U.S. patent application number 11/225141 was filed with the patent office on 2006-01-12 for method and apparatus for measuring strain using a luminescent photoelastic coating.
Invention is credited to James P. Hubner, Peter G. Ifju, David A. Jenkins, Shujun Jiang, Yao Liu, Kirk S. Schanze.
Application Number | 20060007424 11/225141 |
Document ID | / |
Family ID | 32045214 |
Filed Date | 2006-01-12 |
United States Patent
Application |
20060007424 |
Kind Code |
A1 |
Hubner; James P. ; et
al. |
January 12, 2006 |
Method and apparatus for measuring strain using a luminescent
photoelastic coating
Abstract
A method and apparatus for measuring strain on a surface of a
substrate utilizes a substrate surface coated with at least one
coating layer. The coating layer provides both luminescence and
photoelasticity. The coating layer is illuminated with excitation
light, wherein longer wavelength light is emitted having a
polarization dependent upon stress or strain in the coating. At
least one characteristic of the emitted light is measured, and
strain (if present) on the substrate is determined from the
measured characteristic.
Inventors: |
Hubner; James P.;
(Gainesville, FL) ; Ifju; Peter G.; (Newberry,
FL) ; Schanze; Kirk S.; (Gainesville, FL) ;
Jiang; Shujun; (Xian, CN) ; Liu; Yao;
(Gainesville, FL) ; Jenkins; David A.;
(Gainesville, FL) |
Correspondence
Address: |
AKERMAN SENTERFITT
P.O. BOX 3188
WEST PALM BEACH
FL
33402-3188
US
|
Family ID: |
32045214 |
Appl. No.: |
11/225141 |
Filed: |
September 13, 2005 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
10407602 |
Apr 4, 2003 |
6943869 |
|
|
11225141 |
Sep 13, 2005 |
|
|
|
60416105 |
Oct 4, 2002 |
|
|
|
Current U.S.
Class: |
356/34 |
Current CPC
Class: |
G01L 1/241 20130101;
G01B 11/18 20130101 |
Class at
Publication: |
356/034 |
International
Class: |
G01B 11/16 20060101
G01B011/16 |
Claims
1. A method for measuring strain, comprising the steps of:
providing a substrate surface coated with at least one coating
layer, said coating layer providing both luminescence and
photoelasticity; illuminating said coating layer with excitation
light, wherein longer wavelength light is emitted having a
polarization dependent upon stress or strain in said coating layer;
measuring at least one characteristic of said emitted light, and
determining strain on said substrate surface from said
characteristic.
2. The method of claim 1, said coating layer includes at least one
luminophore for providing said luminescence.
3. The method of claim 2, wherein said luminophore is a
polarization preserving material.
4. The method of claim 1, wherein said coating layer is
polarization generating, wherein said excitation light is
non-polarized light.
5. The method of claim 1, wherein said excitation light comprises
polarized light.
6. The method of claim 5, wherein said polarized light comprises
elliptically polarized light.
7. The method of claim 1, wherein said characteristic comprises at
least one selected from the group consisting of the maximum
principal strain, the minimum principal strain and the maximum
shear strain on said substrate surface.
8. The method of claim 1, wherein said characteristic comprises
directions of maximum principal strain and the minimum principal
strain on said substrate surface.
9. The method of claim 1, wherein said coating layer consists of
only a single layer.
10. The method of claim 1, wherein said at least one coating layer
comprises at least two layers, said at least two layers comprising
a luminescent layer disposed on said substrate surface and a
photoelastic layer disposed on said luminescent layer.
11. The method of claim 10, wherein said luminescent layer includes
a first luminophore and said photoelastic coating includes a second
luminophore, wherein said first and second luminophore provide
different emission wavelengths.
12. The method of claim 11, wherein an emission wavelength of said
first luminophore corresponds to an absorption spectrum of said
second luminophore.
13. The method of claim 1, further comprising the step of optical
filtering to selectively pass said higher wavelength light and
reject said excitation light.
14. The method of claim 1, further comprising the steps of
providing said strain on said substrate surface to an analytical
model and updating said analytical model based on differences
between said strain on said substrate surface and strain data
generated by said analytical model.
15. The method of claim 1, wherein said illuminating step comprises
a process comprising oblique incidence, said determining step
providing individual values for maximum principal strain and
minimum principal strain on said substrate exclusively using said
method.
16. The method of claim 1, further comprising the step of scanning
said excitation light across said substrate surface, wherein
regions of high strain are identified.
17. An apparatus for measuring strain, comprising: an excitation
light source for illuminating a surface of a substrate, said
substrate including a coating which provides both luminescence and
photoelasticity; a detector for measuring light emitted by said
substrate surface responsive to said excitation light, said emitted
light being at a longer wavelength and having a polarization
modified as compared to said excitation light based upon stress or
strain on said coating, and a computer for processing to determine
strain on said substrate surface from said emitted light.
18. The apparatus of claim 17, wherein said excitation light source
provides polarized light.
19. The apparatus of claim 17, wherein said computer provides at
least one selected from the group consisting of the maximum
principal strain, the minimum principal strain and the maximum
shear strain on said substrate surface.
20. The apparatus of claim 17, wherein said wherein said computer
provides directions of maximum principal strain and the minimum
principal strain on said substrate surface.
21-40. (canceled)
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application is a continuation of U.S. patent
application Ser. No. 10/407,602, now U.S. Pat. No. 6,943,869 issued
on Sep. 13, 2005, which was filed in the United States Patent and
Trademark Office on Apr. 4, 2003 and claims the benefit of U.S.
Provisional Application No. 60/416,105 entitled METHOD AND
APPARATUS FOR MEASURING STRAIN, filed on Oct. 4, 2002, the
entireties of which is incorporated herein by reference.
STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
[0002] Not applicable.
FIELD OF THE INVENTION
[0003] The invention relates to the field of strain measurement,
more particular, to strain sensitive coatings which provide both
photoelasticity and luminescence and methods for determining strain
on surfaces coated with the same.
BACKGROUND
[0004] In the field of structural analysis, the ability to
determine the stresses which a structural body experiences can
provide important feedback in the design and construction, as well
as the integrity during the service life of such structural bodies.
Typically, surface strain on the structural member can provide
information regarding the stresses that the body is experiencing.
This information can lead to the identification of stress
concentrations, over stressed areas, and general stress mapping for
comparison to and calibration of predictive and/or analytical
methods. Currently, a number of methods exist for measuring surface
strain, including point and full-field methods.
[0005] Point methods include electrical resistance strain gauge
methods and methods employing electro-optic sensors and optical
methods. These methods typically require affixing a plurality of
sensors at various locations on a structural body, or stepping the
sensor across the structural body. When the structure is stressed
each sensor or step indicates the surface strain at individual
discrete points. In order to determine the strain over an entire
body, numerous sensors must be located at critically stressed
points on the surface or numerous iterative steps are required when
movable sensors are used. Accordingly, these point methods can be
cumbersome, making it difficult to determine the stresses over an
entire surface of a structure.
[0006] A number of surface measurement techniques have been
developed to overcome the limitations of the point methods,
including brittle coatings, photoelastic coatings, Moire,
interferometric, thermoelastic and digital image correlation
methods. Each of these methods can be useful for certain
applications, but each have certain characteristics which limit
their usefulness. Brittle coatings typically provide good
qualitative information about the principal stress directions on
objects. However, conventional brittle coating techniques can only
test a part in one loading configuration and can only provide
limited quantitative information. Moreover, methods for automated
data collection using conventional coating techniques are not
presently available.
