BRITISH STANDARD BS EN

705:1995
BS 2782-12:
Methods
1220A to C:
1995

Plastics piping
systems —
Glass-reinforced
thermosetting plastics
(GRP) pipes and
fittings —
Methods for regression
analyses and their use

The European Standard EN 705:1994 has the status of a

British Standard
BS EN 705:1995

Committees responsible for this

British Standard

The preparation of this British Standard was entrusted to Technical

Committee PRI/61, Plastics piping systems and components, upon which the
following bodies were represented:

British Gas plc

British Plastics Federation
British Plumbing Fittings Manufacturers’ Association
British Valve and Actuator Manufacturers’ Association
Department of the Environment (British Board of Agrément)
Department of the Environment (Building Research Establishment)
Department of the Environment (Property and Buildings Directorate)
Department of Transport
Electricity Association
Federation of Civil Engineering Contractors
Health and Safety Executive
Institute of Building Control
Institute of Materials
Institution of Civil Engineers
Institution of Gas Engineers
Institution of Water and Environmental Management
National Association of Plumbing, Heating and Mechanical Services
Contractors
Pipeline Industries Guild
Plastics Land Drainage Manufacturers’ Association
Society of British Gas Industries
Society of British Water Industries
Water Companies Association
Water Services Association of England and Wales

This British Standard, having

been prepared under the
direction of the Sector Board
for Materials and Chemicals,
was published under the
authority of the Standards
Board and comes
into effect on Amendments issued since publication
15 November 1995
Amd. No. Date Comments
© BSI 01-2000

The following BSI references

relate to the work on this
standard:
Committee reference PRI/61
Draft for comment 92/41338 DC

ISBN 0 580 24136 X

 BS EN 705:1995

Contents

Page
Committees responsible Inside front cover
National foreword ii
Foreword 2
Text of EN 705 5

© BSI 01-2000 i
BS EN 705:1995

National foreword

This British Standard has been prepared by Technical Committee PRI/61, and is
the English language version of EN 705:1994 Plastics piping systems —
Glass-reinforced thermosetting plastics (GRP) pipes and fittings — Methods for
regression analyses and their use, published by the European Committee for
Standardization (CEN).
It is incorporated into BS 2782 Methods of testing plastics — Part 12. Reinforced
plastics pipes, fittings and valves, as Methods 1220A to C:1995 as follows.
Method 1220A corresponds to method A (Covariance method) of EN 705:1994.
Method 1220B corresponds to method B (Regression with time as the
independent variable) of EN 705:1994.
Method 1220C corresponds to method C (Second order polynomial
relationships) of EN 705:1994.
These methods are incorporated into BS 2782 for association with related test
methods for plastics materials and plastics piping components.
This test method has been prepared for reference by other standards under
preparation by CEN for specification of reinforced plastics piping systems and
components. It has been implemented as a national standard to enable experience
of the method to be gained and for use for other fresh applications.
It is also for use for the revision or amendment of other national standards as
practicable, but it should not be presumed to apply to any existing standard or
specification which contains or makes reference to a different method until that
standard/specification has been amended or revised to make reference to this
method and adjust any requirements as appropriate.
A British Standard does not purport to include all the necessary provisions of a
contract. Users of British Standards are responsible for their correct application.
Compliance with a British Standard does not of itself confer immunity
from legal obligations.

Summary of pages
This document comprises a front cover, an inside front cover, pages i and ii,
the EN title page, pages 2 to 31 and a back cover.
This standard has been updated (see copyright date) and may have had
amendments incorporated. This will be indicated in the amendment table on the
inside front cover.

ii © BSI 01-2000
EUROPEAN STANDARD EN 705
NORME EUROPÉENNE
April 1994
EUROPÄISCHE NORM

UDC 621.643.2-036.067.5:621.643.06:620.1:519.2

Descriptors: Pipelines, plastic tubes, pipe fittings, thermosetting resins, reinforced plastics, glass, data, statistical analysis,
computation, design inspection

English version

Plastics piping systems — Glass-reinforced thermosetting

plastics (GRP) pipes and fittings — Methods for regression
analyses and their use

Systèmes de canalisations plastiques — Tubes Kunststoff-Rohrleitungssysteme — Rohre und

et raccords plastiques thermodurcissable Formstücke aus glasfaserverstärkten
renforcés de verre (PRV) — duroplastischen Kunststoffen (GFK) —
Méthodes pour une analyse de régression et Verfahren zur Regressionsanalyse und deren
leurs utilisations Anwendung

ｗｗｗ．ｂｚｆｘｗ．ｃｏｍ
This European Standard was approved by CEN on 1994-04-11. CEN members
are bound to comply with the CEN/CENELEC Internal Regulations which
stipulate the conditions for giving this European Standard the status of a
national standard without any alteration.
Up-to-date lists and bibliographical references concerning such national
standards may be obtained on application to the Central Secretariat or to any
CEN member.
This European Standard exists in three official versions (English, French,
German). A version in any other language made by translation under the
responsibility of a CEN member into its own language and notified to the
Central Secretariat has the same status as the official versions.
CEN members are the national standards bodies of Austria, Belgium,
Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy,
Luxembourg, Netherlands, Norway, Portugal, Spain, Sweden, Switzerland and
United Kingdom.

CEN
European Committee for Standardization
Comité Européen de Normalisation
Europäisches Komitee für Normung
Central Secretariat: rue de Stassart 36, B-1050 Brussels

© 1994 Copyright reserved to CEN members

Ref. No. EN 705:1994 E
EN 705:1994

Foreword Contents
This standard was prepared by CEN/TC 155, Page
Plastics piping systems and ducting systems. Foreword 2
This standard is based on document N 197 Introduction 5
Glass-reinforced thermosetting plastics (GRP) pipes
1 Scope 5
and fittings — Standard extrapolation procedures
and their use, prepared by working group 1 of 2 Principle 5
Subcommittee 6 of Technical Committee 138 of the 3 Procedures for determining the
International Organization for Standardization functional relationships 5
(ISO). It is a modification of 3.1 Linear relationships —
ISO/TC 138/SC6/WG 1 N 197 for reasons of possible Methods A and B 5
applicability to other test conditions and alignment
3.2 Second order polynomial
with texts of other standards on test methods.
relationships — Method C 14
The modifications are:
4 Application of methods to product
— examples have been introduced to enable design and testing 16
validation of alternative calculation facilities;
4.1 General 16
— material-dependent requirements are not
given; 4.2 Design 17
4.3 Examples for validation of
— editorial changes have been introduced.
calculation procedures for design 20
The material-dependent test parameters and/or
4.4 Procedures for verifying conformance
performance requirements are incorporated in the
to product design and performance
referring standard.
values 23

