The Video Extensometer

Introduction

When determining the mechanical properties of today's
varied range of materials and samples it is essential to
use accurate testing machines and instrumentation
which do not influence any results obtained. Stress
and strain are the two main parameters from which
most mechanical properties are derived from uniaxial
testing.

Stress is normally calculated by measuring the initial
specimen cross sectional area and relating this to the
measured load obtained from a calibrated load cell
whilst loading the specimen via grips or adapters.
Care must always be taken to ensure the fixation
method, alignment or specimen's shape does not
influence the results obtained.

For rigid materials, strain can be measured by using
conventional 'clip-on' mechanical extensometers or
foil strain gauges bonded to the specimen.
These devices however are not usually suitable when
testing delicate materials such as fibres, films, foams
or soft plastics as their weight and method of
attachment can influence both the results obtained
and the rupture point.

In many instances it is required to establish the
material properties over large strain ranges up to
rupture point. Most mechanical devices have limited
travel and require removing from the specimen before
fracture. Testing specimens under environmental
conditions i.e. elevated temperature restricts the use
of many mechanical units.

Various non-contacting systems such as mechanically
driven optical followers and laser extensometers have
been available over recent years, but these systems
do not posses the necessary resolution for accurate
determination of material properties at low strain levels,
also these devices operate on a single measuring line
between targets and hence require a second system
if transverse strain is to be simultaneously measured.

In order to overcome these limitations the ME-46 Full
Image Video Extensometer was developed by Messphysik
GmbH, an Austrian company, based on the concept
that "If it can be Seen, then it can be Measured."

 

ME-46 Video Extensometer fitted on
Universal Testing Machine


Extensometer Equipment
The system comprises a Video Camera capable of allowing
all in-built features such as gamma correction and filters
to be easily adjusted in order to obtain the best image.
The camera is fitted with a high precision CCD (Charge
Coupled Device) chip having its many photo-sensitive
elements arranged as an accurate linear grid.

Interchangeable, high precision lenses are fitted to the
camera, these can be fixed focal length units chosen to
suit specific applications or 'zoom' lens which cover a far
wider range of specimen sizes and deflections without
having to adjust the camera location.

The camera is connected to a 'frame-grabber' interface
card fitted in a PC running Windows 9x/2000/NT© based
software.

The PAL video signal is converted into 8 Bit digital format
whilst simultaneously generating a 640 x 480 pixel image
on the monitor.

The interface card is capable of resolving 256 shades of
grey for each pixel, resulting in a minimum resolution better
that 1:131,073 (i.e. 17bit) of the camera's field of view.
The actual working resolution is considerably enhanced by
software interpolation techniques whilst simultaneously
measuring the values of multiple scanned lines. The
camera's field of view depends upon the focal length of
the lens fitted and the distance away from the specimen.

System inaccuracies are confined to lens defects, the
linearity of the camera CCD chip and in maintaining a
constant distance between camera and specimen during
testing.

The video extensometer PC can be directly connected to
the testing machine via either serial or IEEE interfaces,
or alternatively by connecting an analogue load signal in
order that load and strain data values are saved
simultaneously. Facilities are provided to scale the
analogue input signal so the monitor also provides a real
time digital display of applied load.

Control of the extensometer can be automatic from the
testing machine computer or manual under Windows
9x/2000/NT© based software using the extensometer
PC keyboard and mouse, thus allowing the system to
operate as a stand-alone unit with most makes of testing
machines. When operating manually, the facilities are
provided to set both the start and end test load values
between which the test data is automatically saved in
a tab delimited ASCII format.


Principle of Operation
The camera is rigidly fixed to the machine frame or tripod
mounted and focused on contrasting targets marked on
the specimen. It is imperative the distance between
camera and specimen remains constant during testing as
any movement will alter the image size which in turn will
be erroneously interpreted by the software as a change
in specimen size. Provided the testing machine frame is
of adequate stiffness and the specimen grips accurately
aligned then this problem should not occur. The specimen
should also be illuminated at a constant level during testing
and this is best achieved using a separate external source.


