NABL 141

NABL 141
NABL
NATIONAL ACCREDITATION
BOARD FOR TESTING AND
CALIBRATION LABORATORIES
GUIDELINES FOR ESTIMATION

and
EXPRESSION OF
UNCERTAINTY IN
MEASUREMENT
ISSUE NO : 02
ISSUE DATE: 02.04.2000
AMENDMENT NO : 03
AMENDMENT DATE: 18.08.2000
GOVT. OF INDIA
MINISTRY OF SCIENCE & TECHNOLOGY
Technology Bhavan, New Mehrauli Road,
New Delhi - 110016
PROFESSOR V.S. RAMARMURTHY

SECRETARY
FOREWARD
The expression of Uncertainty in Measurements is an integral component of the
accreditation certificate being issued to the calibration laboratories. Globalization of trade and
technology implies the need for interchangeability of components, which must be produced
with a high degree of exactness in measurement system. This concept is equally true for all
other fundamental units of measurement. The International Bureau of Weights and Measures
(BIPM), in consultation with various international bodies, have arrived at a new ISO standard
on Expression of Uncertainty in Measurements, in 1995.
I am glad to dedicate the document of NABL on Guidelines for Estimation and Expression of
Uncertainty in Measurement to the cause of calibration laboratories in the country. I take this
opportunity to congratulate the scientists who have made handsome contributions in bringing
out this document based on the latest ISO standard.
New Delhi
2nd April, 2000
V. S. Ramamurthy,
Chairman, NABL
and Secretary, DST
AMENDMENT SHEET
Sl
no
1
Page
No.
28
Clause
Date of
Amendment made
Reasons
No.
Amendment
Appen- 18.08.00 Not deleted from Printing
dix B
Type B evaluation Mistake/
in Note: /
APLAC
evaluation
Signature
QM
Sd/-
Signature
Director
Sd/-
29
Appen- 18.08.00
dix B
Reference to GUM- APLAC

evaluation
additional
information for high
precision
measurement
Sd/-
Sd/-
29-30 Appen- 18.08.00

dix B
Interpretation
to For Better clarity
effective degrees of on selection of
freedom is added
effective degrees
of freedom /
APLAC
evaluation
Sd/-
Sd/-
10
National Accreditation Board for Testing and Calibration Laboratories

Doc. No: NABL 141
Issue No: 02
Guidelines for Estimation and Expression of Uncertainty in Measurement

Issue Date: 02.04.00
Amend No: 03
Amend Date: 18.08.00
Page No: ii
Contents
Sl.
Section
Page No.
Introduction
Uncertainty concept, sources and measures
Definitions of related terms and phrases
Evaluation of standard uncertainty in input estimates
10
Evaluation of standard uncertainty in output estimates
16
Expanded uncertainty in measurement
18
Statement of uncertainty in measurement
19
Apportionment of standard uncertainty
20
Step by step procedure for calculating the uncertainty in measurement
21
10
Appendix A : Use of relevant probability distribution
22
11
Appendix B : Coverage factor and effective degrees of freedom
27
12
Appendix C : Solved Examples
32

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: ii
1.
Introduction
1.1
Purpose
The purpose of the document is to harmonize procedures for evaluating uncertainty in
measurements and for stating the same in calibration certificates as are being followed
by the NABL with the contemporary international approach. The document will apprise
calibration laboratories of the current requirements for evaluating and reporting
uncertainty and will assist accreditation bodies with a coherent assignment of test
measurement capability to calibration laboratories accredited by them. The document
will also provide broad guidelines to all those who are concerned with measurements
about uncertainty in measurement, estimation and apportionment of uncertainty and
interpretation of uncertainty. In fact, the purpose is to provide guidelines to users about
contemporary requirements for global acceptance of various kinds of measurements.
Attempts have been made to make the provisions of this document easy to understand
and ready for implementation. The present document will replace NABLs document
141 (1992).
1.2
Scope
Provisions of this document apply to measurements of all sorts as are carried out in
calibration laboratories. For specialized measurements, these may have to be
supplemented by more specific details and, in some cases, appropriately modified
forms of the concerned formulae. Measurements which can be treated as outputs of
several correlated inputs have been excluded from the scope of this document. The
document covers the following topics:
-
Uncertainty concept, sources and measures

Evaluation of standard uncertainty in input estimates
Evaluation of standard uncertainty in output estimates
Step by step procedure for calculating the uncertainty in measurement
Appendix A: Use of relevant probability distribution
Appendix B: Coverage factor and effective degrees of freedom
Appendix C: Solved examples showing the application of the method outlined
here to eight specific problems in different fields.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 1/ 70
1.3
Normative References :
This document is based primarily on the Guide to the expression of uncertainty in
measurement (1993) jointly prepared by BIPM, IEC, ISO and OIML for definition of
various terms and phrases. One should refer to ISO 3534-I (1993) part I probability
and general statistical terms.
1. Guidelines for estimation and statement of overall uncertainty in measurement
results, NABL 141, Department of Science and Technology, New Delhi (India),
(1992).
2. Guide to the expression of uncertainty in measurement, International Bureau of
Weights and Measures (BIPM), International Organization for Standardization (ISO)
et. al., Switzerland, 1995.
3. International vocabulary of basic and general terms in metrology, International
Bureau of Weights and Measures (BIPM), International Organization for
Standardization (ISO) et. al. ., Switzerland , 1993.
4. Expression of the uncertainty of measurement in calibration, European Cooperation
for Accreditation of laboratories (EAL R-2), 1997
5. Guidelines on the evaluation and expression of the measurement uncertainty,
Singapore Institute of Standards and Industrial Research, Singapore 1995.
6. International standard ISO 3534 I, statistics vocabulary and symbols Part I.
Probability and general statistical terms, first edition, International Organization for
Standardization (ISO) , Switzerland ,1993.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 2/ 70
2.
Uncertainty Concept, Sources and Measures
2.1
Concept
2.1.1
Quality of measurements has assumed great significance in view of the fact that
measurements (in a broad sense) provide the very basis of all control actions.
Incidentally, the word measurement should be understood to mean both a process and
the output of that process.
2.1.2
It is widely recognized that the true value of a measurand (or a duly specified quantity to
be measured) is indeterminate, except when known in terms of theory. What we obtain
from the concerned measurement process is at best an estimate of or approximation to
the true value. Even when appropriate corrections for known or suspected components
of error have been applied, there still remains an uncertainty, that is, a doubt about how
well the result of measurement represents the true value of the quantity being
measured.
2.1.3
A statement of results of measurement (as a process) is complete only if it contains

both the values attributed to the measurand and the uncertainty in measurement
associated with that value. Without such an indication, measured results can not be
compared, either among themselves or with reference values given in a specification or
standard.
2.1.4
The uncertainty of measurement is a parameter, associated with the result of a

measurement, that characterizes the dispersion of the true values, which could
reasonably be attributed to the measurand. The parameter may be, for example, the
standard deviation (or a given multiple of it), or the half-width of an interval having a
stated level of confidence.
2.2
Source
2.2.1
Errors in the observed results of a measurement (process) give rise to uncertainty

about the true value of the measurand as is obtained (estimated) from those results.
Both systematic and random errors affecting the observed results (measurements)
contribute to this uncertainty. These contributions have been sometimes referred to as
systematic and random components of uncertainty respectively.
2.2.2
Random errors presumably arise from unpredictable and spatial variations of influence
quantities, for example:
-
the way connections are made or the measurement method employed
uncontrolled environmental conditions or their influences
inherent instability of the measuring equipment
personal judgement of the observer or operator, etc.
These cannot be eliminated totally, but can be reduced by exercising appropriate

controls.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 3/ 70
2.2.3
Various other kinds of errors, recognized as systematic, are also observed. Some
common type of these errors are due to:
-
those reported in the calibration certificate of the reference standards

/instruments used
different influence conditions at the time of measurement compared with those

prevalent at the time of calibration of the standard (quite common in length and
d.c. measurements) etc.
2.2.4
It should be pointed out that errors, which can be recognized as systematic and can be
isolated in one case, may simply pass of as random in another case.
2.3
Measures
2.3.1
Measurands are particular quantities subject to measurement. In calibration one usually

deals with only one measurand or output quantity Y that depends upon a
number of input
quantities Xi (i = 1,2,., N) according to the functional
relationship,
Y = f(X1, X2,., XN).
(2.1)
The model function f represents the procedure of the measurement and the method of
evaluation. It describes how values of the output quantity Y are obtained from values of
the input quantities Xi.
2.3.2
An estimate of the measurand Y (output estimate) denoted by y, is obtained from Eq.

(2.1) using input estimates xi for the values of the input quantities Xi,
y = f(x1,x2, . x N) .
(2.2)
It is understood that the input values are best estimates that have been corrected for all
effects significant for the model. If not, necessary corrections have been introduced as
separate input quantities.
2.3.3
The standard uncertainty of measurement associated with the output estimate y,

denoted by u(y), is the standard deviation of the unknown (true) values of the
measurand Y corresponding to the output estimate y. It is to be determined from the
model Eq. (2.1) using estimates xi of the input quantities Xi and their associated
standard uncertainties u (xi).
The standard uncertainty associated with estimate has the same dimension as the
estimate. In some cases the relative standard uncertainty of measurement may be
appropriate which is the standard uncertainty associated with an estimate divided by
the modulus of that estimate and is therefore dimensionless. This concept cannot be
used if the estimate equals zero.
2.3.4
The standard uncertainty of the result of a measurement, when that result is obtained
from the values of a number of other quantities is termed combined standard
uncertainty.
2.3.5
An expanded uncertainty is obtained by multiplying the combined standard uncertainty

by a coverage factor. This, in essence, yields an interval that is likely to cover the true
value of the measurand with a stated high level of confidence.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 4/ 70
3.
The guide explains explicitly a large number of metrological terms which are used in practice.
A few terms of general interest have been taken from the International Vocabulary of Basic
and General terms in Metrology and EAL document [3-4]. To facilitate the reader, various
terms and phrases are arranged in alphabetical order
accepted reference value
a value that serves as an agreed upon reference for comparison.
accuracy of measurement
the closeness of agreement between a test result and the accepted reference value
arithmetic mean
The sum of values divided by the number of values
combined standard uncertainty (uc)
standard uncertainty of the result of a measurement when that result is obtained from the
values of a number of other quantities, equal to the positive square root of a sum of terms, the
terms being the variances or covariances of these other quantities weighted according to how
the measurement result varies with changes in these quantities
conventional true value (of a quantity)
a value of a quantity which for a given purpose, may be substituted for the true value
correction
value added algebraically to the uncorrected result of a measurement to compensate for
systematic error
correction factor
numerical factor by which the uncorrected result of a measurement is multiplied to
compensate for a systematic error
correlation
the relationship between two or several random variables within a distribution of two or
more random variables
correlation coefficient
the ratio of the covariance of two random variables to the product of their standard deviations.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 5/ 70
covariance
The sum of the products of the deviations of xi and yi from their respective averages divided by
one less than the number of observed pairs:
)(
1 N
s xy =
x i x yi y
n 1 i =1
(3.1)
Where n is number of observed pairs

coverage factor (k)
numerical factor used as a multiplier of the combined standard uncertainty in order to obtain an
expanded uncertainty
coverage probability or confidence level
the value of the probability associated with a confidence interval or a statistical coverage
interval
degrees of freedom ()
the number of terms in a sum minus the number of constraints on the terms of the sum
errors
In general, a measurement has imperfections that give rise to an error in the measurement
result. An error is viewed as having two components, namely, a random component and
systematic component.
error in measurement
result of a measurement minus accepted reference value (of the characteristic)
estimation
the operation of assigning, from the observations in a sample, numerical values to the
parameters of a distribution chosen as the statistical model of the population from which this
sample is taken
estimate
the value of a statistic used to estimate a population parameter
expanded uncertainty (U)
quantity defining an interval about the result of a measurement that may be expected to
encompass a large fraction of the distribution of values that could reasonably be attributed to
the measurand

