Sensitive eddy-current testing of tubes

for defects on the inner and outer

surfaces
F. Forster

The eddy-current method is widely used for the non-destructive testing of tubes. 1 This
report describes the extensive progress achieved in the last few years. The author dis-
cusses examples of testing techniques for tubes for nuclear engineering where extreme
sensitivity is of the utmost importance, although the principles are also applicable to
all types of tubes.

Several authors 2,3 report that their endeavours to inspect There are two distinct coil designs associated with the
reactor components by conventional eddy-current methods reflection and transmission methods: the encircling coil
met with negative results. Perhaps because they were un- which encircles the test part completely, and a rotating
aware of the range of test designs and test possibilities with probe which spins around the sample in a circular path,
eddy current systems, they considered such techniques to either in contact with the test piece, or contactless. When
be unsuitable for the inspection of reactor components. the encircling coil is used inside the tube, it is then called
During the past few years however, numerous investigations an internal coil. Both these coil designs can be applied
have been conducted at Institut Dr F6rster in co-operation according to the above reflection and transmission test
with European reactor research institutes and manufacturers principles, and eight possible combinations are listed in
of reactor components. They prove conclusively that the Table 1. The highly developed technology associated with
eddy-current method is eminently suitable to solve these coil designs and with the external and internal
numerous test problems with extreme defect resolution. rotating heads result in optimum test properties which will
Non-destructive testing of tubes has been materially be the subject of a further report.
advanced by this essential achievement and progress.
The examples discussed are from a range of improved eddy-
current tube-testing systems, and they cover eight different Applications of eddy-current methods to tube
eddy-current test designs (Fig.l) (Table 1). For each design inspection
examples are quoted of test problems from technical
spheres where a safe and sensitive detection of quite small The reflection method where both the exciting and
defects is required. For each of these test examples the receiving coils are outside (method 1 of Table 1) is applied
particular eddy-current test design listed in Table 1 which for the underwater testing of nuclear fuel cans (with or
offers the most favourable test result is indicated. without the pellets). In such a case it is possible to fully
exploit the versitility of an eddy-current test system whereby
When considering eddy-current tube testing, two different
contactless testing under difficult conditions (testing under
test principles must be distinguished: the reflection method
water or at high temperatures etc) is achieved even when
and the transmission method. With the reflection method
long test coil cables are needed. Method 1 is now well
the primary coil which excites the electromagnetic field and
established industrially for testing thick-walled uranium
the secondary coil which indicates the defect are both
tubes for holes and coarse grain formation (tube dimen-
arranged on the same side of the tube to be tested, ie these
sions: 45 mm diameter, 11 mm wall thickness).
exciting and receiving coils are arranged either both at the
outer or both at the inner surface. With the transmission
method, however, the exciting and the receiving coil are The use of a test-coil pistol
placed on different sides of the tube wall. Furthermore
A compressed-air pistol is placed at the opening of the tube
with transmission techniques the receiving coil is affected
only by those electromagnetic fields that have passed to be tested (Fig.2). By pressing a button, the test coil
through the complete wall of the test part. As a result the with its guide head and connecting cable are propelled to
the other end of the tube. A return motor incorporated in
transmission method is eminently suitable for indicating
the compressed air pistol automatically brings the test coil
tube defects of the same magnitude on the inner and outer
back to it. Three leads at the lower end of the handle are
surfaces with the defect signal amplitudes of the same
for compressed air, current supply for the return motor, and
height.
for the cable for the test coils which slides through the
pistol. This new method of internal tube testing satisfies
Dr F. Forster is with the Institute Dr FiSrster, Reutlingen,
W. Germany. This article is published with the cooperation of hitherto unsolved test problems. When internally inspect-
Wells-Krautkramer Ltd, Letchworth, Herts, UK, which is the UK ing heat exchanger tubes it is necessary to detect defects
agent for F(irster equipment. in the supporting stay areas. It is furthermore essential

28 NON-DESTRUCTIVE TESTING . FEBRUARY 1974

,,t I,I \ /

Fig. 1 Several arrangements for eddy-current tube testing are listed by number in Table 1: these drawings show the disposition of exciting
coils - heavy black dots; receiving coils -- dashes; exciting rotating probes - dotted black; and receiving rotating probes -- hatched black

