http://www.npl.co.uk/upload/pdf/cop04.

pdf

SM & T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

Manual of Codes of Practice for the Determination of Uncertainties in

Mechanical Tests on Metallic Materials
Code of Practice No. 04

The Determination of Uncertainties in Critical Crack Tip

Opening Displacement (CTOD) Testing

L.M. Plaza
INASMET
Centro Tecnolgico de Materiales
Camino de Portuetxe, 12
ES-20009 San Sebastin
SPAIN

Issue 1
September 2000

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

CONTENTS
1

SCOPE

SYMBOLS AND DEFINITIONS

INTRODUCTION

A PROCEDURE FOR THE ESTIMATION OF UNCERTAINTY IN CTOD

PARAMETER DETERMINATION
Step 1- Identifying the parameters for which uncertainty is to be estimated
Step 2- Identifying all sources of uncertainty in the test
Step 3- Classifying the uncertainty according to Type A or B
Step 4- Estimating the standard uncertainty for each source of uncertainty
Step 5- Computing the combined uncertainty uc
Step 6- Computing the expanded uncertainty U
Step 7- Reporting of results

REFERENCES
ACKNOWLEDGEMENTS
APPENDIX A
Mathematical formulae
determination
APPENDIX B
A worked example
determination

for

for

calculating

calculating

uncertainties

in

CTOD parameter

uncertainties

in

CTOD parameter

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

1.

SCOPE

This procedure covers the evaluation of uncertainty in the determination of Critical

Crack Tip Opening Displacement (CTOD) of metallic materials according to the testing
Standards
British Standard, BS 7448. Part 1-1991: Amd 1: August 1999.
Fracture Mechanics Toughness Tests. Part 1. Method for determination of
KIC, critical CTOD and critical J values of metallic materials.
ASTM E1290-93 Crack-Tip Opening Displacement (CTOD) Fracture
Toughness Measurement
These standards give a method for determining critical crack tip opening displacement
(CTOD) for metallic materials. The method uses fatigue precracked specimens. The
tests are carried out in displacement control with monotonic loading, and at a constant
rate of increase in stress intensity factor within the range 0.5 3 MPa m s-1 during the
initial elastic deformation. The specimens are loaded to the maximum force associated
with plastic collapse. The method is especially appropriate to materials that exhibit a
change from ductile to brittle behaviour with decreasing temperature. No other
influences of environment are covered.

2.

SYMBOLS AND DEFINITIONS

For a complete list of symbols and definitions of terms on uncertainties, see Reference
1, Section 2. The following are the symbols and definitions used in this procedure.

a
B
ci
CoP
CT
dv
E
F
f
g
k
K

n
p
q

nominal crack length

specimen thickness
sensitivity coefficient
Code of Practice
Compact Tension Test Specimen
divisor associated with the assumed probability distribution, used to
calculate the standard uncertainty
Youngs modulus of elasticity
Applied applied Force
mathematical function of (a/W)
Poissons ratio
coverage factor used to calculate expanded uncertainty
Stress Intensity Factor: the magnitude of the stress field near the
crack tip for a particular mode in a homogeneous, ideally linearelastic body.
number of repeat measurements
confidence level
random variable

arithmetic mean of the values of the random variable q

Page 1 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

s
S

span between outer loading points in three point bend test

experimental standard deviation (of a random variable) determined
from a limited number of measurements, n
Sy
proof strength
SE (B) three-point bend specimen
u
standard uncertainty
uc
combined standard uncertainty
U
expanded uncertainty
V
value of a measurand
Vp
plastic component of notch opening displacement
W
effective width of test specimen
xi
estimate of input quantity
Y
measurand
y
test (or measurement) mean result
z
distance of the notch opening gauge location above the surface of
the specimen

crack tip opening displacement

3.

INTRODUCTION

It is good practice in any measurement to evaluate and report the uncertainty associated
with the test results. A statement of uncertainty may be required by a customer who
wishes to know the limits within which the reported result may be assumed to lie, or the
test laboratory itself may wish to develop a better understanding of which particular
aspects of the test procedure have the greatest effect on results so that this may be
monitored more closely. This Code of Practice (CoP) has been prepared within
UNCERT, a project to simplify the way in which uncertainties are evaluated funded by
the European Commissions Standards, Measurement and Testing programme under
reference SMT4-CT97-2165. The aim is to produce a series of documents in a common
format that is easily understood and accessible to customers, test laboratories and
accreditation authorities.
This CoP is one of seventeen produced by the UNCERT consortium for the estimation
of uncertainties associated with mechanical tests on metallic materials. Reference 1 is
divided into six sections as follows with all individual CoPs included in Section 6.
1.
2.
3.
4.
5.
6.

Introduction to the evaluation of uncertainty

Glossary of definitions and symbols
Typical sources of uncertainty in materials testing
Guidelines for estimation of uncertainty for a test series
Guidelines for reporting uncertainty
Individual Codes of Practice (of which this is one) for the estimation of
uncertainties in mechanical tests on metallic materials

This CoP can be used as a stand-alone document. For further background information
on measurement uncertainty and values of standard uncertainties of the equipment and
instrumentation used commonly in materials testing, the user may need to refer to
Section 3 in Reference 1. The individual CoPs are kept as simple as possible by

Page 2 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

following the same structure:

The main procedure
Quantifying the major contributions to the uncertainty for that test type
(AppendixA)
A worked example (Appendix B)
This CoP guides the user through the various steps to be carried out in order to estimate
the uncertainty in the determination of the CTOD parameter.