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 6/ 70
expectation
the expectation of a function g(z) over a probability density function p(z) of the random
variables z is defined by
E[g(z)] = g(z)p(z)dz
(3.2)
the expectation of the random variable z, denoted by z and which is also termed as the
expected value or the mean of z. It is estimated statistically by z , the arithmetic mean or
average of n independent observations zi of the random variable z, the probability density
function of which is p(z)
z=
1 n
zi
n i =1
(3.3)
experimental standard deviation [s(qj)]

for a series of n measurements of the same measurand, the quantity s(qj) characterizing the
dispersion of the results and given by the formula :
s (q j ) =
(q
n
j =1
n 1
(3.4)
qj being the result of the jth measurement and q being the arithmetic mean of the n results
considered.
measurand
a quantity subject to measurement
probability distribution
a function giving the probability that a random variable takes any given value or belongs to a
given set of values
probability density function
the derivative (when it exits) of the distribution function :
f(x) = dF(x) /dx
(3.5)
f(x)dx is the probability element

f(x) dx = Pr(x < X < x + dx)
(3.6)
probability function
a function giving for every value x, the probability that the random variable X takes value x :
F(x) = Pr(X = x)
(3.7)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 7/ 70
random error
result of a measurement minus the mean that would result from an infinite number of
measurements of the same measurand carried out under repeatability conditions
Notes :
1.
random error is equal to error minus systematic error
2.
because only a finite number of measurements can be made,
it is possible to determine only one estimate of random error

random variable
a variable that may take any of the values of a specified set of values and with which is
associated a probability distribution
repeatability (of results of measurements)
closeness of the agreement between the results of successive measurements of the same
measurand carried out under repeatability conditions
repeatability conditions
conditions where independent test results are obtained with the same method on identical test
items in the same laboratory by the same operator using the same equipment within short
interval of time.
reproducibility (of results of measurements)
closeness of the agreement between the results of the measurements of the same measurand
carried out under reproducibility conditions .
reproducibility conditions
conditions where test results are obtained with the same method on identical test items in
different laboratories with different operators using different equipment
results of a measurement
value attributed to a measurand, obtained by measurement
Note:
Complete statement of the result of a measurement includes information about uncertainty in
measurement
sensitivity coefficient associated with an input estimate (ci)
the differential change in the output estimate generated by the differential change in that input
estimate
standard deviation ()
the positive square root of the variance

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 8/ 70
standard uncertainty
uncertainty of the result of a measurement expressed as a standard deviation
systematic error
mean that would result from an infinite number of measurements of the same measurand
carried out under repeatability conditions minus acceptance reference value of the measurand.
Note:
Systematic error is equal to error minus random error
true value (of a quantity)
the value which characterized a quantity perfectly defined in the conditions which exist when
that quantity is considered
Note:
The true value is a theoretical concept, and, in general, can not be known exactly
Type A evaluation (of uncertainty)
Method of evaluation of uncertainty by the statistical analysis of series of observations
Type B evaluation (of uncertainty)
Method of evaluation of uncertainty by means other than the statistical analysis of series of
observations.
uncertainty (in measurement)
parameter, associated with the result of a measurement that characterizes the dispersion of the
values that could reasonably be attributed to the measurand.
variance
A measure of dispersion, which is the sum of the squared deviations of observations from their
average divided by one less than the number of observations.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 9/ 70
4.
Evaluation of standard uncertainty in Input estimates
4.1
General considerations
4.1.1
The uncertainty of measurement associated with the input estimates is evaluated

according to either a Type A or a Type B method of evaluation. The Type A
evaluation of standard uncertainty is the method of evaluating the uncertainty by the
statistical analysis of a series of observations. In this case the standard uncertainty is
the experimental standard deviation of the mean that follows from an averaging
procedure or an appropriate regression analysis. The Type B evaluation of standard
uncertainty is the method of evaluating the uncertainty by means other than the
statistical analysis of a series of observations. In this case the evaluation of the
standard uncertainty is based on some other scientific knowledge.
Examples:
Case I : Digital multimeter (DMM)
Let us consider, an experiment in which a high accuracy reference standard e.g. a 6
digit stable meter calibrator is used to calibrate a device of much lower accuracy
like 4 digit DMM . The readings of the test DMM may remain unchanged or
undergo flicker 1 count due to its digitizing process. In this case, the Type A
evaluation of the uncertainty may be taken to be negligible, and the uncertainty on
account of repeatable observations can be treated as Type B on the basis of the
resolution error of the test DMM.
Case II : Length Bar

While calibrating a length bar by comparison method, one has to include the
component of uncertainty associated with the thermal expansion coefficient [ = ]
in the uncertainty budget. Usually, for the test and standard is taken from
handbook or as per manufacturers specification, in this case, although the estimation
of uncertainty in temperature measurement is Type A but the estimation of
uncertainty in is Type B. However, in a special case where high precision is
needed, in situ measurement of thermal expansion is carried out. In such a case, the
evaluation of uncertainty in both temperature and are of Type A.
4.2
Type A evaluation of standard uncertainty
4.2.1
Type A evaluation of standard uncertainty applies to situation when several

independent observations have been made for any of the input quantities under the
same conditions of measurement. If there is sufficient resolution in the measurement
process, there will be an observable scatter or spread in the values obtained.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 10/ 70
4.2.2
Let us denote by Q the repeatedly measured input quantity Xi. With n statistically
independent observations (n > 1), the estimate of Q is q, the arithmetic mean of the
individual observed values qj (j = 1, 2,.n).
q=
1 n
qj
n j =1
(4.1)
The uncertainty of measurement associated with the estimate q is evaluated according

to one of the following methods
4.2.3
An estimate of the variance of the underlying probability distribution of q is the

experimental variance s2 (q) of values qj given by,
s 2 (q ) =
1 n
qj q
n 1 j =1
(4.2)
The positive square root of s2 (q) is termed experimental standard deviation. The best
estimate of the variance of the arithmetic mean q is given by
s 2 (q )
s q =
n
2
()
(4.3)
Table 4.1: Data for calculation of mean and standard deviation of temperature:
Observation
numbers
Temperature 0C
(t
t 10 2
0
(t
t 10 4
( C)
0
1
2
3
4
5
6
7
8
9
10
90.68
90.83
90.79
90.64
90.63
90.94
90.60
90.68
90.76
90.65
-4
11
7
-8
-9
22
-12
-4
4
-7
16
121
49
64
81
484
144
16
16
49
Total
907.2
1040
The positive square root of s2 (q) is termed as estimated standard error of the mean.
The standard uncertainty u (q) associated with the input estimate q is the standard
error.
() ()
uq =sq
(4.4)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 11/ 70
4.2.4
For a measurement that is well-characterized and under statistical control a combined

or pooled estimate of variance s2p may be available from several sets of repeat
measurements that characterizes the dispersion better than the estimated variance
obtained from a single set of observations. If in such a case the value of the input
quantity Q is determined as the arithmetic mean q of n independent observations, the
variance of the mean may be estimated by
()
s2 q =
s 2p
(4.5)
The standard uncertainty is deduced from the value given by Eq. (4.4)
Example:
Table (4.1) is shown the data from a temperature measurement. We now
estimate different parameters as follows:
Mean Temperature:
t
j
j =1
= 90.72
t=
n
(4.6)
t = 90.72 C
(4.7)
The best estimate of temperature is therefore:
Standard Deviation:
s (t ) =
1 n
(t j t
n 1 j =1
1
1040 10 4 = 10.75 10 2
9
(4.8)
Standard error of the mean:
()
st =
s (t ) 10.75 10
=
n
10
2
= 3.40 10 2 C
(4.9)
Standard uncertainty:
()
u t = 3.40 10 2 C
(4.10)
Degrees of freedom ()
= n 1 = 10 1 = 9
(4.11)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 12/ 70
4.3
Type B evaluation of standard uncertainty
4.3.1
The Type B evaluation of standard uncertainty is the evaluation of the uncertainty

associated with an estimate xi of an input quantity Xi by means other than the statistical
analysis of a series of observations. The standard uncertainty u(xi ) is evaluated by
scientific judgment based on all available information on the possible variability of Xi.
Values belonging to this category may be derived from
4.3.2
previous measurement data ;
experience with or general knowledge of the behaviour and properties of

relevant materials and instruments ;
manufacturers specifications ;
data provided in calibration and other certificates;
uncertainties assigned to reference data taken from handbooks.
The proper use of the available information for a Type B evaluation of standard
uncertainty of measurement calls for insight based on experience and general
knowledge. It is a skill that can be learned with practice. A well-based Type B
evaluation of standard uncertainty can be as reliable as a Type A evaluation of standard
uncertainty, especially in a measurement situation where a Type A evaluation is based
only on a comparatively small number of statistically independent observations. The
following cases must be discerned:
(a)
When only a single value is known for the quantity Xi, e.g. a single measured
value, a resultant value of a previous measurement, a reference value from the
literature, or a correction value, this value will be used for xi. The standard
uncertainty u (xi) associated with xi is to be adopted where it is given. Otherwise
it has to be calculated from unequivocal uncertainty data. If data of this kind are
not available, the uncertainty has to be evaluated on the basis of experience
taken as it may have been stated (often in terms of an interval corresponding to
expanded uncertainty).
(b)
When a probability distribution [see Appendix A] can be assumed for the

quantity Xi, based on theory or experience, then the appropriate expectation or
expected value and the standard deviation () of this distribution have to be
taken as the estimate xi and the associated standard uncertainty u (xi),
respectively.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 13/ 70
Examples:
In cases, where the uncertainty is quoted to be particular multiple of standard deviation (), the
multiple becomes the specific factor (see Appendix A).
Case I:
A calibration certificate states that the mass of a given body of 10 kg is 10.000650 kg. The
uncertainty at 2 (at confidence level of 95 .45 %) is given by 300 mg. In such a case, the
standard uncertainty is simply,
u(m) = 300 / 2 = 150 mg
(4.12)
and estimated variance is
u2(m) = 0.0225 g2
(4.13)
Case II:
Suppose in the above example, the quoted uncertainty defines an interval having a 90% level
of confidence. The standard uncertainty is then
u(m) = 300 / 1.64 = 182.9 mg
(4.14)
Where we have taken 1.64 as the factor corresponding to the above level of confidence,
assuming the normal distribution unless otherwise stated.
Case III:
A calibration certificate states that the resistance of a standard resistor, Rs of nominal
value 10 is 10.000742 129 at 23 0 C and that the quoted uncertainty of 129
defines an interval having a level of confidence of 99%. The standard uncertainty of the
resistor may be taken as
u(Rs ) = 129 / 2.58 = 50
(4.15)
Therefore, in this case, specific factor is 2.58. The corresponding relative standard uncertainty
u(Rs )/ Rs = 5 10-6
(4.16)
The estimated variance is

u2 = (50 )2 = 2.5 10-9 2
(4.17)
Case IV:
A calibration certificate states that the length of a standard slip gauge (SG) of nominal value 50
mm is 50.000002 mm. The uncertainty of this value is 72 nm, at confidence level of 99.7 %
(corresponding to 3 times of standard deviation). The standard uncertainty of the standard slip
gauge is then given by
u(SG) = 72 nm /3 = 24 nm
(4.18)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 14/ 70
(c)
If only upper and lower limits a + and a - can be estimated for the value of the
quantity Xi (e.g. manufacturers specifications of a measuring instrument, a
temperature range, a rounding or truncation error resulting from automated data
reduction), a probability distribution with constant probability density between
these limits (rectangular probability distribution) has to be assumed for the
possible variability of the input quantity Xi . According to case (b) above this
leads to
1
(a + + a - )
2
xi =
(4.19)
for the estimated value and
u 2 (x i ) =
1
(a + a - )2
12
(4.20)
for the square of the standard uncertainty . If the difference between the limiting values
is denoted by 2a , Eq. (4.20) yields
u 2 (x i ) =
1
(a )2
3
(4.21)
Examples:
The specifications of a dial type pressure gauge are as follows :
Range : 0 to 10 bar,
Scale : 1 division = 0.05 bar,
Resolution : division = 0.025 bar
Accuracy : 0.25 % Full Scale Deflection
Assuming that with the above specifications, there is an equal probability of the true
value lying anywhere between the upper (a + ) and lower (a ) limits. Therefore, for
rectangular distribution,
a=
Here ,
and
(a +
a )
2
a + = (0.25% 10) bar = 0.025 bar

a = (0.25% 10) bar = 0.025 bar
a = 0.05 / 2 = 0.025 bar .
(4.22)
(4.23)
(4.24)
(4.25)
Hence the standard uncertainty is given by ,
u=
a
3
0.025
3
= 0.0144 bar
(4.26)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 15/ 70
5.
Evaluation of standard uncertainty in output estimate
5.1
For uncorrelated input quantities the square of the standard uncertainty associated with
the output estimate y is given by,
n
u 2 ( y ) = ui2 ( y )
(5.1)
i =1
The quantity ui(y) (i = 1, 2,, n) is the contribution to the standard uncertainty

associated with the output estimate y resulting from the standard uncertainty associated
with the input estimate xi,
ui (y) = ci u(xi)
(5.2)
where ci is defined as sensitivity coefficients associated with the input estimate xi i.e.
the partial derivative of the model function f with respect to Xi , evaluated at the input
estimates xi .
c i = (f /xi ) = (f /Xi ) at Xi = xi
(5.3)
5.2
The sensitivity coefficient ci, describes the extent to which the output estimate y is
influenced by variations of the input estimate xi. It can be evaluated from the function f
by Eq. (5.3) or by using numerical methods, i.e. by calculating the change in the output
estimate y due to a change in the input estimate xi of + u(xi) and -u(xi) and taking as the
value of ci the resulting difference in y divided by 2u (xi) . Sometimes it may be more
appropriate to find the change in the output estimate y from an experiment by repeating
the measurement at e.g. xi u (xi).
5.3
If the model functions is a sum or difference of the input quantities Xi,