Monitor display showing axial targets
and Grey Scale histogram


Targets are marked by producing straight lines on the
specimen using paint, a felt marker pen of a contrasting
colour or by attaching self adhesive targets, alternatively
the specimen outline can be used for specimen width
targets and also for axial deformations with ridged
specimens. It is essential the targets create a sharp
and as great a contrast difference as possible to ensure
correct automatic target recognition and tracking. The
target position is detected at the edge of a contrast
transition and is hence not affected by changes in target
width.

Conventional extensometers must be attached to specimens
at a known gauge length and the extension signal obtained
is converted into a strain value. This method can result
errors as normally it is not possible to verify the exact gauge
length once the extensometer is fitted on the specimen.
The video extensometer operates as a 'strain meter' by
directly calculating the measured extension as a percentage
of the original length and only requires exact knowledge
of the initial gauge length if actual extension values are
required. A calibration facility for axial and transverse axes
is provided to enable actual distances between targets to
be displayed and saved.

The camera viewed image is usually rotated through 90°
when displayed on the monitor as specimens are normally
tested in a vertical mode and this allows the screen
aspect ratio to be used to the best effect.

The viewed image is digitised and the resulting grey scale
values (0..255) for each pixel stored in a continuously
refreshed Frame Buffer memory location.
From this information it is possible to produce a grey scale
(contrast) diagram for every horizontal scan line or
vertically at any horizontal pixel position.

Due to the large amount of data obtained from each picture
scan, it is essential this is efficiently processed in order
that the extensometer has a dynamic capability suitable
for use with conventional static testing machines. Using
the Frame Buffer information the software automatically
detects the gauge marks and follows them during testing.

Facilities are provided in the configuration set-up menu
to select the target edge colour e.g. black to white, white
to black, black or white object outer edges and with these
combinations it is possible to accommodate most specimen
types.

Targets are automatically detected by differentiating the
grey scale data along the scan axis to ascertain the rate
of change in values and then selecting the maximum values
as the targets.

The software automatically detects the number of peaks
corresponding to the number of marks known to have
been fitted and selects these as the targets.

Reference marks are shown on the video monitor
indicating the selected targets found within manually
selected window and the operator is given the opportunity
to manually select alternatives in the event they have
been incorrectly detected.

In order to assist the operator in setting the camera
aperture to obtain the correct target identification,
provision is made to display the grey scale level
of the datum scan line as a histogram on the video
monitor.

Having established the differential signature pattern,
memory zones are defined around each target values
in the Frame Buffer so that only data within these
locations is further processed.

The target datum points for a scan line are selected
at the mean grey scale level of the maximum and
minimum values of the selected gradients as shown
below.

In order to enhance the accuracy and resolution of
measurement, the mean grey scale point is carried
out over an operator selected band of scan lines
either side of the datum and the mean length computed
from these values.

The crossover points are dynamically adapted throughout
the testing processso that targets are not lost if the
target edge contrast reduces due to the specimen
stretching. This method of measurement makes it
possible to obtain resolutions far better than the
theoretical minimum value of 1:131,073 of the field of
view.

For the study of axial strain distribution, up to 10 targets
can be fitted, and the measuring process is automatically
carried for each zone with the individual lengths being
saved.

Using the same technique as previously described, the
image can be scanned at right angles to the specimen
axis and the data obtained used to measure width.
The specimen width outline can be used as the width
target or alternatively marks can be placed on the
surface. The width can be measured at multiple
equispaced distances between the outer axial targets
with the individual values being saved, thus allowing
the study of transverse strain distribution. A 'Neck-tracking'
facility can be selected, whereby the specimen width
is continuously scanned and the minimum value detected
and measured through to rupture.

All length and width values are transmitted to the machine
controlling PC or stored to disk in tab?delimited format
together with details of the elapsed test time and load
values, making it easy for post test processing using
Messphysik software or conventional spreadsheets.