ｗｗｗ．ｂｚｆｘｗ．ｃｏｍ
Annex A, which is informative, describes procedures
for solving the given set of equations (see 3.2.3) on a 4.5 Examples for validation of calculation
procedures for design or product
mathematical basis using the example shown
performance verification 26
in 3.2.6.
Annex A (informative) Mathematical
No existing European Standard is superseded by
procedures 30
this standard.
This standard is one of a series of standards on test A.1 Matrix system 30
methods which support System Standards for A.2 Substitution system 30
plastics piping systems and ducting systems. Figure 1 — Regression line from the
This standard shall be given the status of a national results in Table 4 11
standard, either by publication of an identical text Figure 2 — Regression line from the
or by endorsement, at the latest by October 1994, results in Table 7 16
and conflicting national standards shall be Figure 3 — Derived values 18
withdrawn at the latest by October 1994.
Figure 4 — Extrapolated values 19
In accordance with the CEN/CENELEC Internal
Regulations, the following countries are bound to Figure 5 — Example on pressure 21
implement this European Standard: Austria, Figure 6 — Example for ring stiffness 24
Belgium, Denmark, Finland, France, Germany, Figure 7 — Property test values 25
Greece, Iceland, Italy, Luxembourg, Netherlands, Table 1 — Minimum values for the squared,
Norway, Portugal, Spain, Sweden, Switzerland, r2, and linear coefficient of correlation, r, for
United Kingdom. acceptable data from n pairs of data 7
Table 2 — Percentage points of Student’s t
distribution (upper 2,5 % points; two
sided 5 % level of confidence; tv for 97,5 %) 8
Table 3 — Basic data for example calculation
and statistical analysis validation 10
Table 4 — Estimated mean values, Vm, for V 11
Table 5 — Basic data for example
calculation and statistical validation 13

2 © BSI 01-2000
 EN 705:1994

Page
Table 6 — Estimated mean values, Vm, for V 13
Table 7 — Estimated mean values, Vm, for V 16
Table 8 — Initial burst test results 20
Table 9 — Initial ring stiffness test results 22
Table 10 — Initial burst test failure
pressures for the sample 26
Table 11 — Times to failure of burst tests of
the sample 27
Table 12 — Initial ring stiffness results for
the sample 28
Table 13 — Stiffness results at 1 000 h for
the sample 28

ｗｗｗ．ｂｚｆｘｗ．ｃｏｍ

© BSI 01-2000 3
 ｗｗｗ．ｂｚｆｘｗ．ｃｏｍ

4 blank
 EN 705:1994

Introduction
This standard has been prepared to describe the procedures intended for analysing the regression of test
data, usually with respect to time, and the use of the results in design and assessment of conformity with
performance requirements. Its applicability has been limited to use with data obtained from tests carried
out on samples. The referring standards require estimates to be made of the long-term properties of the
pipe for such parameters as circumferential tensile strength, deflection and creep.
The committee investigated a range of statistical techniques that could be used to analyse the test data
produced by tests that were destructive. Many of these simple techniques required the logarithms of the
data to
a) be normally distributed;
b) produce a regression line having a negative slope; and
c) have a sufficiently high regression correlation (see Table 1).
Whilst the last two conditions can be satisfied, analysis has shown that there is a skew to the distribution
and hence this primary condition is not satisfied. Further investigation into techniques that can handle
skewed distributions resulted in the adoption of the covariance method for analysis of such data for this
standard.
The results from non-destructive tests, such as creep or changes in deflection with time, often satisfy these
three conditions and hence simpler procedures, using time as the independent variable, can also be used
in accordance with this standard.

1 Scope
This standard specifies procedures suitable for the analysis of data which, when converted into logarithms

ｗｗｗ．ｂｚｆｘｗ．ｃｏｍ
of the values, have either a normal or a skewed distribution. It is intended for use with the test methods
and referring standards for glass-reinforced plastics pipes or fittings for the analysis of properties as a
function of, usually, time. However it can be used for the analysis of any other data.
For use depending upon the nature of the data, three methods are specified. The extrapolation using these
techniques typically extends the trend from data gathered over a period of approximately 10 000 h, to a
prediction of the property at 50 years.

2 Principle
Data are analysed for regression using methods based on least squares analysis which can accommodate
the incidence of a skew and/or a normal distribution and the applicability of a first order or a second order
polynomial relationship.
The three methods of analysis used comprise the following:
— method A: covariance using a first order relationship;
— method B: least squares with time as the independent variable using a first order relationship;
— method C: least squares with time as the independent variable using a second order relationship.
The methods include statistical tests for the correlation of the data and the suitability for extrapolation.

3 Procedures for determining the functional relationships

3.1 Linear relationships — Methods A and B
3.1.1 Procedures common to methods A and B
Use method A (see 3.1.2) or method B (see 3.1.3) to fit a straight line of the form
y = a + bx (1)
where
y is the logarithm (lg) of the property being investigated;
a is the intercept on the y axis;
b is the slope;
x is the logarithm (lg) of the time, in hours.

© BSI 01-2000 5
EN 705:1994

3.1.2 Method A — Covariance method

3.1.2.1 General
For method A calculate the following variables as necessary in accordance with 3.1.2.2 to 3.1.2.5:

(2)

(3)

(4)

where
Qy is the sum of the squared residuals parallel to the y axis divided by n;
Qx is the sum of the squared residuals parallel to the x axis divided by n;
Qxy is the sum of the squared residuals perpendicular to the line, divided by n;
Y is the arithmetic mean of the y data, i.e.

X is the arithmetic mean of the x data, i.e.

xi, yi, are individual values;

n is the total number of results (pairs of readings for xi, yi).

is negative.
3.1.2.2 Suitability of data ｗｗｗ．ｂｚｆｘｗ．ｃｏｍ
NOTE If the value of Qxy is greater than zero the slope of the line is positive and if the value of Qxy is less than zero then the slope

Calculate the squared, r2, and the linear coefficient of correlation, r, using the following equations:

(5)

(6)

If the value of r2 or r is less than the applicable minimum value given in Table 1 as a function of n, consider
the data unsuitable for analysis.
3.1.2.3 Functional relationships
To find a and b for the functional relationship line
y = a + bx (1)
first set

(7)

then calculate a and b using the following equations:

b = – (¿)0,5 (8)
a = Y – bX (9)
3.1.2.4 Calculation of variances
If tu is the applicable time to failure, then set
xu – lg tu (10)
Using equations (11), (12) and (13) respectively, calculate for i = 1 to n the following sequence of statistics:
— the best fit xi½ for true xi;
— the best fit yi½ for true yi; and

6 © BSI 01-2000
 EN 705:1994

— the error variance, Ö¸2 for x.