[0007] Photoelastic coatings provide the shear stress and principal
stress direction information on objects. Conventional photoelastic
coating techniques are typically cumbersome and time consuming to
apply to large bodies. Moire methods are typically limited to flat
objects and are not used on complex three-dimensional geometries.
Interferometric methods, such as holographic interferometry,
electric speckle pattern interferometry and shearography, require
sophisticated vibration isolation which greatly reduces their
applicability. Thermoelastic methods require cyclical loading of
the specimens to generate surface temperature gradients related to
the stress field. Digital image correlation methods can lack the
sensitivity required to test parts in the material linear range in
regions of high strain gradients.
[0008] U.S. Pat. No. 6,327,030 to Ifju et al. (Ifju '030), which
shares several common inventors with the present application,
discloses a strain-sensitive coating, a strain measurement system,
and a method to determine strains on substrate materials. Ifju '030
is herein incorporated by reference in its entirety. The disclosed
system can include a strain field mapping system which can be used
to create a full-field strain map of the mechanical strain on a
substrate material. The luminescent strain-sensitive coating is
preferably a polymer-based coating, and can incorporate one or more
luminescent compounds, such as luminescent dyes. Appropriate
illumination stimulates the dye, which in turn emits a longer
wavelength luminescence signal. Differences in excitation and
emission wavelengths permit optical filtering of these signals. One
or more characteristics of the luminescence emanating from the
coating can then vary in relation to the strain on the substrate
material. In one embodiment, the change in the morphology of cracks
in the coating can cause variation in one or more characteristics
of the luminescence emanating from the coating such that the strain
on the substrate material can be determined by measuring the
luminescence emanating from the coating.
[0009] U.S. Pat. No. 6,219,139 to Lesniak discloses a structural
specimen coated with or constructed of photoelastic material. When
illuminated with circularly polarized light the coated specimen
will upon stressing, reflect or transmit elliptically polarized
light, the direction of the axes of the ellipse and variation of
the elliptically light from illuminating circular light will
correspond to and indicate the direction and magnitude of the shear
stresses for each illuminated point on the specimen. A preferred
spray coating is formulated to produce a coating which is of a
thickness such that only a quarter wave of birefringence is
produced when the maximum stress is imposed on the coated
specimen.
[0010] Dyes are disclosed by Lesniak, but only non-luminescent
dyes. Specifically, non-luminescent dyes are disclosed exclusively
for providing light attenuation data sufficient to solve for the
thickness of the coating. Since the attenuation difference between
two wavelengths of light are not sufficient to solve for coating
thickness when all three main sources of amplitude variation are
considered, that being coating thickness, the surface of coating
reflection, and reflection from the surface of the specimen, three
dyes are required to provide information to solve the three
unknowns. By adding one or more non-luminescent dyes to a
photoelastic coating so that the attenuation of the three colors
red, green, blue are each substantially different and of a known
amount, a RGB camera can be used to provide data sufficient to
solve for the three main sources of amplitude variation.
SUMMARY OF THE INVENTION
[0011] A method for measuring strain on a surface of a substrate
includes the step of providing a substrate surface coated with at
least one coating layer. The coating layer provides both
luminescence and photoelasticity. The coating layer is illuminated
with excitation light, wherein red shifted (longer wavelength) is
emitted having a polarization which is dependent upon the stress or
strain in the coating based on stress or strain on the underlying
substrate surface. At least one characteristic of the emitted light
is measured, and strain (if present) on the substrate is determined
from the measured characteristic.
[0012] As used herein, the word "strain" can mean shear strain,
normal strain, maximum principal strain, maximum shear strain and
other strains. The photoelastic technique described herein is
sensitive to the maximum shear strain in the plane of propagation
of the light within the coating. If this plane is parallel to the
plane of the surface, then the maximum in-plane shear strain is
equal to the difference of the two in-plane principal strains
.epsilon..sub.1-.epsilon..sub.2. If the plane is not parallel,
which is the case for oblique excitation, the maximum in-plane
shear strain (that is the plane of propagation) is a function of
.epsilon..sub.1, .epsilon..sub.2, and .epsilon..sub.3.
[0013] The coating layer can include at least one luminophore for
providing luminescence. The luminophore is preferably a
polarization preserving material.
[0014] The coating layer can be polarization generating. In this
embodiment, the excitation light can be non-polarized light. In
another embodiment, the excitation light comprises polarized light.
The polarized light can be elliptically polarized light, including
circularly polarized light.
[0015] The measured characteristic can include the maximum
principal strain, the minimum principal strain and the maximum
shear strain on the substrate surface. The measured characteristic
can also include the direction of maximum principal strain and the
direction of minimum principal strain.
[0016] The coating layer can be a single layer or two or more
layers. The two layer embodiment can include a luminescent layer
disposed on the substrate surface and a photoelastic layer disposed
on the luminescent layer. In this embodiment, the luminescent layer
can include a first luminophore and the photoelastic coating can
include a second luminophore, wherein the first and second
luminophore provide different emission wavelengths. The emission
wavelength of the first luminophore can correspond to an absorption
spectrum of the second luminophore. The coating can include one or
more thixotropic additive.
[0017] The method can include the step of optical filtering to
selectively pass the emitted higher wavelength light and to reject
the reflected excitation light. An image can be formed based on
information provided by the emitted higher wavelength light. The
method can comprise the step of scanning the excitation light
across the substrate surface to identify regions of high stress or
strain.
[0018] The method can further comprise the step of providing the
experimentally determined strain on the substrate surface to an
analytical model, such as finite-element analysis (FEA) model. The
model is then updated and its accuracy improved based on
differences between the experimentally determined strain data and
strain data generated by the model.
[0019] The illuminating step can comprise the process of oblique
incidence. In this embodiment, the angle of propagation of light
with respect to the incident surface is oblique to the surface of
the coated substrate. By using multiple a plurality of excitation
angles the determining step can provide the individual strains on
the substrate (.epsilon..sub.1 and .epsilon..sub.2) using only the
claimed method.
[0020] An apparatus for measuring strain includes an excitation
light source for illuminating a surface of a substrate. The
substrate includes a surface coating which provides both
luminescence and photoelasticity. A detector is provided for
measuring light emitted by the substrate surface responsive to the
excitation light, the emitted light being at a longer wavelength
and having a polarization modified as compared to the excitation
light based upon stress or strain on the coating resulting from
stress on or strain on the underlying substrate surface. A computer
provides processing to determine strain in the substrate surface
from analyzing the emitted light.
[0021] The excitation light source preferably provides polarized
light. The apparatus can determine the maximum principal strain,
the minimum principal strain and the maximum shear strain on the
substrate surface, as well as the direction of maximum principal
strain and the direction of minimum principal strain on the
substrate surface. The apparatus can include a linear polarizer or
a combination of a quarter wave plate and linear polarizer disposed
between the object and the detector.
[0022] A coating for indicating strain on an underlying surface
comprises at least one coating layer. The coating layer provides
luminescence and photoelasticity, wherein the coating layer emits
longer wavelength light having an altered polarization responsive
to illumination with excitation light. The coating layer preferably
includes at least one polarization preserving luminophore. The
coating layer can be polarization generating. The coating layer can
comprise a bisphenol A diglycidyl ether or a bisphenol A
glycerolate diacrylate based polymer.
