Local Strain and Temperature Measurement

337 Pages·1999·61.099 MB·English

by  J. Ziebs (Eds.)

LOCAL STRAIN
AND TEMPERATURE MEASUREMENTS
IN NON-UNIFORM FIELDS
AT ELEVATED TEMPERATURES
Proceedings of the Symposium held at
Berlin, Germany,
14-15 March 1996
Edited by
J. ZIEBS*. J. BRESSERS**, H. FRENZ*. D.R. HAYHURST
H. KLINGELHOFFER*, S. FOREST****
*' Bundesanstalt fur Materialforschung und -priifung, Berlin, Germany,
**JRC, Institute for Advanced Materials, Petten, The Netherlands,
Department of Mechanical Engineering, UMIST, Manchester, United Kingdom,
**** Centre des Materiaux, Ecole des Mines de Paris, Evry, France
WOODHEAD PUBLISHING LIMITED
CAMBRIDGE, UK
Published by Woodhead Publishing Limited,
80 High Street, Sawston, Cambridge CB22 3HJ, UK
www.woodheadpublishing.com
Woodhead Publishing, 1518 Walnut Street, Suite 1100, Philadelphia,
PA 19102-3406, USA
Woodhead Publishing India Private Limited, G-2, Vardaan House,
7/28 Ansari Road, Daryaganj, New Delhi - 110002, India
www.woodheadpublishingindia.com
First published 1999
© Bundesanstalt fur Materialforschung und -prufung, 1999
This book contains information obtained from authentic and highly regarded sources. Reprinted
material is quoted with permission, and sources are indicated. Reasonable efforts have been
made to publish reliable data and information, but the authors and the publisher cannot assume
responsibility for the validity of all materials. Neither the authors nor the publisher, nor anyone
else associated with this publication, shall be liable for any loss, damage or liability directly or
indirectly caused or alleged to be caused by this book.
Neither this book nor any part may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, microfilming and recording, or by any
information storage or retrieval system, without permission in writing from Woodhead Publishing
Limited.
The consent of Woodhead Publishing Limited does not extend to copying for general
distribution, for promotion, for creating new works, or for resale. Specific permission must be
obtained in writing from Woodhead Publishing Limited for such copying.
Trademark notice: Product or corporate names may be trademarks or registered trademarks,
and are used only for identification and explanation, without intent to infringe.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library.
ISBN 978-1-85573-424-1 (print)
ISBN 978-0-85709-314-1 (online)
Printed by Lightning Source.
PREFACE
The International Symposium "Local Strain and Temperature Measurements in Non-
Uniform Fields at Elevated Temperatures", held at the Technical University Berlin (TUB), Berlin,
Germany, on March 14 - 15, 1996, was jointly organized by the High Temperature Mechanical
Testing Committee (HTMTC) and by the Federal Institute for Materials Research and Testing
(BAM), Berlin, in cooperation with the German Association of Materials Research and Testing
(DVM).
The symposium was devoted to precision strain and temperature measurements in non­
uniform fields encountered in well-controlled laborarory testpieces and in highly complex
loading situations typical of those encountered in operating engineering components.
Measurement techniques contribute to all areas of experimental mechanics research and play
a critical role in the type of information which can be determined on material response, defects
and residual stresses. All of these parameters being crucial to the accurate prediction and
interpretation of failures at both the material and at the engineering component levels.
Expertise developed for optical methods of stress analysis has been redirected towards
more general optical methods for determining the displacements, strains, and velocities of the
surfaces of bodies in non-uniform fields. Development of these methods has been accelerated
by concurrent advances in laser technology. In addition, innovative applications of interferometry
and holography have led to extremely valuable techniques for monitoring the motion and
deformation of surfaces. Because these measurements can be made in-situ on the material,
and often in the actual component, they can be compared directly with numerical predictions
earned out at the component design stage, which are based on inelastic models for the material
response. Furthermore, because optical techniques do not require mechanical contact with the
moving/deforming body, measurements can be made at high temperatures and at high rates of
deformation. Other developments taking place which have potential for novel applications are
measurement techniques for the micron and submicron range.