(11)

yi½ = a + bxi½ (12)

(13)

Calculate the following quantities:

(14)

(15)

Calculate the variance C of the slope b using the following equation:

C = D(1 + E) (16)
Table 1 — Minimum values for the squared, r2, and
linear coefficient of correlation, r, for
acceptable data from n pairs of data
Minimum values Minimum values
(n – 2) (n – 2)
r2 r r2 r
11 0,6416 0,8010 23 0,3816 0,6177
12 0,6084 0,7800 24 0,3689 0,6074
13 0,5781 0,7603 25 0,3569 0,5974
14 0,5506 0,7420 30 0,3070 0,5541
15 0,5250 0,7246 35 0,2693 0,5189
16 0,5018 0,7084 40 0,2397 0,4896
17 0,4805 0,6932 45 0,2160 0,4648
18 0,4606 0,6787 50 0,1965 0,4433
19 0,4425 0,6652 60 0,1663 0,4078
20 0,4256 0,6524 70 0,1443 0,3799
21 0,4099 0,6402 80 0,1273 0,3568
22 0,3953 0,6287 90 0,1139 0,3375
100 0,1031 0,3211
NOTE In Table 1 and elsewhere in this standard, the equations and corresponding
values for r2 and r are given, for convenience of use in conjunction with reference data
published elsewhere in terms of only r2 or r.

3.1.2.5 Check for the suitability of data for extrapolation

If it is intended to extrapolate the line, calculate T using the following equation:

(17)

If the absolute value|T|(i.e. ignoring signs) of T is equal to or greater than the applicable value for
Student’s t, tv, shown in Table 2 for (n – 2) degrees of freedom then consider the data suitable for
extrapolation.

© BSI 01-2000 7
EN 705:1994

Table 2 — Percentage points of Student’s t distribution (upper 2,5 % points; two sided 5 % level
of confidence; tv for 97,5 %)
Degree of Student’s t value Degree of Student’s t value Degree of Student’s t value
freedom tv freedom tv freedom tv
(n – 2) (n – 2) (n – 2)

1 12,7062 36 2,0281 71 1,9939

2 4,3027 37 2,0262 72 1,9935
3 3,1824 38 2,0244 73 1,9930
4 2,7764 39 2,0227 74 1,9925
5 2,5706 40 2,0211 75 1,9921

6 2,4469 41 2,0195 76 1,9917

7 2,3646 42 2,0181 77 1,9913
8 2,3060 43 2,0167 78 1,9908
9 2,2622 44 2,0154 79 1,9905
10 2,2281 45 2,0141 80 1,9901

11 2,2010 46 2,0129 81 1,9897

12 2,1788 47 2,0112 82 1,9893
13 2,1604 48 2,0106 83 1,9890
14 2,1448 49 2,0096 84 1,9886
15 2,1315 50 2,0086 85 1,9883

16 2,1199 51 2,0076 86 1,9879

17 2,1098 52 2,0066 87 1,9876
18 2,1009 53 2,0057 88 1,9873
19 2,0930 54 2,0049 89 1,9870
20 2,0860 55 2,0040 90 1,9867

21 2,0796 56 2,0032 91 1,9864

22 2,0739 57 2,0025 92 1,9861
23 2,0687 58 2,0017 93 1,9858
24 2,0639 59 2,0010 94 1,9855
25 2,0595 60 2,0003 95 1,9853

26 2,0555 61 1,9996 96 1,9850

27 2,0518 62 1,9990 97 1,9847
28 2,0484 63 1,9983 98 1,9845
29 2,0452 64 1,9977 99 1,9842
30 2,0423 65 1,9971 100 1,9840

31 2,0395 66 1,9966
32 2,0369 67 1,9960
33 2,0345 68 1,9955
34 2,0322 69 1,9949
35 2,0301 70 1,9944

8 © BSI 01-2000
 EN 705:1994

3.1.2.6 Validation of statistical procedures by an example calculation

The data given in Table 3 together with the results given in this example are for use to verify that the other
statistical procedures as adopted by users will produce results similar to those obtained from the equations
given in this standard. For the purposes of example, the property in question is represented by V, the
values for which are of a typical magnitude and in no particular units. Because of rounding errors, it is
unlikely that the results will agree exactly, so for the calculation procedure to be acceptable, the results
obtained for r, r2, b, a, and the mean value of V, Vm, shall agree to within ± 0,1 % of the values given in this
example, as applicable. The values of other statistics are provided to assist checking of the procedure.
Sums of squares
Qx = 0,79812;
Qy = 0,00088;
Qxy = – 0,02484.

Coefficient of correlation
r2 = 0,87999;
r = 0,93808.

Functional relationships
¿ = 0,00110;
b = – 0,03317;
a = 1,62731.

Calculated variances (see 3.1.2.4)

E = 3,5202 × 10–2;
D = 4,8422 × 10–6;
C = 5,0127 × 10–6 (the variance of b);
Ö¸2 = 5,2711 × 10–2 (the error variance for x).

Check of the suitability for extrapolation (see 3.1.2.5)

n = 32;
tv = 2,0423;
T = – 0,03317/(5,0127 × 10–6)0,5 = – 14,8167;
|T| = 14,8167 > 2,0423.
The estimated mean values for V at various times are given in Table 4 and shown in Figure 1.

© BSI 01-2000 9
EN 705:1994

Table 3 — Basic data for example calculation

and statistical analysis validation
n V y Time x
(lg V) h (lg time)
(time in h)

1 30,8 1,4886 5 184 3,7147

2 30,8 1,4886 2 230 3,3483
3 31,5 1,4983 2 220 3,3464
4 31,5 1,4983 12 340 4,0913

5 31,5 1,4983 10 900 4,0374

6 31,5 1,4983 12 340 4,0913
7 31,5 1,4983 10 920 4,0382
8 32,2 1,5079 8 900 3,9494

9 32,2 1,5079 4 173 3,6204

10 32,2 1,5079 8 900 3,9494
11 32,2 1,5079 878 3,9435
12 32,9 1,5172 4 110 3,6138

13 32,9 1,5172 1 301 3,1143

14 32,9 1,5172 3 816 3,5816
15 32,9 1,5172 669 2,8254
16 33,6 1,5263 1 430 3,1553

17 33,6 1,5263 2 103 3,3228

18 33,6 1,5263 589 2,7701
19 33,6 1,5263 1 710 3,2330
20 33,6 1,5263 1 299 3,1136

21 35,0 1,5441 272 2,4346

22 35,0 1,5441 446 2,6493
23 35,0 1,5441 466 2,6684
24 35,0 1,5441 684 2,8351

25 36,4 1,5611 104 2,0170

26 36,4 1,5611 142 2,1523
27 36,4 1,5611 204 2,3096
28 36,4 1,5611 209 2,3201

29 38,5 1,5855 9 0,9542

30 38,5 1,5855 13 1,1139
31 38,5 1,5855 17 1,2304
32 38,5 1,5855 17 1,2304
Means: Y = 1,5301; X = 2,9305

10 © BSI 01-2000
 EN 705:1994

Table 4 — Estimated mean values, Vm, for V

time Vm
h

0,1 45,76
1,0 42,39
10,0 39,28
100,0 36,39
1 000 33,71
10 000 31,23
100 000 28,94
438 000 27,55

Figure 1 — Regression line from the results in Table 4

3.1.3 Regression with time as the independent variable

3.1.3.1 General
For method B calculate the following variables:
Sy = C (yi – Y)2 (18)
(The sum of the squared residuals parallel to the y axis)
Sx = C (xi – X)2 (19)

© BSI 01-2000 11
EN 705:1994

(The sum of the squared residuals parallel to the x axis)

Sxy = C {(xi – X) (yi – Y)} (20)
(The sum of the squared residuals perpendicular to the line)
where
Y is the arithmetic mean of the y data, i.e.