[0023] A coated structural member includes a substrate having a
coating layer on its surface. The coating layer indicates strain on
the substrate surface and includes provides both luminescence and
photoelasticity. The coating layer emits longer wavelength light
having a polarization dependent upon stress or strain on the
coating layer resulting from stress or strain on the underlying
substrate upon illumination with excitation light.
[0024] A method for real-time monitoring of strain in surfaces of
mechanical components, comprises the steps of providing a
mechanical element having at least one surface, the surface
including at least one coating layer for indicating the strain on
the surface. The coating layer provides both luminescence and
photoelasticity, wherein the coating layer emits longer wavelength
light having a polarization dependent upon stress or strain on the
coating resulting from stress or strain on the underlying substrate
responsive to illumination with excitation light. The mechanical
element is utilized in a stress inducing application, which can
include its intended use. Strains developed on the mechanical
element are monitored during the utilizing step, the monitoring
step comprising illuminating the coated surface with the excitation
light, measuring at least one characteristic of the emitted light,
and determining strain on the surface from the measured
characteristic.
BRIEF DESCRIPTION OF THE DRAWINGS
[0025] The invention is pointed out with particularity in the
appended claims. The above and further advantages of this invention
may be better understood by referring to the following description
taken in conjunction with the accompanying drawings, in which:
[0026] FIG. 1 illustrates a system for determining strain in
objects coated with a luminescent photoelastic coating, according
to an embodiment of the invention.
[0027] FIG. 2A is a plot of normalized intensity and phase of an
emitted signal from a substrate coated with a luminescent
photoelastic coating with a load applied as compared to no load
applied, the data compiled by rotating an analyzer in 45 degree
increments.
[0028] FIG. 2B is a vector form presentation of the data shown in
FIG. 2A.
[0029] FIG. 3 illustrates a coated substrate including a
luminescent photoelastic coating comprising a photoelastic layer
disposed on a luminescent layer, the luminescent layer disposed on
a substrate, according to another embodiment of the invention.
[0030] FIG. 4A illustrates the spectral response (emission
intensity versus wavelength) for a luminescent photoelastic coating
with the luminescent probe in the undercoat. The relative change in
intensity is not the same for all wavelengths due to a detected
luminescence interaction of the photoelastic overcoat.
[0031] FIG. 4B illustrates the trend between the ratio of the
emission intensity at 600 nm to 560 nm with respect to coating
thickness for the data shown in FIG. 4A. This information can be
used to calibrate for coating thickness.
[0032] FIG. 5A illustrates the spectral response (emission
intensity versus wavelength) for a luminescent photoelastic coating
with a luminescent probe in the undercoat and a different
luminescent probe in the photoelastic overcoat. The relative change
in intensity is not the same for all wavelengths due to the
interaction in luminescence between the two luminescent probes.
[0033] FIG. 5B illustrates the trend between the ratio of the
emission intensity at 700 nm to 600 nm with respect to coating
thickness for the data shown in FIG. 5A. This information can be
used to calibrate for coating thickness.
[0034] FIG. 6 is a flow chart which lists steps to measure the
full-field shear strain and strain orientation in substrate
surfaces.
[0035] FIG. 7 illustrates normalized absorption and emission
spectra as well as the excitation and emission dependent anisotropy
for a luminescent polyurethane undercoat with perylene dye. A
positive value for emission anisotropy indicates that the
luminescent undercoat retains the polarization state of the
excitation.
[0036] FIG. 8A is a shear strain image results for an aluminum
specimen loaded in a three-point bend configuration. The grayscale
images indicate low strains for black to dark gray and high strains
for light gray to white. This example demonstrates the measurement
system response to a shear strain field with spatial gradients.
[0037] FIG. 8B is a comparison of the measured shear strain using
the luminescent photoelastic coating (as shown along the white
horizontal line in FIG. 8A) and the calculated shear strain using
simple beam theory showing close agreement.
[0038] FIG. 8C is an image showing processed principal strain
orientation results for an aluminum specimen loaded in a
three-point bend configuration where 0 (medium gray) would indicate
compression along the horizontal axis and .pi. (white) or -.pi.
(black) would indicate tension along the horizontal axis.
[0039] FIG. 9A illustrates a cylinder having a thinned section to
provide a stress concentration area.
[0040] FIG. 9B is an image showing processed strain results for the
cylinder shown in FIG. 9A with an applied torsional load. Results
from two overlapping regions by two different CCD imagers are shown
where low strain is indicated by dark gray and high strain is
indicated by light gray.
[0041] FIG. 9C illustrates strain measurement with respect to
azimuthal position (.theta.-deg) and axial position (x-in) for two
torsional loads applied to the cylinder shown in FIG. 9A.
[0042] FIG. 10A illustrates a three-dimensional component coated
with a luminescent photoelastic coating. The component is attached
to a loading apparatus to apply a force and induce a strain.
[0043] FIG. 10B is an image showing processed strain results for
the three-dimensional component shown in FIG. 10A under an applied
force.
DETAILED DESCRIPTION OF THE INVENTION
[0044] A method and apparatus for measuring strain on a substrate
utilizes a substrate surface coated with at least one coating
layer. The coating layer provides both luminescence and
photoelasticity. The coating layer is illuminated with excitation
light, wherein longer wavelength light is emitted having a
polarization state dependent upon the stress or strain present on
the coating. Thus, the invention utilizes emitted light, not
reflected light from the coating. At least one characteristic of
the emitted light is measured, and strain (if present) on the
substrate is determined from the measured characteristic.
[0045] Photolelasticity relies on birefringence, and as used herein
generally refers to stress or load induced birefringence where the
material becomes birefringent, or experiences a change in
birefringence, under the influence of stress or loading. The stress
in the coating layer can be provided by a substrate (e.g. object)
under stress, whether natural or stress induced, beneath the
coating. Stress can originate from a variety of sources including
thermal and mechanical forces.
[0046] By calibrating the relationship between strain in the
substrate and the light emitted from the coating as compared to
light incident on the coating, the coating can be used to provide
quantitative measurements relating to the strain on the substrate,
including a full field map of the strain. The measured
characteristic can be based on one or more of the polarization,
intensity, spectral position, band shape, and emission decay time
of the light emanating from the coating. The invention allows
determination of the magnitude of the difference of or the
respective values of the two in-plane principal strains, the
maximum shear strain, as well as the principal strain direction on
the substrate.
[0047] The coating can be applied to mechanical or structural
components, such as steel, aluminum, polymeric or composite
components, and can be used for testing of prototypes to assist in
shortening product design time. In addition, the invention can be
applied to components during actual field use to provide strain
data. The technology is non-contact, non-destructive and permits
measurement of full-field strain, and is even applicable to objects
with substantially non-planar (3D) substrate surfaces.
[0048] As used herein, the term "light" refers to electromagnetic
radiation having wavelengths both within the visible spectrum and
outside the visible spectrum. For example, the invention can
generally be practiced with visible, infrared and/or ultraviolet
light provided appropriate luminophores and detectors are
provided.
[0049] Birefringent materials provide two different refractive
indices, generally referred to as the extraordinary index
.eta..sub.e and the ordinary index .eta..sub.o. Light traveling
through a birefringent material has a velocity, v, which is
dependent upon its polarization direction given by Eq. 1: v=c/.eta.
