Note: Descriptions are shown in the official language in which they were submitted.
3l~LG~5~1
Description
Technical Field
.. . ..
This invention relates to an ultrasonic flaw
detecting deviGe for determinin~ abnormalities
within a pipe or similar structure. Nondestructive
ultrasonic testing of objects such as pipes is
well known. With such testing an ultrasonic beam
oE energy is sent into an ohject by a transducer
and detection o~ reflections or echoes off internal
structure within the object permits determination
of characteristics of that internal structure.
More particularlyr if a piezoelectric crystal
; is pulsed with an electrical energy signal, that
electric pulse causes an ultrasonic signal to be
e~itted. It is also known that if the ultrasonic
sic3nal reflects off an object within its path and
returns to the piezoelectric crystal, that crystal
responds by producing an electric si~nal. It is
possible, therefore, to send ultrasonic signals
into a test object whose internal structure is
of interest and to develop information from cry~tal
output signals which result from the reflections
ofE the internal structure of the object.
BacXground Art
Ultrasonic techniques for examining the in-
ternal structure of a pipe or other cylindrical
objects are known. Wben the object of interest
is a pipe and the area of interest within the pipe
i5 a longitudinally extending weld area, it has
been more effective in studying the internal struc-
ture of the pipe weld area to send the ultrasonic
signals into the pipe transverse to the weld area
rather than along its length.
One example of a proposed mechanism for de-
termining the internal structure o~ a pipe or
cylindrical object is the patent No. 3,924,453
to Clark et al. The angles of incidence of the
ultrasonic beams into the pipe in the Clark device
: ~ ,
:
5~
are une~ual. If this technique of ultrasonic flaw
detection is utilized, the correlatlon of flaw
severity to reflected signal strength becomes
diEficult due to the non-monotonic variations in
the strength o the received echo as a function
of distance.
Patent No. 3,693,~15 to Whittington proposes
another orm of ultrasonic Elaw detection for use
with a cylindrical obiect. The Whittington patent
teaches the use of an array of u:ltrasonic transducers
which must be positioned about the pipe in a circular
or cylindrical arrangement. Means must be provided
for sequentially pulsing the transducers which
make up this array in order that ùltrasonic beams
of the proper phase arrive at a given point in
the pipe structure and then enter that struature
to be reflected by flaws or irregularities within
the pipe.
Patent No. 3,916,675 to Perdljon propos~s
still another method for ultrasonically testing
the internal structure of a cylindrical device.
The Perdijon device, utilizes a complex deElector
means which receives parallel ultrasonic beams
from an ultrasound transducer and reflects these
beams into the pipe structure. As can be seen
by the complexity of the Perdijon proposal much
care must be taken in designing the deflector means
in order that the angle of incidence of the deflected
beam strikes the pipe in the proper angle. The
complex design reguired for the proper reflector
shape must be repeated for pipes of different sizes
and shapes and the complexity ;s significantly
increased if any variation in detection capability
is to be achieved.
Disclosure of Invention
The present invention comprises an ultrasonic
flaw detector for scannina an annular section of
an object whose internal structure is of interes~.
S~
The use of a novel ultrason1c transducer design
causes ultrasonic energy to enter the section and
completely scan that section for flaws and irregu-
larities. When a flaw exists within the section
ultrasound ener~y is reflected from that flaw and
returns along its incident path to the transducer
where the reflected ultrasound energy is converted
to electrical energy. Increased scanning reliability
is achieved by causing the ultrasound energy to
initiall~ impinge the annular section of interest
along a waveform whose angles of incidence with
- that section are substantially equal. Equal angles
of beam incidence result in equal angles of bea~
refraction when the ultrasound enters the object
oE interest. If the angles of refraction are equal
and the transducer dimensions properl~ determined,
the ultrasonic beam will completely scan the in-
terior o the annular section and no flaws wi:ll
~e missed.
Equal angles of incidence can be achieved
by the utilization of an ultrasonic transducer
device whose shape coincides or substantially
coincides with that of an involute. An involute
is a curve traced by the end of a taut strin~ which
when wound or unwound upon a fixed cuxve creates
a certain configuration. In theory, the ultrasonic
transducer surface could comprise a series of
planar involutes of infinitesimal width which when
summed together would create a surface involute.
Engineering considerations, however, have dictated
that instead of an actual involute being utilized,
a portion of a circle is used in creating the
transducer.
In designing the actual transducer surface
three points are chosen on a theoretical involute
and a circle is traced through those three points.
The transducer which is actually constructed comprises
a section of a cylinder-the radius of which coincides
with the radius of the circle which appraxima~es
a real involute.
5 ~1
It can be shown ~rom geometrical considerations
that if an ultrasonic beam's angles of incidence
are to impinge upon an angular cross-sectlon such
as that of a pipe the involute must be generated
from a generating curve or evolute which coincides
with a circle. It can also be shown that the center
of the circular evolu~e which generates the involute
must coincide with the axis of the pipe to be scanned.
With the present invention the exact shape
and size of the evolute, i.e., the generating curve
is determined based on considerations of the pipe
dimension to be studied. In determining the evolute's
radius it is of primary importance to determine
at what angle oE incidence the ultrasonic beam
lS is to be projected into the pipe. The two constraints
oE pipe radius and beam angle o~ incidence deine
a constant radius evolute Eor generating a theo~etical
transducer surface oE the proper involute shape.