Table 1 Properties and applications o f e d d y - c u r r e n t tube-testing techniques

Tech 5 Reflection rotating Very high defect resolution

No Procedure Remarks probe, exciting and for defects in and underneath
receiving probe outside the outer surface of the tube.
Applications: inspection of
Reflection, encircling Preference towards od and od defects on smooth tubes.
coil, exciting and transverse defects. With a Smallest blemishes on the
receiving coil outside large wall thickness faint or surface. Testing of top end
no id defect indication, In plugs of cans for longitudinal
case of differential coils the shrinkages; test of voids in
alteration of defect depth is sodium of the fast breeder
indicated. Application: reactor fuel element.
Testing of radioactive cans
under water. 6 Reflection rotating Very high defect resolution
probe, exciting and for id defects of tubes inde-
Reflection encircling Preference towards id and receiving probe inside pendent of the wall thickness.
coil, exciting and transverse defects of tubes. Indication of the absolute
receiving coil inside In case of differential coils depth of defect. Applications:
the alteration of defect depth testing of longitudinal and
is indicated. Applications: transverse defects inside the
internal investigation of tube, testing of U-shaped
incorporated heat exchanger heat exchanger tubes incor-
tubing. Internal investigation porated in the reactor for
of reactor tubes with medium corrosion and cracks.
defect resolution.
7 Transmission: Com- Measurement of wall thick-
Transmission encir- Increased defect sensitivity bination exciting ness and eccentricity on thin
cling coil, exciting compared with (1) id defects rotating probe outside and thick-walled tubes with
coil outside, receiving preferred with an improved receiving encircling high sensitivity of measure-
coil inside od defect sensitivity com- coil inside ment. Application: measure-
pared with (1). Application: ment of wall thickness.
defect inspection of 6-fin-
8 Transmission: Com- With small wall thickness id,
tubes when the tubes pass
bination: receiving intermediate and od defects
over the test coils.
rotating probe out- of the same size are recorded
Transmission encir- Like (3), but od defects pre- side, exciting en- with the same amplitude.
cling coil, exciting ferred by improved id defect circling coil inside Simultaneous measurement
coil inside, receiving sensitivity. Application: of wall thickness and
coil outside defect inspection of 6-fin- eccentricity. Applications:
tubes during the return pass fast investigation of reactors
of the tubes. Hereby with high defect resolution
changing over from (3) to and indication of the position
(4), of the defect in the tube wall.

that internal corrosion is reliably recorded as well as that defects which can be satisfactorily indicated beneath such
corrosion on the outer side of the condenser tube occur- supporting stays. Extensive practical experience has shown
ring at the edges of boreholes in the supporting stays. that for the indication and identification of the different
tube defects that occur, absolute and differential coils must
Intratest records were obtained from a test tube 4 in which
be used simultaneously. Therefore the records in Fig.3 are
defects were inserted such that they could not be indicated
the traces o f absolute and differential coils obtained for a
satisfactorily by existing ndt methods (Fig.3).
given test. The absolute channel reproduces area corrosions
In particular recordings were taken at the supporting stays, as well as annular corrosion at the edges o f the supporting
with and without defects. It is essential to determine those stay. s,6,7,s The indication from the supporting stay itself

NON-DESTRUCTIVE TESTING . FEBRUARY 1974 29

Fig.2 A compressed-air gun is used to shoot the test coil through
the tube; retraction is motor-driven