(1) where c is the speed of light in a vacuum and .quadrature. is
the refractive index parallel to that polarization direction. By
definition, .eta..sub.e is greater than .eta..sub.o for a positive
uniaxial birefringent material. For a positive uniaxial
birefringent material, the extraordinary axis is referred to as the
slow axis, while the lower refractive index ordinary axis is
referred to as the fast axis.
[0050] Light polarized parallel to the fast axis travels at a
higher velocity than light parallel to the orthogonal slow axis.
Thus, when plane (linearly) polarized light passes through a
birefringent material in which the fast axis is tilted with respect
to the axis of the polarized light, the polarized light is resolved
into two orthogonal components, a first component being along the
fast axis and a second component along the slow axis. This produces
two components of the linearly polarized light that travel in the
birefringent material and emerge from the birefringent material
separated in time. When the fast axis and slow axis components are
viewed through a polarizing filter, commonly referred to as an
analyzer, which is properly oriented with respect to the emerging
radiation, a component of each of the first and second components
will be able to pass through the analyzer. Each component is
resolved into a portion that is parallel to the analyzer, and the
emanating light can be imaged.
[0051] FIG. 1 illustrates a system 100 for measuring the surface
strain of a substrate 140, in accordance with one embodiment of the
invention. The substrate 140 can be opaque or transparent. Although
shown as a back-emission based system in FIG. 1, the invention can
also be practiced in a corresponding forward-emission
(transmission) mode for substantially transparent substrates
140.
[0052] Substrate 140 is coated with at least one coating layer 145
and together are referred to as the coated substrate 135. The
coating layer 145 provides both luminescence and photoelasticity.
In the embodiment shown in FIG. 1, a single coating layer 145
provides both photoelasticity and luminescence. However, the
coating layer 145 can include at least two layers, such as a
photoelastic layer disposed on a luminescent layer (see FIG. 3). In
addition, coating 145 can include more than one different
luminophore species.
[0053] The luminescence provided by coating layer 145 generates an
emitted signal responsive to excitation radiation, as opposed to
reflective signals produced by prior art strain sensitive coating
systems, such as U.S. Pat. No. 6,219,139 to Lesniak. The use of an
emitted signal in the present invention which is wavelength shifted
relative to the excitation light provides several advantages over
reflection based systems, such as optical filtering which permits
separation of the desired emitted signal from the reflected signal,
blockage of specular reflections, diffuse emission fields on
complex geometries, and measurements on surfaces not normal to the
propagation of the excitation or emission.
[0054] The light source 110 can be coherent or incoherent. For
example, a non-polarized excitation light source 110 can be
provided, such as one or more LEDs or laser diodes. The choice of
light source 110 depends on the absorption wavelength(s) of the
luminophore(s) in coating layer 145. For example, most commonly
used luminophores provide absorption in the visible spectrum. For
example, many laser dyes can be excited by blue light (400-500 nm).
To regain the ground state electronic energy level, the laser dye
emits fluorescent light at a longer wavelength, such as at about
650 nm.
[0055] A second luminophore which can fluoresce or absorb in a
region of the visible spectrum that is distinct from the first
luminophore can be used along with the first luminophore. This
arrangement is referred to herein as a multi-luminophore coating
which comprises at least two (2) different luminophore species. The
added dye(s) is (are) also preferably soluble in the non-polar
solvents used to deliver the coating. Suitable choices could
include, for example, aromatic hydrocarbons and transition metal
complexes. Examples of dyes which can be used with the invention
include, but are not limited to anthracene, pyrene, rubrene,
coumarins, and metal complexes such as rhenium and osmium. The
additional emission information can provide assistance in measuring
the strain accurately such as an emission signal that enables
correction for coating thickness.
[0056] The first polarizing filter 120 is shown converting
non-polarized light from light source 110 into plane (linearly)
polarized light. System 100 is configured so that light output by
polarizing filter 120 is oriented at an angle, such as 45 degrees,
relative to the fast axis of the quarter wave plate 130. Quarter
wave plate 130 is based on a birefringent or similar material. The
quarter wave plate 130 decomposes the incident plane polarized
light into two orthogonal components shown in reference 130
vibrating along both the fast and slow axes. The orthogonal
polarization components travel through the quarter wave plate 130
with different velocities (due to birefringence) and are phase
shifted relative to each other producing a modified polarization
state. The transmitted light leaving the quarter wave plate can be
considered to be two linearly polarized beams, each having an
amplitude and phase. When the quarter wave plate 130 is
wavelength-matched with the light source 110 and its extraordinary
or ordinary axis is offset 45 degrees with respect to the first
polarizer 120 alignment, a circularly polarized beam 132 emanates.
Circularly polarized beam 132 is received by the coated substrate
135.
[0057] Coating 145 includes at least one photoluminescent material
which provides absorption of the wavelength of light generated by
excitation light source 110 and in response emits a red-shifted
luminescent signal. The surface of coating 145 also produces a
reflected signal 148, the reflected signal having a wavelength
equal to the wavelength of the excitation radiation 132.
Luminescence in the coating 145 can be generated by providing at
least one luminophore, such as a luminescent dye, or other
photoluminescent material. The photoluminescent dye is preferably
selected from dyes which substantially retain the polarization of
the incident light upon emission.
[0058] As noted above coating 145 also provides photoelasticity.
Thus, light emitted by luminophores in coating 145, as well as the
light generated by the illumination sources, is modified during
travel though coating 145 due to the birefringence of coating 145
when strain is present in the coating 145, such as due to strain on
the underlying substrate 140. For example, the strain in the
coating 145 alters the state of polarization of emitted light, and
the intensity of light as it traverses coating 145 as a function of
the magnitude and direction of the strain field in coating 145.
Emitted light 147 is altered by the birefringent coating 145 and is
then processed by a second polarizing filter 150, which is commonly
referred to as an analyzer, where it can be recombined. An emission
filter 160, such as a bandpass or highpass emission filter (e.g.
650 nm bandpass filter), is preferably included to reject the
reflected excitation radiation 148 and pass the longer wavelength
luminescent signal 147. Although shown disposed post-analyzer 150,
emission filter 160 may be disposed pre-analyzer 150.
[0059] By recombining the two components using an optical
instrument such as analyzer 150, the corresponding intensity is
related to the principal strain magnitude, .quadrature., and
direction, 2.quadrature., as shown below in Eqs 2A-2D: I I o = 1 +
.delta. .times. .times. sin .function. ( 2 .times. .alpha. - 2
.times. .theta. ) ( 2 .times. A ) .delta. = .PHI. .times. .times.
sin .function. ( .DELTA. ) ( 2 .times. B ) .DELTA. = 2 .times. .pi.
.times. .times. Kh .lamda. * .times. .gamma. ( 2 .times. C )
.lamda. * = .lamda. ex .times. .lamda. em .lamda. ex + .lamda. em (
2 .times. D ) ##EQU1## where I is the intensity, I.sub.o is the
average intensity over one analyzer period (180 degrees), .alpha.
is the analyzer angle, .delta. is the amplitude response, .phi. is
the coating polarization efficiency, .DELTA. is the angular
retardation, K is the optical sensitivity of the photoelastic
coating, h is the photoelastic coating thickness for normal
excitation and emission, and .lamda.* is an effective wavelength
based on the filtered excitation and emission wavelengths.