From the theoretical transducer surface a quasi-
involute which coincides with a portion of a circleis chosen and frorn that circle a cylindrical trans-
ducer surface is constructed.
A transducer surface in this configuration
causes ultrasonic waves to impinge upon a pipe's
surface with equal or substantially equal angles
of incidence. When refracted within the pipe
structure these waves tend to travel in paths which
neither concentrate nor diffuse beam energy.
Equal angles of incidence also provide more
uniform energy transferral to the pipe. It is
known that varying the angle of incidence of the
ultrasonic energy varies the amount of energy trans-
mitted to the pipe. If the waveform incident- on
the pipe varies along its length in angle of inci-
dence, the transmitted waveform therefore var;esin the amount of energy transmitted through the
pipe. This non-uniform energy distribution will
,
S ~L
provide non-uniform echo signals from pipe flaws
which produce peaks and valleys in signal amplitude.
The constant energy content of rays coming from
an involute transducer provides a signal response
which minimizes the peaks and valleys.
The transducer surface, once constructed,
must be maintained in a proper geometric relatiorlship
with the pipe structure in order that the proper
angles of incidence are maintained. To achieve
this proper correlation between the transducer
surface and the pipe surface a wedge structure
which transmits ultrasonic beams is placed between
the pipe and the transducer. To achieve the proper
correlation the wedge structure is designed to
contain two important surfaces. One sur~ace coacts
with the pipe structure and the other sur~ace coacts
with the ultrasonic transducer surEace. The same
quasi-involute used to construct the tran~ducer
can be used to create one wedge surEace. For the
other surface it is only necessary that the outside
diameter of the pipe be known in order that the
second surface of the wedge structure coi~cides
with that diameter.
In addition to the constraints placed upon
the shape of the transducer surface, a requirement
exists with regard to a dimension of the transducer
structure Xn the embodiment of the transducer
surface which comprises a segment of a cylinder,
this dimension refers to the circumferential extent
or size of the segment of the cylinder or the number
of pi radians that segment intersec-ts.
To understand this constraint one must examine
the propa~ation of ultrasonic beams within -tne
pipe structure. Once the ultrasonic beam enters
the pipe it is re~racted at the outer surface tthe
angle of refraction of course depends upon the
refractive index of the wedge and pipe material)
and then travels to the inside surface of the pipe.
3l~ Si~
The angle oE incidence on this surface is such
that there is substantially total internal re-flection.
The ultrasonic beam of substantially undiminished
energy is then transmitted again to the outside
sur~ace and again substantially totally internally
reflectedO Internal reflection continues for a
number of reflections until the beam is gradually
attenuated.
It can be seen that to be certain a flaw
within the pipe struc-ture is detected, the multiple
reflections must cause the ultrasonic beam to sweep
the entire section oE interest within the pipe.
If the quasi-involute surface is of an insufficient
circumferential extentl it is conceivable that
there will exist se~ments within the pipe structure
which the ultrasonic beam never intersects. For
this reason there is a minimum transducer dimsnsion
which insures the pipe to be examined is adequately
swept by the reflected beams within that pipe
structure. It also should be noted that any climension
beyond this minimum is excess and serves no useful
function. Since the cos~ of fabricating the ultra-
sonic transducer surEace increases as the size
of the transducer increases, the minimum dimension
should not be greatly exceeded.
If areas other than that in direct contact
with the ul~rasonic transducer are of interest
the plexiglass wedge and transducer surface can
be moved circumferentially about the pipe structure
in order that other areas of the pipe structure
are tested. Thus it is seen that the present inven-
tion does not require the design of an ultrasonic
transducer SUL Eace which completely surrounds the
pipe structure nor is a complex deflecting device
necessary.
Accordingly one ob~ect of the present invention
is to provide a flaw scanning device and ~ethod
which completely scans an area of interest in an
~:~6~
annular object.- These and other objects, features
and advantages of this invention hecome more apparent
from the detailed description that follows when
considered in connection with the accompanying
drawings.
Brief Description oE the Drawin~s
Figure 1 i5 a perspective view oE an ultrasonic
~law detector scanning a pipe section for flaw~;
Fic3ure 2 is a cross-sectional view of an
ultrasonic flaw detector in contact with a section
of a pipe;
Figure 3 is a diagrammatic sectional view
showing two flaw detectors in difEerent positions
about the circum~erence of a pipe to be inspected;
Figure ~ shows a 1aw detector whose circum-
ferential length i5 adequate to scan ~he pipe shown
in that figure;
Figure S shows an end elevational view with
parts broken away and removed, oE a flaw det~ctor
assembly that maintains its detector in close
relation to a pipe to be scanned;
Figure 6 shows a lower plan view from the
pipe of the flaw detector assembly of Fig. 5;
Figure 7 shows a schematic electrical diagram
of the electronics for controlling the sending
and receiving of ultrasonic signals.
Best Mode for Carrying Out the Invention
Figure 1 shows the ultrasonic flaw detector
assembly of the present invention positioned along
a pipe 10. The pipe 10 has a center axis 11 and
a longitudinal weld 14. It can be seen that the
weld 14 forms a substantially straight line along
a length of the pipe. The pipe has inside and
outside surfaces lS, 17 whose spacing determine
the thickness of the pipe to be scanned.