4.

Step 1.

A PROCEDURE FOR ESTIMATING THE UNCERTAINTY IN THE

DETERMINATION
OF
CRITICAL
CRACK
TIP
OPENING
DISPLACEMENT (CTOD), USING A THREE POINT BEND TEST
SPECIMEN (SE (B))
Identifying the Parameters for Which Uncertainty is to be Estimated

The first step is to list the quantities (measurands) for which the uncertainties must be
calculated. Table 1 shows the parameter that is usually reported in the test. This
measurand is not measured directly, but is determined from other quantities (or
measurements).
Table 1 Measurand and Measurements, their units and symbols
Measurand
Crack tip opening displacement
Measurements
Thickness of the specimen
Width of the specimen
Crack length
Applied force
Plastic component of notch opening displacement
Span between outer loading points in three point bend test
Distance of the notch opening gauge location above the
surface notch specimen

Step 2.

Units
mm

Symbol

mm
mm
mm
N

B
W
a
F (F c, Fu, Fm)

mm
mm

Vp
s

mm

(See Fig A1)

Identifying all Sources of Uncertainty in the Test

In the Step 2, the user must identify all possible sources of uncertainty that may have an
effect (either directly or indirectly) on the test. The list cannot be identified
comprehensively beforehand, as it is associated uniquely with the individual test
procedure and apparatus used. This means that a new list should be prepared each time a
particular test parameter changes (e.g. when a plotter is replaced by a computer). To
help the user list all sources, five categories have been defined. The following table
(Table 2) lists the five categories and gives some examples of sources of uncertainty in
each category for the test method applied.

Page 3 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

It is important to note that Table 2 is not exhaustive and is for guidance only. Relative
contributions may vary according to the material tested and the test conditions.
Individual laboratories are encouraged to prepare their own list to correspond to their
own test facility and assess the associated significance of the contributions.
Table 2 Sources of Uncertainty, their Type and their likely contribution to
Uncertainties on Measurands and Measurements
(1 = major contribution, 2 = minor contribution, blank = no influence, * = indirectly affected)

Sources of uncertainty
1. Apparatus
Load cell
Extensometer
Plotter X
Plotter Y
Caliper
Knife edges thickness 2
2. Method
Span
Alignment
Perpendicularity
Speed
3. Environment
Laboratory ambient temperature and
humidity
4. Operator
Graph interpretation
Distance between knife edges 2
Error in measuring specimen dimensions
Crack length measurement
5. Test Piece
Specimen thickness
Specimen width
1
2

Type 1

Measurand and Measurements

d B W a F VP s z

B
B
B
B
B
B

*2
*2
*2
*2
*2
*2

B
B
B
B

*2
*2
*2
*2

2
2
2
2

2
2
2
2

*2

A
A
A
A

*1
*2
2
*1

B
B

*2
*1

1
1
1
1

1
1

1
1

1
2
1
1
1

2
2

2
2

2
2

See Step 3
The specimen must be provided with a pair of accurately machined knife-edges that support the
gauge arms and serve as the displacement reference points

Step 3. Classifying the Uncertainty According to Type A or B

In this third step, which is in accordance with Reference 2, Guide to the Expression of
Uncertainties in Measurement, the sources of uncertainty are classified as Type A or B,
depending on the way their influence is quantified. If the uncertainty is evaluated by
statistical means (from a number of repeated observations), it is classified Type A, if it is
evaluated by any other means it should be classified Type B (see Table 2).
The values associated with Type B uncertainties can be obtained from a number of
sources including a calibration certificate, manufacturers information, or an experts
estimation. For Type B uncertainties, it is necessary for the users to estimate for the
Page 4 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

most appropriate probability distribution for each source (further details are given in
Section 2 of Reference 1).
It should be noted that, in some cases, an uncertainty could be classified as either Type
A or Type B depending on how it is estimated.
Step 4. Estimating the Standard Uncertainty for each Source of Uncertainty
In this step the standard uncertainty, u(xi), for each measurement is estimated (see
Appendix A). The standard uncertainty is defined as one standard deviation and is
derived from the uncertainty of the input quantity divided by the parameter dv, which is
associated with the assumed probability distribution. The divisors for the distributions
most likely to be encountered are given in Section 2 of Reference 1.
The significant sources of uncertainty and their influence on the evaluated quantity are
summarised on Table 3.
This table is structured in the following manner:
column :
sources of uncertainty
column -:
sources value. There are two types:

(1)
permissible range for the measurement according to the test standard

(2)
maximum range between measures on the same test made by several
trained operators
column :
measurements affected by each source
column :
measurement values obtained in a real test
column :
source of uncertainty type
column :
assumed probability distribution
column :
correction factor for Type B sources (dv)
column :
effect on the measurement uncertainty produced by the input quantity
uncertainty
This column is obtained by two different ways:
if the influence of the source of uncertainty on the measurement is
proportionally direct.
(column -).(column ) /(column )
if the influence is not direct it should be obtained by calculating
the measurement for both the maximum and minimum values of
the range in column - without variation in the rest of the sources
of uncertainty and applying the appropriate correction factor for
the probability (column ).

Page 5 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

Table 3 Example Worksheet for Uncertainty Calculations in CTOD Tests

Column No.

