in single measurement
• Distribution of manufacturing tolerances
Quantification of individual systematic error
components
Environmental condition errors
• If a measurement is subject to unpredictable environmental conditions,
the usual course of action is to assume midpoint environmental
conditions
• Specify the maximum measurement error as ±x% of the output
reading to allow for the maximum expected deviation in
environmental conditions away from this midpoint
• Of course, this only refers to the case where the environmental
conditions remain essentially constant during a period of measurement
and become random error in case of fluctuations
Quantification of individual systematic error
components
Calibration errors
• All measuring instruments suffer from drift in their characteristics
over a period of time
• The maximum error just before the instrument is due for recalibration
becomes the basis for estimating the maximum likely error
• The best way to express this is to assume some midpoint value of
calibration error and compensate all measurements by this midpoint
error
• The maximum error can be expressed as ±x% of the output reading
Quantification of individual systematic error
components
System disturbance errors
• Disturbance of the measured system by the act of measurement itself
introduces a systematic error that can be quantified for any given set
of measurement conditions
• If the quantity being measured or the conditions of measurement can
vary than calculate the maximum likely error under worst case system
loading
• Then express the likely error as a plus or minus value of half this
calculated maximum error
Calculation of overall systematic error
• If there are three components of systematic error with a magnitude
of ±1% each, a worst case prediction error would be the sum of the
separate errors, that is, ±3%
• It is very unlikely that all components of error would be at their
maximum or minimum values simultaneously
• Usual course of action is therefore to combine separate sources of
systematic error using a root sum squares method
• Applying this method for n systematic component errors of
magnitude ± x1 %, ± x2 %, ± x3 %,………,± xn %
Sources and treatment of random errors
• Typical sources of random error are
• measurements taken by human observation of an analogue meter, especially
where this involves interpolation between scale points
• electrical noise
• random environmental changes, for example, sudden draught of air
• The process of averaging over a finite number of measurements only
reduces the magnitude of random error to a small value
• The degree of confidence that the calculated mean value is close to
the correct value of the measured quantity can be indicated by
calculating the standard deviation or variance of data
Mean and median values
• For any set of n measurements x1, x2, ……,xn of a constant quantity, the
most likely true value is the mean given by

• This is valid for all data sets where the measurement errors are
distributed equally about the zero error value
• The median, an approximation to the mean, is he middle value when
measurements in the data set are written down in ascending order of
magnitude, for a set of n measurements
• For a set of 9 and 10 measurements median is x5 and (x5 + x6)/2
Mean and median values
• Suppose that the length of a steel bar is measured by a number of
different observers and the following set of 11 measurements are
recorded

• For set A mean = 409.0 and median = 408

• For set B mean = 406.0 and median = 407
• If we increase the number of measurements to 23 measurements
Standard deviation and variance
• A much better way of expressing the distribution about mean value is
to calculate the variance or standard deviation of the measurements
• The starting point for calculating these parameters is to calculate the
deviation of each measurement from mean value
• The variance (Vs) of the set of measurements is defined formally as the
mean of the squares of deviations
• The standard deviation (σs) is

• A better prediction of the V and σ of the infinite population is

Calculate σ and V for measurement sets A, B,
and C given earlier
• We have observed so far that random errors can be reduced by taking
the average (mean or median) of a number of measurements
• As we can only make a finite number of measurements in a practical
situation, the average value will still have some error
• This error can be quantified as the standard error of the mean
Frequency Distributions
• Graphical techniques are a very useful way of analyzing the way in
which random measurement errors are distributed
• The simplest way of doing this is to draw a histogram, in which bands
of equal width across the range of measurement values are defined and
the number of measurements within each band is counted
• A useful rule for defining the number of bands (bins) is known as the
Sturgis rule, which calculates the number of bands as

• where n is the number of measurement values

Draw a histogram for the 23 measurements in
set C of length measurement data
Probability density function

• Cumulative distribution function is

defined as the probability of
observing a value less than or equal
to Do
Gaussian distribution
• The shape of a Gaussian curve is such that the frequency of small
deviations from the mean value is much greater than the frequency of
large deviations
• A Gaussian curve is defined formally as a normalized frequency
distribution that is symmetrical about the line of zero error and in
which the frequency and magnitude of quantities are related by the
expression

• Error frequency distribution can be defined by deviation D = x - m

Gaussian distribution
• If the standard deviation is used as a unit of error, the
Gaussian curve can be used to determine the probability
that the deviation in any particular measurement in a
Gaussian data set is greater than a certain value

• Solution of this expression is simplified by the

substitution z = D/σ
• The effect of this is to change the error distribution curve
into a new Gaussian distribution that has a standard
deviation of one and a mean of zero
Standard Gaussian tables (z distribution)
• It tabulates the area
under the Gaussian
curve F(z) for various
values of z

• For negative values of z,

we can make use of the
following relationship
Standard error of the mean
• The means of the subsets will form a Gaussian distribution about the
mean of the infinite data set
• The standard deviation of mean values of a series of finite sets of
measurements relative to the true mean is defined as the standard error
of the mean, α
• If we use the mean value of a finite set of measurements to predict the
true value of the measured quantity, what is the likely error in this
prediction?
• The measurement value obtained by calculating the mean of a set of n
measurements can be expressed as with 68% certainty for error α
Standard error of the mean
• The problem of expressing the error with 68% certainty is that there is
a 32% chance that the error is greater than α
• Such a high probability of error may not be acceptable in many cases
• We can use with 95.4% certainty that the
magnitude of the error does not exceed 2α
• If we wish to express the maximum error with even greater probability
that the value is correct, we could use 3σ limits (99.7%confidence)
• Referring to set C of length measurements 3σ = 5.64, 3α = 1.17
Estimation of random error in single
measurement
• If only one measurement can be made, some means of estimating the
likely magnitude of error in it is required
• The normal approach to this is to calculate the error within 95%
confidence limits that correspond to a deviation of ±1.96σ
• It is necessary to maintain the measured quantity at a constant value
while a number of measurements are taken in order to create a
reference measurement set from which σ can be calculated
• The maximum likely error with 95% confidence limits in a single
measurement can be expressed as
Distribution of manufacturing tolerances
• Many aspects of manufacturing processes are subject to random
variations caused by factors similar to those that cause random errors
in measurements
• In most cases, these random variations in manufacturing, which are
known as tolerances, fit a Gaussian distribution
• Analysis of random measurement errors can be applied to analyze the
distribution of these variations in manufacturing parameters