[0060] In one embodiment of the invention, the photoelastic coating
thickness is limited such that the angular retardation for the
strain range of interest is less than .pi./2. For such an
embodiment, the system operates in the sub-fringe environment,
eliminating the need for phase-unwrapping and fringe counting
necessary for conventional photoelastic techniques which use thick
coatings or transparent photoelastic components.
[0061] The coating thickness can be designed for the expected
strain level. A thinner coating is generally used for higher
strains. Currently obtainable practical coating thickness are from
about 10 microns to 1000 microns (1 mm) using conventional spraying
techniques.
[0062] An imager 170, such as a CCD array, preferably including a
plurality of independent sensing pixels which each correspond to
different areas on substrate 140, receives the luminescent signal
147 (I in Eq. 2A) and forms an image. For example, a suitable
computer or similar device (not shown) for receiving and processing
image data compiled by imager 170 can be used to determine the
various strain information based on Eqs. 2A-2D and related
equations described hereafter. In the case where excitation
radiation 132 incident on coated object 135 is elliptically
polarized, the luminescent signal will be altered to another
elliptical state based on strain in the coating 145 resulting from
strain on the substrate 140.
[0063] In operation, a sequence of images can be acquired for a
given state of strain by rotating the analyzer 150 through a
plurality of analyzer angles, such as by -45, 0, 45 and 90 degrees
with measurements made at each rotation angle. Other rotation
angles can clearly be used. Alternatively, polarizing filter 120
and quarter wave plate 130 may be rotated together. Thus, through
appropriate rotations, a prescribed set artificial intensity
changes can be induced.
[0064] System 100 can be used with other strain measurement systems
which measure the addition of the first and second in-plane
principal strains (.epsilon..sub.1+.epsilon..sub.2), for example
Ifju '030, to decouple principal strains. By combining the addition
of the first and second in-plane principal strains
(.epsilon..sub.1+.epsilon..sub.2) to the subtraction of the first
and second in-plane strains (.epsilon..sub.1-.epsilon..sub.2)
determined as shown above, the respective in-plane principal
strains (.epsilon..sub.1, .epsilon..sub.2) can each be determined.
In one embodiment, in performing the strain decoupling process,
both tests can be run using the same testing component. Using the
same part can eliminate the manufacturing variability induced by
using different test parts. The strain decoupling capability of the
subject invention can allow further calibration of the analytical
and computation models, and allow the performance of
fatigue/durability service life predictions as crack initiation is
typically driven by the maximum principal strain,
.epsilon..sub.1.
[0065] In another embodiment of the invention, the principal
in-plane strains can be measured individually using only the
luminescent photoelastic coating technique described below by
employing a process of oblique excitation and/or emission. The
general theory of oblique incidence applied to conventional
photoelastic coatings to resolve the maximum principal strain is
disclosed in a book entitled "Photoelastic Coatings" by Zandman, et
al., Iowa State University Press, Ames, 1979. In this inventive
embodiment, the coated substrate 135 is illuminated by a light
source 110 with a linear polarizer 120 and quarter wave plate 130
such that the angle of propagation of light with respect to the
incident surface is not perpendicular but oblique to the surface of
the coated substrate 135. The range of generally effective oblique
angles is greater than about 30 degrees, preferably from about 45
to 60 degrees. The excitation wave traveling through the
photoelastic overcoat is retarded with respect to the plane normal
to the propagation direction. This plane orientation is a function
of the incidence angle relative to the substrate surface and the
refractive index of the coating.
[0066] The relative retardation includes the effect of the in-plane
principal strains for the coating as well as the out-of-plane
principal strain, .epsilon..sub.3, experienced by the coating such
that the maximum in-plane shear strain in the new plane, .gamma. in
Eqs. 2C, is a function of .epsilon..sub.1, .epsilon..sub.2, and
.epsilon..sub.3. Using a plurality of oblique excitation angles
with a light source or multiple light sources, sufficient
information can be attained to solve for both .epsilon..sub.1 and
.epsilon..sub.2. In general, the greater the angle of oblique
incidence, the more sensitive the measured response is to the
out-of-plane strain on the coating.
[0067] Additionally, the imager 170 can be positioned such that it
is obliquely positioned with respect to the surface of the
substrate 140, and hence, the retardation of the emission from the
luminescent dye will be a function of the out-of-plane principal
strain. Similarly, a plurality of imager 170 positions could be
used to acquire sufficient images to resolve the two in-plane
principal strains, independently.
[0068] When an image sequence is acquired for a specified loading
state, the state of strain information will be a combination of
both the residual stresses of the coating 145 and the true
"load-applied" stresses of the substrate 140. In some applications,
it is desirable for these residual stresses to be accounted for.
Calibration of emission from the coating 145 with no load applied
to substrate 140 is preferably done a priori to the measurement.
Thus, a sequence of images in the unloaded state can be taken and
utilized to account for residual stress state by analyzing the
image set as described above so as to yield a full-field
measurement. At each load condition, the measured light intensity
can be referred to as the combined signal. The residual signal,
acquired at no load, can be decoupled from the combined signal.
[0069] As shown in FIG. 2A a sinusoidal function can be fit through
multi-point analyzer data for a given load application. In FIG. 2A
a sinusoidal function is fit to the ratioed intensity response
I/I.sub.o for the cases of a load applied 210 and no load applied
220. The variables .delta..sub.c and .theta..sub.c are the
amplitude response and phase with load applied 210, while
.delta..sub.r and .theta..sub.r are the amplitude response and
phase with no load applied 220, respectively. Alternatively, as
illustrated in FIG. 2B, the loaded and unloaded intensity responses
can be equivalently presented in vector form.
[0070] Vector decoupling can be utilized to determine the
load-applied stresses. Decoupling equations which can be used are
provided below as Eqs. 3A-3D: .DELTA. = sin - 1 .function. (
.delta. .PHI. ) ( 3 .times. A ) .DELTA. t .times. e I2.theta. t =
.DELTA. c .times. e I2.theta. c - .DELTA. r .times. e I2.theta. r (
3 .times. B ) .DELTA. t = ( .DELTA. c .times. sin .function. ( 2
.times. .theta. c ) - .DELTA. r .times. sin .function. ( 2 .times.
.theta. r ) ) 2 + ( .DELTA. c .times. cos .function. ( 2 .times.
.theta. c ) - .DELTA. r .times. cos .function. ( 2 .times. .theta.
r ) ) 2 ( 3 .times. C ) 2 .times. .theta. t = tan .times. .times. 2
- 1 .times. ( .DELTA. c .times. sin .function. ( 2 .times. .theta.
c ) - .DELTA. r .times. sin .function. ( 2 .times. .theta. r )
.DELTA. c .times. cos .function. ( 2 .times. .theta. c ) - .DELTA.
r .times. cos .function. ( 2 .times. .theta. r ) ) ( 3 .times. D )
##EQU2## where the subscripts c, r, and t represent the
combination, residual, and true states of strain. Both .DELTA. and
2.theta. are functions of pixel location. Prior to decoupling but
after the residual and combination vector maps are determined, the
combination amplitude response and phase maps can be registered to
the residual amplitude response and phase maps to account for rigid
body movement and/or deformation. At each load or unloaded case,
the strains can be defined using the series of analyzer images at
that condition. The image at each analyzer angle for a set load
condition can be acquired under the same excitation field, such
that there is no movement amongst the image set.