in the trace of the absolute channel is clearly identifiable in

relation to the other defects. This represents important
progress in the internal tube testing because formerly
supporting stays paralysed eddy-current indications in most
tubes by overmodulating the amplifier.
For the two upper records, al (absolute coil) and a2 (differ-
ential coil), the supporting stay 9 was slid to a sound part
of the tube at the beginning of the al test. The outer
reductions of wall thickness (0.5 mm depth, 100 mm length),
Fig.3 Intratest recordings b y differential and absolute eddy-
and (0.4 mm depth and 100 mm length) simulate outside current methods f r o m a tube of 25 mm diameter and 1 mm wall
surface corrosion and can be quantitavely evaluated in the thickness showed defects along the length of the tube: 1 -- hole
absolute trace. On the other hand, holes 1 - 4 on the tube 0.9 mm diameter on tube circumference; 2 -- t w o holes as in 1;
circumference are recorded only in the differential trace a2, 3 -- three holes as in 1 ; 4 -- f o u r holes as in 1 ; 5 - o u t e r notch,
width 2.5 ram, depth 0.4 ram; 7 -- outer reduction of the wall
and here there is an excellent signal-to-noise ratio. In
thickness, length 100 mm, depth 0.5 mm; 8 - outer reduction of the
recording al the supporting stay was not between the two wall thickness, length 100 mm, depth 0.4 mm; 9 -- supporting stay,
notches s,6 so that the simulated annular corrosion could 20 mm thick; 10 -- dent, depth 0,5 mm
be independently recorded; in recording b 1 the support-
ing stay was slid between the two notches. The absolute
trace recording b 1 shows that the indication of simulated
edge corrosion of the tube at the supporting stay position
is minimally influenced by the affect of the stay itself. In
recording c of Fig.3 the supporting stay 9 was shifted
exactly over the outer notch 6 thus simulating relative
positions met with in practice. Although the notch indica-
tion is reduced in recording cl it can be clearly identified.
T
Moreover, if the supporting stay indication 9 in cl with
associated defect is compared with the indication al of the
supporting stay 9 without a defect it can be concluded that
much smaller corrosion effects in the area of the supporting
stay can be indicated and identified. The development of I
eddy-current units, the technology of coils and transport
systems have achieved material progress in the field of
internal tube testing. This conclusion is confirmed by ex-
tensive practical tests carried out in traditional power stations
nuclear power stations and in chemical plants. . . . . o-----o . . . . 0 - - - - - - 0

Eddy-current inspection of tubes with irregular

surfaces L•
4mrn~'
I

II ....~
2 3

I] ~
4

........
5

-FV...
In the reactor industry, methods 3 and 4 are combined for
use in the inspection of 6 and 3-fin tubes of larger diameter Fig.4 Finned tube may be inspected b y an encircling-coil
as well as for those with irregularly shaped surfaces. The technique as in this example where artificial defects are drilled in
different parts of the tube: 1 -- 0.1 mm diameter hole through
external coils are manufactured by a specially developed
tube wall; 2 -- 0.1 mm diameter hole, 0.1 mm deep; 3 - 0.2 mm
technology to fit exactly - according to the section shape - diameter hole, through the wall; 4 -- 0.2 mm diameter hole,
around the finned tube. Fig.4 shows the defect resolution 0.2 mm deep; 5 -- 0.2 mm diameter hole, through wall near fin

30 NON-DESTR UCTI VE TESTI NG . FEBRUAR Y 1974

obtained on a 6-fin tube, as well as the position of the test Eddy-current inspection of tubes using an external
defects between the fins. rotating probe
It is remarkable that defect 2, which has a volume of less Procedure 5 using a rotating external probe gives a near
than a ten thousandth of a mm 3 is recorded with a signal- surface defect resolution hitherto unachieved by any other
to-noise ratio of more than 6 : 1. Moreover tubes can be ndt method. Furthermore this extreme defect resolution
tested 10 times faster with this transmission method using is accomplished at relatively high test speeds. Within the
an encircling coil than by ultrasonic techniques. rotating heads four probes spin round the test parts at
100 rev s"1 . The circumference of the test part is thus
scanned 400 s"1 (Fig.5). Different types of spinning probes
can be used according to the required application (Fig.6).
The smaller the defects to be detected in the test material,
the more point-like the eddy-current probes must be so that
these small defects occupy a sufficient area of the focused
eddy-current field. Although for normal defects a micro-
probe is sufficient, recordings show that the micro-gap
probe has a range of action five times larger, and that the
signal-to-noise ratio of about 50 to 1 results from point
defects of microscopic extent. For small surface blemishes
in nuclear fuel cans, method 5 is more reliable than the
human eye assisted by optical magnification (Fig.7).