X is the arithmetic mean of the x data, i.e.

xi, yi are individual values;

n is the total number of results (pairs of readings for xi, yi).
NOTE If the value of Sxy is greater than zero the slope of the line is positive and if the value of Sxy is less than zero then the slope is
negative.

3.1.3.2 Suitability of data

Calculate the squared, r2, and the linear coefficient of correlation, r, using the following equations:

(21)

r =|(r2)0,5| (22)
If the value of r2, or r, is less than the applicable minimum value given in Table 1 as a function of n, consider
the data unsuitable for analysis.
3.1.3.3 Functional relationships
Calculate a and b for the functional relationship line [see equation (1)], using the following equations:

(23)

a = Y – bX (24)
3.1.3.4 Check for the suitability of data for extrapolation
If it is intended to extrapolate the line, calculate M using the following equation:

(25)

where
tv is the applicable value for Student’s t determined from Table 2.
If M is equal to or less than zero consider the data unsuitable for extrapolation.
3.1.3.5 Validation of statistical procedures by an example calculation
Use the data given in Table 5 for the calculation procedures described in 3.1.3.2 to 3.1.3.4 to ensure that
the statistical procedures to be used in conjunction with this method will give results for r, r2, a, b and Vm
to within ± 0,1 % of the values given in this example.

12 © BSI 01-2000
 EN 705:1994

Table 5 — Basic data for example calculation

and statistical validation
n Time x V y
h (lg t) (lg V)
(t in h)
1 0,10 – 1,0000 7 114 3,8521
2 0,27 – 0,5686 6 935 3,8410
3 0,50 – 0,3010 6 824 3,8341
4 1,00 0 6 698 3,8259
5 3,28 0,5159 6 533 3,8151
6 7,28 0,8621 6 453 3,8098
7 20,0 1,3010 6 307 3,7999
8 45,9 1,6618 6 199 3,7923
9 72,0 1,8573 6 133 3,7877
10 166 2,2201 5 692 3,7552
11 219 2,3404 5 508 3,7410
12 384 2,5843 5 393 3,7318
13 504 2,7024 5 364 3,7295
14 3 000 3,4771 5 200 3,7160
15 10 520 4,0220 4 975 3,6968
Means: X = 1,4450 Y = 3,7819

Sums of squares
Sx = 31,6811;
Sy = 0,0347;
Sxy = – 1,0242.
Coefficient of correlation
r2 = 0,9556;
r = 0,9775.
Functional relationships (see 3.1.3.3)
a = 3,8286;
b = – 0,0323.
Check of the suitability for extrapolation (see 3.1.3.4)
tv = 2,1604;
M = 942,21.
Table 6 — Estimated mean values, Vm, for V
Time Vm
h
0,1 7 259
1,0 6 739
10,0 6 256
100,0 5 808
1 000 5 391
10 000 5 005
100 000 4 646
438 000 4 428

© BSI 01-2000 13
EN 705:1994

3.2 Second order polynomial relationships — Method C

3.2.1 General
This method fits a curved line of the form
y = c + dx + ex2 (26)
where
y is the logarithm (lg) of the property being investigated;
c is the intercept on the y axis;
d, e are the coefficients to the two orders of x;
x is the logarithm (lg) of the time, in hours.
3.2.2 Variables
For method C calculate the following variables:
Cxi (sum of all individual x data);
Cxi2 (sum of all squared x data);
Cxi3 (sum of all x data to the third power);
Cxi4 (sum of all x data to the fourth power);
Cyi (sum of all individual y data);
(Cyi)2 (squared sum of all individual y data);
Cyi2 (sum of all squared y data);
C(xiyi) (sum of all products xiyi);
C(xi2 yi) (sum of all products xi2 yi);
Sx = C (xi – X)2 (sum of the squared residuals parallel to the x axis for the linear part);
Sxx = C (xi2 – X2)2 (sum of the squared residuals parallel to the x axis for the quadratic part);
Sy = C (yi – Y)2 (sum of the squared residuals parallel to the y axis);
Sxy = C[(xi – X)(yi – Y)] (sum of the squared residuals perpendicular to the line for the linear part);
Sxxy = C[(xi2 – X2)(yi – Y)] (sum of the squared residuals perpendicular to the line for the quadratic
part).
where
Y is the arithmetic mean of the y data, i.e.

X is the arithmetic mean of the x data, i.e.

3.2.3 Solution system

Determine c, d and e (see 3.2.1) using the following matrix:
Cyi = cn + dCxi + eCxi2 (27a)
C(xiyi) = cCxi + dCxi2 + eCxi3 (27b)
C(xi 2yi) = cCxi2 + dCxi3 + eCxi4 (27c)
NOTE Examples showing the procedures that can be used are detailed in Annex A.
3.2.4 Suitability of data
Calculate the squared, r2, and the linear coefficient of correlation, r, using the following equations:

(28)

r =|(r2)0,5| (29)

14 © BSI 01-2000
 EN 705:1994

If the value of r2, or r, is less than the applicable minimum value given in Table 1 as a function of n, consider
the data unsuitable for analysis.
3.2.5 Check for the suitability of data for extrapolation
If it is intended to extrapolate the line, calculate M using the following equation:

(30)

If M is equal to or less than zero consider the data unsuitable for extrapolation.
3.2.6 Validation of statistical procedures by an example calculation
Use the data given in Table 5 for the calculation procedures described in 3.2.2 to 3.2.5 to ensure that the
statistical procedures to be used in conjunction with this method will give results for r, r2, a, b and Vm to
within ± 0,1 % of the values given in this example (n = 15).
Sums of squares and other variables
Cxi = 21,671;
Cxi2 = 62,989;
Cxi3 = 180,623;
Cxi4 = 584,233;
Cyi = 56,728;
(Cyi)2 = 3218,09;
Cyi2 = 214,574;
Cxiyi = 80,932;
Cxi2yi = 235,175;
Sx = 31,681;
Sxx = 386,638;
Sy = 0,347;
Sxy = – 1,0242;
Sxxy = – 3,0418.
Solution system (see 3.2.3)
c = 3,8288;
d = – 0,0262;
e = – 0,0022.
Coefficient of correlation (see 3.2.4)
r2 = 0,9647;
r = 0,9822.
Check of the suitability for extrapolation (see 3.2.5)
tv = 2,1604;
M = 15859,6.
The estimated mean values, Vm, for V at various times are given in Table 7 and shown in Figure 2.