As a foundation for the numerous applications of these techniques, it is essential to
maintain a strength and continuing effort in the development and standardization of basic
measurement concepts. But principal goals are to enable the verification of numerical/computer
predictions of component behaviour against high quality experimental data; and, to promote
improved standards and better Quality Assurance.
The objective of the symposium was also to bring together scientists and engineers with
different backgrounds and perspectives, but with a common interest in measurement
techniques. Also, the symposium aimed to identify problems which have been encountered with
current techniques, and to identify how future research may be directed towards overcoming
these.
Symposium sessions were held on State of the Art and Development Trends in Local
Strain Measurements, Local Temperature Control and Measurement, Link between
Specimen/Component Testing and Computer Simulation of Specimen/Component Behaviour,
and Areas of Standardization and Need for Future Research. The Presentations were of three
kinds: 4 keynote lectures, 20 contributions and 10 poster contributions. These Proceedings
contain the text of the 34 papers.
The editors wish to thank all the authors and delegates for their contributions. Thanks
are also due to the keynote speakers and session chairmen who all played decisive roles in
stimulating a rewarding discussion. Finally the editors would like to thank colleagues from BAM,
who contributed behind the scenes to the efficient organisation of the symposium and the
compilation of these proceedings.
Editors
International Advisory Board
J. Y. Guedou, Evry
D. Lohe, Karlsruhe
P. McCarthy, Leatherhead
D. Munz, Karlsruhe
D. Nouailhas, Chatillon
A. Pineau, Paris
R. Ritter, Braunschweig
F. Schubert, Julich
D.J. Smith, Bristol
H. Wollenberger, Berlin
Programme and Organizing Committee
J. Bressers, Petten
S. Forest, Paris
H. Frenz, Berlin
D.R. Hayhurst, Manchester
R.C. Hurst, Petten
H. Klingelhoffer, Berlin
J. Olschewski, Berlin
J. Ziebs, Berlin
Conference Chairman
J. Ziebs, Berlin
STRAIN MEASUREMENT
IN MATERIAL TESTING
BY OPTICAL FIELD METHODS
R. RITTER
Abteilung Experimentelle Mechanik
Institut fur Technische Mechanik
Technische Universitat Braunschweig
SchleinitzstraBe 20, D-38106 Braunschweig
Abstract
A summary will be given of the characteristics and properties of optical field methods. The
methods have achieved a high technological level during the last few years, due to the advance­
ments in modern optoelectronic components and the development of digital image processing of
optical signals. These signals are the information carriers of the searched geometrical characte­
ristics of the considered object. The paper is divided into three parts. First, the object grating
method and the electronic speckle pattern correlation interferometry are presented. In general
these methods lead directly to displacement fields. The associated strain values have to be de­
termined from these fields numerically. Several related procedures, which are practicable and
of high precision, will be introduced. Finally, three examples will demonstrate the applicability
and the advantages of these field methods related to various problems in material testing.
Keywords
Optical field method, Object grating method, Electronic speckle pattern correlation interfero­
metry, Digital image processing, Displacement measurement, Numerical strain determination
1. INTRODUCTION
The fundamental principle of optical field methods consists of generating and imaging patterns
as grey level distributions in the image plane of recording cameras. These distributions are
the information carriers of the searched geometrical object characteristics, such as local vector,
displacement, contour or slope. The image values of the object are determined using digital
image processing of the patterns. The object values are then retransformed from the image
data into the object space [1], [2].
The optical field methods can be classified by models, describing the properties of light. In case
of incoherent methods or ray optics the amplitude of the light is the information carrier. Their
sensitivity depends on the lateral resolution of the recording CCD camera and the accuracy of
1
the image processing algorithm. Interferometric methods are all based on coherent optics. Here
the information carrier is the phase difference of the light and the sensitivity depends on the
order of the wavelength [3].