Sources of Uncertainty
Source

Value
or (2)

(1)

Measurements
Measurement
Affected

Nominal or
Averaged
Value (Units)

Uncertainties
Type

Probabl.
Distribt.

Divisor
(dv)

Effect on
Uncertainty in
Measurement

Apparatus
Load cell

(kN)

Rectang.

u(load cell)

Extensometer

VP

(mm)

Rectang.

u(extensom)

Plotter Y

VP

(mm)

Rectang.

3
3

Plotter X

u(plotterY)

The influence on Uncertainty is Negligible

z
VP
W
B
a

(mm)

Rectang.

u(knife edges)

(mm)
(mm)
(mm)

Rectang.

u(caliper)

s
F
VP
F
VP
F
VP

(mm)
(N)
(mm)
(N)
(mm)
(N)
(mm)

Rectang.

u(span)

Rectang.

u(alignm)

Rectang.

u(perpen)

Distance between knife

edges

VP

(mm)

Rectang.

u(distance)

Speed

F
VP

(N)
(mm)

Rectang.

u(speed)

F
VP

(N)
(mm)

Rectang.

u(ro om temp)

VP
B
W

(mm)
(mm)
(mm)

A
A
A

normal
normal
normal

1
1
1

u(graph)
u(B msrment.)
u(W msrment.)

(mm)

normal

u(a msrment.)

(mm)

Rectang.

u(thickness)

(mm)

Rectang.

u(width)

Knife edges thickness

Caliper

Method
Span
Alignment
Perpendicularity

Environment
Room temperature
Operator
Graph interpretation
B Measurement
W Measurement
Crack length
measurement
Test Piece
Specimen thickness
Specimen width

(1)
(2)

permissible range for the source according to the test standard

maximum range between measures made on the same test by several trained operators

Step 5. Computing the Combined Uncertainty uc

Assuming that individual uncertainty sources are uncorrelated, the
uncertainty of the measurand, uc(y), can be computed using the root sum squares:

Page 6 of 24

combined

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

uc ( y) =

[c
N

i =1

u ( xi )]

(1)

where ci is the sensitivity coefficient associated with the measurement xi. This
uncertainty corresponds to plus or minus one standard deviation on the normal
distribution law representing the studied quantity. The combined uncertainty has an
associated confidence level of 68.27%.

Step 6. Computing the Expanded Uncertainty U

The expanded uncertainty, U, is defined in Reference 2 as the 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. It is obtained by
multiplying the combined uncertainty, uc, by a coverage factor, k, which is selected on
the basis of the level of confidence required. For a normal probability distribution, the
most generally used coverage factor is 2, which corresponds to a confidence level
interval of 95.4% (effectively 95% for most practical purposes). The expanded
uncertainty, U, is, therefore, broader than the combined uncertainty, uc. Where the
customer (such as aerospace and electronics industries) demands a higher confidence
level, a coverage factor of 3 is often used so that the corresponding confidence level
increases to 99.73%.
In cases where the probability distribution of uc is not normal (or where the number of
data points used in Type A analysis is small), the value of k should be calculated from
the degrees of freedom given by the Welsh-Satterthwaite method (see Reference 1,
Section 4 for more details).

Step 7. Reporting of Results

Once the expanded uncertainty has been estimated, the results should be reported in the
following format:
V = y U
where:

V
y
U

is the estimated value of the measurand

is the test (or measurand) mean result
is the expanded uncertainty associated with y

An explanatory note, such as that given in the following example should be added
(change where appropriate):
The reported expanded uncertainty is based on a standard uncertainty multiplied by a
coverage factor, k=2, which for a normal distribution corresponds to a coverage
probability, p, of approximately 95%. The uncertainty evaluation was carried out in
accordance with UNCERT COP 04:2000.

Page 7 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

5.

REFERENCES

1.

Manual of Codes of Practice for the determination of uncertainties in

mechanical tests on metallic materials. Project UNCERT, EU Contract
SMT4-CT97-2165, Standards Measurement & Testing Programme, ISBN 0
946754 41 1, Issue 1, September 2000.

2.

BIPM, IEC, IFCC, ISO, IUPAC, OIML, Guide to the expression of

Uncertainty in Measurement. International Organisation for Standardisation,
Geneva, Switzerland, ISBN 92-67-10188-9, First Edition, 1993. [This Guide
is often referred to as the GUM or the ISO TAG4 document after the ISO
Technical Advisory Group that drafted it.]

3.

British Standard, BS 7448. Part 1-97. Fracture Mechanics Toughness Tests.

Part 1. Method for determination of KIC, critical CTOD and critical J values
of Metallic Materials.

4.

ASTM E1290-93 Crack-Tip Opening Displacement (CTOD) Fracture

Toughness Measurement

5.

ASTM E399-90 Plane-Strain Fracture Toughness of Metallic Materials

ACKNOWLEDGEMENTS
This document was written as part of project Code of Practice for the Determination of
Uncertainties in Mechanical Tests on Metallic Materials. The project was partly funded
by the Commission of European Communities through the Standards, Measurement and
Testing Programme Contract No. SMT4-CT97-2165.
The author gratefully acknowledges the efforts of Mrs. Yeyes San Martn of INASMET
in the development of this Code of Practice.
Many thanks are also due to many colleagues from UNCERT for invaluable helpful
comments.