A detector mounting assembly or structure 16
is shown in Figure 1 resting atop the pipe 10.
This mounting structure is drawn along the pipe
length in a direction indicated by -the arrow A
by ally suitable means, not shown. Alternatively
the pipe 10 is moved relatlve to the detector
mounttng structure 16 since relative motion is
to be achieved.
The detector mounting structure 16 is an
outriyger arrangement which includes a cross-member
- 26 which carries a pair of end pieces 32. Each
end piece 32 includes a pair of outrigger arms
28, only three of which are visible in Fi~ure 1.
Each outrigger arm 28J carries an outrigger roller 30.
lS A rotatable mounting ro~ 34 extends between the
two end pieces 32. r~wo detector mount ar~s 18
are mounted to ~his rod 34.
~ s relative motion oE the pipe and detector
occurs the outrigger crosspiece 26 maintains ~he
four outrigger arms 28 in rigid relatlonship, one
with the other. The detector mounting arms 18
on the contrar~ are rotatably mounted with the
rod 34 and are arranged to rotate as the detector
mounting structure encounters small variations
in the pipe's surface. ~s the detector mounting
structure 16 is drawn along the pipe in the direction
indicated by arrow A~ the outrigger rollers 30
are maintained in symmetric relationship to each
other relative to the weld area 14. Thus, the
assembly 16 is intended to be opera~ed such that
a plane located by the center axis 11 of the pipe
and the weld area will bisect the rod 34.
Two ultrasonic transducer de~ecting de~ices
20 ar~e mounted to the detector mount arms 18~
As the ultrasonic detecting devices ~0-move alony
- the pipe 10 signals are sent from an electronic
signal module 24 by means of two diagrammatically
illustrated electrical interconnects 22 and 23.
s~
An electronic signal processor contained within
the module sends sign~ls to the ultrasorlic transducer
detecting devices 20 causing them to send ultrasonic
; sound waves ;nto the pipe to scan for flaws and
; 5 defects as the detecting mounting structure moves
along the pipe. The electronic signal module 24
also interprets reflected siqnals from wi-thin the
-pipe structure when those signals rebound or echo
of variations in density within the pipe structure.
As will be seen with reference to Figure 7, suitable
means are connected to the electronic signal module
to suitably mark the pipe at locations in which
defects or flaws are found.
To ensure ultrasonic coupling between the
transducer devices 20 an~ the pipe 10 a liquid
coupling medium is provided by two hoses 21. These
hoses typically provide a layer of water which
is forced between the devices 20 and the pipe 10
to couple them for ultrasound transmittal.
By using two opposi~ely positioned transducers,
each transducer can be tested by sending it a pulse
from the other transducer. In this way proper
coupling between pipe and transducer is assured.
Also certain flaws may be difficult to detect for
one transducer but due to the difference in orientation
to the second transducer they will appear on that
second device.
j Figure 2 is a schematic diagram of an ultrasonic
flaw detector 50 with its mechanical mounting structure
removed. The flaw detector 50 has an ultrasonic
transducer 51 having a surface 52. The detector
50 also includes a sound transmitting wedge 54
which coacts with the surface 52, a transducer
housing 56, and an electrical interconnection 58
attached to the transducer 51.
The wedge 54 is shown in direct contact with
- an outside surface 62 of a pipe 60. The wedge
must be transmissive to sound and can be constructed
from a synthetic acrylate resin such as that sold
commercially under the trademark Lucite. Since
the wedge 54 is in direct contact with the outside
surface 62 of the pipe, the wedge obviously has
one surface with a radius of curvature correspondin~
to the pipe outside surface's radius of curvature.
In operation, the signal module shown in
sc~emat:ic ~orm in Figure 1 sencls an electrical
signal along an electrical interconnect 58 to the
transducer Sl. The transducer is comprised of
a material which upon receiving an electrical
signal produces an ultrasonic sound energy wave
over its surface 52. When the ultxasonic txansducer
receives an electrical signal from the electrical
signal modulet ultrasonic energy beams are transmitted
through the wedge 54 unt.il they imp;.n~e upon the
p.ipe 50. It is accura~e to speak o.E the ultrasonic
energy as a volume o~ energy, but, E~r illustration
purposqs, the beam ~ill be considered in a plane
o~ cross-section re~erre~ to as a number o.~ individual
beam elements whose paths Eollow substantially
straighk lines.
Upon entering the pipe 60 the ultrasonic beams
are refracted away from normal to the p.ipe surEace
and travel through the pipe searching for 1aws
or other irregularities in the pipe. If a ~law
or irregularity is found, ultrasonic energy beam
energy is reflected off that flaw and retraces
its incident path to the ultrasonic transducer
surface 52.
Ult.rasonic beams travel from the transducer
surface 52 along straight line paths and strilce
the outside surEace 62 of the pipe. Typical incident
beam elements indicated by ~he dashed lines 70,
72, 74 travel from the transducer 51 ancl strike
the pipe 60 at exterior locations 76, 77, 78 along
the pipe's outer surEace 62.
ll
` When typical beam elements 70, 72, 74 strike
. the pipe surface they form angles of incidence
A', A " and A' ". Upon entering the pipe structure
the individual ultrasonic beams are refracted away
from the normal to the sur~ace o~ the pipe along
reEracted beam paths 80, B2, 84 to form re~raction
angles S', S'l and S'''. These refracted beams
travel through the pipe stxucture searching for
flaws and irregularities until they reach internal
points 86, ~7, 88 located upon inside surface 64
of the pipe 60. If this refracted angle of incidence
on the inside surface is greater than a critical
angle (the definition of a critical angle is known
within the art) there occurs substantially total
internal reflection and the beams are sent from
the inside toward the outside sur~ace. The critical
angle is dependen~ upon the index oE refraction
oE the pipe and of the composition contained within
the pipe. In the configuration shown in Figure 1
the intexnal composition is air. The ultrasonic
beams continue to bounce from the inside and outside
surfaces oE the pipe until gradually there occurs
an attenuation which causes the beam strength to
diminish.