© BSI 01-2000 15
EN 705:1994

Table 7 — Estimated mean values, Vm, for V

Time Vm
h

0,1 7 125
1,0 6 742
10,0 6 315
100,0 5 856

1 000 5 375
10 000 4 884
100 000 4 393
438 000 4 091

Figure 2 — Regression line from the results in Table 7

4 Application of methods to product design and testing

4.1 General
The referring standards specify limiting requirements for the properties and performance of a product.
Some of these are based on destructive tests, for example hoop tensile strength, whilst others are based on
actual or derived physical properties, such as stiffness.

16 © BSI 01-2000
 EN 705:1994

Where the physical property being determined enables both method B and method C to be used then both
procedures shall be performed. The value for r2 and/or r determined for each procedure shall be compared
and the value for the property determined using the procedure with the highest r2 or r value shall be taken
as the value for the property. If the referring standard specifies for such properties a method then only that
procedure shall be performed.
In either case, many of these properties need an extrapolated long-term (e.g. 50 years) value for comparison
with the requirement. This extrapolated value is determined by inserting, as necessary, the values
for a, b, c, d and e, determined in accordance with 3.1 or 3.2 as appropriate, into equations (31) or (32)
respectively.
lg y = a + btL (31)
lg y = c + dtL + etL2 (32)
where
tL is the logarithm (lg) of the long-term period, in hours, [for 50 years (438 000 h), tL = 5,64147].
Solving the equation, (31) or (32), for y gives the extrapolated value for comparison with the requirement
specified in the referring standard.
For supplementary procedures, where relevant, for the application of the results to derive design
requirements, see 4.2 and the worked examples given in 4.3 to enable validation of any calculation
facilities used.
For their use for testing products, to predict and verify the ability of a product to conform to a specified
requirement, see 4.4 and the examples given in 4.5 to enable validation of any calculation facilities used.
The wording of 4.2 to 4.5 is appropriate to specification limits in terms of minimum values, long-term
performance at 50 years and short-term performance at 6 min. For limits comprising maximum values or
other time periods, appropriate adjustments are necessary.
In conjunction with sampling requirements and limits on the acceptable levels, if any, on the quantity of
non-conforming products to be specified by referring standards and in certification or quality plans, as
appropriate, these methods can be used for quality control purposes.
4.2 Design
4.2.1 Regression values
4.2.1.1 Derived long-term values
Where it is assumed that, from initial short-term tests on pipe representative of such a design, the mean
value of the property being investigated, V0,m, and the estimated standard deviation, Ö, of the initial test
results are known, the procedures for designing a pipe to conform to a requirement of a referring standard
are as follows.
If a safety factor, FS, is specified, calculate the minimum long-term (50 years) value for the property,
V50, min., (see Figure 4), using the following equation:
V50,min. = FsV50,s, min. (33)
otherwise:
V50,min. = V50,s,min. (34)
where
V50,s,min. is the specified minimum long-term (50 years) requirement.
4.2.1.2 Regression ratio (see Figure 3)
Calculate the regression ratio, RR, using the following equation:

R R = extrapolated long-term (50 year) property value-

-----------------------------------------------------------------------------------------------------------------------------------
extrapolated short-term (6 min) property value

(35)

where both the 50 year and the 6 min extrapolated property values are calculated using equation (31)
or (32), as applicable, except that for the 6 min value the logarithm of 0,1 h (6 min), i.e. – 1, is used in place
of 5,64147.

© BSI 01-2000 17
EN 705:1994

Figure 3 — Derived values

4.2.1.3 Factor C (see Figure 3)

Calculate the factor C, which relates the extrapolated 6 min value to the test result for the initial tests on
the pipes, using the following equation:

(36)

4.2.2 Initial values

4.2.2.1 Minimum initial property value (V0,min.)
Derive the minimum initial property value, V0,min., using the following equation:

(37)

where
C is calculated using equation (36);
V50,min. is calculated using equation (33) or (34);
RR is calculated using equation (35).
4.2.2.2 Property design values
Derive the design value for the initial property, V0,d, using the following equation:
V0,d = V0,min. + 1,96Ö (38)
where
Ö is the standard deviation of the short-term test results for this property.

18 © BSI 01-2000
 EN 705:1994

Figure 4 — Extrapolated values

The factor 1,96 assumes that a 2,5 % failure criteria is acceptable. Where other percentage failure criteria
are specified in the referring standard then the relevant factor obtained from standard statistical
references shall be substituted for 1,96.
NOTE Where the standard deviation can be expected to increase with an increase in mean property values, and vice versa, then the
design property values for different levels of the property can be obtained using a constant coefficient of variation.
Determine the short-term (6 min) and long-term (50 years) design property values, and hence the design
line as follows:
— determine the estimated 6 min design property value, V6,d, using the following equation:

(39)

where
C is determined in accordance with 4.2.1.3;
— determine the estimated 50 years property value, V50,d, using the following equation:
V50,d = RRV6,d (40)
where
RR is the regression ratio determined in accordance with 4.2.1.2;
— if applicable (see 4.1), determine the equation for the linear design line as follows:
lg Vt,d = ad + (b × lg t) (41)
where
Vt,d is the design property value at a given time t;
b is the slope of both the minimum and the mean linear regression lines;

© BSI 01-2000 19
EN 705:1994

ad is the constant of the linear design line;

where
ad = a + ¸d (42)
where
a is the constant of the property value mean linear line [see equation (1)];

(43)

— if applicable (see 4.1), determine the equation for the second order design curve as follows:
lg Vt,d = cd + (d × lg t) + {e × (lg t)2} (44)

where
Vt,d is the design property value at a given time t;
d, e are the coefficients of both the design curve and the mean second order regression curve;
cd is the constant of the second order design curve;
where
cd = c + ¸d (45)
where
c is the constant of the second order mean regression curve [see equation (26)];