L o n g i t u d i n a l and transverse defects

Longitudinal as well as transverse defects must be found

when testing nuclear fuel cans. To assess the suitability of
non-destructive test methods, transverse and longitudinal
standard defects were inserted in a zircalloy tube with an
outer diameter of 14.3 mm and a wall thickness of 0.94 mm.
The depth of the transverse longitundinal defects were
46/am = 5% of the wall thickness, 94/am = 10% wall
thickness, and 187/am = 20% of the wall thickness. The
lengths of the transverse defects were 0.75 up to 0.9 mm.
These transverse longitudinal defects were inserted both in
the outer and inner walls of the tube which was then tested
and recorded by 6 different eddy-current methods of
Table 1 in the following order 1,2, 4, 8, 5 and 6. The
results showed that the combination of methods 5 (external
rotating head) and 6 (internal rotating head) offer the
optimum results (Fig.8).
Specially shaped rotating probes were developed simultane-
ously to give optimum indications of longitudinal and
Fig.5 The Ci¢cograph is system using an external exciting and transverse defects. During the rotation of the probes the
receiving probe: In this case instrumentation for high resolution is
used and the signals on the screen are from three standard defects
individual defect impulses were connected to a profile
at depths o f 5%, 10% and 20% of the wall thickness o f a reactor record by a peak recorder. For industrial tube investigation
tube the external and the internal rotating heads were switched

I*-,mm-J
Fig.6 Two different spinning probe configurations were used to record signals f r o m a hole 0.1 mm deep in a reactor t u b e w a l l 1 mm thick:
l e f t -- microprobe; right -- micro-gap probe

NON-DESTRUCTIVE TESTING . FEBRUARY 1974 31

 cladding is essential. The intermediate sodium layer must
be free of voids or holes in order to avoid hot spots, ie areas
of the fuel rod not cooled by the sodium. The problem of
sodium void indication has been dealt with in the USA using
both radiographic inspection and eddy-current probes. 6
During these investigations the sodium layer between pellet
and can was simulated by an aluminium alloy of about the
same electrical conductivity as solid sodium. An aluminium
tube containing holes of different diameters was inserted
inside a can of wall thickness 0.25 mm. American authors
concluded from their experiments using eddy-current probes
outside the can that the minimum size defect detectable in
the aluminium tube is approximately 0.8 mm in diameter.
Similar investigations were carried out at Institut Dr F6rster
using can walls twice as thick ie 0.5 mm, in order to estab-
lish the limits of detectability of sodium voids under the
most unfavourable conditions (Fig.9). The indication
amplitude of a sodium void with a diameter of 0.8 mm,
that is a diameter corresponding to the 'minimum size
defect detectable' by Cwynar and McGeary was designated
by the relative value 1 000. Fig.9 gives the relative ampli-
tudes of the sodium void with diameters of 0.6, 0.3, and
0.1 mm, and shows that the F6rster eddy-current tech-
niques can detect a sodium void 50 times smaller through

/I
cans with double the wall thickness. Although this sensi-
tivity may not be required in practice the facts shown in
Fig.9 demonstrate the progress in defect detectability
achieved within the last few years.

Fig.7 Eddy-current results using an external probe method are

better than visual inspection even with magnification s

Inspection of top and bottom plugs

Inner s u r f a c e O u t e r surface
t J ]li I "l A further application of method 5 for reactor components
is for the inspection of top and bottom end plugs. The
problems of testing such zircalloy end plugs is reported by
Nukem 7 as follows 'The testing of top and bottom end

I
T L T L T L i T L T L T L
I
5 5 20 20 IO IO 5 5 20 20 IO IO
[O/o]

Fig.8 Recordings of transverse (T) and longitudinal defects (L) L--

of various depths (as percentages of wall thickness) were plotted
along the length for defects on the inner and outer surfaces of a
zircalloy tube 14.3 mm in diameter with walls 0.g4 mm thick; the
recorder overloaded for defects thicker than 2% wall thickness
;70 200 15

to the same electronic system. The internal rotating probe

was then in action during the forward movement, the
external rotating probe during the return motion. The
defect resolution of the 5% transverse and longitudinal
defects in the outer and inner surface of the tube indicates
the ability of these adjusted eddy-current probes to fulfill
the prescribed American Test specifications.
- - F - .-,,-.4--
Indication o f v o i d s in s o d i u m between the pellet 570 210 20
and the can