Page 8 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

APPENDIX A
Mathematical Formulae for Calculating Uncertainties in Crack Tip Opening
Displacement (CTOD) Parameter Determination Testing, using SE (B) specimens

The formula for the calculation of the Crack Tip Opening Displacement (CTOD)
Fracture Toughness for a SE (B) specimen is:

K 2 (1 g 2 ) 0.4(W a) V P
=
+
2SY E
0.4W + 0.6a + z

(A1)

where K is a stress intensity factor, calculated from equation

Fs
K=
f
(A2)
BW 1.5
a
where f is a mathematical function of
, given in next equation
W
a 1
a
a
a
a
3 ( ) 2 1.99 ( ) (1 ) 2.15 3.93( ) + 2.7( ) 2
a
W
W
W
W
W

f( )=
(A3)
3
a
a 2
W
2(1 + 2 )(1 )
W
W
where:
g
is the Poissons ratio
Sy
is the 0.2% proof strength at the temperature of the fracture test
E
is the Youngs modulus at the test temperature
VP is the plastic component of notch opening displacement. It is obtained by drawing
a line parallel to the tangent of the initial linear part of the record from point F (Fc,
Fu or Fm); the VP value is the distance between the origin and the intersection of
this parallel with X-axis (see Figure A1).
Load, F

Fc

Fu

Fc

Fu
Fm

Fc
Vc

Vu

Vc

Vu
Vm

Vc

Vp

Vp

Vp

Vp

Vp

Displacement, V

Figure A1. Definition of F (F c, Fu or Fm ) and VP

The derived measurand (y) is a function of eight measurements F, f, VP, z, s, a, B and
Page 9 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

W (x i), and each x i is subject to uncertainty u(xi). The general combined standard
uncertainty uc(y) is expressed by equation (1) in main procedure:
u c ( y) =

[c iu (x i )]2

(A4)

i =1

y
xi

ci =

where

(A5)

Using these formulae, it is possible to write the combined standard uncertainty of

parameter:
[u c ()] 2 = c F 2 u ( F ) 2 + c f 2u ( f ) 2 + cW 2u (W ) 2 + ca 2 u (a ) 2 +
+ c B 2 u ( B) 2 + cVP 2u (V P ) 2 + c s 2u ( s ) 2 + c z 2 u ( z ) 2

(A6)

The sensitivity coefficients (ci )for equation A6 are given by:

K
= AK
F
F

cF =

K
=
F

where

A=

and

(A7)

sf
3

BW

(A8)
2

1 g 2
S yE

(A9)

K
= AK
f
f

cf =

(A10)

K
F s
=
3
f
B W 2

where

cW =

0.4VP ( a + z )

K
= AK
+
W
W ( 0.4W + 0.6 a + z ) 2
K
1.5Fsf
=
5
W
BW 2

where

ca =

0.4VP (W + z )

=
a ( 0.4W + 0.6a + z ) 2

Page 10 of 24

(A11)

(A12)

(A13)

(A14)

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

K
= AK
B
B

cB =

where

K
=
B
cV P =
cs =

where

(A15)

Fsf
3

B W

(A16)
2

0.4(W a )
=
V P 0.4W + 0.6a + z

K
= AK
s
s

K
=
s
cz =

UNCERT COP 04: 2000

(A18)

Ff
3

BW

(A17)

(A19)
2

0.4(W a)V P

=
z
(0.4W + 0.6a + z ) 2

(A20)

A6 equation is composed by eight terms that will be analysed in next paragraphs.

A.1

UNCERTAINTY IN LOAD ( u (F ) )

F is a measurement affected by different sources of uncertainty (see Table 2 and Table

3 in the main procedure), but its influence can be considered negligible except for the
load cell. So:
u ( F ) = u (load cell )

(A21)

u(load cell) can be estimated using Table 3 in this procedure.

A.2

UNCERTAINTY IN f ( u ( f ) )

f is a mathematical function of a/W (equation A3).

Assuming that this source of uncertainty (f) is considered as Type A, the contribution to
total uncertainty can be calculated from the standard deviation of the arithmetic mean:

s( f j )
u( f ) = s( f ) =
(A22)
n
Where:
s( f j ) =

_
1 n
(
f

f )2

j
n 1 j =1

Using the maximum and the minimum values to calculate s(f j), (n=2):
Page 11 of 24

(A23)

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

u (f ) =

_
_
1

2
(
f

f
)
+
(
f

f
)2
max
min

2 1

(f max f ) 2 + (f min f ) 2
2

(A24)

Where:

and

a
f max = f ( max )
and
W min
a max = a 0 + 2 u (a )
a min = a 0 2 u( a )
Wmax = W0 + 2 u ( W)

a
f min = f ( min )
W max

(A25)

(A26)

Wmin = W0 2 u ( W )
(assuming that both a and W have a normal probability distribution, k=2)

A.3

UNCERTAINTY IN SPECIMEN WIDTH ( u (W ) )

W is a measurement affected by three sources of uncertainty (see Table 2 and Table 3).
Assuming that individual uncertainty sources are uncorrelated, the combined
uncertainty in W can be computed using the root sum squares:
u (W ) = u ( s i ) 2 = u ( caliper ) 2 + u(W measurement ) 2 + u( width ) 2

A.4

(A27)

UNCERTAINTY IN CRACK LENGTH ( u (a ) )

a is a measurement affected by two sources of uncertainty (see Table 2 and Table 3).
Assuming that individual uncertainty sources are uncorrelated, the combined
uncertainty in a can be computed using the root sum squares:
u (a ) = u ( si ) 2 = u( caliper ) 2 + u (crack length measurement ) 2

(A28)

The term u(si) can be calculated using Table 3 in this procedure.