N
f (X1 ,X2 , XN) =
p X
i =1
(5.4)
the output estimate according to Eq. (2.2) is given by the

difference of the input estimate
corresponding sum or
y = pi x i
i =1
(5.5)
whereas the sensitivity coefficients equal to pi and Eq. (5.1) converts to

N
u 2 (y ) = p i2u 2 (x i )
i =1
(5.6)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 16/ 70
5.4
If the model function f is a product or quotient of the input quantities XI

N
f ( X 1, X 2 ,......., X N ) = c X ipi
(5.7)
i =1
the output
estimate again is the corresponding product or quotient of the input
estimates
N
y = c X ipi
(5.8)
i =1
The sensitivity coefficients equal piy/xi in this case and an expression analogous to Eq.
(5.6) is obtained from Eq. (5.1), if relative standard uncertainties w(y) = u(y)/y and w (xi)
= u (xi) / xi are used,
w 2 (y ) =
p w (x )
i =1
2
i
(5.9)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 17/ 70
6.
6.1
Calibration laboratories shall state an expanded uncertainty in measurement (U),

obtained by multiplying the standard uncertainty u(y) of the output estimate y by a
coverage factor k,
U = ku (y)
(6.1)
In cases where a normal (Gaussian) distribution can be attributed to the measurand

and the standard uncertainty associated with the output estimate has sufficient
reliability, the standard coverage factor k = 2 shall be used. The assigned expanded
uncertainty corresponds to a coverage probability of approximately 95 %.
6.2
The assumption of a normal distribution cannot always be easily confirmed

experimentally. However, in the cases where several (i.e. N 3) uncertainty
components derived from wellbehaved probability distributions of independent
quantities, e.g. normal distributions or rectangular distributions, contribute to the
standard uncertainty associated with the output estimate by comparable amounts, the
conditions of the central limit theorem are met and it can be assumed to a high degree
of approximation that the distribution of the output quantity is normal.
6.3
The reliability of the standard uncertainty assigned to the output estimate is determined
by its effective degrees of freedom (see Appendix B). However, the reliability criterion is
always met if none of the uncertainty contributions is obtained from a Type A evaluation
based on less than ten repeated observations.
6.4
If one of these conditions (normality or sufficient reliability) is not fulfilled, the standard
coverage factor k = 2 can yield an expanded uncertainty corresponding to a coverage
probability of less than 95 %. In these cases, in order to ensure that a value of the
expanded uncertainty is quoted corresponding to the same coverage probability as in
the normal case, other procedures have to be followed. The use of approximately the
same coverage probability is essential whenever two results of measurement of the
same quantity have to be compared, e.g. when evaluating the results of an
interlaboratory comparison or assessing compliance with a specification.
6.5
Even if a normal distribution can be assumed, it may still occur that the standard
uncertainty associated with the output estimate is of insufficient reliability. If, in this
case, it is not expedient to increase the number n of repeated measurements or to use
a Type B evaluation instead of the Type A evaluation of poor reliability, the method
given in Appendix B should be used.
6.6
For the remaining cases, i.e. all cases where the assumption of a normal distribution
cannot be justified, information on the actual probability distribution of the output
estimate must be used to obtain a value of the coverage factor k that corresponds to a
coverage probability of approximately 95 %.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 18/ 70
7.
7.1
In calibration certificates the complete result of the measurement consisting of the

estimate y of the measurand and the associated expanded uncertainty U shall be given
in the form
(y U). To this an explanatory note must be added which in the general
case should have the following content: The reported expanded uncertainty in
measurement is stated as the standard uncertainty in measurement multiplied by
the coverage factor k = 2, which for a normal distribution corresponds to a
coverage probability of approximately 95 %.
7.2
However, in cases where the procedure of Appendix A has been followed, the
additional note should read as follows: The reported expanded uncertainty in
measurement is stated as the standard uncertainty in measurement multiplied by
the coverage factor k which for a t-distribution with eff effective degrees of
freedom corresponds to a coverage probability of approximately 95 %. (See
Appendix B).
7.3
The numerical value of the uncertainty in measurement should be given to at most two
significant figures. The numerical value of the measurement result should in the final
statement normally be rounded to the least significant figure in the value of the
expanded uncertainty assigned to the measurement result. For the process of
rounding, the usual rules for rounding of numbers have to be used. However, if the
rounding brings the numerical value of the uncertainty in measurement down by more
than 5 %, the rounded up value should be used.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 19/ 70
8.
8.1
The uncertainty analysis for a measurement-sometimes called the Uncertainty Budget

of the measurement-should include a list of all sources of uncertainty together with the
associated standard uncertainties of measurement and the methods of evaluating them.
For repeated measurements the number n of observations also has to be stated. For
the sake of clarity, it is recommended to present the data relevant to this analysis in the
form of a table. In this table all quantities should be referenced by a physical symbol Xi,
or a short identifier. For each of them at least the estimate xi, the associated standard
uncertainty in measurement u (xi), the sensitivity coefficient ci and the different
uncertainty contributions ui(y) should be specified. The degrees of freedom have to be
mentioned. The dimension of each of the quantities should also be stated with the
numerical values in the table.
8.2
A formal example of such an arrangement is given as Table (8.1) applicable for the
case of uncorrelated input quantities. The standard uncertainty associated with the
measurement result u(y) given in the bottom right corner of the table is the root sum
square of all the uncertainty contributions in the outer right column. Similarly, eff has to
be evaluated as mentioned in Appendix B.
Table 8.1: Schematic view of an Uncertainty Budget

Source of
Uncertainty
Xi
Estimates Limits
xi
xi
Probability
Distribution
- Type A or
B
Standard
Sensitivity
Uncertainty coefficient
u(xi)
ci
Uncertainty
contribution
ui(y)
Degree
of
freedom
i
X1
X2
X3
x1
x2
x3
x1
x2
x3
-Type A or B
-Type A or B
-Type A or B
u(x1)
u(x2)
u(x3)
c1
c2
c3
u1(y)
u2(y)
u3(y)
1
2
3
XN
Y
xN
y
xN
-Type A or B
u(xN)
cN
uN(y)
uc(y)
N
eff

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 20/ 70
9.
Step-by-step procedure for calculating the uncertainty in

measurement
The following is a guide to the use of this document in practice:

Step 1
Express in mathematical terms the dependence of the measured (output

quantity) Y on the input quantities Xi according to Eq. (2.1). In the case of a
direct comparison of two standards the equation may be very simple, e.g.
Y = X1 + X2
(9.1)
Step 2
Identify and apply all significant corrections to the input quantities.
Step 3
List all sources of uncertainty in the form of an uncertainty analysis in

accordance with Section 8.
Step 4
Calculate the standard uncertainty u q for repeatedly measured quantities in

accordance with sub-section 4.2.
Step 5
For single values, e.g. resultant values of previous measurements, correction

values or values from the literature, adopt the standard uncertainty where it is
given or can be calculated according to paragraph 4.3.2(a). Pay attention to the
uncertainty representation used. If no data are available from which the
standard uncertainty can be derived, state a value of u (xi) on the basis of
scientific experience.
Step 6
For input quantities for which the probability distribution is known or can be
assumed, calculate the expectation and the standard uncertainty u (xi) according
to paragraph 4.3.2 (b). If only upper and lower limits are given or can be
estimated, calculate the standard uncertainty u (xi) in accordance with paragraph
4.3.2(c).
Step 7
Calculate for each input quantity Xi the contribution ui (y) to the uncertainty
associated with the output estimate resulting from the input estimate xi according
to Eqs. (5.2) and (5.3) and sum their squares as described in Eq. (5.1) to obtain
the square of the standard uncertainty u(y) of the measurand.
Step 8
Calculate the expanded uncertainty U by multiplying the standard uncertainty

u(y) associated with output estimate by a coverage factor k chosen in
accordance with Section 6.
Step 9
Report the result of the measurement comprising the estimate y of the

measurand, the associated expanded uncertainty U and the coverage factor k in
the calibration certificate in accordance with Section 7.
()

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 21/ 70
Appendix A
Probability distribution
A.1
Normal distribution
The probability density function p(x) of the normal distribution is as follows:
P (x ) =
1
2
exp (x ) / 2 2 ,
2
-<x <+
(A.1)
Where is the mean and is the standard deviation. Figure (A.1) represents such a
distribution.
Figure A.1: Schematic view of the normal (Gaussian) distribution
A.1.1
When to use normal distribution

In some situation, the quoted uncertainty in an input or output quantity is stated
along with level of confidence. In such cases, one has to find the value of coverage
factor so that the quoted uncertainty may be divided by this coverage factor to
obtain the value of standard uncertainty. The value of the coverage factor depends
upon the distribution of the (input or output) quantity. In the absence of any specific
knowledge about this distribution, one may assume it to be normal. Values of the
coverage factor for various level of confidence for a normal distribution are as
follows:

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 22/ 70
Table A.1: Confidence Level and the corresponding Coverage factor (k)
Confidence 68.27 %
level
Coverage 1.000
factor (k)
90 %
95 %
95.45 %
99 %
99.73 %
1.645
1.960
2.000
2.576
3.000
If based on available information, it can be stated that there is 50 % chance that the value of
input quantity Xi lies in the interval between a and a + and also it is assumed that the
distribution of Xi is normal, then the best estimate of Xi is:
x i = a = (a + a + ) / 2 , with u(x i ) = 1.48 a
(A.2)
If based on available information, it can be stated that there is 68% chance that the value of
input quantity Xi lies in the interval of a and a + ; and also it is assumed that the distribution of
Xi is normal, then the best estimate of Xi is :
x i = a, with u(x i ) = a
A.2
(A.3)
Rectangular distribution
The probability density function p(x) of rectangular distribution is as follows:
P (x ) =
1
, a < x < a + , where a = (a + a ) / 2
2a
(A.4)
Figure (A.2) represents such a distribution. The expectation of Xi is given as xi
E (X i ) = x i = (a + + a ) / 2
(A.5)
and its variance is
Var (X i ) = a 2 / 3 , where a = (a + a ) / 2
(A.6)
A.2.1 When to use rectangular distribution

In cases, where it is possible to estimate only the upper and lower limits of an input
quantity Xi and there is no specific knowledge about the concentration of values of Xi
within the interval , one can only assume that it is equally probable for Xi to lie
anywhere within this interval . In such a situation rectangular distributions is used.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 23/ 70
- a /3
+ a /3
Figure A.2: Schematic view of the rectangular distribution
A.3
Symmetrical Trapezoidal Distribution

The above rectangular distribution assumes that Xi can assume any value within the
interval with the same probability. However, in many realistic cases, it is more
reasonable to assume that Xi can lie anywhere within a narrower interval around the
midpoint with the same probability while values nearer the bounds are less and less
likely to occur. For such cases, the probability distribution is represented by a
symmetric trapezoidal distribution function having equal sloping sides (an isosceles
trapezoid), a base of width a + a = 2a, and a top of width 2 a , where 0 1
is used .
The expectation of Xi is given as:
E (X i ) = (a + + a ) / 2 ,
(A.7)
and its variance is
Var (X i ) = a 2 (1 + )2 /6
(A.8)
When 1, the symmetric trapezoidal distribution is reduced to a rectangular

distribution.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 24/ 70
A.3.1
Triangular Distribution
When = 0, the symmetric trapezoidal distribution is reduced to a triangular
distribution. Figure (A.3) shows such a distribution. When the greatest concentration of
the values is at the center of the distribution, then one must use the triangular
distribution.
The expectation of Xi is given as,
E (X i ) = (a + + a ) / 2 ,
(A.9)
Var (X i ) = a 2 / 6
(A.10)
and its variance is
Figure A.3: Schematic view of the triangular distribution

