ISO/TTA 2:1997

TECHNOLOGY lSO/TTA 2
TRENDS
ASSESSMENT
First edition
1997-04- 15
Tensile tests for discontinuously reinforced
metal matrix composites at ambient
temperatures
Essais de traction pour composites 2 matrice renforcke de manike
discontinue de mktal h temperatures ambiantes
Reference number
ISO/TTA 2:1997(E)

---------------------- Page: 1 ----------------------
ISO/TTA 2: 1997(E)
CONTENTS
Page
. . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
FOREWORD 111
EXECUTIVE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .~.~.
iv
INTRODUCTION - VALIDATION EXERCISE
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V
MATERIALS AND TESTPIECES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
V
l’ARTiClPATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
V
DISCUSSIOlV OF RESULTS . .0.e.~.~.
vi
CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
xi
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
REFERENCES xii
1 . SCOPE . . . . . . . . . .0.~.~.~~~~.
1
2 . PRINCIPLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .~.
1
. DEFINITIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3 1
4 . SYMBOL DESIGNATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4
5 . TESTPIECES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5
e DETERMINATION OF ORIGINAL CROSS-SECTIONAL AREA (SJ
6 . . . . . . . . . . . . . . . . . . . . . . . . . . .
5
7 e MARKING THE ORIGINAL GAUGE LENGTH (L,) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6
8 . ACCURACY OF TESTING APPARATUS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
.
9 CONDITIONS OF TESTING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6
10 . DETERMINATION OF PERCENTAGE ELONGATION AFTER
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
FRACTURE (AP) 7
11 . DETERMINATION OF PROOF STRENGTH (NON-PROPORTIONAL
EXTENSION) (Rp) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8
DETERMINATION OF PROOF STRENGTH (TOTAL EXTENSION)
12 . (R,) . . . . . . . . . . . . . . . . . . . . 8
13 . DETERMINATION OF YOUNG ’S MODULUS (E) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9
14 . DETERMINATION OF PROPORTIONAL LIMIT (PL)
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
15 . DETERMINATION OF TENSILE STRENGTH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10
16 . TEST REPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10
ANNEXES
A TESTPIECE DIMENSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
11
B TEST REPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
13
C BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
16
0 ISO 1997
All rights reserved. Uniess otherwise specified, no patt of this publication may be reproduced or utilized in any form or by any means, electronie
or mechanical, including photocopying and microfilm, without Permission in writing from the publisher.
International Organization for Standardization
Case postale 56 l CH-121 1 Geneve 20 l Switzerland
Internet central @ iso.ch
x.400 c=ch; a=400net; p=iso; o=isocs; s=central
Printed in Switzerland
ii