[0071] The system and method can allow measurement of full-field
in-plane shear strain maps (diameter of the Mohr circle) on the
surface of substrates, such as test prototypes, having a variety of
sizes and shapes. The method can obtain strain data which would
have required thousands of strain gauges applied onto the object
for test can be used as a predictive tool to identify the design
critical regions, and can provide data for evaluating and
calibrating analytical models (such as finite-element analysis
(FEA)) for prototype design optimization.
[0072] Although the system 100 shown in FIG. 1 uses a polarizer 120
together with a quarter wave plate 130 to produce circularly
polarized light as generally described above, circularly or
ordinary elliptically polarized light can be produced in other
known fashions. Although a quarter wave plate 130 is described, and
may be convenient, other apparatus can be utilized to create
elliptically or circularly polarized light.
[0073] In addition, although the invention preferably utilizes
circularly polarized light incident on the coated substrate
surface, circularly polarized light is not required to practice the
invention. For example, non-circularly polarized elliptical
incident light can also be used. However, non-circularly
elliptically polarized incident light can add complexity to the
strain analysis as compared to when circularly polarized incident
light is used.
[0074] It is also possible to utilize linearly polarized incident
light with the invention. For example, in one embodiment, linearly
polarized light can also be used by directing the first light beam
having a linear polarization and a second light beam, from the same
or different light source, having a linear polarization which is
orthogonal or non-parallel to the polarization of the first light
beam on the coating. The response of the coating emission for this
embodiment would have an ellipticity relating to the state of
strain of the coating and the orientation of linear polarization of
the excitation. The ellipticity of the emission response could be
quantified with the appropriate optics and imager, for example
using analyzer 150, filter 160, and imager 170 shown in FIG. 1. An
appropriate process would be used to determine the strain magnitude
and orientation.
[0075] Although the invention is generally practiced by providing
polarized incident excitation light, it may also be possible to use
non-polarized incident excitation light. For example, this
embodiment could eliminate the need for both polarizer 120 and
quarter wave plate 130 from the system shown in FIG. 1. In one
embodiment, luminophore molecules can be selectively oriented in
the coating 145, such as in a luminescent polarizing undercoat
layer with a photoelastic overcoat layer. This can permit
luminophores to act as a back lighting source which can emit
polarized radiation responsive to non-polarized excitation
radiation. In this embodiment the photoelastic overcoat layer
modifies the polarization state of the polarized radiation emitted
by the luminophores, but not the excitation for the light sources,
based on strain in the substrate underlying the coating.
[0076] As noted above, the coating 145 can comprise two or more
discrete layers. FIG. 3 illustrates a coated substrate 300
including a luminescent photoelastic coating comprising a
photoelastic layer 330 disposed on a luminescent layer 320, the
luminescent layer 320 disposed on a substrate 310. The luminescent
layer 320 provides a back lighting source. The substrate 310 is
shown having a uniform stress 355 (.sigma.) therein.
[0077] Excitation light 340 is shown directed toward the surface of
coated substrate 300. Excitation light 340 includes two orthogonal
polarization components which interact with luminophores (not
shown) in luminescent coating 320 to produce an emission signal
350. Emission signal 350 includes two orthogonal polarization
components which are altered relative to their relationship in
excitation light 340 due to strain induced birefringence in the
photoelastic coating 330.
[0078] In a two-layer coating embodiment, the coating can comprise
a luminescent coating layer 320 containing a visible light
fluorescent compound in a polymer binder as a luminescent undercoat
applied over the substrate 310. The luminescent coating layer can
have a thickness of from about 20 to 80 .mu.m. Examples of visible
light luminescent dyes are cyanine, rhodamine, coumarin, stilbene,
perylene, rubrene, perylene diimide, phenylene ethynylene,
phenylene vinylene, and tetraphenylporphyrin. Examples of the
polymer binder include, but are not limited to, polyurethane,
polyacrylate, cellulose acetate and poly(dimethylsiloxane).
[0079] The substrate surface can be cleaned with acetone and hexane
and/or lightly sanded (preferably >240 grit). The substrate is
then preferably coated with a thin layer (about 10-20 .mu.m) of
black primer to provide a uniform, diffuse background. The black
primer layer is generally cured for about 1 hour at 25.degree. C.
The substrate is then coated with the layer of polymer including
one or more dissolved luminescent dye.
[0080] A sample formulation includes luminescent dye rhodamine B,
(0.02-1% by weight) a polyurethane binder cone part, along with one
part per volume ethanol. The mixture can be stirred for about 10
minutes then is ready for application. The rhodamine-polyruethane
coating can then be applied directly onto the black primer layer
using an airbrush. The polyurethane coating can be cured at ambient
temperature overnight.
[0081] Photoelastic coating 330 comprising an epoxy overcoat
including an epoxy binder having a thickness of about 50 to 500
.mu.m can be disposed over the luminescent layer 320. The epoxy
coating is typically applied in sequential coatings (ca. 50-75
.mu.m each) followed by 5 minutes exposure to a near-UV light
source to effect curing. The final coating can be cured by exposure
to near-UV light for about 1-2 hrs at about 25.degree. C.
[0082] One exemplary epoxy photoelastic overcoat 330 which can be
used with the invention is derived from the BGM monomer. The
structure for the BGM monomer is shown below as Structure 1.
##STR1##
[0083] The BGM monomer has the following specifications:
TABLE-US-00001 Formula weight: 312.37 g-mol.sup.-1, mp. -15.degree.
C. Density: 1.19 g-mL Viscosity (25.degree. C.) 2000-3000 cps.
[0084] The BGM based epoxy coating can be cured using acid or base
catalysts. For example, a photogenerated acid catalyst system based
on a triarylsulfonium hexafluorophosphate salt (TAS) 50% by weight
in a propylene carbonate solution can be used. Alternatively, the
coating can be cured by exposing the coating to ultraviolet (UV)
radiation. In one embodiment, the curing of the epoxy coating can
take place or near room temperature, for example 20-25.degree. C.
for 2 hrs and at 50-60% relative humidity (RH), hence not requiring
curing equipment such as an oven. The cured coating, an epoxy,
exhibits the phenomenon of birefringence.
[0085] Another exemplary photoelastic material which can be used
with the invention is formed from the curing of the bisphenol-A
glycerolate diacrylate monomer. The structure for this monomer is
shown in Structure 2. This monomer is quite viscous and can be
cured by typical acrylate initiators. This epoxy monomer is an
acrylate ester and generally shares properties with other acrylate
coatings. Use of this epoxy monomer can produce an easily applied
acrylate coating which has reduced flow after air brush deposition.
The structure for the bisphenol-A glycerolate diacrylate monomer is
shown below as Structure 2. ##STR2##
[0086] In one embodiment, a specific photoelastic coating
formulation can include bisphenol-A glycerolate diacrylate
(40-60%), chloroform (20-30%), toluene (10-20%) and benzoin ethyl
ether (1-8%), where all values are listed in % by weight. The epoxy
coating can be applied to the luminescent undercoat and cured by
exposure to UV light for about 1 hour at ambient temperature.
[0087] A variety of other optically transparent photoelastic
materials can be used with the invention, such as polycarbonate or
polymethylmethacrylate. Preferred materials are optically
transparent in the wavelength range of interest, provide high
polarization sensitivity, provide high optical sensitivity, have
low surface roughness, have low viscosity or alterable viscosity
with additives, have good adhesion qualities, and have reasonable
curing times and conditions.