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 25/ 70
A.4
U-Shaped Distribution
This U-shaped distribution is used in the case of mismatch uncertainty in radio and
microwave frequency power measurements (shown in figure (A.4). At high frequency
the power is delivered from a source to a load, and reflection occurs when the
impedances do not match. The mismatch uncertainty is given by 2s L where s and
L are the reflection coefficients of the source and the load respectively. The standard
uncertainty is computed as:
u2 (xi) = (2 S L)2 / 2
(A.11)
Figure A.4: Schematic view of the U-shaped distribution

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 26/ 70
Appendix B
Coverage factor derived from effective degrees of freedom

B.1
To estimate the value of a coverage factor k corresponding to a specified coverage

probability requires that the reliability of the standard uncertainty u(y) of the output
estimate y is taken into account. That means taking into account how well u(y)
estimates the standard deviation associated with the result of the measurement. For an
estimate of the standard deviation of a normal distribution, the degrees of freedom of
the estimate, which depends on the size of the sample on which it is based, is a
measure of the reliability. Similarly, a suitable measure of the reliability of the standard
uncertainty associated with an output estimate is its effective degrees of freedom eff,
which is approximated by an appropriate combination of the i for its different
uncertainty contributions u i (y).
B.2
The procedure for calculating an appropriate coverage factor k :
B.3
Step 1
Obtain the standard uncertainty associated with the output estimate.
Step 2
Estimate the effective degree of freedom eff of the standard uncertainty u(y)
associated with the output estimate y from the Welch-Satterthwaite formula.
Step 3
Obtain the coverage factor k from the table of values of student t distribution.
If the value of eff is not an integer, it is truncated to the next lower integer and
the corresponding coverage factor k is obtained from the table.
Welch-Satterthwaite formula is as follows:
eff
u4 (y)
= N 4
ui ( y )
i =1
(B.1)
where u i (y) (i = 1,2,3,..N) defined in Eqs. (5.1) and (5.2) , are the contributions to
the standard uncertainty associated with the output estimate y resulting from the
standard uncertainty associated with the input estimate x i which are assumed to be
mutually statistically independent , and the i is the effective degrees of freedom of the
standard uncertainty contributions u i (y).
Note :
The calculation of the degrees of the freedom for Type A and Type B of the evaluation
may be as follows:

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 27/ 70
Type A Evaluation
For the results of direct measurement (Type A evaluation), the degree of freedom is related to
the number of observations (n) as,
i = n - 1
(B.2)
Type B Evaluation
For this evaluation, when lower and upper limits are known
i
(B.3)
It is suggested that i should always be given when Type A and Type B evaluations of
uncertainty components are documented.
Where high precision measurements are undertaken, the accredited calibration laboratories
shall be required to follow ISO Guide to the expression of uncertainty in Measurement (1995).
Concerned laboratories should refer to Annexure G (with special emphasis on table G-2 )
and Annexure H for related examples.
However, assuming that i is not necessarily unrealistic, since it is a common practice to
chose a- and a+ in such a way that the probability of the quantity lying outside the interval a- to
a+ is extremely small.
Further interpretation on the above is given on page 29 & 30

Doc. No: NABL 141
Issue No: 02

Amend No: 01
Page No: 28/ 70
Interpretation on Effective Degrees of Freedom

Whilst the reason for determining the number of degrees of freedom associated with an
uncertainty component is to allow the correct selection of value of students t, it also gives an
indication of how well a component may be relied upon. A high number of degrees of freedom
is associated with a large number of measurements or a value with a low variance or a low
dispersion associated with it. A low number of degrees of freedom corresponds to a large
dispersion or poorer confidence in the value.
Every component of uncertainty can have an appropriate number of degrees of freedom,,
assigned to it. For the mean, x, for example = n - 1, where n is a number of repeated
measurements. For other Type A assessments, the process is also quite straightforward. For
example, most spreadsheets provide the standard deviation of the fit when data is fitted to a
curve. This standard deviation may be used as the uncertainty in the fitted value due to the
scatter of the measurand values. The question is how to assign components evaluated by Type
B processes.
For some distributions, the limits may be determined so that we have complete confidence in
their value. In such instances, the number of degrees of freedom is effectively infinite. The
assigning of limits, which are worst case, leads to this instance, namely infinite degrees of
freedom, and simplifies the calculation of effective degrees of freedom of the combined
uncertainty.
If the limits themselves have some uncertainty, then a lesser number of degrees of freedom
must be assigned. The ISO Guide to the expression of uncertainty in measurement (GUM)
gives a formula that is applicable to all distributions. It is equation G.3 that is:
[ u(xi)/ u(xi) ]-2
Where :
u(xi)/ u(xi) is the relative uncertainty in the uncertainty
This is a number less than 1, but may for convenience be thought of as a percentage or a
fraction. The smaller the number, the better defined is the magnitude of the uncertainty.
For example, if relative uncertainty is 10%, i.e.
u(xi)/ u(xi) = 0.1
Then it can be shown that the number of degrees of freedom is 50. For a relative uncertainty of
25 % then = 8 and for relative uncertainty of 50 %, = is only 2.
Rather than become seduced by the elegance of mathematics, it is better to try to determine
the limits more definitely, particularly if the uncertainty is a major one. It is of the interest to note
that equation (1) tells us that when we have made 51 measurements and taken the mean, the
relative uncertainty in the uncertainty of the mean is 10%. This shows that even when many
measurements are taken, the reliability of the uncertainty is not necessarily any better than
when a type B assessment is make. Indeed, it is usually better to rely on prior knowledge
rather than using an uncertainty based on two or three measurements. It also shows why we
restrict uncertainty to two digits. The value is usually not reliable enough to quote to better than
1 % resolution.
Doc. No: NABL 141
Issue No: 02

Amend No: 02, 03
Page No: 29/ 70
Once the uncertainty components have been combined, it remains to find the number of
degrees of freedom in the combined uncertainty. The degrees of freedom for each component
must also be combined to find the effective number of degrees of freedom to be associated
with the combined uncertainty. This is calculated using the Welch-Satterthwaite equation,
which is:
n
eff = [uc4 (y)/ { ui4 (y)/ i}]
1
. 2
Where:
eff
is the effective number of degrees of freedom for uc the combined uncertainty
is the number of degrees of freedom for u i , the ith uncertainty term
u i (y) is the product of c i u i

The other terms have their usual meaning.
Adopted from NATA document on Assessment of uncertainties in Measurement, 1999.

Doc. No: NABL 141
Issue No: 02

Amend No: 03
Page No: 30/ 70
Table B.1: Student t-distribution for degrees of freedom . The t-distribution for defines an
interval -t p () to + t p () that encompasses the fraction p of the distribution. For p = 68.27 %,
95.45 %, and 99.73 %, k is 1, 2, and 3, respectively.
Degrees
Freedom
Fraction p in percent
()
68.27
90
95
95.45
99
99.73
1.84
6.31
12.71
13.97
63.66
235.80
1.32
2.92
4.30
4.53
9.92
19.21
1.20
2.35
3.18
3.31
5.84
9.22
1.14
2.13
2.78
2.87
4.60
6.62
1.11
2.02
2.57
2.65
4.03
5.51
1.09
1.94
2.45
2.52
3.71
4.90
1.08
1.89
2.36
2.43
3.50
4.53
1.07
1.86
2.31
2.37
3.36
4.28
1.06
1.83
2.26
2.32
3.25
4.09
10
1.05
1.81
2.23
2.28
3.17
3.96
11
1.05
1.80
2.20
2.25
3.11
3.85
12
1.04
1.78
2.18
2.23
3.05
3.76
13
1.04
1.77
2.16
2.21
3.01
3.69
14
1.04
1.76
2.14
2.20
2.98
3.64
15
1.03
1.75
2.13
2.18
2.95
3.59
16
1.03
1.75
2.12
2.17
2.92
3.54
17
1.03
1.74
2.11
2.16
2.90
3.51
18
1.03
1.73
2.10
2.15
2.88
3.48
19
1.03
1.73
2.09
2.14
2.86
3.45
20
1.03
1.72
2.09
2.13
2.85
3.42
25
1.02
1.71
2.06
2.11
2.79
3.33
30
1.02
1.70
2.04
2.09
2.75
3.27
1.000
1.645
1.960
2.000
2.576
3.000

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 31/ 70
Appendix C
Solved examples showing the applications of the method outlined here to
eight specific problems in different fields
C.1 Micrometer calibration using 0 grade slip gauge at 25mm

Introduction
The instrument under calibration is a micrometer of 0 25 mm range with a slip gauge of 25
mm of nominal size.
The detailed specifications of the slip gauge are as follows:
Range = 25 mm,
Standard reference temperature (Tr e f) = 20 0 C,
Actual calibrating temperature (Tc) = 23 0 C,
Calibrated value = 25.00010 0.00008 mm, and
Least count of thermometer used = 1 0 C.
Mathematical model
YGUT =
XSTD +
(C.1)
Where YGUT is the micrometer reading [Gauge under test (GUT), XSTD is the gauge block size
and X is the error or the difference between the micrometer reading and gauge block size.
Uncertainty equation
The combined standard uncertainty equation is given by,
uc (YGUT ) = GUT {u ( STD )} + GUT {u (X )}
STD

(C.2)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 32/ 70
Measured results
Type A evaluation
Five readings are taken and the deviation from the nominal value is as follows.
Mean Deviation:
n
x = ( x j ) /n = 0.0006mm
(C.3)
j =1
Standard deviation :
s( x ) =
1 n
xj x
n 1 j =1
1
(
12 10 7 ) = 3 10 7 mm = 5.47 10 4 mm
4
(C.4)
Table C.1 : Data for calculation of mean and standard deviation
Observation
numbers
Deviation from nominal value (xj)

(mm)
1.
2.
3.
4.
5.
0.001
0.000
0.001
0.000
0.001
x
(mm)
(xj - x)2 10-7

(mm)
0.0006
1.6
3.6
1.6
3.6
1.6
Standard deviation of the mean:
()
sx =
s 2 ( x ) 5.47 10 4
=
= 2.446 10 4 mm = 0.2446m
n
5
(C.5)
Standard uncertainty:
()
u x = 0.2446m
(C.6)
Degrees of freedom (i)

= n1 =
5 1 = 4
(C.7)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 33/ 70
Type B evaluation
The uncertainty quoted in the gauge block calibration certificate is considered to be Type B
uncertainty of normal distribution.
Standard uncertainty (u1) due to the temperature measurement 1 0 C.
Standard thermal expansion coefficient of the gauge block is 11.510-6/ 0C
u 1 = 25 111.5 10-6 mm = 287.5 10-6 mm = 0.2875 m
(C.8 )
Standard uncertainty (u2) due to difference in temperature of micrometer and slip gauge
Assuming the temperature of the slip gauges and micrometer are the same but still it can have
a difference 1 0 C. Hence, again uncertainty component
u2 = 0.2875 m
(C.9)
Standard uncertainty (u3) due to difference in thermal expansion coefficient of the slip
gauge and micrometer
It is assumed that the difference in thermal expansion coefficient of standard slip gauge and
the micrometer screw is amounting to 20 %, hence the uncertainty component
[T = Tc Tref = 30 C],
u3 = 25 3 11.5 10-6 (20 /100) mm = 0.1725 m
(C.10)
Standard uncertainty (u4) due to the flatness of micrometer faces
say 1 fringe u 4 = 0.5 m

2
(C.11)
Standard uncertainty (u5) due to the parallelness of micrometer faces
say1 fringe u 5 = 0.5 m

2
(C.12)
Standard uncertainty (u6) due to the Standard used for calibration

The uncertainty in the value of the standard is taken from the calibration certificate say 0.08
m. Assuming rectangular distribution, the standard uncertainty is
u6 =
0.08
m = 0.046m
3
(C.13)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 34/ 70
The sensitivity coefficients (ci) are 1 and degree of freedom is i = [Type B components ] in
all six cases .
Degrees of freedom (eff )
eff =
uc (y )
4
n (u c (y ))
4
i =1
Veff =
(0.531)4
(0.2446)4 + (0.165 )4 + (0.165 )4 + (0.099)4
(0.288 )4
(0.288 )4
(0.046)4
(C.14)
0.07950
0.00358
4
89
Combined uncertainty
Combined uncertainty [uc (YGUT)] is ,
(0.2446)2 + (0.165)2 +(0.165)2 +(0.099 )2 + (0.288)2 + (0.288)2 + (0.046)2 m

uc (YGUT )
0.531 m
(C.15)
(C.16)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 35/ 70
Table C.2: Uncertainty Budget:

Source of
Uncertainty
Xi
Estimate
s
xi
(m)
Limits
x i
(m)
Probability
Distribution
Type A or B
- factor
Standard
Sensitivity Uncertainty Degree
Uncertainty coefficient contribution of
u(x i)
u i (y)
ci
freedom
(m)
(m)
i
u1
0.287
0.143
0.165
1.0
0.165
u2
0.287
0.143
0.165
1.0
0.165
u3
0.172
0.086
0.099
1.0
0.099
u4
0.5
0.25
0.288
1.0
0.288
u5
0.5
0.25
0.288
1.0
0.288
u6
0.08
0.04
0.046
1.0
0.046
Repeatability
u(x)
Rectangular
- Type B
- 3
Rectangular
- Type B
- 3
Rectangular
- Type B
- 3
Rectangular
- Type B
- 3
Rectangular
- Type B
- 3
Rectangular
- Type B
- 3
Normal
- Type A
- 5
0.547
1.0
0.2446
0.531
1.062
uc (YGUT )
Expanded
uncertainty
k=2
Expanded Uncertainty (U)

From the students distribution table, for the confidence level of 95.4 % and for
eff = , the coverage factor k = 2.
U = k uc (YGUT ) = 2 0.531 m = 1.062 m
Reporting of results
The value at 25 mm is
25.00010 mm 1.062 m
with coverage factor k = 2 for confidence level of 95.4 % and for
eff
= .