---------------------- Page: 2 ----------------------
@ ISO
ISO/lTA 2: 1997(E)
FOREWORD
ISO (the International Organization for Standardization) is a worldwide federation of national
Standards bodies (ISO member bodies). The work of preparing International Standards is
normally carried out through ISO technical committees. Esch member body interested in a
subject for which a technical committee has been established has the right to be represented
on that committee. International organizations, govemmental and non-govemmental, in
liaison with ISO, also take part in the work. ISO collaborates closely with the International
Electrotechnical Commission (IEC) on all matters of electrotechnical Standardisation.
To respond to the need for global collaboration on standardization questions at early stages
of technological innovation, the ISO CounciI, following recommendations of the ISO/IEC
Presidents’ Advisory Board on Technological Trends, decided to establish a new series of ISO
publications named ‘Technology Trends Assessments” (ISO/ ‘ITA). These publications are
the results of either direct cooperation with prestandardization organizations or ad hoc
Workshops of experts concemed with standardization needs and trends in emerging fields.
Technology Trends Assessments are thus the result of prestandardization work or research.
As a condition of publication by ISO, ISO/TTAs shall not conflict with existing International
Standards or draft International Standards (DIS), but shall contain information that would
normally form the basis of standardization. ISO has decided to publish such documents to
promote the harmonization of the objectives of ongoing prestandardization work with those
of new initiatives in the Research and Development environment. It is intended that these
publications will contribute towards rationalization of technological choice Prior to market
entry.
This Technology Trends Assessment, ISO/TTA 2, has been developed by the Versailles
Project on Advanced Materials and Standards (VAMAS) and is published under a
Memorandum of Understanding concluded between ISO and VAMAS. It reports the results
of the Technical Working Area (TWA) 15 of VAMAS, which has the task of investigating
mechanical test methods for metal matrix composites and which retains the responsibility for
the technical content of this ISO/TTA. Users of this ISO/TTA who would like information
on the research project should refer to a recent report of VAMAS TWA 15 which was
prepared by Dr B Roebuck, Dr L N McCartney and Dr J D Lord of the NPL under the
leadership of Dr Steve J Johnson at Georgia Tech., Atlanta, USA. The ISO Technical Board
approved the publication of this classification as an ISO/ ‘ITA in late 1995.
Whilst ISO/TTAs are not Standards, it is hoped that they will be used as a basis for
Standards development in future national and international standardization processes. In the
particular case of ISO/TTA 2, the publication has been brought, in the first instance, to the
attention of ECISS/TCl, Tensile Testing Standards, for use in its Standardisation work.

---------------------- Page: 3 ----------------------
ISO/TTA 2: 1997(E)
EXECUTIVE SUMMARY
There is a need for a tensile testing Standard for discontinuously reinforced metal
matrix composites (MMC). Use of the current ISO Standard for metals EN 10002 leads
to unsatisfactory uncertainties in the property values measured, particularly for
Young ’s modulus and proportional limit.
The measurement of Young ’s modulus in
MMC is important for several reasons:
Improvements in specific stiffness are an important driver in increasing the use
a>
of MMC over conventional materials.
An accurate knowledge of the
engineering value of Young ’s modulus is vital for preliminary design studies.
Proof stress measurements require a Prior knowledge of the Young ’s modulus.
W
If the material of interest has a high work hardening rate in the early Stage of
yield then inaccuracies in the Young ’s modulus tan lead to significant
inaccuracies in proof stress.
MMC have low proportional limits because of internal residual Stresses. It is
4
important to be able to measure the proportional limit accurately and to assess
the extent of yield at low strains. An accurate value of Young ’s modulus is
required to obtain reliable values for the proportional limit.
Accurate measurements of Young ’s modulus are required to give good fits to
4
the constitutive expressions for the stress/strain data.
Following analysis of the results of a UK exercise to examine the sources of uncertainty
in the measurement of the tensile properties of Sic particulate reinforced Al alloys a
draft procedure was written for tensile tests on particulate MMC at ambient
temperatures. The draft procedure recommends appropriate testpiece dimensions,
testing rates, methods of gripping and strain measurement techniques. It also defines
methods for measuring Young ’s modulus, proportional limit, proof stress, tensile
strength and elongation to failure. Significantly it contains a recommended proforma
for the test report in anticipation of future database requirements. The draft procedure
forms the basis of this ISO/TTA document. It was validated by two interlaboratory
exercises, one through VAMAS (internationally) and one in the UK (led by NPL). The
outcome of this Validation exercise is also summarised in the Introduction to the
ISO/TTA document.
The style of the draft procedure is similar to that adopted for the current EN tensile
testing Standards, EN10002 pt 1 (tensile tests for metals) and its sister document for
Aerospace materials EN20024 part 1.