4.3 Examples for validation of calculation procedures for design

4.3.1 General
Use the data given in 4.3.2 and 4.3.3, as applicable, to verify that the calculation procedures used will give
results for a, b, c, d, e, r2, r, V50,min., V6,min., V0,min. and V0,d to within ± 0,1 % of the example values.
4.3.2 Example 1. Linear relationship; destructive test failure behaviour
4.3.2.1 The problem
Assuming that
a) the data given in Table 3 are burst test pressure results, in bars;
b) Table 4 comprises the results of the regression analysis of these burst test data for a pressure pipe;
c) initial burst test data are available in Table 8 for a pipe with a comparable wallbuild-up, as given
under a) and b),
determine the design values for a PN 10 (nominal pressure) pipe with a safety factor, FS, of 1,8.
Table 8 — Initial burst test results
Values in bars1)
52,0 44,5 49,0 50,3 46,7 51,1
47,3 49,7 53,3 51,1 46,0 50,3
45,9 49,1 48,8 46,7 49,8 53,6
Mean initial burst pressure p0,m = 49,18
Standard deviation Ö = 2,59
1)
1 bar = 105 N/m2 = 0,1 MPa

20 © BSI 01-2000
 EN 705:1994

Figure 5 — Example on pressure

4.3.2.2 Calculations and final results
For the data in Table 3 (see 3.1.2.6):
a = 1,62731;
b = – 0,03317,
and hence (Table 4):
p6 = 45,76 bar;
p50 = 27,55 bar.
Using equation (35):

Using equation (36):

The design values for the PN 10 pipe with a safety factor of 1,8 then become:
using equation (33):

and using equation (35) and (37):

p0,min. = C p6,min. = 1,0747 × 29,9 bar = 32,1 bar;

© BSI 01-2000 21
EN 705:1994

using equation (38):

p0,d = p0,min. + 1,96 Ö = 32,1 + 1,96 × 2,59 = 37,18 bar.
The 6 min and 50 years design values are derived using equations (39) and (40):

p50,d = RR p6,d = 0,6021 × 34,6 bar = 20,83 bar

These equations give rise to the plots shown in Figure 5.
4.3.3 Example 2. Second order relationship; non-destructive test (creep) behaviour
4.3.3.1 The problem
Assuming that
a) the data given in Table 5 are ring stiffness test results, in newtons per square metre;
b) Table 6 and Table 7 comprise the results of the regression analysis of these ring stiffness test data;
c) initial ring stiffness test data are available in Table 9 for pipes with a comparable wall build-up as
for a) and b),
determine the design values for a SN 5000 pipe with a minimum long-term ring stiffness of 2 000 N/m2 and
a safety factor, FS, of 1,0.
Table 9 — Initial ring stiffness test results
Values in newtons per square metre
7 540 7 200 6 970 7 190 7 760
7 170 7 100 7 310 6 990 7 180
Mean initial ring stiffness S0,m = 7 241
Standard deviation Ö = 243

4.3.3.2 Calculations and definitive results

Using method B (linear relationship, see 3.1.3.2 and 3.1.3.5):
r2 = 0,9556;
r = 0,9775.
Using method C (polynomial relationship, see 3.2.4 and 3.2.6):
r2 = 0,9647;
r = 0,9822.
Because r2, or r, obtained using method C is greater than that obtained using method B, the line derived
using method C is the more appropriate and therefore the results given in 3.2.6, calculated from the data
in Table 5, are the following:
c = 3,8288;
d = – 0,0262;
e = – 0,0022,
and hence (Table 7):
S6 = 7 125 N/m2;
S50 = 4 091 N/m2.
From these results, it follows that:
using equation (35):

using equation (36) and the results given in Table 9:

22 © BSI 01-2000
 EN 705:1994

The design values for a SN 5000 pipe with minimum long-term ring stiffness of 2 000 N/m2 and a factor of
safety of 1,0 then become:
using equation (34):
S50,min. = S50,s,min. = 2 000 N/m2;

and using the equations (35) and (37):

S0,min. = C S6,min. = 1,0163 × 3 483 N/m2 = 3 540 N/m2.
The value for S0,min from the initial requirement (5 000 N/m2) is greater than the one from the long-term
requirement (3 540 N/m2), therefore the relevant value is:
S0,min. = 5 000 N/m2.
Hence, using equation (38) and the results given in Table 9:
S0,d = S0,min + 1,96 Ö = 5 000 + (1,96 × 243) = 5 476 N/m2.
The 6 min and 50 years design values are derived using equations (39) and (40):

S50,d = RR × S6,d = 0,5742 – 5 388 N/m2 = 3 094 N/m2

Based on the initial requirement (5 000 N/m2) the 6 min and 50 years minimum values are also derived
using equations (39) and (40):

S0,50,min. = RR S0,6min. = 0,5742 × 4 920 N/m2 = 2 825 N/m2

These equations give rise to the plots shown in Figure 6.
4.4 Procedures for verifying conformance to product design and performance values
4.4.1 General
Given the following assumptions:
a) a pipe design is to be verified by short-term tests or by both short-term and long-term tests on a
product;
b) the design value, V0,d, of the property being investigated has been determined in accordance with 4.2,
use the following procedures as applicable to show that the product conforms to the design requirements,
where:
— method 1 (see 4.4.2) is valid for destructive test behaviour;
— method 2 (see 4.4.3) is valid for non-destructive test behaviour.
4.4.2 Method 1 — Verification using destructive test data
4.4.2.1 Relationship between design and/or minimum line and test results of products
Using the property design values determined in accordance with 4.2.2.2 unless otherwise specified in the
referring standard the following requirements shall be used to assess conformity with the standard.
When a large group, i.e. more than 20 consecutive results, is being assessed then the following conditions
shall be satisfied to show conformity with the referring standard:
a) the mean of the results shall be equal to or greater than the design value (see 4.2);
b) no individual result shall be less than 80 % of the minimum value (see 4.2).
When a small sample, i.e. up to and including five consecutive results, is being assessed then the
following conditions shall be satisfied unless the referring standard specifies otherwise:
c) the mean of the sample shall not be less than the design value minus the standard deviation used to
determine the design value (see 4.2);
d) no individual result shall be less than 80 % of the minimum value (see 4.2).

© BSI 01-2000 23
EN 705:1994

Figure 6 — Example for ring stiffness

4.4.2.2 Relationship between design line and product data
Where times other than zero, 6 min or 50 years are specified in the referring standard then equation (41)
(see 4.2.2.2) shall be used to determine the design value, Vt,d (see Figure 7).
4.4.2.3 Relationship between minimum line and product data (see Figure 7)
Where times other than zero, 6 min or 50 years are specified in the referring standard then the following
equation shall be used to determine the minimum required value, Vt,min..
lg Vt,min. = amin. + (b × lg t) (46)

where
Vt,min. is the minimum property value at a given time t;
b is the slope of both the minimum and the mean line;
amin. is the constant of the minimum line;
where
amin. = a + ¸min. (47)
where
a is the constant of the mean property value line [see equation (1)];

24 © BSI 01-2000
 EN 705:1994

Figure 7 — Property test values

4.4.2.4 Conformance to a design property value, Vt,d, after a specified time t (see Figure 7)
Where either the time to failure for the design line, td, or the time to failure for the minimum line, tmin., is
to be verified the appropriate one of the following equations shall be used:

(49)

(50)

4.4.3 Method 2 — Verification using non-destructive test data

NOTE Non-destructive tests for verification of long-term design or physical property behaviour are usually creep tests. It is assumed
that the period of time for testing, tC, and related property requirements are specified in the referring standard (see also 4.4.1).
4.4.3.1 Relationship between design and/or minimum line and test results of products
For these relationships the assumptions and requirements given in 4.4.2.1 are applicable.
4.4.3.2 Relationship between design line and product data
Where times other than zero, 6 min or 50 years are specified in the referring standard then equations (41)
and (44) as well as equations (46), (47) and (48) (see 4.2.2.2) shall be used to determine the design value,
Vt,d (for principle see Figure 7).
4.4.3.3 Relationship between minimum line and product data
4.4.3.3.1 Linear relationship (see Figure 7)
For the linear relationship, the assumptions, procedures and equations given in 4.4.2.3 are applicable.