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 36/ 70
C.2 Calibration of weight of nominal value 10 kg

Introduction
The calibration of a weight of nominal value 10 kg of OIML class M1 is carried out by
comparison to a reference standard (OIML class F2 ) of the same nominal value using a mass
Comparator whose performance characteristics have previously been determined.
Mathematical model
The unknown conventional mass mx, is obtained as
mx = ms+ mD + m + mc + A
where ms
mD
m
standard
mc
A
(C.17)
conventional mass of the standard,

drift of the value of the standard since its last calibration,
observed difference in mass between the unknown mass and the
correction for eccentricity and magnetic effects,

correction for air buoyancy.
For uncorrelated input quantities, the combined standard uncertainty is given by

n
f
uc2 ( y ) = u 2 ( xi )
i =1 x
2
(C.18)
Details of the specifications

1.
Reference standard (ms): The calibration certificate for the reference standard gives a
value of 10,000.005 g with an associated expanded uncertainty of 45 mg (coverage
factor k = 2)
2.
Drift of the value of the standard (mp): The drift of the value of the reference standard
is estimated from the previous calibrations to be zero within 15 mg.
3.
Comparator (m): A previous evaluation of the repeatability of the mass difference

between two weights of the same nominal value gives a pooled estimate of standard
deviation of 25 mg.
4.
Eccentricity and magnetic effect (mc): The variation of mass due to eccentric load and
magnetic effect is found to be 10 mg.
5.
Air buoyancy (A): The limit of air buoyancy correction is found to be 10 mg.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 37/ 70
Table C.3 : Observations

No.
A reading
(g)
B reading
(g)
B reading
(g)
A Reading
(g)
1.
2.
3.
4.
5.
0.015
0.010
0.020
0.020
0.010
0.020
0.030
0.045
0.040
0.030
0.025
0.020
0.040
0.030
0.020
0.010
0.010
0.015
0.010
0.010
Observed
difference
(g)
0.010
0.015
0.025
0.020
0.015
Type A evaluation
Five observations of the difference in mass between the unknown mass (B) and the standard
(A) are obtained using the substitution method and the ABBA weighing sequences:
Arithmetic mean m = 0.017 g,
pooled estimate of standard deviation sp ( m) = 0.025 g (obtained from prior evaluation)
Standard uncertainty uA = u(m) = s(m ) =
25
mg = 11.18 mg,
5
Degrees of freedom = 5 1 = 4
Type B evaluation
1.
From the calibration certificate of the standard (A), the expanded uncertainty (U) is
certified as 45 mg with a coverage factor k = 2. the standard uncertainty
u (ms ) =
2.
Drift in the value of the standard is quoted as 15 mg.

distribution, the standard uncertainty
u (m D ) =
3.
45
mg = 22.5mg
2
Assuming a rectangular
15
mg = 8.66mg
3
Due to eccentricity and magnetic effects, the variation of mass is found to be 10 mg.
Assuming a rectangular distribution, the standard uncertainty,
u (m c ) =
10
mg = 5.77mg
3

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 38/ 70
4.
Error in air buoyancy correction is reported to be 10 mg.

distribution, the standard uncertainty
u ( ) =
uB =
Assuming rectangular
10
mg = 5.77mg
3
(22.5)2 + (8.66)2 + (5.77 )2 + (5.77 )2 mg = 25.45mg
(C.19)
Degrees of freedom (i): In all these four cases, the degree of freedom is i =
Degrees of freedom eff
v eff =
(27.8)4
(11.8)4 + (22.5)4 + (8.66)4 + (5.77 )4
4
(
5.77 )
+
598029.44
15623.1
4
~ 153.1
Combined standard uncertainty

Combined standard uncertainty is
uc = u A2 + u B2 =
(11.18)2 + (25.45)2
= 27.8mg
(C.20)
Expanded uncertainty
U = ku c (m) = 2 27.8 mg = 55.6 mg

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 39/ 70
Reported result
The measured mass of the nominal 10 kg weight is 10.000025 kg 56 mg. The reported
expanded uncertainty of measurement is stated as the standard uncertainty in measurement
multiplied by the coverage factor k = 2, which for a normal distribution corresponds to a
coverage probability 99.5 %.
Source of
Uncertainty
Estimates
xi
Xi
Limits
x i
ms
10.000005 45 mg
kg
m d
15 mg
7.5 mg
m c
10 mg
5 mg
10 mg
5 mg
Repeatabilit
y
Probability
Distribution
-Type A or B
- Factor
Standard
Sensitivity Uncertainty
Uncertainty coefficient contribution
u(x i)
(mg)
ci
Normal
- Type B
- 2
Rectangular
- Type B
- 3
Rectangular
- Type B
- 3
Rectangular
- Type B
- 3
Normal
- Type A
- 5
22.5
1.0
22.5
8.66
1.0
8.66
5.77
1.0
5.77
5.77
1.0
5.77
11.18
11.18
27.8
55.6
uc (m )
Expanded
uncertainty
k=2
u i (y)
(mg)
Degree
of
freedom
vi

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 40/ 70
C.3 Calibration of a Industrial Dead Weight Tester

An industrial Dead Weight Tester (DWT) up to 60 MPa is calibrated against a standard piston
cylinder (SPC) assembly, which is used as the secondary standard. The specifications of the
industrial DWT and SPC are as follows:
Dead Weight Tester
Range: 0.1 60 MPa
Serial No: XXXXXXX
Make: XXXXXXXX
Manufacturers data :
Resolution - = 0.1 MPa
Accuracy - 0.1 % of full scale pressure
SECONDARY STANDARD STANDARD PISTON CYLINDER ASSEMBLY
Range 0.2 100 MPa
Class S
The data obtained from the characterization of the piston gauge is as follows:
Resolution - 0.01 MPa
Standard uncertainty 0.01 MPa in the whole studied range of pressure
Calibration Procedure
The calibration is carried out by crossfloat method. This method starts with the following steps:
Leveling: - the instrument has been leveled so that the axis of rotation of the piston is vertical.
Rotation: - during the calibration, the piston of the secondary standard gauge (SPC) as well as
the industrial DWT have been rotated at a constant rpm with a synchronous motor to relieve
friction.
Temperature: - the temperature of SPC and DWT has been maintained near 230C and
suitable temperature correction has been incorporated to make all the results at the same
temperature, that is, at 23 0C.
Repetition: - the calibrations have been carried out at least five times under identical condition
both at the increasing pressure as well as decreasing pressure.
Acceleration due to gravity (g) correction: - The correction has been incorporated whenever
there is a difference in g- value. For example, gNPL is 9.791241 m/s2 therefore the correction
factor is (9.791241/gManufac).

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 41/ 70
Reference level: - The pressure generated at the reference level both at the secondary
standard (SPC) and test gauge (DWT) has been maintained constant.
Mathematical Model
The mathematical relationship can be modeled as:
PDWT = PSPC + P,
Where PDWT is the pressure as measured by the industrial DWT , PSPC
is the average
standard pressure as obtained from the standard gauge PSPC . The average sign indicate the
arithmetic mean of several repetitive measurement under identical condition.
P is the
difference between the two readings of the PDWT and the PSPC . Table (C.4) represents the data
where represents the standard uncertainty in each set of readings at a given pressure.
In the simplest case and also in this limited pressure range up to 100 MPa, it is normally
observed that the P is a linear function of PSPC , which is,
P = Po +
where P0 and S 1
S1
PSPC
(C.22)
are assumed to be constants ,
The combined standard uncertainty equation is then given by

2
uc (PDWT ) = DWT {u (PSPC )} + DWT {u (O )} + DWT {u (S1 )} (C.23)
SPC
O
S1
where the bracketed quantities are the standard uncertainties due to the repeatability of the
readings and the standard gauge, respectively. It may be mentioned here that the partial
derivatives of PDWT with PSPC and PDWT with Po is 1 , therefore, the sensitivity coefficients
also for both these cases equal to one. However partial derivative of PDWT with S1 is PSPC .
Therefore, we have to take into account the value of
standard uncertainty.
PSPC
for the estimation of combined

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 42/ 70
Experimental Results
Table (C.4) shows the data as obtained from the experiment. As mentioned, we have taken
five readings (n = 5) of the same pressure point while increasing / decreasing the pressure
cycle. The PSPC is the average of these five readings. P is the error or the difference
between PDWT and PSPC . PW represents Piston and Weight hanger.
Type A evaluation of standard uncertainty
(1)
Repeatability
We have taken 5 repeatable readings at each and every pressure or n= 5. The maximum
standard uncertainty () from the table (C.5) is,
s(qk)
= 0.0026827 MPa
(C.24)
Hence, the standard uncertainty is given by,
u1(s ) = 0.0026827
5 = 0.001199MPa
(C.25)
Degree of freedom i = 5 - 1 = 4
Table C.5: Increasing and Decreasing Pressure
Weight
used
Nominal
Pressure
PDWT
PW
PW , 7-8
PW,1
PW 1-2
PW , 1-3
PW,1-4
PW, 1-4 ,
6
PW , 1-5
PW , 1-6
PW, 1-9
PW , 1-9
PW , 1-6
PW, 1-5
PW, 1-4 ,
6
PW , 1-4
PW, 1-3
PW , 1-2
PW , 1
PW , 7-8
PW
Reading of SPC
(MPa)
1.0
5.0
10.0
20.0
30.0
40.0
45.0
0.9989
4.9965
9.9931
19.9861
29.9791
39.9694
44.9665
50.0
55.0
60.0
60.0
55.0
50.0
45.0
40.0
30.0
20.0
10.0
5.0
1.0
PSPC
(MPa)
0.9992
4.9967
9.993
19.9865
29.9787
39.9706
44.9654
(MPa)
0.9988
4.9964
9.9933
19.9867
29.9782
39.971
44.9659
0.9978
4.9971
9.9931
19.9871
29.9799
39.9701
44.9671
0.9991
4.9961
9.9935
19.9861
29.9787
39.9691
44.9669
0.9987
4.9965
9.9932
19.9865
29.9789
39.9700
44.9663
(kPa)
0.619
0.422
0.162
0.477
0.728
0.896
0.675
(kPa)
1.24
3.44
6.80
13.50
21.08
29.96
33.64
49.96
54.9548
59.9494
59.9497
54.9568
49.9641
44.9675
49.9611
54.9576
59.9532
59.9552
54.9576
49.9611
44.9654
49.9606
54.9556
59.9498
59.9549
54.9556
49.9606
44.9659
49.9604
54.9561
59.9512
59.9512
54.9561
49.9604
44.9671
49.9595
54.9545
59.9532
59.9532
54.9545
49.9595
44.9669
49.9603
54.9557
59.9513
59.9528
54.9561
49.9611
44.9665
0.626
1.364
1.819
2.682
0.780
0.911
1.000
39.68
44.28
48.64
47.16
43.88
38.86
33.44
39.9704
29.9799
19.9871
9.9939
4.9975
0.9997
39.9706
29.9787
19.9875
9.993
4.9967
0.9992
39.971
29.9782
19.9867
9.9933
4.9964
0.9988
39.9701
29.9801
19.9871
9.9931
4.9971
0.9978
39.9691
29.9787
19.9861
9.9935
4.9961
0.9991
39.9702
29.9791
19.9869
9.9933
4.9967
0.9989
0.603
0.943
0.425
0.392
0.467
0.116
29.76
20.88
13.10
6.64
3.24
1.08