A.5

UNCERTAINTY IN B ( u (B) )

B is a measurement affected by three sources of uncertainty (see Table 2 and Table 3 in

the procedure). Assuming that the individual uncertainty sources are uncorrelated, the
combined uncertainty in B can be computed using the root sum squares:
u ( B) = u ( s i ) 2 = u( caliper ) 2 + u ( B measurement ) 2 + u ( thickness ) 2
The terms u(si) can be calculated using Table 3 in this procedure.
Page 12 of 24

(A29)

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

A.6

UNCERTAINTY IN THE
DISPLACEMENT ( u (VP ) )

PLASTIC

COMPONENT

UNCERT COP 04: 2000

OF N OTCH OPENING

VP is a measurement affected by five sources of uncertainty (see Table 2 and Table 3 in

the procedure). Assuming that the individual uncertainty sources are uncorrelated, the
combined uncertainty in VP can be computed using the root sum squares:
u (VP ) = u ( si ) =
2

u( extensometer ) 2 + u( plotter Y ) 2 + u ( graph interpretation) 2 +

+ u ( knife edges thickness) 2 + u (distance between knife edges ) 2
(A30)

(u(si) is the standard uncertainty of each source that contributes to F combined

uncertainty).
The terms u(si) can be calculated using Table 3 in this procedure.
A.7

UNCERTAINTY IN THE SPAN ( u (s) )

It is directly obtained from data in Table 3.

A.8

UNCERTAINTY IN THE KNIFE EDGES THICKNESS ( u (z ) )

It is directly obtained from data in Table 3

The combined uncertainty of each measurement is shown in Table A1.
Table A1

Formulae for calculating Combined Uncertainties

Measurement

Sources of uncertainty (si )

Applied force
F
Plastic
component of
notch opening
displacement
VP
Knife edges
thickness
z
Specimen
width
W
Specimen
thickness
B
Crack length a
Span
s

Load cell
Extensometer
Plotter Y
Graph Interpretation
Knife Edges Thickness
Distance between Knife
Edges
Knife Edges Thickness

u (si )
(Units)
(kN)
(mm)
(mm)
(mm)
(mm)

u ( F ) = u (load cell)
u (extensometer )2 + u ( plotterY )2 +
u (VP ) =

(mm)
(mm)
(mm)
(mm)
(mm)
(mm)
(mm)
(mm)
(mm)

+ u ( graph interpretation) 2 +
+ u ( knife edges thickness) 2 +
+ u ( distancebetweenknife edges) 2

(mm)
(mm)

Caliper
W measurement
Specimen Width
Caliper
B measurement
Specimen Thickness
Caliper
Crack Length Measurement
Span

Uncertainty of Measurements u(xi )

u ( z ) = u ( knife edges thickness)

u (W ) =

u (B ) =

u (caliper)2 + u(W measuremen

t )2 +
+ u (width)2
u (caliper)2 + u( B measuremen
t )2 +
+ u(thickness
)2

u (a) = u(caliper) 2 + u (cracklengthmeasuremen

t )2

u(s) = u(span)

Page 13 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

APPENDIX B
A Worked Example for Calculating Uncertainties in Crack Tip Opening
Displacement (CTOD) Parameter Determination Testing

B1.

Introduction

A customer asked a testing laboratory to carry out a fracture test to determinate the
Crack Tip Opening Displacement (CTOD) at room temperature, according to British
Standard, BS 7448. Part 1-1991, on three point bend specimens. The mechanical
properties of the material were:
proof strength at 0.2% at the test temperature: 602 MPa
Youngs modulus: 210,000 MPa
Poissons ratio: 0.3.
The laboratory has considered the sources of uncertainty in its test facility and has found
that the sources of uncertainty in the test results are identical to those described in Table
2 of the Main Procedure.

B2.

Estimation of Input Quantities to the Uncertainty Analysis

All tests were carried out according to the laboratorys own procedure using an
appropriately calibrated tensile test facility. The test facility was located in a
temperature-controlled environment, at room temperature.

The specimen was a three point bend type, and its dimensions were:
thickness B = 18 mm 0.5%
width
W = 36 mm 0.5%
The dimensions of the specimen were measured using a caliper with an
uncertainty of 0.05 mm, typical for calipers used in the laboratory.

The crack length (a) was measured with the same caliper and the value obtained
was 17.57 mm.

The test was carried out on a universal test machine using a strain rate of
0.4 kNs-1 . (A constant loading rate of 0.3 1.5 kNs-1 for a standard bend
specimen corresponds to a rate of increase in stress intensity factor within the
range 0.55 2.75 MPa m s-1 , according to ASTM E 399 Standard, Clause
A3.4.2.1).

The machine was calibrated to Grade 1 of BS 1610.

Strain was measured using a clip gauge extensometer with a nominal gauge
length of 10 mm. The extensometer complied with Class 0.5, specification
according to EN 10002-4:1994.

Page 14 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

The thickness of the knife-edges attached to the specimen were 1.5 mm (BS
7448 Clause 5.1.3 permits a thickness of 1.5 - 2 mm) and were attached at a
distance of 10 mm. This distance was determined by the extensometer, because it
is impossible to calibrate the extensometer if the distance between edges is not
within the nominal value 1%

The line of action of the applied force passed midway between the centres of the
rollers within 1% of the distance between these centres. The crack tip midway
squared to the roller axes within 2.