As best seen in Fig. 2 the ultrasound producing
transducer surface 5? comprises substantially an
involute. A generating curve or evolute shown
as a circle S9 is located coaxial with the pipe
and has a diameter less than the pipe's inside
surface 64. The transducer surface 52 is constructed
such that for any point on its surEace there exists
.
an imaginary line to that point which intersects
the evolute 59 at a point and in a direction tangent
to that evolute. Choosing a point 53 in the mid-region
of transducer surface 52 it is possible to trace
a path 72 to the surface 52 that intersects the
evolute 59 at a point 92. The path 72 intersects
the evolute 89 in a direction that is perpendicular
~: "
to a radius ~6 to the point 92 of intersection.
Similarl~ it can be seen that for other points
55 and 57 on the transducer surface 52 there exists
perpendicular paths 70 and 74 that intersect evolute
59 at points 90 and 94 and with directions tangential
to the evolute 59. -
One method oE tracing an involute comprises
the technique oE unwinding a string with a pencil
or other marker at its end from around the generating
curve or evolute. Thus, referring to Figure 2
a ~aut string with a pencil or marker tied to its
end could be wound around evolute 5~. As the
string is unwound it will coinci~e with the paths
70, 72 and 74 and ~he pencil would trace out the
involu~e cross-sec~ional shape oE the surEace 52.
The si~e of angles oE incidence A', A ", and
A''' hetween the pipe 60 and the ultrasonic beams
will depend upon the size oE the generating circle
or evolute 59~ ~s the radius 95 oE the evolute
approaches the radius o~ the pipe'~ inside surEace
64 those angles oE incidence will increase. As
the radius 95 of the evolute becomes smaller angles
o~ incidence A', A'', and A''' will also becom2
smaller. As the size of the evolute 59 approaches
the limit of a point centered at the pipe axis
the angles of incidence shrink to zero and the
typical rays 70, 72, and 74 become radial to the
pipe's outside sur~ace 62.
It can be shown that regardless of the size
of the evolute 59, as long as it comprises a circle,
the angles of incidence A', A'' and A''' must be
equal. In fact, any parts of an ultrasonic beam
emitted by a transducer having a transducer surface
generated ~y a circular evolute, will strike the
annular cross-section of a pipe at equal angles
along the outside surface of that pipe.
The proof oE.this propos;tion is straightforward.
It requires the showing of congr~ency bet~een two
triangles shown in Figure 2. One triangle contains
vertices de~ine~ by the center 99 of the evolute,
S the point of tangency 90 to the evolute oE a typical
ray 70, and the point the ray 70 intercepts the
pipe's outside sur~ace 7G. The second triangle
contains vertices de~ined hy the center 99 of the
evolute, a second point 92 of tangency to the
evolute and a second point 77 oE interception of
that tangency with the pipe's outside sur~ace 62.
By hypothesis the angles between the typical rays
70 and 72 and typical evolute radii 95 and 96 are
right angles. Since radii 95 and 96 are radi.i
to the same circle those sides o the two trianc~les
are e~ual in .l.ength. Also the distarlce from t:he
center o~ the evolute 99 to the two points oE
interception 77 and 76 are equal since they a~e
merel~ the outside radius Oe the pipe 60. Th~refo.re
the two above de~ined triangles have two equal
sides and one equal angle and therefore must be
congruentO Since this is true the ang}e A'' defined
by typical ray 72 and the normal to the pipe at
the poin-t 77 typical ray 72 intercepts the pipe
must equal the angle A' defined by a second typical
ray 70 ana the normal at the point 76 that ray
70 strikes the outer surface 62 of the pipe 64.
This completes the proof that an involute will
send typical beam rays to impinye upon the pipe .
. 30 with equal non-normal angles or lncidence.
By utilizing equal angles of incidence Al,
A'' and A''' a detector made according to the present
invention completely scans the pipe 60. It is
instruc-tive to examine the three typical beam paths
70, 72, 74 as they enter the pipe 60. These typical
beams enter the pipe at points 76, 77, 78 which
are equally,spaced about the outside surface 62
of the pipe 60. That is, the circumferential distance
14
between the lower point 76 an~l the midpoint 77 is ap-
proximately equal to the circumferential distance from
the midpoint 77 to the uppermost point 78.
Upon entering the pipe the ultrasonic beam
is reEracted due to the diEferent indexes oE refraction
of the pipe ~0 and the wedge 54. Upon refraction
typical beam paths 70, 72, 74 are bent to ~orm equal
angles of refractiQn S', S'' and Sl'l, respectively.