---------------------- Page: 4 ----------------------
@ ISO
ISOKT ’A 2:1997(E)
INTRODUCTION - VALIDATION EXERCISE
Two Validation exercises were carried out to tonfirm the Utility of the draft procedure:
VAMAS
An intercomparison using the tensile testing draft procedure [l] was instigated under
the guidance of the VAMAS Technical Working Area 15 on Metal Matrix Composites.
One of the important objectives of VAMAS is to harmonise testing procedures
intemationally. The current exercise included Organkations from the UK, USA, Japan,
France, Spain and Germany.
UK MMC Forum
Another intercomparison was organised by NPL through a sub-committee of the UK
FORUM on TEST METHODS for MMC. It included a subset of the organisations
involved in the first UK exercise [2] which were Chosen to be representative of
industry, academia and research organisations.
Appropriate testpieces were distributed by NPL to the participating organisations in
each exercise together with copies of the draft tensile testing procedure. Esch
Organisation tested 3-4 testpieces. The results were retumed to NPL for collation and
analysis.
MATERIALS AND TESTPIECES
VAMAS:
The MMC was provided by ACMC Ltd (USA) and was in the form of extruded 2009
A1/20% Sic,.
It was machined into dogbone rectangular testpieces (Type Tl [l] -
6 mm x 3 mm Cross section; 25 mm gauge length) by NRIM, Japan.
UK Forum:
An MMC and an unreinforced Al matrix alloy were included in this study. The
MMC was provided by AMC Ltd (UK) as rolled plate 2124 Al/20% SiCp. The Al
alloy was provided by Alcan International Ltd as extruded bar (Alcan Cospray
2618). Both materials were machined at NPL into similar geometry testpieces as
those used in the VAMAS exercise (Type Tl [ 11). All the testpieces were machined
using diamond (PCD) Tooling.
PARTICIPATIORI
VAMAS:
NPL UK Bordeaux Univ France
DRA (Farnborough) UK BMW
Germany
BAe (Warton) UK DLR Germany
NIST USA TUHH Germany
NASA USA Honda a an
JP
Inasmet Spain
NRIM a an
JP

---------------------- Page: 5 ----------------------
ISO/lTA 2: 1997(E)
UK Forum:
ERA
NPL
DRA (Famborough) BAe (Warton)
Oxford Univ
Lucas
Sheffield Univ
Hi-Tee
In reporting the results, alI the VAMAS participants were identified (by agreement); in the
UK exercise participants remained anonymous and coded.
DISCUSSION OF RESULTS
GENERAL COMMENTS
It is significant that alI the participants were able to use the draft procedure and results
proforma without any major Problems and this clearly validated the draft procedure as a
satisfactory written document. A number of comments were made on the tests and results
by some of the participants and these remarks were used to make small changes to the
procedure outlined in this document.
YOUNG ’S MODULUS AND STRAIN MEASUREMENT METHOD
The draft procedure for tensile testing [l] allowed three different types of analysis method
to be used to calculate Young ’s modulus. These are referred to as Ml, M2 and M3 and there
are two subsets of M2 - M2A and M2B. These methods tan be summarised as follows.
Ml - Graphical
From a straight line drawn parallel to the initial Portion of a load/strain curve, idealIy
plotted as close as possible to 45O to the strain axis on A3 Paper.
M2 - Chorda1 (using Computer Software)
From a straight line between two arbitrarily Chosen limits on the initial Portion of the
stress/strain curve.
M2A -
direct straight line between the two Points.
M2B -
linear regression fit to the data between the Points.
M3- Tangent Computer Software)
wng
This is the NPL recommended method [3], based on the derivative of the quadratic
polynomial fitted locally to the stress/strain data.
All three methods were used by the various participants. Data were obtained using either
Single or double sided strain measurement with either strain gauges or extensometers.
VAMAS
It was clear that for the most part the use of double sided strain measurement Systems gave
more reproducible and more accurate results.
Typically the Standard deviations (SD) obtained using double sided strain gauges were less
than 1% and less than 2% for the double sided extensometry. However, for the Single sided
Systems the Standard deviations were much larger, sometimes significantly greater than 5%.
Vi