The span for the test method was 144 mm 0.5%, according to the test standard.

10

The accuracy of the plotter used to record the load-displacement curve was
within 0.5% in both axes.
The curve obtained in this test is represented in Figure B1. The maximum load
value recorded by the machine FC was 33,800 N. The value indicated in
Figure B1 as VP was obtained graphically by the operator (VP = 0.42 mm),
according to the procedure described in Appendix A of this CoP.
Load, kN

Fc

35

Tg (F)

30

II Tg (F)
25

20

15

10

0
0

0.2

0.4
0.6
0.8
Displacement, mm

1.0

1.2

Figure B1. Test Record

11

An indication of the uncertainty associated with interpreting the graph was

obtained using another test record (Figure B2). Four trained operators analysed
the record and obtained the values indicated in Table B1.

Page 15 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

Load, kN
35

Tg (F)
30

25

20

15

10

II Tg (F)
0
0

0.1

0.2
0.3
0.4
Displacement, mm

0.5

0.6

Figure B2. Test Record used to establish the Uncertainty in Interpreting the Graph B1

Table B1 Uncertainty in Interpreting the Graph B2

Operator 1
Operator 2
Operator 3
Operator 4
Mean

VP (mm)
0.273
0.274
0.265
0.265
0.269

The uncertainty for this input value is its standard deviation:

u=

_
1 4
( Vj V) 2 =

n 1 j=1

1
[(0.273 0.269) 2 + ( 0.274 0.269) 2 + 2 (0. 265 0.269) 2 ] =
3

= 0.005 mm 1.829%
12

The uncertainties associated with the error in measurements (B, W and a) were
obtained in the same manner as above (using a compact tension test specimen
with the same nominal dimensions than the used in this example B = 18 mm and
W = 36 mm)
Page 16 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

Table B2 Uncertainty in Dimensions Measurement

B (mm)
18.02
17.98
18.01
17.98

W (mm)
35.95
36.04
35.97
36.05

a (mm)
19.14
19.30
19.27
19.30

Mean 17.998
Standard Deviation 0.021
Uncertainty* 0.115%
(*)
(Percentage)
s ( q i ) 100
(*) uncertainty (%) =

36.003
0.050
0.139%

19.253
0.076
0.396%

Operator 1
Operator 2
Operator 3
Operator 4

mean

The highest uncertainty associated with all these measurements was in the crack
length, so an uncertainty of 0.4% has been selected for both B, W and a.
13

The influence of the knife-edges thickness was studied using a Computer

Assistant Design Program:

0.617898

9.900000

10.000000

10.100000

0.617898

10.420000

10.320006
10.32006-9.9=0.420006

Fig. B3.1

Fig. B3.2

10.519994
10.519994-10.1=0.419994

Fig. B3.3

Figure B3. Uncertainty in Distance between Knife Edges

First of all, with a distance of 10 mm between knife edges, the angle for a VP of
0.42 mm is calculated (Figure B3.1).
Then putting the knife edges at both the maximum and the minimum distances
allowed, and with the angle calculated in first step, maximum and minimum VP
values are obtained (Figures B3.2 and B3.3).
Page 17 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

Then, the influence of the knife edges distance on VP is:

u (distance ) =
14

Vmax Vmin
0.420006 0.419994
=
= 6 10 6 mm 0.0014%
2
2

The influence of the knife edges thickness was obtained using a CAD program
too:
First of all, with knife edges of 1.5 mm thickness, the angle for a VP of 0.42 mm
is calculated (Figure B4.1).
Then with knife edges of 2 mm thickness (maximum allowed), and with the
angle calculated in first step, VP value is obtained (Figure B4.2).
Then, the influence of the knife edges thickness on VP is:
u (thickness) =

Vmax Vmin
0.430784 0.42
=
= 0.005 mm 1.284%
2
2

0.617898
2.0

1.5

10.00000

10.00000

1.5

2.0
10.43078

10.42000

Fig. B4.1

Fig. B4.2

Figure B4. Uncertainty in Knife Edges Thickness

Page 18 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

B3

Example of Uncertainty Calculations and Reporting of Results

B3.1

Calculations

Firstly, Table 3 in main procedure should be completed.

Column
Sources of Uncertainty
Source

Type

Proba bl.
Distribt.

Measurements
Nominal or
Averaged
Value (Units)

Uncertainties
Divisor
(dv)

Effect on
Uncertainty in
Measurement

Value
(1)
or (2)

Measurement
Affected

1% ( 1)

33,800 N

Rectang.

195 N(A)

0. 5% (1)
0. 5% (1)

VP
VP

0.42 mm
0.42 mm

B
B

Rectang.
Rectang.

3
3

0.001 mm(A)
0.001 mm(A)

1.5-2 mm(1)
1.284 % ( B )

z
VP
W
B
a

Apparatus
Load cell
Extensometer
Plotter Y
Plotter X
Knife edges
thickness
Caliper

The influence on Uncertainty is Negligible

0. 05mm (1 )

1.5 mm
0.42 mm
36 mm
18 mm
17.57 mm

Rectang.

0.144 mm(E)
0.005 mm(B)

Rectang.