The extreme test sensitivity of the latest eddy-current probe Fig.9 Voids in sodium reactor coolant were simulated by holes
in an aluminium alloy tube; an eddy-current recording made for hole
method is illustrated by the ability to detect voids in the diameters from left -- 0.6, 0.3, 0.1 mm; the top indication were
sodium of sodium-cooled fuel elements. Here continuous through a can 0,5 mm thick and the bottom indications with only
contact of the cooling sodium between the pellets and the an air gap of 0.5 ram. The numbers are relative amplitudes

32 NON-DESTRUCTIVE TESTING. FEBRUARY 1974

 v
Imm 1
Fig.lO These eddy-current indications from a filiform micro-hole of diameter 0.08 mm are typical; indications (middle) from a microprobe
and (right) a micro-gap probe are shown side by side

plugs regarding f'tliform holes takes up much time and is

difficult. The ultrasonic methods cannot be used because
l Circogroph]__
Re:order ]
of the different grain b o u n d a r i e s . . . The table of the test
report shows that in case of a finely smooth surface all
~ ProbetransportT
I Cable drive synchronizotion
defective end plugs were de£mitely detected by the
Circograph (Fig. 10).'
Further interesting applications in the field of non-
destructive testing of reactor components are given by
method 6 with rotating probes inside the tube. Fig.8 shows
the records of longitudinal and transverse defects inside
a tube with a depth of 5,10 and 20% of the wall thickness.
No other single test-head ndt method indicates longitudinal
and transverse tube defects which are 5% of the wall thick-
ness with such a good signal to noise ratio.

Eddy-current inspection of reactor heat-exchanger

tubes
A particularly interesting application of the internal rotating
eddy-current probe is the examination for cracks and
n9 heod
corrosion in U-shaped, water fdled, incorporated reactor
heat-exchanger tubes. Here the area of the tube fastening
in the pressure vessel as well as the bend in the tubes is of cod ~
great interest. Defects and corrosion in the heat exchanger
tube must be determined and their position on the tube
circumference -- whether on the inner or outer surface -
must be indicated (Fig.11).

Internal tube testing with extreme defect resolution

Fig.11 An internal rotating eddy-current probe can be used to
The following example shows the extreme defect resolution inspect water-filled heat-exchanger tubes fitted in a reactor; this
arrangement is typical 8
of the internal roating eddy-current probe for the investiga-
tion of reactor heating tubes. In such tubes small inner
defects can occur of up to 1 or 2% of the wall thickness and
could cause internal corrosion due to their shape. In a brass rotational axes, spinning within a protecting tube, which
tube for example, slight surface blemishes proved to be also contains nylon brush guidance. The rotating axes
scaly cleavages 10-20/am deep (Fig.12). These blemishes carrying on their centres the feed wires for the eddy-current
can be detected and measured by eddy-current methods assembly, can easily be fitted together for desired lengths.
(Fig. 13). The possibility of rapid recording such defect Fig.15 shows the internal rotating head operating 15 m
depth profiles of tube defects on the inner surface within away from the test unit. The tube to be tested passes over
the rotating axis, and its inner surface is inspected by close,
a depth range of a few hundredth of a mm is another
spiral scanning. Such internal rotating heads find numer-
example of the advance in eddy-current techniques to
ous industrial applications, often being used in conjunction
satisfy extreme resolution applications.
with an external rotating head whose probes spin around
Fig.14 shows two eddy-current internal rotating heads with the outside of the tube. The electronic test system can
guiding nylon brushes. These heads are attached on thereby be used twice, for during the forward movement

NON-DESTRUCTIVE
TESTING.FEBRUARY1974 33
 of the tube the internal rotating head is switched to the
electronic system and during the retraction of the tube it
is switched over to the external rotating head. Methods 7
and 8, the transmission methods with rotating eddy-current
probes are used for the quick absolute measurement of wall
thickness and eccentricity of thin wailed and thick walled
tube. The measurement and recording of geometrical
properties of tubes such as wall thickness, eccentricity,
inner diameter etc will be the subject of an extensive pub-
lication on the theory and practical application of the
transmission method for the measurement of geometrical
E
properties of tubes, this is being prepared.

Fig.12 Slight blemishes on the inside surface of a brass tube

(diameter 23 mm, wall thickness 2 mm) are shown on a photograph
(left), photomicrographs (right) show the blemishes to be scaly
cracks 18, 15 and 13 #m from top to bottom

I! ~ I h '
' i t t:'~: !'