---------------------- Page: 6 ----------------------
@ ISO
ISOmA 2: 1997(E)
The Ml method in general gave less scatter than the M2 (computer-based) method.
However, this was not true in every case because the NASA results obtained using the M2
method were as repeatable and accurate as the results from NPL using the M3 method. The
reason for this discrepancy tan possibly be explained through examination of the upper and
lower limits used bv the different participants:
I I
Method of Upper and lower Standard Deviation Deviation from
Analysis limits kN mmo2 mean
Participant
N mni2 kN mmB2
ll
M2 0-275 04 . +0.2
NASA
M2 0-100 14 . -49
Inasmet
M2 54 . +2:4
BMW M2 1500250,175-350 66 . +7.5
M2B 25-125 24 . + 5.6
BAe
Clearly there is a wide range in the values Chosen for the upper and lower limits and this
may have contributed to greater uncertainties.
Another possible reason for the accurate and repeatable results from the NASA data set was
the use of a class 0.5 extensometer. The draft procedure allows the use of two tes(piece
geometries with nominal gauge lengths of 25 or 50 nun. It might be prudent to reco&end,
where possible, the use of the larger testpiece (Type T2) for measurements using double
sided extensometry. For example, for measurements using the M2 method (between 50 and
250 N mmm2) the equivalent strains are about 0.05 and 0.25%. On a gauge length of 25 mm
these strains correspond to displacements of 12.5 and 62.5 Pm respectively. As tan be seen
in the following table increasing the gauge length to 50 nun brings about a useful potential
increase in accuracy.
Gauge Displacement, Fm Uncertainty (extensometer Estimated uncertainty in E, %
length M2 method class*), Pm
(50-250 N mm- ‘)
mm
I
UPPer Lower Class 0.5 type Class 1.0 type Class 0.5 type Class 1.0 type
25 12.5 62.5 0.5 1.0 32% f4%
I
50 25 125 0.5 1.0 fl% It2%
* estimates have been used because of the difficulty of comparing values from different
available Standards.
UK Forum
For the UK FORUM exercise the outcome and uncertainties associated with the different
methods were very similar to those reported above for the VAMAS exercise. For example,
the measurements made using Single sided Systems were more likely to be in error than with
double sided Systems. Also, double sided strain gauges gave more repeatable resuhs than
double sided extensometry. However, the use of strain gauges did not always give accurate
values for the modulus. Some organisations which used double sided strain gauges had the
same systematic deviation (approximately -5 and +5 kN mmo2 respectively) for tests on both
the MMC and Al matrix, thus indicating a common Cause. The most likely reason for this
is uncertainty in the value of the gauge factor. In a separate exercise [4] it has been shown
that differentes of 5% tan easily be reported from this Source. The report format should
therefore have a suitable entry for recording the gauge factor if strain gauges are used and
vii

---------------------- Page: 7 ----------------------
ISO/lTA 2: 1997(E)
@ ISO
to what accuracy this is known. Clearly gauges of different tost are available and in general
the eheaper the gauge the less accurate is the gauge factor.
As in the VAMAS exercise method Ml gave more accurate results than method M2, possibly
for similar reasons since the proportional limit for these materials was even lower (-250 cf
-300 N mmo2). Method M3 gave the most acc-wate and repeatable results, as had been found
in the previous UK intercomparison exercise [2].
Summary (Young ’s Modulus and Strain Measurement Method)
A number of conclusions tan be drawn from the two exercises (VAMAS and UK FORUM)
conceming the measurement of Young ’s modulus.
.
1 The most accurate values were obtained at NPL using a double sided strain
measurement System together with the M3 method of analysis. This procedure
resulted in Standard deviations of about K).5% (1 SD) in the measurement of
modulus.
2 . In general, the use of double sided strain measurement Systems resulted in
uncertainties of less than ti% (1 SD) in the measurement of modulus; Single sided
Systems were generally significantly worse, with uncertainties of &5% (1 SD) or
greater.
3 . Overall, except for two organisations, the exercise reported uncertainties of less than
EJ% (1 SD) in the measurement of modulus. This compares very well with the
previous UK exercise where a significant number of uncertainties greater than HO%
(1 SD) were reported. With some modification the use of the draft procedure should
ensure that in future tests uncertainties should be kept within -t3% (1 SD) for all
methods. The potential exists within the Standard procedure for uncertainties to be
as low as fo.5% (1 SD).
4 .
The results were more dependent on the use of a double sided strain measurement
System than on the method of analysis. The Chorda1 method could possibly be
modified to specify bounds for the upper and lower limits for the data fit. These
limits are likely to be material dependent and necessary guidelines would need to be
investigated through collaborative projects between users and suppliers. For example,
in aluminium alloy matrix MMC it would be unwise to use values for the upper limit
much greater than 250 N mmB2 because of the low proportional limit in these
materials.
5 . The finalised test procedure should recommend the use of the larger testpiece
(Type T2) where the most accurate measurements are required (to better than fl%)
and where only extensometry is available for the tests.
6 . The test procedure should also request users to include and use an accurate value for
the gauge factor if strain gauges are used.
. . .
VIII