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 43/ 70
(2)
Data Analysis
As mentioned in the mathematical model, there is a difference in pressure (P) at each
pressure of PSPC [(Eq. (C.22)] . Thus, this P can be fitted with PSPC
in a linear fitting
program.
In the present case, we have 20 data points as are shown in Table (C.5), the fitted
equation reduces to,
= 0.000802 PSPC (MPa ) 0.0013(MPa )
(C.26)
The different fitting parameters with standard uncertainty are shown in Table (C.6). From
the above Eq.(C.26), P is equal to 0.04682 MPa at 60 MPa. It is therefore P is
maximum at 60 MPa but reduces as we decrease the pressure.
Table C.6: Regression Output:
Source Parameters
Fitted Value
Po
U(Po)
(1 )
S1
u(S1 )
(1 )
Degrees of Freedom
-0.001304 (MPa)
0.0012762 (MPa)
0.000802
0.000014
19
The standard uncertainty in P is evaluated from Eq. (C.23) with the sensitivity
coefficient, as mentioned earlier, equals to PSPC .
( )
u2 ( ) = u (o ) + PSPC u (S1 )
2
(C.27)
At the maximum pressure [60 MPa], the standard uncertainty reduces to,
u2 ( ) =
(0.0012762)2 + (60 0.000014)2

= 0.001527 MPa
(C.28)
Therefore, estimated Type A standard uncertainty is
u A = u1 (s ) + u ( )
2
(0.001199)2 + (0.001527)2
= 0.00194 MPa
(C.29)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 44/ 70
Type B Evaluation of Standard uncertainty

From the specifications of the standard piston cylinder assembly, there is an equal probability
of the value lying anywhere between the lower (a-) and upper (a+) limits, the boundaries of this
rectangular distribution is given by
a =
a+ a
= 0.01MPa
2
(C.30)
Hence, the Type B component is given by,
uB =
a
0.01
=
= 0.0057MPa
3
3
(C.31)
Degree of freedom i =
Table C.7: Summary of standard uncertainty components
Source of
uncertainty
(X i)
Estimates
( xi )
Limits
x i
(MPa)
Probability
Distribution
Type A or B
- factor
(MPa)
Standard
uncertainty
u(x i )
Sensitivity
coefficient
(MPa)
uB
0.01
0.01
Rectangular
- Type B
3
Normal
-Type A
uA
uc (PDWT )
Expanded
uncertainty
Uncertainty
contribution
u i (y)
Deg. of
freedom
(i )
(MPa)
0.00570
1.0
0.0057
0.00194
1.0
0.00194
19
0.00602
0.012
k = 2.00

The combined standard uncertainty is then given by
uc (PDWT ) = u A2 + u B2
(0.00194)2 + (0.0057)2 MPa = 0.00602 MPa
(C.32)
(C.33)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 45/ 70
Effective Degree of freedom (eff)

The effective degree of freedom of the combined standard uncertainty is given by
eff =
(uc )4
u14 u24 u B4
+
+
v1 v1
(0.00602)4
(0.001199)4 + (0.001527)4 + (0.0057)4
4
19
(C.34)
Using the students t-distribution table, k = 2.0 for a confidence level of approximately 95.45%.
Therefore, the expanded uncertainty is given by
k
U=
uc (PDWT)
= 2.00 0.00602 MPa
0.012 MPa
For the range 0- 60 MPa, the uncertainty U is 0.012 MPa which is approximately 0.02 % of
the full scale pressure. This is determined from a combined standard uncertainty uc = 0.00602
MPa and a coverage factor k = 2.00 based on students distribution for = degrees of
freedom and estimated to have a level of confidence of 95.45 %.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 46/ 70
C.4 Estimation of measurement uncertainty in luminous flux

measurement
For Luminous flux measurement of light sources with integrating sphere, a substitution method
is applied in which a test lamp substitutes a luminous flux standard and the luminous flux of a
test source is evaluated by comparing the indirect illuminance in the two cases.
The luminous flux measurement has to be conducted as follows:
1.
2.
3.
4.
5.
6.
7.
Switch on the measuring equipment and let the auxiliary lamp warm up for 15 minutes.
Mount the standard lamp into the sphere center.
After burning in period, the indirect illumination by standard lamp Es is measured.
Turn the supply voltage down and switch off the standard lamp.
The switched on auxiliary lamp is moved into the sphere. It should remain switched on
always to avoid warm up period. The indirect illuminance EAS is measured.
The standard lamp is moved out of the sphere and the test lamp to be measured is
mounted into the sphere center with the auxiliary lamp still burning, the indirect
illuminance EAT is measured.
Turn on the test lamp to be measured. After burning in period the indirect illuminance
ET is measured.
From the above it is clear that the luminous flux of the test lamp is a function of the luminous
flux of the standard lamp S, the indirect illuminance from the auxiliary lamp EAS and EAT with
standard lamp inside the integrating sphere and test lamp inside the sphere respectively and
the indirect illuminance ES and ET produced by the standard lamp and the test lamp
respectively. The functional dependence of T, can be written as
T, = f (S, E S, E T, E AS, E AT)
(C. 35)
The luminous flux T, of the lamp can be calculated from the luminous flux S of the standard
lamp according to the following relation
T , = s
E s E AS
ET E AT
(C.36)
The factor E AS E AT considers the effect of different sizes and types of test lamps and the
standard lamps.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 47/ 70
Since the quantities on the RHS of the Eq. (C.36) are in product form, the equation for the
combined standard uncertainty can be expressed as an estimated relative combined variances
u 2 ( T ) u 2 ( S ) u 2 (E S ) u 2 (ET ) u 2 (E AS ) u 2 (E AT )
=
+
+
+
+
2
2
T2
2S
E S2
ET2
E AS
E AT
(C.37)
If S for standard source is given to be 1045 m and the standard uncertainty is 9.12 m,
from the measurement of expectation value of ES / E T and E AS / E AT, the value T can be
calculated from Eq. (C.36).
Table C.8: Observations
S. No.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
E AS
E AS
10.296
10.279
10.254
10.290
10.271
10.272
10.286
10.285
10.254
10.277
10.276
(E AS - E AS)2
400 10-6
9 10-6
484 10-6
196 10-6
25 10-6
16 10-6
100 10-6
81 10-6
484 10-6
441 10-6
For identical standard and test lamp of identical size, shape, same electrical parameters and
same colour temperature, the expectation value of ES / ET and EAS / EAT will be close to 1.
Hence the value of T will be almost equal to S and the uncertainty u (T) can be calculated
from Eq. (C.37). However, we will calculate value of T and standard uncertainty in u (T) by
the following example in which the uncertainties in the measurements of EAS and EAT and ES
and ET are calculated by statistical method and is an example of Type A evaluation of standard
uncertainty.
In the case study , for identical standard lamp and test lamp of identical shape and size , if the
expectation or the average values of the EAS , EAT , ES and ET are 10.276 lux , 10.20 lux, 81.14
lux and 83.76 lux , respectively , the value of T from Eq. (C.36) comes out to be
T = 1045
10.276 83.76
= 1086.6 m
10.20 81.14
(C.38)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 48/ 70
Type A Evaluation
We will illustrate by an example the calculation of the standard uncertainty u(E AS ) and relative
standard uncertainty u(E AS) / E AS for one of the parameters, e.g. E AS by Type A evaluation.
(E
10
i =1
Variance
s (E AS
2
AS
E AS
(
)= E
10
AS
i =1
= 2326 10 6
E AS
9
(C.39)
2326 10 6
9
(C.40)
Standard Deviation
s (E AS ) = 16.08 10 3
(C.41)
Standard deviation of the mean which is known as standard uncertainty is
( )
u E AS
16.08 10 3
=
= 5.08 10 3
10
(C.42)
The relative standard uncertainty is
( )
(C.43)
The degree of freedom in this case is

i = n 1 = 10 1 = 9
(C.44)
u E AS
= 5 10 4
E AS
Similarly the relative uncertainties for E AT, E S and E T, can be evaluated by Type A evaluation.
The degree of freedom in each case is also 9 as the total number of observations made in
each case are 10.
The values of the relative uncertainties are
( )
u E AT
E AT
= 6 10 4 and degree of freedom in this case is = i = n - 1 = 10 1 = 9
( )
u ES
= 1.2 10 2 and degree of freedom in this case is = i = n - 1 = 10 1 = 9
ES
u (ET )
= 1.6 10 2 and degree of freedom in this case is = i = n - 1 = 10 1 = 9
ET

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 49/ 70
Type B Evaluation
The uncertainty in the value of S is 9.12 m and is taken from the certificate of the
calibration of the standard lamp. Assuming rectangular distribution, the standard uncertainty in
the value of S is :
u ( s ) =
9.12
= 5.3m
3
(C.45)
degree of freedom in this case is i =

The relative standard uncertainty is
u ( s ) 5.3
=
= 5 10 3
1045
s
(C.46)

The value of the relative combined uncertainty
uc ( T )
for the value of T is calculated using
T
Eq. (C.37).
uc ( T )
= 25 10 6 + 144 10 6 + 256 10 6 + 0.36 10 6 + 0.25 10 6
T2
2
Therefore,
(C.47)
uc ( T )
= 2.06 10-2
T
Effective degrees of freedom
u c ( T )

T
=
4
4
4
u ( S )
u (E S )
u (ET )

E
E
S
S
+
+ T +
9
9
4
eff
u (E AS )
E
AS +
9
4
u (E AT )
E
AT
9
(C.48)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 50/ 70
eff =
[2.06 10 ]
] + [1.6 10 ] + [6 10 ] + [5 10 ]
2 4
[5 10 ] + [1.2 10
3 4
2 4
2 4
4 4
(C.49)
5 4
eff = 18.8 = 19
(C.50)
Expanded uncertainty (U)

From Students t distribution, for confidence level of 95 %, and for
eff = 18.8 = 19 , the coverage factor k = 2.09.
Reporting results
The value of T = (1086.6 22.4) m and expanded uncertainty with coverage factor k =
2.09, the value of T = (1086.6 46.8) m
Table C.9: Details of the uncertainty budget
Source of
Uncertainty
Estimates
xi
Xi
ES
u(ES )
ET
u(ET )
EAS
u(EAS )
EAT
u(EAT )
u(S )
Combined
uncertainti
es
uc(T )
Expanded
uncertainti
es
Limits
xi
Probability
Distribution
- Type A or B
- Factor
Standard
Uncertainty
u(xi )
coefficient contribution
ci
ui(y)
Normal
- Type A
- 10
Normal
- Type A
- 10
Normal
- Type A
- 10
Normal
- Type A
- 10
Rectangular
- Type B
1.2 10 2
1.0
1.2 10 2
1.6 10 2
1.0
1.6 10 2
5 10 3
1.0
5 10 3
1.6 10 3
1.0
1.6 10 2
5.3 lm
1.0
5.3 m
22.4 m
19
46.8 m
19
k = 2.09
Deg. of
freedom
i

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 51/ 70
C.5 Temperature measurement using thermocouple

Introduction
A digital thermometer with a Type K thermocouple was used to measure the temperature
inside a temperature chamber. The temperature controller of the chamber was set at 5000C .
Digital thermometer specification
Resolution: 0.10 C
Type K accuracy (one year): 0.6 0 C
Thermocouple
The Type K thermocouple is calibrated every year. The last calibration report provided an
uncertainty of 2.0 0 C at confidence level of 99 %. The correction for the thermocouple at
500 0 C is 0.50 C.
Measurement record
When the temperature chamber indicator reached 500 0 C, the readings were taken after a
stabilization time of half an hour. Ten measurements were taken as recorded in Table (C.10)
Mathematical Model
The mathematical model is represented as follows:
T = D + Correction
(C.51)
Where
T= Temperature measured.
D = Displayed temperature of the digital thermometer.
Correction = Correction due to the digital thermometer and Type K thermocouple
Uncertainty evaluation
The combined standard uncertainty (uc ) includes uncertainties of the repeatability of the
displayed readings, the digital thermometer and the thermocouple. This can be represented in
the equation below:
uC = u12 + u22 + u32
(C.52)
where, uc = combined standard uncertainty in the measurement,

u1 = standard uncertainty in the repeatability of measured readings,
u2 = standard uncertainty in the digital thermometer,
u3 = standard uncertainty in the thermocouple

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 52/ 70
Table C.10: Measurements record

Measurement (i)
1
2
3
4
5
6
7
8
9
10
T in 0 C
500.1
500.0
501.1
499.9
499.9
500.0
500.1
500.2
499.9
500.0
Analysis of measurement uncertainty components
Type A evaluation
(A) Standard uncertainty in the readings (u1)
Table (C.10) shows the data obtained from the experiment. Mean value,
T=
1 10
Ti = 500.02
10 i =1
(C.53)
where Ti are the 10 measurements taken as listed in Table (C.10). The temperature of the
chamber after taking into consideration the correction of the thermocouple is 500.50C.
The variance is calculated as follows:
s 2 (Ti ) =
1 n
Ti T
n 1 i =1
1
(0.096) = 0.0106 0 C 2
9
(C.54)
Standard deviation
s (Ti) = 0.103 0 C
(C.55)
Standard deviation of the mean is as follows:
()
u1 = s T =
s (Ti ) 0.103 0C
=
= 0.030 C
n
10
(C.56)
Thus the standard uncertainty (u1) is equal to 0.03 0 C.