- - I

t I ]

II II
m
Fig.13 Circograph indications from the brass tube in Fig.12 shows
defective and sound parts of the tube; top - a peak amplitude
recorder shows defects on the inner circumference along the tube
length; centre -- Circograph screen displays show variations on a
single inside circumference, the amplitude is proportional to the
defect depth and the position and width is proportional to the
defect position and width on the circumference; the two lower
screen displays show sound tube and the peak amplitude recording Fig.15 The internal rotating head may be operated at 15 m away
at the bottom shows a length of sound tube from the test head

Fig.14 Two internal rotating heads for high-resolution eddy-current testing of tubes of different diameters

34 NON-DESTRUCTIVE TESTING . FEBRUARY 1974

Transmission method with an external rotating The defects which are faintly indicated by method 5 are
probe distinctly indicated by method 8. By superimposing the
recordings obtained by method 5 and method 8 in Fig. 16
Finally, an example to illustrate the transmission defect the exact defect position in the tube wall can be deduced
inspection of metallic tubes, including tubes made of by the signal amplitude ratio of the signals obtained from
austenitic steels. Here method 8 of Table 1 (transmission the reflection method and the transmission method. A
with external rotating probe is used) whereby a static coil defect on the outer surface is for instance recorded by
of a certain configuration creates an electromagnetic method 5 and 8 with amplitudes of the same height. The
alternating field inside the tube which is focused through nearer the defect is situated to the inner tube wall, the
the wall on to an external probe, rotating at 100 rev s-1 . higher will be the ratio with the defect amplitudes of
Methods 8 and 5 can be used to give complementary method 8 (transmission) compared with method 5
information (Fig.16). (reflection). The quotient of the defect amplitudes from
method 8 and method 5 shows the position of a defect in
the tube wall whereas the height of the amplitude of
Reflection
method 8 indicates the defect size.

Comparison between ultrasonic and eddy-current

recordings
The comparison of test results by ultrasonic methods with
those from eddy-current transmission methods obtained
Transmission from identical tubes is of particular interest. The zirc-
alloy tube (diameter 11 ram, wall thickness 0.85 mm)
was tested by eddy-current transmission methods and
showed characteristic defects. The same tube was carefully
investigated by ultrasonic techniques in the testing depart-
ment of one of the largest European manufacturing plants
for reactor components. Three different test-head designs
were used for the detection of longitudinal and transverse
defects using ult.rasonic transmission method (Fig.17). The
comparison shows that the eddy-current transmission
Fig.16 The reflection method 5 (upper recording) and the
transmission method 8 (lower recording) which both use rotating method indicates all the defects that are found by ultra-
probes give useful Complementary information about the depth of sonic methods using 3 different ultrasonic heads. This is to
a defect in a zircalloy tube 11 mm in diameter and 0.85 mm in wall be expected because with the eddy-current transmission
thickness method the electromagnetic alternating field must pene-
trate the whole tube wall, and therefore faults in this wall
Eddy current "¢estin9 F will cause corresponding disturbances in the eddy-current
TransmissionA BC distribution. Moreover, the eddy-current method detected
additional defects, eg W in Fig.17, and recorded them with
high amplitude. These defects were not indicated by the
three ultrasonic test heads. Furthermore, because the eddy-
current method 8 scans at 100 lines s"t this results in a test
speed which cannot be achieved by any other non-
destructive testing method.
Ultra sonic testin9
Transmission
References
1 F6rster,F. The non-destructive inspection of tubings for dis-
continuities and wall thickness using electromagnetic test
methods. Materials Evaluation 28 (1970)
"rransverse
2 Sharpe, R. S., Aveyard,S. The inspection of thin walled stainless
steel reactor grade tubing. Journal o f the Iron and Steel
al Institute 201 (1973) October 856-862
l 3 Ktohm,G. Prufung van Hullrohren fur Brennstabe. Material-
Longitudinal Cl
prfifung 8 (6) (1966) 224
Ai ! D~ 4 The recording of Fig.3 where obtained from a test tube of the
! t F
I IlL I I --- .. ~ - L Jl Rontgen Technische Dienst NV Rotterdam by the Prufdienst
of Magnetische Prufanlagen GmbH, Reutlingen
5 These standard defects were prepared by General Elec!tric