---------------------- Page: 8 ----------------------
@ ISO ISO/TTA 2: 1997(E)
PROPORTIONAL LIMZT
The uncertainty in the measurement of proportional limit was fairly high as the following
summary indicates
Exercise Proportional Limit Standard Deviation
(Mean value)
N mni2 N mni2 (f%)
366 58 (16)
VAMAS
UK FORUM (MMC) 268 48 (18)
UK FORUM (Matrix) 298 72 (24)
These uncertainties were however considerably better than had been observed in the first UK
intercomparison [2] where the Standard deviation in results had been about SS%. For most
of the organisations using double sided measurement Systems the measurements were
reasonably repeatable with uncertainties (1 SD) typically about fl%. However, the
reproducibility, between organisations, was less good, increasing the uncertainties to typically
klO%. It was suggested by the Bordeaux University participants that the reproducibility
could probably be improved by increasing the value of plastic strain at which the
proportional limit is defined to that equivalent to the measurement of a 0.02% proof stress.
The data from one test was analysed to examine the Variation in proportional limit with a
A
range of selected values of proof stress with the following results A
Proof stress, % Proportional limit
N mmo2
NPL procedure, (0.005) 351
.
0.02 354
, d
0.05 395
01 . 416
435
02 .
1
J
1’
Due to the high initial work hardening rate of the MMC there is a very rapid increase in
proportional limit for small increments in plastic deformation. If an alternative definition is
to be adopted from that in the draft procedure along the lines indicated by Bordeaux
University, then 0.002% or 0.005% would be more realistic than 0.02%. It will probably be
useful to rewrite the procedure so that this alternative is allowed provided that the % plastic
strain is not greater than O.Ol%, and that the value Chosen is specified in the results sheet.
It is also likely that better reproducibility would have been observed if the method of analysis
had been more constrained, particularly M2, (where arbitrary values of stress are Chosen,
between which the modulus is fitted). For example, the values of proportional limit
correlated with the analysis method, since the M2 and M3 methods gave smaller values
than Ml.
IX