[0088] A "thixotropic" additive can also be added to the coating. A
thixotropic additive can make the viscosity of the coating material
vary with shearing. For example, the viscosity can be low during
spraying due to the shear stress, but after spraying the viscosity
of the applied and uncured film can be higher. The higher viscosity
of the uncured film can reduce film flow.
[0089] Silica gel can be used as a thixotropic additive. A specific
photoelastic coating formulation can include BGM (40-60%), TAS
(1-10%), AT-924 (1-20%), fumed silica gel (0.1-5%), chloroform
(20-30%) and toluene (10-20%), where all values are listed in % by
weight. At-924 is a commercially available photocured epoxy
formulation from Adherent Technologies, Inc. Albuquerque, N. Mex.
This coating can apply well, resist flowing, and results in a film
that is level.
[0090] The amplitude response, .delta., shown in Eqs. 2B-2C, is
related to the thickness of the coating. For normal excitation and
emission, the effective length of retardation, h.sub.eff, through
the coating is twice the coating thickness, h. This increases for
off-axis (oblique) excitation, .beta..sub.ex, or emission,
.beta..sub.em, and is also related to the index of refraction of
the photoelastic coating such that h.sub.eff=f(.beta..sub.ex,
.beta..sub.em, .eta.)>2h. There are several non-invasive methods
that can be used with the invention to provide a full field
determination of the thickness of the photoelastic layer. For
example, ratiometric fluorescence imaging can be utilized to
extract full-field images that provide a quantitative measure of
the thickness of the coating. Such images can then be used in
matrix algorithms to correct the strain data for coating thickness
non-uniformity.
[0091] One example of a ratiometric method for thickness correction
utilizes the variation of the coating's fluorescence as a function
of coating thickness for a plurality of wavelengths, wherein the
coating exhibits a fluorescence intensity that varies independently
as a function of coating thickness at two or more different
fluorescence wavelengths. This concept is illustrated in FIGS. 4A
and 4B.
[0092] FIG. 4A shows the spectral response of a constant thickness
rhodamine-polyurethane undercoat and a varying thickness epoxy
overcoat (50-400 .mu.m) system when excited by non-polarized blue
illumination (450 nm). As the coating thickness increases, the
fluorescence intensity increases. The relative change in measured
emission intensity with respect to thickness is different at
different wavelengths. At 560 nm, there is little change with
respect to thickness. However, at 600 or 650 nm, there is a greater
change in the raw emission change with respect to thickness. The
relative change in spectral shape is due to the competing emissions
from the luminescent undercoat and the natural luminescences of the
overcoat.
[0093] FIG. 4B compares the ratio of emission intensities collected
with center-wavelength of 560 nm to 600 .mu.m, clearly
demonstrating a trend which can be exploited to measure coating
thickness. Such a method of thickness calibration can be referred
to as dual wavelength imaging.
[0094] A multi-luminophore coating comprising two (2) coating
layers each containing different luminophore species can provide
another method for non-contact thickness measurement. The two layer
multi-luminophore coating can include a luminescent undercoat
including a first luminophore and a photoelastic coating containing
a second (different) luminophore disposed on the undercoat layer to
correct for the effective thickness due to oblique angles and
spatial thickness variations. The emission wavelength of the first
luminophore can match the absorption spectrum of the second
luminophore. For example, non-polarized excitation light can excite
a luminescent undercoat which can in turn excite the luminescent
and photoelastic overcoat to create longer wavelength emission
proportional to the thickness of the photoelastic overcoat,
enabling a means to optically detect coating thickness.
[0095] In an exemplary multi-luminophore embodiment, the undercoat
can comprise of a commercially available binder including a blue
emitting luminophore. The overcoat can be epoxy based and include a
red-emitting luminophore. A black primer coat is preferably first
disposed on the object to be coated. Commercially available flat
black enamel spray paint can be applied to the substrate with a
thickness of about 10 to 20 .mu.m. A luminescent undercoat
comprising is then applied to black layer. Acetone can be used as a
solvent for the undercoat. An exemplary mixing ratio for a
luminescent undercoat is blue emitting dye (1%), cellulose acetate
56 m., and acetone 8 ml.
[0096] The epoxy overcoat, which provides photoelasticity and the
second red-emitting luminophore can be mixed. These components are
preferably stirred in a sealed flask for 8 hours. The two
luminophore coatings can provide a polarization efficiency, .phi.,
of about 20% and an optical sensitivity, K, of about 0.12 at T=295
K.
[0097] Results for a dual-luminophore system which provides
thickness correction as described above is shown in FIGS. 5A and
5B. FIG. 5A shows the spectral response of a constant thickness
undercoat and a varying thickness photoelastic overcoat (50-400
.mu.m) system when excited by non-polarized blue illumination (450
nm). As the coating thickness increases, the emission intensity
decreases between the wavelength range of 500-650 nm and increases
for wavelengths over 650 nm. For the range less than 650 nm, the Pe
emission is dominant. It in turn stimulates or pumps the red-dye's
emission, which is proportional the photoelastic overcoat
thickness, in the range of 700 nm. The opposite trends of the
emission intensity response at the different wavelengths can be
utilized to measure coating thickness.
[0098] FIG. 5B compares the ratio of emission intensities collected
at center-wavelength of 600 nm to 700 nm. The linear trend can be
used to measure coating thickness. Such a method of thickness
calibration can be referred to as multi-luminophore imaging.
[0099] In an alternate embodiment of the invention, the
experimentally collected strain data regarding the coating can be
provided to an analytical model (e.g. FEA) for model comparison,
evaluation, and update. Two methods are described below to connect
experimentally derived strain data to analytic models. In both
methods, local absolute differences in values between
experimentally determined and model derived data is used for model
modification and update. Generally, as the number of data points
used to modify the analytical model increases, the results
generated by the model become increasingly accurate.
[0100] The first method comprises discrete point analysis where
dots are dispersed randomly on the surface of the physical parts.
The coordinates of these discrete points are measured using a
coordinate measuring machine (CMM). Also, the measured strain
values from the coating are experimentally determined at these
points. Using the coordinates of these discrete points, nodes are
built in analytic models whose calculated strains are extracted for
direct comparison with the experimentally collected strain values
and modification of the analytical model.
[0101] The second method comprises the step of measuring full-field
strain data in the coating using a digitizing imager, such as
pixilated camera. Accordingly, each pixel has a set of coordinates
with reference to the camera coordinate system. The image
resectioning matrix method is used to map the 2D camera data to a
new set of 3D coordinates. The new 3D coordinate value for each
pixel is then compared to the coordinates of nodes included in the
analytical model. The pixels and nodes having similar coordinates
can be used for direct comparison, evaluation, and modification of
the analytical model. The second method is generally preferred as
it permits a larger number of data points to be transferred into
the analytical model as compared to the first method.
EXAMPLES
[0102] The present invention is further illustrated by the
following specific examples. The examples are provided for
illustration only and are not to be construed as limiting the scope
or content of the invention in any way.
Example 1
Method Flow Chart
[0103] FIG. 6 is a flow chart which lists steps generally used to
measure the full-field shear strain and strain orientation on
substrate surfaces. The flow chart includes descriptions for
specimen preparation, coating, imaging, and analysis, using a
typical luminescent photoelastic coating and a measurement system,
such as system 100 shown in FIG. 1.