Fig.17 A recording from the rotating-probe method using the

1
transmitted eddy-current (method 8) was compared with recordings
Nuclear Energy Division, USA
6 Cwynar, J. C., McGeary, R. D. Eddy-current and radiographic
evaluation of sodium-cooledfast breeder reactor fuel element.
Materials Evaluation 28 (1970) (3)
7 Nukem. NuclearChemie u Metallurgi. Test report of 11 April
1969 (Experience with the F6rster Circograph in testing end
caps with f'flliformholes)
from ultrasonic testing using three separate probes. The defects 8 The layout and results of this unit for internal investigation will
are identified by letter be the subject of another report

NON-DESTRUCTIVE TESTING. FEBRUARY 1974 35

I n f o r m a t i o n 2 on non-destructive material testing Advertisement

High defect resolution now also for inhomogeneous

material surfaces
Ferro-magnetie tubes, bars and round billets During test the parts are scanned by heli-
with relatively inhomogeneous surfaces can cally rotating stray flux probes. Eight fiirste ir
be non-destructively tested by the magnetic channels are used for the test to ensure the
stray flux method with high defect resolution highest possible transit speeds. In other means
using the ROTOMA~® 6.048. Even as-rolled words, eight test probe pairs simultaneously
surfaces are tested by the ROTOMAT® with- scan the material surface.
than teslin 9 ......
out a noticeable decrease in the fault de-
tection sensitivity. Investigations on round billets and tubes
have shown that faults on and under the
Quality assurance and increased efficiency surface are detected and interpolated with
in production have placed special emphasis high resolution even in hot-rolled material.
on the testing of mill products such as
drawn or welded steel tubes and ferro- Principally, test pieces in the diameter range Fully automatic test installation
magnetic bars and round billets. Material between 20 and 450 mm can be tested; the
separation defects such as cracks on and two standard models of the ROTOMAT® Already a number of large establishments
below the surface, cavities, laps, inclusions are designed for 20 to 140 mm and 100 to in the steel industry has integrated FOR-
etc. which are largely detrimental to or 450 mm material diameters. STER ROTOMAT® test installations into
exclude the further use of the material must
be found by these tests.

The method which is proven to be most

suitable for this duty is the non-destructive
magnetic stray flux method. The ROTO-
MAT® 6.048 is based on this method.

Probe assembly of R O T O M A T ® with five

probe fingers.

Reliable testing, faster testing,

rational testing
The FORSTER ROTOMAT® equipped with A sample of the high defect resolution for a
eight channels will allow test speeds up to test with the F O R S T E R R O T O M A T ® :
90 m/rain to be reached. These high test
Rotating head for R O T O M A T ® , testing of a speeds permit of a continuous material flow indication of a material defect of 0,6 mm
round billet. and thus optimum use of capacity. depth and 0,1 mm width on a hot-rolled round
billet.
The aim in developing this test instrument Clear defect indication
with a view to absolutely dependable
The ROTOMAT® electronics process the the production line. The separation of test
quality inspection was set at the reliable
fault signals from the test probes so that pieces, their transport through the test
detection of smallest faults in the material.
these can be made visible on the screen of installation, the actual test, the analysis,
Longitudinal cracks are primarily detected
an oscillograph and can simultaneously be the sorting and/or marking of the tested
but also transverse cracks if these have a
recorded. parts, all these processes are fully auto-
longitudinal component. The ROTOMAT®
mated. They are controlled by FORSTER
will reliably detect faults on or under the Above all, however, these signals are inter- Automation Devices.
internal surface of tubes with wall thick- polated with the final aim of sorting the
nesses up to approximately 8 or 10 mm. l)etailed information is available from Wells-
test pieces according to quality classes or
In this instance the fault detection sensiti- respectively automatically paint marking Krautkramer Ltd., Castle Vale Industrial
vity corresponds to API specifications. the defect positions on the test piece. A Estate, Minworth, Sutton Coldfield, Warks.
distinction between inside and outside de- (or their authorized representatives).
® Registered trade-mark fects can be done by different color.

36 NON-DESTR U C T I V E TESTI NG . F E B R U A R Y 1974