In traversing the pipe 30 the typical beams follow
paths 80, 82, 84 and strike the inside surface 54
at nearly equally spaced points 86, 87, 88. Thus,
the circumferen~ial distance between the lower posi-
tion 86 and the mid-position 87 is approximately
equal to the circumEerential distance between the
mid-position 87 to ~he uppermost position 88. Thus,
these typical beams tencl to maintaiJl their ~epar~ion
without diverging or converging as th~y travel
throughout the pipe 60.
If proper angles of incidence are chosen ~he
~o beam paths 80, ~2, 8~ will strike the inslde surE~ce
64 at anyles sufficiently large to result in nearly
total internal reflection. As seen in Fi~ures
3 and 4, if these angles of incidence are properly
chosen the beam will enter the pipe and be reflected
o~E the inside and outside pipe wall a number of
; times before they are attenuated. In one embodiment
oE the invention, incident angles of from 33 degrees
to 45 degrees prove effective to achieve the required
performance with an angle of incidence of approximately
35 degrees producing the best results. As noted
previously it is possible to selectively choose
the desired angle of incidence by changing the
radius ol the evolute or generating curve,
The preceding geometrical proof and discussions
of performance all relate to a transducer surEace
with a cross-section coincident with an involute.
While engineering considerations do not preclude
the pos.sibility of the construction of an involu~e
. ~
transducer surface they do suggest the use of a
transducer su~Eace which approximates an involute.
In practice tllree points such as the three points
53, 55, 57 chosen in Figure 2 are chosen on a real
involute. Using these three points it is possible
to construc~ a circle passing throuyh these points
whic~ closely approximates the real involute.
The circle in turn is the cross-sectional representation
of a cylinder. The actual transducer ~omprises
a segment of a cylinder whose cross-section coincides
with a segment of that circle, three points of
whic~ coincide with an optimum involute shape.
Although this quasi-involute or circle cannot produc2
exac~ly equal angles ~f incidence, the results
approximate e~ual an~l~s to the degree of accurac~
requirecl to ade~uately scan a weld area for flaws
or irregularities.
Figures ~ and 4 show how ultrasonic scanning
performance can vary depencling on the physical
dirnensions of the transducer surf~ce. The ultrasonic
flaw detector 110 shown on the right in Figure 3
- is a schematic representation of a flaw detector
constructed using the design described in Fiyure 2.
It comprises a transducer housing 112, a signal
cable llI, a wedge 122 and a transducer 115 having
a surface 116. The ~ucite wedge 122 coacts with
a Fipe 160 having inside I64 and outside 162 surfaces.
As shown in the figure, the size of the transducer
surface 116 is small relative to the width ~f the
pipe segment. The width of the ultrasonic beam
propagating from the transducer is indicated by
the two boundary ultrasound beams 118 and 120.
An ultrasound beam 124 emi~ted from the surface
116 travels to a location 121 on the outside-surLace
162 of the pipe 160 where it is refracted. The
beam 124 then travels throug-h the pipe 160 to a
location 126 on the inside surface where nearly
total internal reflection occurs. Similar internal
16
reflections occur at locations 12~, 130 along the
pipe's cross-section. To illustrate these multiple
reflections that can occur, the ultrasonic beam
124 as pictured is internally reflected five times
S before the bea~ strilces a flaw 100 and similarly
rebounds five times before re-entering the Lucite
wedge .
The transducer 115 is constructed of a piezo-
electric crystal. This crystal exhibits the property
that when it receives physical energy in the form
of sound waves, it conver-ts-that energy to an
electrical signal. Thus, when the wave retraces
its incident path and strlkes the transducer 115
an electrical signal is sent to a cable 111 which
transmlts that signal to a suitable electronic
signal moclule tnot shown in Figure 3~. Throu~h
electronic diagnostic techniques known within th~
art (and to be described) it is possible to deduc~
the existence and locatiQn of the ~law 100 within
the pipe 160 through interpretation of the reflected
signals.
As seen Erom the repreSentatiQn of the flaw
detector 150 when moved to the left of Fig. 3,
a narrow beam transducer surface may inadequately
2S scan the entire pipe structure. The flaw detector
150 comprises a transducer housing 152, signal
cable ISl, Lucite wedge 153 and transducer 155
having a surface 156. The sur~ace 156 produces
an ultrasonic heam 167 with outermost boundaries
166, 168 defining a relatively narrow beam. The
beam 167 enters the pipe at a location 161 on the
outside surface 152 and is refracted. When the
beam strikes the inside surEace 164 total tor nearly
total) internal reflection occurs and the beam
CQntinues its travel along the pipe's interior.
Due to the narrow width of the beam 167, however,
the beam never strikes the flaw 100. Instead the beam
continues along its path until it is totally attenuated.
s'~
It is appaxen~ from ~his discussion that in
one position the detector 110 receives reflected
signal~ from the flaw 100 and that in the second
position the detector 150 receives no reElected
signal from the flaw. rrhus, an ultrasonic signal
with width dimensions which are relatively too
narrow Eor the pipe 160 may rniss à Elaw or irreg~llarity
in the pipe's internal s~ructure.