---------------------- Page: 9 ----------------------
ISORTA 2:1997(E)
PROOF AND TENSZLE STRESS
The values for proof stress showed the least scatter of all the measured properties, with
typical uncertainties of 1-2.3% (1 SD) for all participants. The tensile strength values had
slightly more scatter with uncertainties of 3.5%. However a trend of increasing tensile
strength with increasing elongation to failure was noted, particularly in the VAMAS exercise.
Thus, with more consistent elongations to failure it might be expected that the uncertainties
in tensile strength resulting from the method of measurement could be as low as k 1%.
ELONGATION TO FAILURE
The elongation to failure values showed considerable Variation in the MMC tests, ie about
2.7% in both the VAMAS and UK FORUM exercises. Even the tests on the Cospray Al alloy
showed variations of about 342%. Much ’of this Variation was due to testpieces failing
outside the gauge length. For example in the VAMAS exercise about 50% of the failures
were at or close to the Position where the extensometers were attached to the testpieces. The
Overall uncertainty on elongation including these “invalid tests” was about tiS%. The spread
in elongation values was much less, about HO%, for those tests in which testpieces failed
within the gauge length.
STRAlN RATE EFFECTS
The draft test procedure specifies a maximum stressing rate of 10 N mni2 so1 in the elastic
range; this corresponds to a strain rate for the MMC tested in this exercise of about 10”’ sa1
and is a compromise between sufficient time for data Capture and test convenience. Beyond
the elastic limit, for measurements of proof Stresses, the strain rate tan be increased to
2~10~ so ’. The draft procedure does not indicate an appropriate strain rate for testing
between the proof stress and tensile strength in those cases where Young ’s modulus, proof
stress and tensile strength are all required to be measured. It only specifies a strain rate of
10°3 s ’l in the plastic range in those cases where modulus is not required to be measured.
Clearly the draft procedure requires some modification to Section 9 to include an upper limit
-1
of 10. s for testing in the plastic range in those cases where all the tensile properties are
required to be measured.
The procedure does allow other strain rates to be used if specified in a product Standard.
RESULTS PROFORh4A
The intercomparisons have underlined the usefulness of making a number of small changes
to the results proforma.
These have been included in the modified procedure which form
the basis of this TTA.

---------------------- Page: 10 ----------------------
@ ISO
ISO/lTA 2: 1997(E)
UNCERTAZNTIES
Typical values for the uncertainties (1 SD) associated with each property measurement tan
be summarised as follows in comparison with the uncertainties associated with the previous
UK intercomparison exercise.
Intercomparison Uncertainties (1 SD)
VAMAS and UK FORUM results UK intercomparison
Property
(New MMC procedure) (Existing Standards for metals)
double sided strain measurement
Young ’s modulus It 2%’ * 7%
Proportional limit * 20%+ zk 28%
Proof stress * 2% i4%
* 4%$
Tensile strength *3%
Elongation to Fracture * 25(10)%- * 35%
* Potentially better than & 1% with the M3 method of analysis and strain gauges with
accurately known gauge factors
*’ For all tests; (+ 10%) for tests failed in gauge length
+ Could possibly be reduced further by the use of a x% plastic strain specification for the
proportional limit, where x should be less than 0.01 and specified by agreement
$ Probably better than & 1% for those testpieces that failed in the gauge length.
CONCLUSIONS
The VAh4AS and UK FORUM intercomparisons have validated the draft procedure [l] for
tensile testing of particulate reinforced MMC at ambient temperatures. Analysis of the results
has indicated the need for a small number of changes to the procedure, including the results
proforma (Appendix). The original draft procedure has been modified to take account of
these changes (proportional limit, strain rate) and will be submitted to the appropriate
Standards bodies for approval when this ‘ITA has been published and circulated and after
taking into account additional comments that this wider dissemination might generate. For
example, some changes have been made already as a result of peer review by ISO member
countries - on the use of strain gauges, machine grips and testing rate.
The intercomparisons demonstrated that measurement uncertainties were very much reduced
by the use of the new test procedure when compared with the first UK intercomparison
exercise, which in general followed existing Standards for metals. Much of the improvement
has clearly been due to the use of double sided strain measurement Systems.
Xi

---------------------- Page: 11 ----------------------
@ ISO
ISO/TTA 2: 1997(E)
REFERENCES
1 . B Roebuck and J D Lord. NPL Report DMM(A)lOO, December 1993. Particulate
MMC - Draft Procedure for Tensile Tests at Ambient Temperature.
2 .
B Roebuck, L N McCartney, P M Cooper, E G Bennett, J D Lord and L P Orkney.
NPL Report DMM(AI77, November 1992. UK Interlaboratory Tensile Tests on Al
Alloy Sic Particulat
...