After the above processes have taken place the frame
information is refreshed and the exercise repeated. The
time between refresh cycles depends not only upon the
speed of the computer and frame grabber card but also
upon the amount of data being processed. In practice
using the standard Messphysik extensometer to just
measure axial deformation over a 10 scan line band the
refresh frequency is in the order of 100Hz, however if
width is also simultaneously measured and averaged at
3 positions the refresh frequency falls to approximately
30Hz.


Method of Operation
Windows© based software is used to control all
extensometer operations and procedures.

Firstly, the specimen is marked and then placed in the
testing machine. The camera which is ideally attached
to the machine frame is positioned and adjusted so that
the video image clearly shows the targets and has a
sufficient field of view to display them throughout the
test.

The position of the axial datum is set using the mouse
and the histogram displays the grey scale values along
this scan line scan line, allowing the lens iris and external
lighting to be adjusted for the correct contrast condition.

The extensometer can be calibrated by placing a known
distance piece adjacent to the specimen face or measuring
the specimen width/diameter with a micrometer. These
targets points are then selected using the computer mouse
and their known values entered, this results in a recalculation
and display of the size of the monitor's field of view. Provided
the camera is not moved or the lens adjusted between tests,
subsequent specimens will be automatically measured without
recalibration.

If transverse strain is required for the calculation of true
stress, n value, Poisson's ratio etc. the operator selects
the number of zones between the axial targets over which
it is to be measured and whether the mean or minimum value
is to be recorded. When the minimum value is selected the
system automatically searches and tracks the neck value.

The specimen image is continuously displayed on the monitor
during a test together with lines indicating the axial and
transverse points being measured.

Several calculated values can also be displayed in real time
having been selected by the operator from a configuration
menu. These can be axial and transverse values, Poisson's
ratio, r-value, operating frequency etc.

All extensometer and target settings are held in a named
parameter file which can be easily recalled when testing
similar specimens at a future date.

When starting a test, the software automatically identifies
the targets and indicates these on the monitor. The operator
confirms the targets have been correctly identified, or
alternatively points to the correct ones using the mouse.

All targets are automatically tracked during testing with
the measured values being transmitted to the testing
machine control computer, or when configured as a stand
alone unit, the data is saved to disk in a tab delimited
format with a unique file name. The following data is saved:
elapsed time, load and crosshead/piston displacement
(obtained from an A/D card connected to testing machine),
axial extension and transverse extension.

The stored data can be evaluated using Messphysik or
conventional spreadsheet and statistical software packages.

Application Examples
The following stress/strain diagrams illustrate the suitability
of the ME46 extensometer for testing elastic and high
strength metals.

Some plastics standards specify the material modulus must
be calculated between low fixed strain values (i.e. 0.05 to
0.25%) and that extensometer resolution must be better
than 1mm irrespective of gauge length or measuring stroke.
It is also usually necessary to test and record strain to
rupture, which for plastics can be in the order of 800%,
consequently on a specimen with a 50mm gauge length
the extensometer would require a resolution better than
1/400,000,000 to meet the standard. Although the
extensometer provides adequate data for accurate modulus
calculation it is recommended that comparative checks
for modulus values are carried out with the field of view
set to cover the prescribed strain range to comply with
the standard.


Optional Configurations
The system can be fitted with a Digital/Analogue interface
to provide simultaneous analogue outputs of both axial and
transverse strain values and these signals can be used for
closed loop control of servo machines.



Video extensometer set up for 'dot measuring'
for strain distribution measurement over
a flat surface.





The video extensometer can be configured to operate
with optional 'Dot Measuring' software to enable strain
distribution of a flat surface to be measured. In this
mode the specimen can be marked with up to 100 dot
targets and during testing the centre of area of each
target is calculated and the XY coordinates relative to
the displayed field of view are saved together with
elapsed time and load data.