As shown in Figure 4 it is possi~le to construct
an ultrasonic flaw detector 180 according to the
present invention with a transducer 182 large enougl
to completely scan the pipe 160. ~t should be
noted that the pipe 160 shown in Figure 4. has the
same inside 162 ancl outside 16~ su.rface as the
pipe shown in Figure 3. The detector 180 compris~s
a transducer housing 184l cabling 186, a Lucite
wedge 138 and an ultrasonic transducer 189 having
a surface 190. Three typical ultrasonic beams
segmen~s 191~ 192, 193 are schemRti.call.y shown
emerging ~rom the t.ransducer surEAce 190. If the
ultrasonic beam is shortened to produce only one
such beam 192 of width appro~imately equal to the
beam width o~ Figure 3, the ~law 100 might be
missed as the beam inadequately scanned the pipe.
With the.transducer wiclth increased to produce
a beam de~ined by the two outside limits 191, 193
shown in Figure 4, any change in densit~ dùe to
the existence of a flaw, regardless of its location
within the pipe, will be detected. As seen in
Figure 4 either o~ two representative beam segments
191 or 193 will strike tne flaw 100 producing a.
reflection which will return to the ultrasonic
transducer and be interpreted by the electronic
signal module. (Not shown in Figure 4).
. The two representative beam paths 191, 193
show the requisite conditions for adequate beam
scanning for a given pipe dimension. If there
exists one beam path 193 which is first reflected
~6~S ~l
1~
fror~l the outside surface 162 at a point 194 within
the boundary 191 of the incident ultrasonic energy
then con~plete ultrasonic scanning will occur.
Thus~ in Figure 4 either oE the representative
paths 191 or 193 will strike the flaw 100. Due
to the non-converging and non-diverging nature
o the beams, they will comple~ely scan the pipe
160 in the vicinity of the detector 180 and will
continue to scan as they travel in a circumEerential
direction about the pipe.
For pipe of a given outside and inside diameter
it is possible to determine optimum transducer
widths. It is apparent tha~ as pipe thickness
increases transducer width also must increase.
For wall thickness o three-eights inch it has
been deter~ined khat a transducer surEace width
oE two inches scans the pipe for Elaws. For other
pipe thicknesses it is desirable to select tr~nsducer
dimensions selected to scan as completely as desired
Eor flaws or other iLregul~rities within the pipe
s~ucture. Since due to engineering considerations
an arc of a circle is used instead of a true involute,
the circumferential extent o the arc is the dimension
which is varied to match the thickness of the pipe
to be scanned.
Shown in Figures 5 and 6 is one method of
mounting an ultrasonic tran~ducer according to
the present invention. Figuce 5 is an end view
of the detecting mounting structure shown in less
detail in Figure 1.
An end piece 202 is provided which has two
circumferentially extending outrigger arms 20~.
The outrigger arm 20~ (on the left) has been partially
sectione~ to indicate how roller bores 20& carries
a roller 208 for contact with the pipe 210. The
r~ller 208 comprises a rodlike cross-member 212
whi~h extends into the bores 206 found in the out-
rigger arm 204. By means of two bearings 21~ the
5 ~
roller 208 rotates about the rodlike cross-member.
The outrigger arm 20~ .is representcltive o~ the
other three outrigger arms which are not shown
in Fi~ure 5.
1'he outrigger arm 204 to the right in Figure 5
has been brokerl away to show an ultrasonic detector
mounting arrangement. 216. This cletector mounting
arrangeme~t is adius~ably mounted to a detecto~
mounting arm 218. The detector mounting arm 218
comprises a slotted arm which serves as a mount
- for two detector mounting brackets 220, only one
. - of which is shown. A crosspiece not shown in
Figure 5 maintains the two detector mounting brackets
220 a fixed d.istance apart longitudinally of the
pipe. ~ach mounting bracket has an attachecl threaded
stud 222 which extends through the slot .in the
detector mounting arm 218.
The position oE the bracket 220 rel~tive to
the arm 218 can be a~jus~ed by ~lns~rewing a threaded
knob 22~ which coacts w.ith the threaded knob 222.
The knob further acts against a tightening washer
226 which creates a friction joint with the slotted
arm 218. When the knob 224 is loosenea the ~riction
joint between the arm 218 and the tightening washer
25. 226 is lost and the bracket 220 may be moved along
the detecting mount arm 218 circumferentially of
the pipe to be inspected. When the proper adjustment
is achieved the knob 224 is retightened and the
frictional joint ree$tablished~ The adjusting
. 30 capability of t'ne bracket 220 allows the ultrasonic
testing device embodied by the present invention
to be adjustably mounted upon pipes of different
diameters.
The detector mounting bracket 220 c~rries
a pivot 228. An intermediate mounting bracket
230 is pivotally mounted by the pivot 228. This
pivot arrangement allows the intermediate bracket
230 to rotate about an axis 232 of the pivot 228
t5.~
; tFi9ure 6). The ultrasonic detector is mounted
to rotate witll the intermediate bracket 230 about
the axis 232 as a part o~ a gimbal arrangement
to mount the ultrasonic detector.
A support member ~34 i5 rotatably moullted
to the intermediate bracket 230. Two shoe support
: members 236 are attachecl to the support member 234.
Each shoe support member 236 comprises a U-shaped
member with inner portion 238 and outer portion 240.
These portions serve to main-tain a number of sliding
shoes 242 in fixed relationship to the shoe support
member 236. These sliding shoes 242 rest upon
. the pipe and as the detector mounting structure
200 is drawn along the len~th of the pipe ~hese
shoes slide along the pipe and maintain the detector
mount in the proper relationship to the pipe.