These optical field methods have achieved a high technological level during the last few years
by applying modern optoelectronic elements and digital image processing algorithms. The mea­
surements are possible on surfaces within regions varying in size from a few square micrometers
to several square meters [4]. They take place without contact and interaction. Because of the
compact construction and small dimensions of the related devices they can be adapted directly
at testing machines and arrangements [5].
For dimensioning of objects, the strain distribution must be known. In general, it is determined
numerically from the measured local vectors, related to two different deformation states of the
object. The first derivatives of displacement can also be measured directly using the well known
shearing principle [6].
In the following two examples, the object grating method and the electronic speckle pattern
correlation interferometry are presented in more detail. Furthermore, a new numerical procedu­
re for determining the strain distribution is discussed. If an area-based algorithm is applied for
determining the displacement field through digital image processing combined with an affine
transformation between the two different intensity distributions associated with the correspon­
ding deformation states of the object, it can be shown that the parameters of this transformation
describe the searched strain values [7].
The advantages of optical field methods are demonstrated by three practical applications in
material testing. The first example is related to the strain analysis around the tip of a crack in
a fracture mechanical specimen by means of the object grating method. The next test describes
the applicability of the object grating method to the case of high temperatures and mechanical
loading of a notched specimen. Finally the electronic speckle pattern correlation interferometry
was applied for the in-plane displacement measurement of a vibrating turbine blade. In all
cases, the strain distributions were determined numerically from the measured local vectors or
displacements.
2. OPTICAL FIELD METHODS
In the following examples two optical field methods will be described in more detail.
2.1 Object Grating Method
The object grating method is based on incoherent optics. A grating of optical features consi­
sting of a geometrically invariant (deterministic) or geometrically variable pattern is applied to
the surface of the considered object, which is called grating, Fig. 1. By loading the object, the
deformation at its surface is equivalent to the change in position of the applied features. The
features are recorded using the photogrammetric method, Fig. 2, and then global coordinates
are determined. The primary information of the measurement is the field of all local vectors.
This method is applicable to inelastically deforming material. One suitable procedure for pro­
ducing high temperature resistant gratings is to spray an agglomeration of a volatile solvent
and titanium dioxide on the polished surface of the specimen through a mesh, as used in screen
printing. In this way, line densities up to 80 lines per millimeter can be achieved easily.
2
MM»«9«lltt«&ll&l*»«&ft8;
;g»«ftsiiaffss&«s**»*i;*j
(*SftBftH«««lf*«»SS8«ttSJ
8g««»«««tt«8«&«*fts*i8j
fftaBBS8B8S8BBB*SB««tfi
t«BB8B«BS8B*i»*BSffi;*8i
lBBBBBBB8B8B$£t!SiBS!»Bi
[B*BBttift*a»tt»iigiS8«ii£iii
BBBBB*BBB88IS8»tt*9»;g)
[|K»iBSii«»»«iiiii(ttH»«eig}i
tttBBBB*BBBB»BBBft»saiii
BBBBBBBBBBBBBBSSlBHSil
[SBBaBaB»BB8&8BBBB8ss!
IB3*B«@BBB8BiiB33BBgiti
BBBa8SBB*^BS8i8»ttlSBBi
(a) BBB«B««KB«a8B»8BttBBi (b)
iBBBBBKBBB»s8»BsiaB£$i
Fig. 1. Deterministic (a) and stochastic (b) grating, consisting
of features of an invariant or variable geometrical structure.
Z
Fig.2. Photogrammetric method.