Accuracy
Current standards for classifying most extensometers used
in metals testing (EN 100024, ASTM E83 etc.) have been
formulated upon conventional devices which have fixed
gauge lengths and as a result all values for accuracy,
resolution have been based around percentages of the
gauge length or operating range.

The ME46 Videoextensometer has neither a fixed gauge
length nor operating range as it depends upon where
specimen targets are located and the selected field of view.

 Axial Field of View

 Minimum Resolution

 50 mm

 < 0.4 µm

 250 mm

 < 2 µm

 500 mm

 < 4 µm

 1000 mm

 < 8 µm


The minimum extension resolution which can be obtained
due to number of pixels and shades of grey based upon a
single scan line is related to the field of view and examples
are indicated in the adjacent table. These values are greatly
improved when utilising several scan lines and interpolation
techniques.

The software provides a real time digital display of the
extension value to a resolution of 0.0001mm and it is this
value which must be recorded when carrying out comparisons
with established standards.

System linearity and errors can be checked using a
calibration rig with the camera set for a standardised field
of view based upon approximately 150% of the initial target
length plus extension range to be calibrated. As can be
seen from the graph below, the results obtained are excellent
when compared with national standards.

Post Test Evaluation Software
A complete range software packages based upon National
Testing Standards (EN 10002-2, ISO 527, DIN 53815 etc.)
are available.

These packages can be run concurrently using a second
PC allowing real time graphing whilst the test is in progress
with the operator able to pre select the displayed axis and
also allow automatic calculation of an extensive range of
results in addition to producing individually configured test
reports.

If the extensometer is used as a stand-alone extensometer
then the data can be processed in a similar manner using
the same software after a test has been completed.

Because of its high resolution and simple operation the
ME 46 Extensometer can be used over a very wide range
of testing with the following data processing software
options based upon National Test Standards.


Metals Testing (Tension & Compression) allowing
Modulus, Proof Stress, Max. Load, Poisson's Ratio, r value,
strain at break, and many more values to be automatically
calculated.

Plastics Tensile Testing allowing Modulus, Proof Stress,
Max. Load, Poisson's Ratio, energy under curve, strain at
break, and many more values to be automatically calculated.

Textile Testing allowing results to be calculated in
accordance with DIN 53815

3 & 4 Point Bending where the strain, stress values
and radius of curvature can be calculated from the
deflections obtained.

Software packages are continuously being developed
to meet additional testing requirements.


Advantages of ME46 Videoextensometer

Non-contacting and therefore does not influence
rupture point.

Directly measures strain.

Can measure strain up to failure.

Arbitrary choice of gauge length and operating
range. (Specimens can be any length from millimetres
to metres).

Can be used in hostile environments and with
temperature cabinets, provided a clear window is available.
Refractive distortion can be corrected by calibrating at
multiple points.

Allows simultaneous measurement of transverse strain
and vertical anisotropy.

Allows automatic detection and measurement of
specimen necking.

Allows testing of problematic materials such as cables,
thin foils, fibres, foams, composites etc.

Entire testing procedure can be observed on a monitor
together with real time calculated values.

No moving parts to wear out.

Option to store and print image at moment of rupture.

Able to operate as a stand-alone unit in conjunction
with any make of testing machine provided an analogue
output proportional to load is available.

Option for axial strain distribution, allows the extension
to be simultaneously measured between a number of
targets (max. 10).

Option for dot-measurement, to evaluate the X & Y
ordinates of up to 100 dot markers during a test, enabling
strain distribution of flat surfaces to be studied.

Although the system has been described for use with
conventional tensile, compressive or bending tests, it can
be used to accurately determine the movement of any object.
i.e. movement of a bridge under load relative to ground,
sideways movement of a wall relative a fixed reference line,
extension of nerves, muscles or bones during surgical
operations etc.


  
 
 
 

For details of Sensor Suppliers, click here...


Home - Website - Search - Suppliers - Links - New Products - Catalogues - Magazines
Problem Page - Applications - How they work - Tech Tips - Training - Events - Jobs