Eacll suppor~ member 234 is rotatably mounted
on i~s int~rmediate bracket 230 ~or rota~ion about
an axis 244 perpendicular to the axis 232. This
2D ro~ational axis is the second axis of the gimbal
arrangement for the shoe support members 236.
- Thus, the member are free to rotate about two axes
to allow the supporting shoes 242 to contact the
pipe regardless of variations in the pipe.
The ultrasonic flaw detector is mounted within
a detector mount 250. The detector mount 250 i5
rigidly attached to the shoe support member 236.
A transducer housing structure 252 is adjustably
mounted within the detector mount 250. The position
of the transducer housing.structure 252 relative
to the detector mount 250 can be adjusted until
the mount's relationship to the pipe is optimum
for sending ultrasound waves with equal an~:les
: of incidence into the pipe.
A first pair of screws 254 coact with two
slots 25~ within the detector mount 250. These
screws 254 are screwed into the transducer housing
structure and can be loosened and their position
acljusted by slicling the~ along the two slots.
Once this adjustment has been made for a particular
pipe these screws are tightened and the transducer
housing structure remains Eixed relative to the
detector mount.
A Lucite wedge 260 is interposec~ between the
transducer, which is mounted to tlle ~ransclucer
housing, and the pipe. Since the shape of the
Lucite wedge like the transducer depends on the
~0 pipe to b2 scanned, means are provided for removing
- the wedge and replacing it with a wedge of different
shape. The Lucite wedge 260 is attached to the
transducer housing structure 252 by means of two
countersunk screws 262.
15When pipes oE different dimansions are to
be scanned a new transducer is mounted within the
housing 252 by any suitahle means to maintain the
ho~lsing and the transducer in constan~ physical
relation t~ one another. The irst pair oE screws
254 are then adjusted to maintain the transducer
surEace (not shown in ~igure 5) in proper relation
to the pipe 210. When this is done, a proper Lucite
wedge with a radius curvature the same as the pipe
under study is attached to the housing 252 by the
countersunk screws 262.
Figure 6 shows the mounting arrangement 216
as seen from the surface of -the pipe. There are
five sliding shoes 242 on either side of the transducer
housing structure 252. Although the countersunk
Lucite mounting screws canno-t be seen from th-is
view, the Lucite wedge 260 can be seen attached
to the transducer housing structure. Located near
the sides Qf the Lucite wedge 260 are water spouts
264 throu~h which water is sprayed. This water
serves to ultrasonically couple the Lucite wedge
260 to the pipe 210. Typically a gap oE between
.010 to .035 inches is maintained between the wedge
and the pipe wall. Water is forced through the
'~ ,S~
` 22
spou~s to fill this yap to couple the wed~e to
the wall. The water may contain additives, such
as aerosol which act as wett;ng agents.
An ultrasonic transducer surface 266 corresponding
to those which have been described in detail, is
shown in phantom in Figure 6. ~hile the description
o~ Figure ~ characterized the sue~ace as an involute,
: or as a ~uasi-involute, the view from the pip2
is one o~ a rectangular transducer. The longer
f the two sides are actually involute or quasi-involute
in shape. The shorter of the two sides are lines
:
in both this view and in any other possible view
oE the transducer surface. In one embodiment used
for testin~, a pipe whose inside and outside surfaces
form a 2 inch cross-section, the dimension of this
rectangular transducer is 1~ x 2-1/4 inches.
~ As the detector mounting structure 216 kravels
~ alonc3 the pipe, the transducer mountecl within th~
detector mount 250 emits ultrasonic signals which
scan the pipe for irregularities witllin the weld
structure. The timing o these emitted signals
is controlled by the electronic signal module 24
of Figure 1. A block diagram illustrating a typical
circuit 300 which might be used Eor controlling
the sending and receiving of signals by the ultrasonic
transducer is shown in Figure 7. The circuit 300
is electrically connected to two transducers 302,
304 by means of electrical interconnects 306, 308.
These two transducers correspond to the two transducers
arranged on either side o the pipe as shown in
Figure 1.
The transaucers are activated by an acti~ation
signal which is sent from a pulser circuit 310.
The pulser circuit 310 is located within the electronlc
signal module and connected to each o ~he two
transducers. The electrical energy sent from the
pulser is converted to ultrasonic energy by the
transduoers.
s a:~
23
As the emitted signal rebounds off flaws or
irregularities within the pipe structure the returning
or reflected signals impinge upon the transducer
and are reconverted to electrical signals. These
return signals return along cabling 306, 308 to
a receiver unit 312~ These returned electrical
signals are processed by the receiver and sent
to a gating circuit 31~ or 316 wh:ich sends a signal
by means of electrical interconnection to a recording
device which records the existence oE the flaw
or defect within the pipe. The system shown in
Figure 7 includes a redundant recording system
which includes both a strip chart recorder 3~0
and a paint marker 322 Eor marking the pipe surEace
lS with a paint spot at the site of -the Elaw.
The signal sent from the pulser 310 to ~he
transduceL~s 302, 304 is an electronic pulse which
resembles a voltage spike ollowed by a damped
volta~e sine wave. The dclmped portion of the signal
is due to ringing within the circuit.