0.029 mm(C)

Method
0.5%

Span

(1)

1% s

Alignment
Perpendicularity

2 (1 )

1%
Distance between
0. 0014 %( B )
knife edges

Rectang.

0.416 mm(A)
Nglg. (D)
Nglg. (D)

Rectang.

Nglg. (D)

Rectang.

Nglg. (D)

0.42 mm

Rectang.

6 10 4 mm (B)

33,800 N
0.42 mm

Rectang.

Nglg.(D)

s
F
VP
F
VP
F
VP

144 mm
33,800 N
0.42 mm
33,800 N
0.42 mm
33,800 N
0.42 mm

VP
F
VP

(1)

0.31.5 kNs -1

Speed

(1)

Environment
Room
temperature
Operator
Graph
Interpretation
B Measurement
W Measurement
A Measurement
Test Piece
Specimen
thickness
Specimen width
(1)
(2)
Nglg.
(A toE)

2C (1)

F
VP

33,800 N
0.42 mm

Rectang.

Nglg.

1.829%(2)

VP

0.42 mm

normal

0.008 mm(A)

0.4% ( 2)
0.4% ( 2)
0 .4% (2 )

B
W

18 mm
36 mm

A
A

normal
normal

1
1

0.072 mm(A)
0.144 mm(A)

17.57 mm

normal

0.070 mm(A)

0 .5% (1)

18 mm

Rectang.

0.052 mm(A)

0 .5 % (1 )

0.104 mm(A)
3
permissible range for the source according to the test standard (see next paragraphs)
maximum range between measures made on the same test by several trained operators
= negligible if compliant to standard (see next paragraphs)
see next paragraphs (Column )
36 mm

Where:
Page 19 of 24

Rectang.

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

Column -:

Load Cell:

Extensometer:

Plotter:

Knife edges:

Caliper:

Span:

Alignment:

Perpendicularlity:

Distance
between edges:

Room
temperature:

Graph
Interpretation:

Specimen
Thickness:

according to the BS 7448, the force sensing

device shall comply with grade 1 of BS 1610
(accuracy within 1% )
our extensometers comply with grade 0.5
(accuracy within 0.5% )
our plotter complies with grade 0.5 (accuracy
within 0.5% )
according to the test standard the knife edges
thickness must be between 1.5 and 2 mm.
Their
influence
has
been
calculated
geometrically using a CAD program (see B2.
Estimation of input quantities to the
uncertainty analysis)
generally uncertainty of 0.05 mm is typical for
calipers used to measure both test piece
dimensions and crack length
according to the test standard, span must be
adjusted to 0.5%
according to the test standard, the line of
action of the applied force must pass midway
between the centres of the rollers within
1% of the distance between these centres.
according to the test standard, the crack tip
midpoint must be perpendicular to the roller
within 2
this
distance
is
determined
by
the
extensometer, because it is impossible to
calibrate the extensometer if the distance
between edges is not within the nominal value
1%
Their
influence
has
been
calculated
geometrically using a CAD program (see B2.
Estimation of input quantities to the
uncertainty analysis)
this influence is considered negligible for
metallic materials, because only small
variations 2 C are allowed by the test
standard
this is an important contribution to total
uncertainty because of the manual method
used to obtain VP value from the test record

according to the test standard, the tolerance for

Page 20 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

Column :
Column :

the specimen thickness is 0.5%

Specimen
Width:

according to the test standard, the tolerance for

the specimen width is 0.5%

are the values obtained in the real test

(A)
(B)

(C)
(D)

(E)

(column) = (column-) (column) / (column)

it has been calculated geometrically using a CAD
program (see B2. Estimation of input quantities to the
uncertainty analysis)
(column ) = (column-) / (column)
These sources have a minor influence on both F and
VP, and these influences are opposed mutually (if F
tends to be greater, VP tends to be smaller), so the
influence on d is compensated and negligible
(column) = (column-) /[2 (column)]

Filling in Table 4 in main procedure:

Measurement
F
VP

z
W

a
s

Sources of uncertainty (xi )

Load cell
Extensometer
Plotter Y
Graph Interpretation
Knife Edges Thickness
Distance between Knife
Edges
Knife Edges Thickness
Caliper
W Measurement
Specimen Width
Caliper
B Measurement
Specimen Thickness
Caliper
Crack Length Measurement
Span

u (xi )
Uncertainty of Measurands u(Xi )
u ( F ) = 195 N
195 N
0.001 mm
0.001 mm
( 0 .001) 2 + ( 0.001) 2 + ( 0 .008) 2 +
= 0 .021 mm
0.008 mm u (VP ) =
+ ( 0 .005) 2 + ( 6 10 4 ) 2
0.005 mm
6 10-4 mm
0.144 mm
0.029 mm
0.144 mm
0.104 mm
0.029 mm
0.072 mm
0.052 mm
0.029 mm
0.070 mm
0.416 mm

u ( z) = 0.144 mm
u (W ) = (0.029)2 + (0.144) 2 + (0.104 ) 2 = 0.180 mm

u ( B) = (0.029) 2 + (0.072) 2 + (0.052 ) 2 = 0.093 mm

u ( a) = (0. 029) 2 + ( 0.07) 2 = 0.076 mm
u ( s) = 0.416 mm

To calculate the uncertainty of the f factor, equations A24 to A26 are applied:
a max = a 0 + 2 u( a ) = 17.57 + 2 0.076 = 17.722
a min = a0 2 u( a ) = 17.57 2 0.076 = 17.418
W min = W 0 2 u (W ) = 36 + 2 0.180 = 35.64
W min = W 0 2 u (W ) = 36 2 0.180 = 36.36