Degrees of freedom (i ) = n - 1 = 10 1 = 9.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 53/ 70
Type B evaluation
Standard uncertainty (u2)
From specifications, the uncertainty in the digital thermometer is 0.60 C. Assuming
rectangular distribution, the standard uncertainty in the digital thermometer (u2) is,
u2 =
0.6
= 0.35 0C
3
(C.57)
degree of freedom (i ) =
Standard uncertainty (u3)
From calibration report, with a confidence level of 99 % (k = 2.58), the uncertainty in the
thermocouple is 2.0 0 C.
u3 =
2.0
= 0.78 0 C
2.58
(C.58)
degrees of freedom (i ) =
The value of the combined standard uncertainty is calculated using Eq. (C.52)
uc =
(0.03)2 + (0.35)2 + (0.78)2
= 0.85 0 C
(C.59)
Effective degrees of freedom
eff =
(0.85)4
(0.03)4
(C.60)
9
U = k uc = 2 0.85 = 1.7 0 C
(C.61)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 54/ 70
Table C.11
Source of
Uncertainty
Estimates Limits
xI
xI
Xi
Digital
Thermometer
Thermocoupl
e
C
0.6
C
0.3
2.0
Repeatability
Combined
uncertainty
Expanded
uncertainties
: Statement of the uncertainty budget
Probability
Distribution
- Type A or B
- Factor
Rectangular
- Type B
- 3
Normal
- Type B
- 2.58
Normal
- Type A
10
Standard
Uncertainty
u(xi )
0
C
0.35
coefficient contribution
ci
ui(y)
0
C
1.0
0.35
Deg. of
freedom
i
0.78
1.0
0.78
0.03
1.0
0.03
0.85
1.7
uc
U
k=2
The temperature of the chamber was measured to be 500.5 0 C. The measurement uncertainty
is 1.7 0 C. The reported measurement uncertainty is estimated at a level of confidence of
approximately 95 % with a coverage factor k of 2.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 55/ 70
C.6 Calibration of a 6 digit DMM on its 1 Volt AC range

Introduction
We discuss the method of calibration of a 6 digit DMM on its 1 Volt AC range at a nominal
0.5 V level at 1 kHz using 0.5 V calibrated thermal voltage converter (TVC) . AC voltage from
highly stable AC Calibrator is applied to both DMM (DUC) and the standard (TVC) connected
in parallel via a coaxial switch and a Tee adaptor for an indication of 0.500000 V on the DMM
and the emf ex indicated by the nanovoltmeter is noted. The AC Calibrator is replaced by a
calibrated DC calibrator and the DUC is disconnected. A DC voltage of positive polarity is
applied to TVC and is adjusted so as to repeat a reading of ex on the nanovoltmeter. The
output of the DC calibration is noted as V+. The polarity of the DC voltage is reversed and
above process is repeated and DC calibrator output voltage V- is recorded. The whole
measurement process is repeated several times.
Mathematical model
The mathematical model used is
VAC = ( VDC + VDC + Vth ) ( 1 + )
(C.62)
VAC is the voltage estimated for an indicated value of 0.500000 V on DUC, V DC = + + 2 is

average of two polarity DC voltage output of calibrator, is AC/DC transfer correction factor of
the TVC at the frequency of calibration. VDC is error of DC calibrator due to its three months
stability from the manufacturers data, as the calibrator was calibrated three months before and
Vth is error due to thermal emf which comes from the fact that the polarity of DC voltage is
reversed. This is very small as compared to 0.5 V being nearly equal to 1 V. So, finally the
equation becomes
VAC = VDC + VDC + VDC + VDC
(C.63)
The product VDC is extremely small and is neglected. The assumption is that the drift in
the values of is small and is also neglected and error of DMM in 1 volt range due to 1 count
is 1V and is also neglected. The precaution is that the interconnecting leads are coaxial
shielded and are kept very small. The reference plane of measurement (mid point of Tee) is
brought close to input plane of DMM (DUC). These precautions minimize the loading as well
as transmission errors. At frequencies up to 10 kHz, with above precautions taken, the error
contribution by above factors are very low ( 2to 3 10-6) and can be neglected.
The inputs are :
1.
The DC calibrator is regularly calibrated at intervals of six months. For the range of 1V
the uncertainty in the calibrator from its calibration certificate is 5.8 10-6 at 95 %
confidence level. Three months stability data from the manufacturers specifications is
5.0 10-6.
2.
The AC /DC transfer correction factor for the thermal converter is + 0.008 %. The
AC/DC transfer uncertainty is 0.01 % at 95 % confidence level.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 56/ 70
The observations are average of two polarity DC voltages.

Table C.12: Experimental observations
Serial Number
1
2
3
4
5
Readings (V)
0.499986
0.499982
0.499991
0.499994
0.499993
VAC = VDC + VDC + VDC
(C.64)
For uncorrelated input quantities, the combined standard uncertainty is

2
f
u = u 2 ( xi ),
i =1 xi
n
2
c
(C.65)
The components of total measurement uncertainty comprise of

u1 (V)
DC calibrators applied voltage uncertainty as mentioned in its calibration

certificate
u2 (V)
DC calibrators uncertainty due to its stability
u3 (V)
Uncertainty in the AC/DC transfer and
u4 (V)
Uncertainty due to repeatability and the corresponding sensitivity

coefficients are
c1 =
V AC
V AC
V AC
= 1, c 2 =
= 1, and c1 =
=1
V DC
V DC
(C.66)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 57/ 70
Type A evaluation
Mean DC Voltage = 0.499989 V,
Standard deviation = 0.0000005 V,
Standard deviation of mean or standard uncertainty
()
sq =
0.000005
= 2.23 10 6V
5
(C.67)
Degrees of freedom
i = 5 1 = 4
(C.68)
Type B evaluation
1.
Uncertainty of DC calibrator from its calibration certificate. The distribution is normal

and the coverage factor for 95 % confidence level is 1.96.
u1 (V ) =
a1
5.8
=
= 2.96 0.5 V = 1.48 V
1.96 1.96
(C.69)
Degrees of freedom =
2.
From DC calibrators specifications, uncertainty due to 3 months stability data a2 = 5.0

10-6. For rectangular distribution, the standard uncertainty
u 2 (V ) =
a2
3
5.0
3
= 2.89 0.5 V =1.44 V
(C.70)
3.
From AC/DC transfer at 95 % confidence level a3 = 100 10-6, distribution is normal and
coverage factor = 1.96
Standard uncertainty
u 3 (V ) =
a3
1 100
=
= 51.02 0.5 V = 25.5 V
1.96
1.96
(C.71)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 58/ 70

There is a dominant factor = 25.5 V.
(1.48)2 + (1.44)2 + (2.23)2
uc = 25.5 + u12 + u22 + u42 = 25.5 +
25.5 + 3.04 = 28.5 V
(C.72)
(C.73)
Effective degrees of freedom (veff)
eff =
[uc ]4
(u1 )4 + (u2 )4 + (u3 )4 + (u4 )4
eff =
(C.74)
[28.5]4
(1.48)4 + (1.44)4 + (2.23)4 + (25.5)4
(C.75)
For 95.45 % level of confidence, the coverage factor k = 2, thus
U = kuc (V) = 2 x 28.5 = 57V
(C.76)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 59/ 70

Source of
Uncertainty
Xi
Estimates Limits
xi
xi
u1
V
0.5
V
2.9
u2
0.0
2.5
u3
0.0
50.0
Repeatabilit
y
uc(Vac)
Expanded
uncertainty
Probability
Standard
Distribution Uncertainty
- Type A or B u(xi)
- Factor
V
Normal
1.48
-Type B
-1.96
Rectangular
1.44
-Type B
-3
Normal
25.5
-Type B
-1.96
Normal
-Type A
Sensitivity Uncertainty Degree

coefficient Contribution of
freedom
ci
ui (y)
vi
V
1.0
1.48
1.0
1.44
1.0
25.5
28.5
28.5
57.0
k=2.0
Reporting of result
The measured average AC Voltage corresponding to 0.500000 V indicated by the DMM,
VAC = 0.499989 (1 + 0.000008 ) V 57 V = 0.499993V 57 V
(C.77)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 60/ 70
C.7 Calibration of a RF Power Sensor at a Frequency of

18 GHz
Introduction
The measurement involves the calibration of an unknown power sensor against a standard
power sensor as reference standard by substitution on a stable, monitored source of known
reflection coefficient. The measurement is made in terms of calibration factor which for a
matched source is defined as the ratio of incident power at the calibration frequency to the
incident power at the reference frequency of 50 MHz under the condition that both incident
powers give equal power sensor response.
The calibration factor Kx of the unknown power sensor is determined by
Kx
Where Ks
Ds
DC
M
REF
=
=
=
=
=
= (Ks + Ds) DC M REF,
(C.78)
calibration factor of the standard sensor,

drift in standard sensor since the last calibration,
ratio of DC voltage outputs, and
ratio of mismatch losses,
ratio of reference power source (short-term stability of 50 MHz).
Method of measurement
Five separate measurements were taken which involved disconnection and reconnection of
both the unknown sensor and the standard sensor on a power transfer system. All
measurements are made in terms of voltage ratios that are proportional to calibration factor.
None of uncertainty contributions is considered to be correlated.
Mismatch uncertainty:
As the source is not perfectly matched and the phase relation of the reflection coefficients of
the source, the unknown and the standard sensors are not known, there is an uncertainty due
to mismatch for each sensor at the calibration frequency as well as reference frequency. The
corresponding limits of deviation is calculated from the well known formula:
Mismatch uncertainty =
2GS for the standard sensor.
(C.79)
Mismatch uncertainty =
2Gx for the unknown.
(C.80)
G, s and x are the reflection coefficients for source, the standard and the unknown,
respectively.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 61/ 70
For this case

G at 50 MHz = 0.02; at 18 GHz = 0.06,
s at 50 MHz = 0.02 ; at 18 GHz = 0.09, and
x at 50 MHz = 0.02 ; at 18 GHz = 0.10
(C.81)
C.82)
(C.83)
The long-term stability from the results of five annual calibrations was found to have limits not
greater than 0.4 % per year.
The values of reflection coefficients are themselves uncertain. This uncertainty has been
accounted for by adding it in quadrature with the actual measured values.
The instrumentation linearity uncertainty has been estimated to lie within 0.1 % from
measurements against a reference attenuation standard up to ratios of 2:1 at confidence level
of 95 %.
Reference source:
The ratio of power outputs of the reference source has been estimated to be 1.000 with
deviations 0.004.
Standard sensor:
The standard sensor was calibrated 6 months ago. The value of calibration factor from its
calibration certificate is
0.965 0.012
at confidence level of 95 %.
Type A evaluation
The measured values of calibration factor for the unknown power sensor are shown in Table
(C.14).
The mean value is
K x = 0.9496 0.950,
(C.84)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 62/ 70
Table C.14: Calibration factor for the unknown sensor