© BSI 01-2000 25
EN 705:1994

4.4.3.3.2 Second order polynomial relationship (for principle see Figure 7)

Where times other than zero, 6 min or 50 years are specified in the referring standard then the following
equation shall be used to determine the minimum required value, Vt,min..
lg Vt,min. = cmin. + (d × lg t) + {e × (lg t)2} (51)

where
Vt,min. is the minimum property value at a given time t;
d, e are the coefficients of both the minimum and the mean regression line;
cmin. is the constant of the minimum line;

where
cmin. = c + ¸min. (52)
where
c is the constant of the mean property value line [see equation (26)];

(53)

4.4.3.4 Conformance to a design property value, Vt,d, after a specified time t (see Figure 7)
To verify that a product conforms to a design value Vt,d, or a minimum value Vt,min., after a time tc specified
in the referring standard the following equations shall be used to determine the relevant values where t is
to be replaced with tc:
for the linear relationship:
lg Vt,d = ad + (b × lg t) (41)
lg Vt,min. = amin. + (b × lg t) (46)
and for the second order polynomial relationship:
lg Vt,d = cd + (d × lg t) + {e × (lg t)2} (44)
lg Vt,min. = cmin. + (d × lg t) + [e × (lg t)2] (51)
4.5 Examples for validation of calculation procedures for design or product performance
verification
4.5.1 General
Use the data given in 4.5.2 and 4.5.3 to verify that the calculation procedures to be used will give results
for pt,d, tmin., St,d and St,min. to within ± 0,1 % of the results given in 4.5.2 and 4.5.3, as applicable.
4.5.2 Example 3. Linear relationship; destructive test failure behaviour
4.5.2.1 The problem
A series of pipes has been produced based on the design described in 4.3.2. The properties of test pieces
have been measured by short-term (initial burst) and long-term (100 h) pressure resistance tests. The
short-term test results are given in Table 10. The long-term test results, obtained from three test pieces
subjected to a test pressure, pt,d, determined as described in 4.5.2.3, are given in Table 11. Do the pipes
conform to 4.4.2.1 and 4.4.2.4?
Table 10 — Initial burst test failure pressures
for the sample
Values in bars
38,6 38,5 31,9 38,4
36,7 39,8 39,2 40,1
32,6 37,5 34,9 38,3
36,9 40,0 39,4 41,7
38,6 38,5 31,9 38,4
36,7 37,7 39,2 37,8
Mean initial burst pressure p0,m = 37,6
Minimum initial burst pressure p0,min. = 31.9

26 © BSI 01-2000
 EN 705:1994

Table 11 — Times to failure of burst tests of

the sample
Values in hours

125,3 101,3 97,4

Mean time to failure tm = 108
Minimum time to failure tmin. = 97.4

4.5.2.2 Calculations and final results

4.5.2.2.1 Determination of conformance to initial burst pressure requirements
From Table 10, p0,m (sample) = 37,6 bars and p0,min. (sample) = 31,9 bars;
from 4.3.2.2, p0,d = 37,36 bars and p0,min. (design) = 32 bars;
therefore p0,m (sample) > p0,d and p0,min. (sample) > 0,8 × p0,min.(design) (= 25,6 bars) and the product sample
conforms to 4.4.2.1 a) and 4.4.2.1 b).
4.5.2.2.2 Determination of conformance to long-term (100 h) pressure resistance requirements
Determine the equation of the design line as follows:
a) using equation (43):

b) using equation (42):

ad = a + ¸d = 1,62731 + (– 0,1178) = 1,5095;
c) using equation (41):
lg pt,d = 1,5095 – (0, 03317 × lg t). (58)
Determine the equation of the minimum line as follows:
d) using equation (48):

e) using equation (47):

amin. = a + ¸min. = 1,62731 + (– 0,1851) = 1,4423.
Determine the equation of the minimum time to failure for a given pressure as follows:
f) using equation (50):

(55)

Assuming that the referring standard requires a 100 h pressure test, determine the test pressure, pt,d,
to be used to test a sample and the minimum time to failure, tmin., as follows:
g) using equation (58):
lg pt,d = 1,44316
hence:
pt,d = 27,74 bar;
h) using equation (55):
lg tmin. = – 0,02593
hence:
tmin. = 0,942 h.
When three test pieces were tested using the test pressure of 27,74 bar, i.e. pt,d, the results given in
Table 11 were obtained.

© BSI 01-2000 27
EN 705:1994

From Table 11, tm (sample) = 108 h and tmin. (sample) = 97,4 h; from 4.5.2.1 td = 100 h and from 4.5.2.2 h)
tmin. = 0,942 h; therefore tm (sample) > td and tmin. (sample) > tmin. and the product conforms to 4.4.2.1 c)
and 4.4.2.1 d).
4.5.3 Example 4. Polynomial relationship; creep behaviour
4.5.3.1 The problem
A series of pipes has been produced based on the design described in 4.3.3. The properties of test pieces
have been measured by short-term (initial ring stiffness) and long-term (1 000 h creep) tests. The
short-term test results are given in Table 12 and the long-term test results in Table 13. Do the pipes
conform to 4.4.3.1 and 4.4.3.3?
Table 12 — Initial ring stiffness results for the sample
Values in newtons per square metre

5 600 4 980 5 720 5 620 5 690 5 695

5 550 5 980 5 020 5 720 560v0 5 295
5 560 5 680 5 550 5 620 5 795 5 570
Mean initial ring stiffness S0,m = 5 569
Minimum initial ring stiffness S0,min. = 4 980

Table 13 — Stiffness results at 1 000 h for the sample

Values in newtons per square metre

4 012 4 205 4 161

Mean long-term ring stiffness Sm = 4 126
Minimum long-term ring stiffness Smin. = 4 012