2.2 Electronic Speckle Pattern Correlation Interferometry
If an optically rough surface is illuminated by a coherent laser, the scattered light rays interfere
and a granular intensity effect (speckle effect), caused by different amplitude and phase distri­
butions, comes into existence, Fig. 3. It can be observed for example in the image plane of a
recording camera. The phase distribution depends mainly upon the geometrical and imaging
conditions of the optical arrangement. As this distribution can not be determined using optical
sensors in interferometry, it is decoded by a reference beam, which interferes with the carrier
beam. The resulting intensity distribution is proportional to the phase and is usually recorded
using CCD cameras.
The correlation of two distributions related to two different deformation states of the consi­
dered object, leads to a fringe pattern, Fig. 4. This contains the information concerning the
deformation of the object surface. The intensity correlation can be achieved using a computer
and a video signal subtraction or addition.
3
Fig.3. Speckle effect.
Fig. 4. Speckle pattern correlation fringes.
In the case of 3D-measurement by speckle interferometry, two in-plane and one out-of-plane
measuring devices have to be distinguished.
For measuring one of the in-plane components of the displacement, the plane surface is illu­
minated by two object waves, OWi and OW2, at the same angle of incidence with respect to
the normal of the surface, Fig. 5. Contours of constant displacement in the X2-direction can be
observed.
The electronic speckle pattern correlation interferometry is suitable for displacement measure­
ment for the case of elastic or inelastic material behaviour.
2.3 Digital Image Processing
Processing of optical patterns always involves digital analysis of pattern images. For this pur­
pose, the image fields are divided into pixels and every image point is described by the grey
value of the related pixel.
In general, the patterns can be classified into features and interference fringes. Features come
into existence by applying grating methods and interference fringes come into existence by
applying interferometric field methods. The moire effect also consists of fringes, however these
fringes are mostly produced by superposition, for example of two line gratings, Fig. 6.
For the case of processing features, feature based and area based algorithms have to be discer­
ned. The structural operators [8], for example, belong to the feature based algorithms. Area
based algorithms are also suitable for the processing of the stochastic grey value distributions
of object textures, [9]. These procedures postulate that all image points can be identified by their
4
Fig.5. Diagrammetric arrangement of an in-plane displacement
sensitive speckle pattern correlation interferometer.
surroundings, called windows. For the case of displacement measurement, the aim of this corre­
lation consists of determining the position of the window, related to two different object states,
under the condition of minimum difference of the grey level distribution belonging to them,
Fig.7, eqn (1)
Fig.6. Moire effect by superposition of two line gratings.
I(xp,y /) = I(xp,y ) (1)
P P
In general, the relation between the coordinates P and P' can be described by the affine trans­
formation, eqns (2) and (3)
xp' = a\ + axp + a yp (2)
2 3
= a + ax + ayp (3)
ypt 4 5 P e
5
The parameters aj in eqns (2) and (3) must be determined under the condition that eqn (1) is
fulfilled.
Fig. 7. Intensity distribution related to the undeformed (a) and
deformed (b) state of an object with corresponding windows.
The best known methods for processing of interference or moire fringes are the phase-shifting
techniques [10]. In this case, the reference phase is shifted by changing the optical path of the
reference beam in three or more steps. For the image point under consideration, the same number
of intensity values can be obtained. Combining the different values leads to a relationship for
determining the relative phase difference. This difference is related directly to the deformation
associated with the corresponding object point.
3. STRAIN DETERMINATION
The different definitions of direct or "normal" strain are specified in terms of an initial fiber
length, I, and a final fiber length, If. For the determination of strain, the definition by Lagrange
is often used, eqn (4)
i
A
Eqn (4) is also the basis for the numerical computation of strain from displacement measured
using optical field methods. For example, the difference between the displacement components
of two neighbouring points in the same direction, related to their initial distance, describes in
a simplified manner the strain.
For the case of plane deformation with the displacement components u and v in the x- and y-
directions of a Cartesian reference coordinate system, the strain in x-direction can be described
by
(I)-
Furthermore, Fig. 7 shows that the relative displacement can be approximated by
6