[0104] In step 610, the specimen to be tested is prepared. The
specimen is sandblasted, degreased and cleaned. A black undercoat
is then applied. In step 620, the specimen is coated. This step can
involve applying a single luminescent-photoelastic coating and then
curing. Alternatively, step 620 can comprise applying a luminescent
undercoat, curing the undercoat, then applying a
luminescent-photoelastic overcoat then curing the overcoat. For
example, a RhoB//polyurethane coating (60-80 .mu.m) can be cured
under normal room conditions (25 C, 50% -60% RH). A BGM overcoat
(200-400 .mu.m) can then be applied and be UV cured at normal room
conditions.
[0105] In step 630, the specimen is imaged, dark field images are
acquired and flat-field images are obtained if necessary. Thickness
correction images are then obtained under no applied load, using
blue excitation, no polarizing optics and narrowband emission
filters. Residual stress images are then obtained under no applied
load, using blue excitation, polarizing optics, and an emission
filter. Finally, combination stress images are obtained under an
applied load, using blue excitation, polarizing optics and an
emission filter.
[0106] Step 640 comprises analysis of the images obtained. Analysis
can comprise the steps of applying dark field and flat field
(shifted images only) corrections, calculating a thickness
correction map, calculating a residual strain map, calculating a
combination strain map. If necessary, register (warp) images are
obtained with the applied load. The thickness correction map is
applied to the residual and combination maps. The true (load
induced) strain map is then calculated.
Example 2
Luminescent Undercoat Polarized Emission
[0107] A single layer of luminescent photoelastic coating
containing a blue emitting dye (0.1%), dissolved in an epoxy
monomer (1 g), along with a photocure agent (1%), a thixotropic
agent, chloroform (0.25 mL) and toluene (0.25 mL) solvents was
sprayed onto a metal surface and cured using UV irradiation (365
nm). A set of four fluorescence spectral scans were performed using
a spectrophotometer to determine the degree in which the
luminescent photoelastic coating retained the polarization state of
the excitation. The excitation (450 nm) was filtered with a linear
polarizer in the vertical and horizontal positions. The coating
emission was detected thorough a second polarizer in the vertical
and horizontal positions relative to each excitation polarizer (a
total of four cases: I.sub.VV, I.sub.VH, I.sub.HV, I.sub.HH). From
the data, the wavelength dependent anisotropy can be calculated as
shown in Eqs 4A and 4B: r = I VV - GI VH I VV + 2 .times. GI VH ( 4
.times. A ) G = I HV I HH ( 4 .times. B ) ##EQU3## A value of r=0
would indicate that the luminescence of the dye does not retain the
polarization of the excitation (the emission is anisotropic).
[0108] FIG. 7 shows results for the luminescent photoelastic
coating as well as spectral absorption and emission curves. The
ability of the coating to retain the polarization state of the
excitation is indicated by values of anisotropy. Anisotropy is
preferably positive and ranging between at least 0.2 and 0.4 at
wavelengths of 600 to 700 nm for a nominal emission wavelength of
650 nm, provided by the blue dye. Thus anisotropy provided by the
coating shown in FIG. 7 maintains the state of polarization of the
excitation. in addition, the spectral emission shows substantial
red-shifting from the excitation.
Example 3
Shear Strain Data on a Flat Specimen
[0109] An aluminum bar (2.0''.times.0.5'' in cross-section) was
degreased and prepared for undercoat and overcoat application of a
two-coating luminescent photoelastic coating as described above.
The specimen was placed in an apparatus such that it was simply
supported from underneath and loaded with a downward force from
above to create a three-point bend as shown in the image provided
by FIG. 8A. The coating was excited using a blue LED lamp
(.lamda.=465 nm) coupled with a linear polarizer and quarter wave
plate. A 600 nm (40 nm bandpass) interference filter was used on
the CCD imager to detect and record the emission intensity
analogous to system 100 shown in FIG. 1.
[0110] Images were acquired at four analyzer positions for each
applied static load condition. FIG. 8A shows processed strain
results for a given applied load. Regions of high shear are
indicated by white and light gray areas. Clearly present is the
spatially varying shear strain field. Beneath the contract point of
the downward applied force is the presense of contact stresses.
[0111] FIG. 8B is a comparison of the measured shear strain using
the luminescent photoelastic coating (as shown along the white
horizontal line in FIG. 8A) and the calculated shear strain using
simple beam theory showing close agreement. Measurements shown were
made for a horizontal line away from the area of contact
stresses.
[0112] FIG. 8C is processed principal strain orientation results
for an aluminum specimen loaded in a three-point bend configuration
where 0 (medium gray) would indicate compression along the
horizontal axis and .pi. (white) or -.pi. (black) would indicate
tension along the horizontal axis.
Example 4
Shear Strain Data on a Cylindrical Specimen
[0113] An aluminum cylinder (0.25'' thickness, 1.5'' OD) was
degreased and prepared for the undercoat and overcoat application
of a two-coating luminescent photoelastic coating as described
above. The specimen was machined with a necked-down region (1.375''
OD) along the center of the specimen to create a stress
concentration. It was placed in an apparatus such that a torsional
load was applied to one end of the cylinder and supported to
constrain rotation at the other end. The coating was excited using
a blue LED lamp (.lamda.=465 nm) coupled with a linear polarizer
and quarter wave plate. A 600 nm (40 nm bandpass) interference
filter was used on the CCD imager to detect and record the emission
intensity (similar to the system 100 shown in FIG. 1). Images were
acquired at four analyzer positions for each applied static load
condition. Two CCD imagers were used to acquire images of an
overlapping area.
[0114] FIG. 9A illustrates a cylinder having a thinned section to
provide a stress concentration area. FIG. 9B shows processed strain
results for the cylinder shown in FIG. 9A with an applied torsional
load taken along the azimuthal (-45 to 30 deg) and axial directions
of the cylinder for two torque values (150 and 300 ft-lbs). As to
be expected, the measured shear strain field is constant, except at
the neck down region were the shear strain is expected to increase.
Results from two overlapping regions by two different CCD imagers
are shown where low strain is indicated by dark gray and high
strain is indicated by light gray. FIG. 9C illustrates strain
measurement with respect to azimuthal position (.theta.-deg) and
axial position (x-in) for two torsional loads applied to the
cylinder.
Example 5
Shear Strain Data on Complex Geometry
[0115] A three-dimension component shown in FIG. 10 A, was
degreased and prepared for the undercoat and overcoat application
of a two-coating luminescent photoelastic coating. The specimen was
excited using a blue LED lamp (.lamda.=465 nm) coupled with a
linear polarizer and quarter wave plate. A 600 nm (40 nm bandpass)
interference filter was used on the CCD imager to detect and record
the emission intensity. Images were acquired at four analyzer
positions for each applied static load condition. FIG. 10B shows
processed strain results for a given applied load. The relative
pixel density is 10/cm. Regions of high shear are clearly visible
as indicated by the light areas. These are generally located near
the regions of high net tensile or compressive regions. Thickness
variation of the coating was corrected by the collecting the
coating emission from nonpolarized light using 10 nm bandpass
interference filters and using an a priori thickness calibration by
ratioing the two intensity images and comparing the response to
pointwise thickness measurements using a eddy-current thickness
probe.
[0116] While the preferred embodiments of the invention have been
illustrated and described, it will be clear that the invention is
not so limited. Numerous modifications, changes, variations,
substitutions and equivalents will occur to those skilled in the
art without departing from the spirit and scope of the present
invention as described in the claims.
* * * * *