Number
1
2
3
4
5
Calibration Factor
0.958
0.946
0.951
0.950
0.943
The experimental standard deviation [s(Kx)],
s(K x ) = 0.0057
(C.85)
Standard uncertainty in u (s )K x is
u( K x ) =
0.0057
= 0.0025
5
(C.86)
Degree of freedom = 5 - 1 = 4.
Type B evaluation
1.
Uncertainty reported in calibration certificate of the standard sensor = 0.012 at the

confidence level of 95 %. The standard uncertainty u(Ks) is
2.
u(Ks) = 0.012 / 1.96 = 0.0061,
(C.87)
Uncertainty in drift in calibration factor since its last calibration is 0.002. This is a
rectangular distribution and the standard uncertainty,
u ( Ds ) =
0.002
= 0.0012 , and
3
(C.88)
3.
Uncertainty due to the stability of 50 MHz reference source is 0.004. This is a

rectangular distribution and the standard uncertainty u (REF)
u( REF ) =
0.004
= 0.0023, and
3
(C.89)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 63/ 70

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 64/ 70
4.
Uncertainty due to the instrument linearity is 0.001. This is a normal distribution and
the standard Uncertainty u (DC)
u( DC ) =
0.001
= 0.0023, and
1.96
(C.90)
5.
Uncertainty due to mismatch:

a)
b)
c)
d)
Standard sensor at 50 MHz = 0.0008

unknown sensor at 50 MHz = 0.0008
Standard sensor at 18 GHz = 0.0108
unknown sensor at 18 GHz = 0.012
This is U-shaped and the corresponding associated standard uncertainty figures are:
u (M s ) =
0.0008
= 0.00056 at 50MHz
2
(C.91)
u (M x ) =
0.0008
= 0.00056 at 50MHz
2
(C.92)
u (M s ) =
0.0108
= 0.0076 at 18GHz
2
(C.93)
u (M x ) =
0.012
2
= 0.0085 at 18GHz
(C.94)
Combined standard uncertainty:

u2c (Kx) = u2 (Ks) + u2 (Ds) + u2 (REF) + u2 () + u2 (MSX) + u2 (Kr)
(C.95)
u2c (Kx) = (0.0061)2 + (0.0012)2 + (0.0023)2 + (0.0005)2 +

2(0.0056)2 + (0.0076)2 + (0.0085)2 + (0.0025)2
or,
u c (K x ) 0.0134
(C.96)
Effective degrees of freedom eff

From Welch-Satterthwaite formula, veff is estimated and is found to be approximately 3301 or .

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 65/ 70

Source of
Uncertainty
Xi
Estimates Limits
Probability
Distribution
- Type A or B
Standard
Sensitivity
Uncertainty Coefficient
u(xi)
ci
Uncertainty Degree of
Contribution Freedom
ui (y)
vi
Normal
-Type B
Rectangular
-Type B
Normal
-Type B
Rectangular
-Type B
0.0061
1.0
0.0061
0.0012
1.0
0.0012
0.0005
1.0
0.0005
0.0023
1.0
0.0023
xi
xi
Ks
0.965
0.012
Ds
0.002
0.002
DC
1.0
0.001
REF
1.0
0.004
0.0
0.0
0.0008
0.0008
-U shaped
-do-
0.00056
0.00056
1.0
1.0
0.00056
0.00056
0.0
0.0
0.0108
0.012
-do-do-Type B
Normal
-Type A
5
0.0076
0.0085
1.0
1.0
0.0076
0.0085
0.0025
1.0
0.0025
0.0134
0.0263
3301
Mismatch at
50 MHz
S
X
Mismatch at
18 GHz
S
X
Repeatabilit
y
uc(Kx)
Expanded
uncertainty
k = 1.96
Expanded uncertainty U (Kx)

U (Kx) = 0.0134 2 = 0.0268 0.027
(C.97)
Reported Result:
The calibration factor of unknown power sensor at 18 GHz is 0.450 0.027
The reported expanded uncertainty of measurement is stated as the combined standard
uncertainty multiplied by the coverage factor 1.96, which for a normal distribution corresponds
to a coverage probability or confidence level of 95 %.

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 66/ 70
C.8 Calibration of 4 digital multimeter for its 100 V range

Introduction
We discuss the method of calibration of a 4 digital multimeter for its 100 V range with the
application of 10 V from calibrated direct volt calibrator.
Mathematical model
The mathematical model used for the evaluation:
Vx = Vs + Vx
Where
Vx
=
Vs
=
Vs
=
(C.98)
Voltage indicated in the DMM

Voltage applied from the calibrator
Error of the multimeter
Some simplifying assumptions have been made:

(i)
(ii)
errors due to loading and wire leads or connections are considered negligible, and
all input quantities are uncorrelated.
We have the following inputs:

(A)
The calibrator is calibrated regularly at the interval of six months. For the range of 10 V,
the specifications are: Resolution = 10 V with the uncertainty at 99 % level of
confidence as (4.5 10-6 of output + 100 V),
Vx = Vs + Vx
(B)
(C.99)
4 digital multimeter specification are: for range 100 V full display is 99.99 with
resolution = 10 mV and the uncertainty at 99 % level of confidence as (10-5 of reading
+ 0.2 10-5 of full scale).
Observation:
Applied voltage from calibrator
10.00000 V
Indicated voltage in DMM

10.01 V
Even after repetition of the readings the multimeter reading indicating the same value or 1
due to digitizing process. This is due to better accuracy of the reference standard (calibrator).

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 67/ 70

For uncorrelated input quantities, the combined standard uncertainty is
f
u =
i =1 xi
N
2
c
2
u ( xi )
(C.100)
The components of the total measurement uncertainty consist of

u1 (V) = the calibrators applied voltage uncertainty
u2 (V) = multimeters random effect uncertainty
Corresponding sensitivity coefficients are:
c1 =
V x
V x
= 1 and c2 =
=1
Vs
V x
(C.101)
Evaluation of uncertainty components:
Type A evaluation:
Even after repetition of the readings the multimeter reading indicating the same value or 1 due
to digitizing process. This is due to better accuracy of the reference standard (calibrator). In this
case, type A uncertainty can be assumed as negligible and the repeatability uncertainty can be
treated as type B uncertainty using the resolution error of the multimeter.
Type B evaluation:
(I) The uncertainty in applied voltage from the calibrator is
a1 =
4.5 10-6 OF OUTPUT + 100 V

=
(4.5 10 + 100) V
145 V
(C. 102)
At 99 % confidence level assuming normal distribution, coverage factor k = 2.58, the standard
uncertainty in applied voltage is
u1 (V) = a1/2.58 = (145 / 2.58) V = 56.20 V
(C.103)
Degree of freedom is = v1 =
(C.104)

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 68/ 70
(II)
From multimeter specification in the 100 V range, resolution is 10 mV (i.e. 1 count).

Since the reading was unchanged and assuming the limit to be half the counts.
a2 =
10
mV = 5mV
2
(C.105)
For rectangular distribution, the standard uncertainty due to the resolution uncertainty of
the multimeter is:
u 2V =
a2
3
5
3
mV = 2886.75 V
(C.106)
Degree of freedom is = v2 =
(C.107)
uc (V ) = u12 (V ) + u22 (V ) = 2.89mV

Effective degree of freedom veff
as v1 = and
(C.108)
v2 =
Expanded uncertainty (U)

For 95.45 % level of confidence the coverage factor, k = 2, thus
U = kuc (V) = 2 x 2.89 mV = 5.78 mV
Result:
The measured average voltage of the unknown cell is 10.01 V 5.78 mV.
The reported expanded uncertainty in measurement is stated as the standard
uncertainty in measurement multiplied by the coverage factor k = 2, which for a normal
distribution corresponds to a coverage probability of approximately 95.45 %

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 69/ 70
Table C. 16: Uncertainty Budget:

Source of
Uncertainty
Xi
Estimates
xi
Vs
10.00 V
Vx
0.01 V
Repeatabilit
y
Limits
xi
5 mV
Probability
Distribution
- type A or B
Standard
Uncertainty
u(xi)
Sensitivity
coefficient
ci
Uncertainty
contribution
ui (y)
Degree of
Freedom
vi
Normal
Type B
- 2.5
56.2 V
1.0
56.2 V
Rectangular
- Type B
-3
2886.25 V
1.0
2886.75 V
Normal
- Type A
uc(Vx)
Expanded
Uncertainty
k=2
2.89 mV
5.78 mV

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 70/ 70
CORE GROUP
Dr. S.P. Mukherjee (University Of Calcutta, Calcutta) Chairman

Dr. A.K. Chakrabarty (NABL, New Delhi)
Dr. P.K. Gambhir (BIS, New Delhi)
Dr. M.C. Gupta (NPL, New Delhi)
Dr. H.C. Kandpal (NPL, New Delhi)
Dr. V.N. Ojha (NPL, New Delhi)
Dr. V.K. Rustagi (NPL, New Delhi)
Dr. Rina Sharma (NPL, New Delhi)
Dr. A.K. Bandyopadhyay (NPL, New Delhi) - Convenor

Doc. No: NABL 141
Issue No: 02

Amend No: 00
Amend Date: -
Page No: 71/ 70

B-4/B-5/B-6 Apartments, Qutab Hotel
New Mehrauli Road
New Delhi 110 016
Tel. : 91-11 26529718 20, 26526864
Fax : 91-11 26529716
Website : www.nabl-india.org

NABL 141 - Guidelines For Estimation and Expression of Uncertainty in Measurement

Uploaded by

Copyright:

Available Formats

NABL 141 - Guidelines For Estimation and Expression of Uncertainty in Measurement

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

NABL 141 - Guidelines For Estimation and Expression of Uncertainty in Measurement

Uploaded by

Copyright:

Available Formats

GUIDELINES FOR ESTIMATION

PROFESSOR V.S. RAMARMURTHY

Reference to GUM- APLAC

29-30 Appen- 18.08.00

National Accreditation Board for Testing and Calibration Laboratories

Guidelines for Estimation and Expression of Uncertainty in Measurement

Uncertainty concept, sources and measures

Definitions of related terms and phrases

Evaluation of standard uncertainty in input estimates

Evaluation of standard uncertainty in output estimates

Expanded uncertainty in measurement

Statement of uncertainty in measurement

Apportionment of standard uncertainty

Step by step procedure for calculating the uncertainty in measurement

Appendix A : Use of relevant probability distribution

Appendix B : Coverage factor and effective degrees of freedom

Appendix C : Solved Examples

National Accreditation Board for Testing and Calibration Laboratories

Guidelines for Estimation and Expression of Uncertainty in Measurement

Uncertainty concept, sources and measures

National Accreditation Board for Testing and Calibration Laboratories

Guidelines for Estimation and Expression of Uncertainty in Measurement

National Accreditation Board for Testing and Calibration Laboratories

Guidelines for Estimation and Expression of Uncertainty in Measurement

Uncertainty Concept, Sources and Measures

A statement of results of measurement (as a process) is complete only if it contains

The uncertainty of measurement is a parameter, associated with the result of a

Errors in the observed results of a measurement (process) give rise to uncertainty

the way connections are made or the measurement method employed

uncontrolled environmental conditions or their influences

inherent instability of the measuring equipment

personal judgement of the observer or operator, etc.

These cannot be eliminated totally, but can be reduced by exercising appropriate

National Accreditation Board for Testing and Calibration Laboratories

Guidelines for Estimation and Expression of Uncertainty in Measurement

those reported in the calibration certificate of the reference standards

different influence conditions at the time of measurement compared with those

Measurands are particular quantities subject to measurement. In calibration one usually

An estimate of the measurand Y (output estimate) denoted by y, is obtained from Eq.

The standard uncertainty of measurement associated with the output estimate y,

An expanded uncertainty is obtained by multiplying the combined standard uncertainty

National Accreditation Board for Testing and Calibration Laboratories

Guidelines for Estimation and Expression of Uncertainty in Measurement

Definitions of related terms and phrases

National Accreditation Board for Testing and Calibration Laboratories

Guidelines for Estimation and Expression of Uncertainty in Measurement

Where n is number of observed pairs

National Accreditation Board for Testing and Calibration Laboratories

Guidelines for Estimation and Expression of Uncertainty in Measurement

experimental standard deviation [s(qj)]

f(x)dx is the probability element

National Accreditation Board for Testing and Calibration Laboratories

Guidelines for Estimation and Expression of Uncertainty in Measurement

random error is equal to error minus systematic error

because only a finite number of measurements can be made,

it is possible to determine only one estimate of random error

National Accreditation Board for Testing and Calibration Laboratories

Guidelines for Estimation and Expression of Uncertainty in Measurement