4.5.3.2 Calculation and final results

4.5.3.2.1 Determination of compliance with initial ring stiffness requirements
From Table 12, S0,m (sample) = 5 569 N/m2 and S0,min. (sample) = 4 980 N/m2;
from 4.3.3.2, S0, d = 5 486 N/m2 and S0,min. (design) = 5 000 N/m2; therefore S0,m (sample) > S0,d and
S0,min. (sample) > 0,8 × S0,min. (design) (= 4 000 N/m2) and the product sample conforms to 4.4.2.1 a)
and 4.4.2.1 b) (through reference by 4.2.2.2).
4.5.3.2.2 Determination of conformance to long-term (1 000 h) ring stiffness requirements
Determine the equation of the design line as follows:
a) using equation (46):

b) using equation (45):

cd = c + ¸d = 3,8288 + – 0,1205 = 3,7083;
c) using equation (44)

lg St,d = 3,7083 – (0, 0262 × lg t) {– 0,0022 × (lg t)2}. (56)

Determine the equation of the minimum line as follows:

d) using equation (53):

e) using equation (52):

cmin. = c + ¸min. = 3,8288 + (– 0,1608) = 3,668;
f) using equation (51):
lg St,min. = 3,668 – (0,0262 × lg t) – {0,0022 × (lg t)2}.

28 © BSI 01-2000
 EN 705:1994

Since the control procedure requires a 1 000 h ring stiffness test, determine the design stiffness, St,d, and
the minimum stiffness, St,min., as follows:
g) using equation (56):
lg St,d = 36 099
hence:
St,d = 4 073 N/m2;
h) using equation (57):
lg St,min. = 3,5696
hence:
St,min. = 3 712 N/m2.
When three test pieces were tested for ring stiffness at 1 000 h, the results given in Table 13 were obtained.
From Table 13, Sm (sample) = 4 126 N/m2 and Smin. (sample) = 4 012 N/m2;
from item g) St,d = 4 073 N/m2 and from item h) St,min. = 3 712 N/m2;
therefore Sm (sample) > St, d and Smin. (sample) > St,min. and the product conforms to 4.4.2.1 c) and 4.4.2.1 d).

© BSI 01-2000 29
EN 705:1994

Annex A (informative)
Mathematical procedures
A.1 Matrix system
Using the equations (27a), (27b) and (27c) given in 3.2.3 the solution using a matrix system is described.
Based on these equations the following matrix equation (A.1) can be constructed:

(A.1)

Using the values given in 3.2.6 for the respective variable the matrix equation (A.1) becomes
equation (A.2):

(A.2)

Following mathematical rules the 3 × 3-matrix on the left hand side of equation (A.2) is inverted which
results in matrix (A.3):

(A.3)

Multiplying both sides of the matrix equation (A.2) by matrix (A.3) results in the new equation (A.4):

(A.4)

Solving matrix equation (A.4) results in equation (A.5):

(A.5)

This rector matrix equation gives the solution for the variables c, d and e shown in 3.2.3 and in 3.2.6.
A.2 Substitution system
With the same set of equations (see 3.2.3) and with the data given in 3.2.6 an alternative way of solving
the problem is as follows.
NOTE Because the following section is worked out with rounded numbers the results may differ slightly when worked out on
different calculators.
From equation (27a):
56,728 = 15,000c + 21,671d + 62,989e (A.6)

From equation (27b):

80,932 = 21,671c + 62, 989d + 180,623e (A.7)
From equation (27c):
235,175 = 62,989c + 180, 623d + 584,233e (A.8)
Solving equation (A.8) to produce an equation for e results in equation (A.9):
e = 0,4025 – 0,1078c – 0,3092d (A.9)

30 © BSI 01-2000
 EN 705:1994

Substituting this equation (A.9) into equations (A.6) and (A.7) results in equations (A.10) and (A.11):
31,375 = 8,210c + 2,195d (A.10)
8,231 = 2,200c + 7,140d (A.11)
Solving equation (A.11) to produce an equation for d results in equation (A.12):
d = 1,1528 – 0,3081c (A.12)
Substituting this equation (A.12) for d into equation (A.10) results in equation (A.13):
7,5337c = 28,8446 (A.13)
and hence:
c = 3,8288 (A.14)
Using the result from equation (A.14) and substituting this for c in equation (A.12) results in
equation (A.15) for d:
d = – 0,0262 (A.15)
Using the values from equations (A.14) and (A.15) for c and d respectively in equation (A.9) gives a value
for e of:
e = – 0,0022 (A.16)
As can be seen the results from both methods are the same (but see note in A.2).

© BSI 01-2000 31
BS EN
705:1995
BS 2782-12: BSI — British Standards Institution
Methods
BSI is the independent national body responsible for preparing
1220A to C: British Standards. It presents the UK view on standards in Europe and at the
1995 international level. It is incorporated by Royal Charter.

Revisions

British Standards are updated by amendment or revision. Users of

British Standards should make sure that they possess the latest amendments or
editions.

It is the constant aim of BSI to improve the quality of our products and services.
We would be grateful if anyone finding an inaccuracy or ambiguity while using
this British Standard would inform the Secretary of the technical committee
responsible, the identity of which can be found on the inside front cover.
Tel: 020 8996 9000. Fax: 020 8996 7400.

BSI offers members an individual updating service called PLUS which ensures
that subscribers automatically receive the latest editions of standards.

Buying standards

Orders for all BSI, international and foreign standards publications should be
addressed to Customer Services. Tel: 020 8996 9001. Fax: 020 8996 7001.

In response to orders for international standards, it is BSI policy to supply the

BSI implementation of those that have been published as British Standards,
unless otherwise requested.

Information on standards

BSI provides a wide range of information on national, European and

international standards through its Library and its Technical Help to Exporters
Service. Various BSI electronic information services are also available which give
details on all its products and services. Contact the Information Centre.
Tel: 020 8996 7111. Fax: 020 8996 7048.

Subscribing members of BSI are kept up to date with standards developments

and receive substantial discounts on the purchase price of standards. For details
of these and other benefits contact Membership Administration.
Tel: 020 8996 7002. Fax: 020 8996 7001.

Copyright

Copyright subsists in all BSI publications. BSI also holds the copyright, in the
UK, of the publications of the international standardization bodies. Except as
permitted under the Copyright, Designs and Patents Act 1988 no extract may be
reproduced, stored in a retrieval system or transmitted in any form or by any
means – electronic, photocopying, recording or otherwise – without prior written
permission from BSI.

This does not preclude the free use, in the course of implementing the standard,
of necessary details such as symbols, and size, type or grade designations. If these
details are to be used for any other purpose than implementation then the prior
written permission of BSI must be obtained.

If permission is granted, the terms may include royalty payments or a licensing

agreement. Details and advice can be obtained from the Copyright Manager.
BSI Tel: 020 8996 7070.
389 Chiswick High Road
London
W4 4AL