Page 21 of 24

(A26)

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

a
17.57
f = f( )= f(
) = 2.564
W
36
a
17.722
f max = f ( max ) = f (
) = 2.639
Wmin
35.64
a
17.418
f min = f ( min ) = f (
) = 2.494
W max
36.36
_

(A25)

( f max f ) 2 + ( f min f ) 2
( 2.639 2.564) 2 + ( 2.494 2.564) 2
u( f ) =
=
= 0.073
2
2
(A24)

Now partial derivatives (equations A7 to A20) are obtained:

K
= AK
F
F

cF =
A=

(A7)

1 g 2
1 0.3 2
=
= 7.198 10 9 MPa 2
Sy E
602 210,000
Fs

K=

BW
K
=
F
cF =

f =

18 36

sf
3

BW

33,800 144

= 0.095 mm

32

(A8)

K
= AK
= 7.198 10 9 3,210.25 0.095 =
F
F
= 2.19 10 6 N 1mm

cf =
K
=
f

K
= AK
f
f
Fs
3

BW

=
2

(A7)

(A10)

33,800 144
18 36

(A2)

144 2.564
18 36

2.564 = 3,210.25 N mm

(A9)

= 1,251.8 N mm

(A11)

cf =

K
= AK
= 7.198 10 9 3,210.25 1,251.8 = 2.89 10 2 mm
f
f

(A10)

cW =

0.4V P (a + z )

K
= AK
+
W
W (0.4W + 0.6a + z ) 2

(A12)

Page 22 of 24

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

5
K
1.5 Fsf
1.5 33,800 144 2.564
2
=
=

133
.
76
N
mm
5
5
W
BW 2
18 36 2

cW =

0.4VP ( a + z )

K
= AK
+
=
W
W ( 0.4W + 0.6a + z ) 2

= 7.198 10 9 3,210.25 ( 133.76) +

0.4 0.42(17.57 + 1.5)
+
= 1.49 10 3
2
(0.4 36 + 0.6 17.57 + 1.5)

(A12)

ca =

0.4VP (W + z )

0.4 0.42( 36 + 1.5)

=
=
= 0.009
2
a (0.4W + 0.6a + z )
(0.4 36 + 0.6 17.57 + 1.5) 2

(A14)

cB =

K
= AK
B
B

(A15)

K
=
B

Fsf
2

B W

=
2

33,800 144 2.564

18 36
2

= 178.347 N mm

(A16)

cB =

K
= AK
= 7.198 10 9 3,210.25 ( 178.347 ) = 4.12 10 3
B
B

(A15)

cV P =

0.4(W a )
0.4( 36 17.57)
=
=
= 0.279
V P 0.4W + 0.6a + z 0.4 36 + 0.6 17.57 + 1.5

(A17)

K
= AK
s
s
5
K
Ff
33,800 2.564
=
=
= 22,293N mm 2
3
3
s
BW 2
18 36 2

K
cs =
= AK
= 7.198 10 9 3,210.25 22.293 = 5.15 10 4
s
s
cs =

cz =

(A13)

0.4(W a)V P

0.4 ( 36 17.57) 0.42

=
=
= 0.004
2
z
(0.4W + 0.6a + z )
( 0.4 36 + 0.6 17.57 + 1.5) 2

The combined standard uncertainty for d will be (equation A6):

Page 23 of 24

(A18)
(A19)

(A18)

(A20)

SM&T
Standards Measurement & Testing Project No. SMT4-CT97-2165

UNCERT COP 04: 2000

[ u c ( )] = c F u ( F) + c f u (f ) + c W u ( W) + c a u (a ) +
2

+ c B u ( B) 2 + c VP u (VP ) 2 + c s u (s) 2 + c z u (z ) 2 =
2

= ( 2.195 10 6 ) 2 195 2 + ( 2.89 10 2 ) 2 0.073 2 +

+ (1.49 10 3 ) 2 0.180 2 + ( 0.009) 2 0.076 2 +
3 2

(A6)

+ ( 4.12 10 ) 0.093 + 0.279 0.021 +

2

+ ( 5.15 10 4 ) 2 0.416 2 + (0.004) 2 0.144 2

= 4.24 10 5 mm 2
u c () = 0.006 mm
The expanded uncertainty with a confidence interval of 95.4% is obtained multiplying
the combined standard uncertainty by a coverage factor of 2:
U ( ) = 0.006 2 = 0.012 mm
The value of d is calculated with equation A1:

K 2 (1 g 2 ) 0. 4 (W a) V P
=
+
=
2SY E
0.4W + 0.6a + z
3,210. 252 (1 0. 32 )
0.4 (36 17.57) 0. 42
=
+
= 0.154 mm
2 603 210, 000
0.4 36 + 0. 6 17.57 + 1.5
B3.2

Reported Results

The Critical Crack Tip Opening Displacement (CTOD) is 0.154 0.012 mm

The above reported expanded uncertainty is based on a standard uncertainty multiplied
by a coverage factor k=2, which for a normal distribution corresponds to a coverage
probability, p, of approximately 95%. The uncertainty evaluation was carried out in
accordance with UNCERT COP 04:2000.

Page 24 of 24