Due to the positioning oE the two transducers
it is necessary that the pulsers alternate their
activating signal to insure the signal from one
~for example 302) does not adversely affect the
~5 receiving of the ultrasonic energy by the other
304. If for example both transducers were simultaneously
activated by the pulser unit the transducers would
not know whether they were receiving a reflected
signal from a flaw or defec~ within the weld struct~e
or whether they were merely receiving a transmitt~d
wave from the other of the two transducers~ To
properly sequence the transducer activation, a
pulser unit contains a sequencing device which
is controlled by a triggering device. The trigger
device acts as a clock or time reference within
the pulser uni. and the sequencer alternately activates
the two transducers to produce the ultrasonic sound
waves of the present invention.
s'~
2~
Pulsers such as the one mention~ed above are
known within the art. One commercially available
device is a Krautkramer-Branson, Inc. unit which
includes a Model TGl triggering unit and a Model
PSI sequencer. This unit uses two SD4 transmitter
devices which are powered by a NE2 power supply
module. It is the NE2 or an equivalent power supply
which provides energy for the remaining elements
of the electronic signal module to be described.
As noted in Figure 7 both transducers are
electrically connected to a receiver unit 312.
The two primary functions of the receiver 312 are
to ampliEy and shape the signals from the transducers
302 and 304. The typical reflection signal from
a flaw within the weld area causes the transducer
to produce an envelope of RF signals oE fairly
small voltage. In order for the su~sequent recorc}ing
apparatus to respond to these signals they must
be signiEicantly ampli~ied. Since the recording
devices 320 and 322 most conveniently respond to
pulses or spikes, the amplified RF envelope must
also be shaped to achieve a single pulse of approxi-
mately 5 microseconds.
Circuitry for achieving this requisite pulse
is known within the art. A Krautkramer-Branson,
Inc. Model ANS 11 and ANS 1, for example, operate
as amplifiers in one typical receiving unit. A
TAl distance amplitude correction unit should also
be included to automatically correct for changes
in reflected beam amplitude due to naturally occurring
attenuation within the pipe. This TAl unit automati-
cally corrects for this attentuation and therefore
provides a uniform signal for the gating circuitry
314, 316.
The gating circuitry 314, 316 receives signals
from the reflected ultrasound energy regardless
of whether the flaws producing these signals are
inside the weld area. In the disclosed embodiment,
~l~
3l
essentially only flaws within the weld area are
of interest to the user. The gates 314, 316 serve
to block out signals coming from flaws or irregulari-
ties outside of the weld area. This capability
is achieved through a knowledge of how long it
takes the pulse to reach the weld area and to return
to the transducer.
Considering Figure 3, for examp:Le, the weld
area may be a range indicated by two boundaries
178, 17~ surrounding the defect 100 as shown in
that igure. Utilizing knowledge of how rapidly
the ultrasonic beam travels through the pipe, it
is possible to program the gating circuitry 314,
316 of Figure 7 to allow signals to activate the
lS recording means only after an initial delay during
which time any reElected signals must be coming
~rom outside the weld area. If, for example, the
gating circuitry allowed earlier signals to ac~ivate
~he recording apparatus, irregularities to the
right of the boundary area 179 in Figure 3 will
be detected. ~hese irregularities are of littla
interest to the user in this application and therefore
are not allowed to control the recording devices
320 or 322. In a like manner irregularities to
the left of the leftmost weld boundary 178 are
of little interest. The gating circuitry therefore
is operative to activate the recording circuitry
for only a short period of time during which the
reflected signals will be coming from within the
weld area.
A gating circuit with the above mentioned capa-
bilities is a Model No. BLl produced by Krautkramer-
Branson, Inc. This circuitry is adjustable to
allow for inspection of pipes of varying diameters
and differing transducer placements relative to
the pipe weld structure. As an illustration, in
one embodiment the BLl gating circuits are closed
and allow no signals to reach the recording device
~,
.
s~
26
for a period of 50 microseconds. These circuits
then open and allow reflected signals to activate
the recording device for a period of 20 microseconds.
During this 20 microsecond open period the reflected
signals would have been scanning the weld area
and are of interest to the user. The BLl then
closes and disregards any returning signals which
would be coming from areas beyond the weld structure
and therefore of no interest.
During the period in which the gates 314, 316
send received signals to a recording device, they
also operate to shape and lengthen the pulse sent
by the receiver. In a typical example the 5 micro-
second pulse sent by the receiver may be lengthened
to a 30 millisecond pulse which is operative to
control the recording devices.
The circuitry of Figure 7 can be used ta activate
a series of different recording devices all of
which indicate the presence of a flaw in the weld
area. As shown in Figure 7 apparatus 320 may be
provided ~or recording permanently upon a strip
chart recorder the presence of a flaw within the
weld structure. It is also possible to connect
the gating circuitry to a paint marking device
322 which automatically produces a spot of paint
on the pipe in the area of the weld flaw. A cathode
ray tube mounted upon a viewing device might also
be used to produce a representation indicative
of a flaw in a workpiece.
As examples of devices known within the art
to produce these results, one could choose a TOl
model by Krautkramer-Branson, Inc. to activate
the painting device, an RVl model by the same manu-
facturer to activate the strip chart recorder,
and a PSl sequencer again by the same manufacturer
could be used as an oscilloscope CRT device for
providing directly readable signal indicating the
presence of a flaw.
~7
While the present invention has been described
with particularity, it should be understood that
various modifications and alterations may be made
therein without departing from the spirit and the
scope of the invention set forth in the appended
claims.