·Table of Contents
·Materials Characterization and testing
Development of Roll Surface Testing Technique by use of Broad Bandwidth Surface Waves
Hajime Takada, Akira Torao, Masanori Kitahama, Takashi Morii, Ryouichi Sugimoto
Kawasaki Steel Corporation
1, Kawasaki-cho, Chuo-ku, Chiba 260-0835, Japan
Email : email@example.com
A heavy damped surface wave probe with high sensitivity suitable for the automatic
roll surface testing has been developed. Surface wave testing is usually applied for the
testing of hot finish rolling rolls made from high-speed tool steel. However, this testing
with a conventional narrow bandwidth probe has problems such as deterioration of
signal-to-noise ratio and false detection owing to high structural noises.
The developed probe is improved in the structure and in its material to get both short
pulse duration of ultrasound and high sensitivity. It has been proved that surface flaws
in rolling rolls can be detected by use of the developed probe with high signal-to-noise
ratio. Moreover, the developed probe is free from false detection of structural noises.
Keywords: Surface testing, Roll, Surface waves, Heavy damped short pulse probe
Surface flaws are occasionally formed in work rolls for rolling under abnormal
rolling conditions. Surface flaws remaining in a rolling roll can cause problems such as
roll spalling and marking on the strip. Rolls are ground by a grinder after rolling of
prescribed amount of strips. Re-profiled roll surface by a grinder is tested by a surface
flaw detector. If a surface flaw is detected, the surface of rolling rolls is re-ground until
the surface flaw is removed.
Surface wave testing is usually applied for the testing of hot finish rolling rolls made
from high-speed tool steel , . However, this testing with a conventional narrow
bandwidth probe has problems as follows:
A new method to resolve the problems mentioned
above was needed.
- Some cracks that cause roll spalling can
not be detected because of high structural noises.
- In some cases, high structural
noises are falsely detected as flaws.
These problems have been resolved by the development of a broad bandwidth surface
wave probe with high sensitivity. In this paper, details of the development will be
2. Conventional technique for surface testing and its problems
Fig 1: Roll surface testing by use of surface
2.1 Method and instruments for surface testing
Surface waves are transmitted in the
circumferential direction of a rolling roll
by use of a surface wave probe as shown
in Fig.1. The surface wave probe with
nominal frequency 2MHz is often used.
An example of automatic surface flaw
detector is shown in Fig.2. Whole surface
of the rolling roll is tested by scanning
the surface wave probe in the axial
direction of rotating roll (in the following
figures, this scanning is called "helical
scan"). The probe transmits the surface
wave in the direction opposite to the
rotation of a roll and receives surface
flaw echoes. Water-gap method is used
for the coupling between the surface
wave probe and the test body as shown
in Fig.3  , and it is said that the water
gap of 0.5mm is suitable. A
considerable quantity of water used as
the coupling medium remains adhering
to the rotating test body owing to the
friction between them. A wiper removes
the water remaining in the path of
surface waves for the prevention of
attenuation of surface waves and
appearance of spurious echoes.
Fig 2: Mechanism for automatic surface wave
Fig 3: Gap method for surface wave testing
2.2 Problems in conventional
technique and their causes
In the surface testing of a rolling
roll, back-scattered waves with small
amplitude are caused by the presence
of coarse grains, a rough surface and a large number of minute fire cracks produced in
hot rolling (these are called "collective fine reflectors" in the following). As the
conventional probe transmits and receives ultrasonic pulse whose duration is several
times longer than its wavelength, back-scattered waves mentioned above tend to
become large in amplitude by superimposing each other. Therefore, deterioration of
detectability and some false indications have happened in surface testing with the
3. Development of a broad bandwidth surface wave probe with high sensitivity
3.1 Reduction of structural noises by use of a broad bandwidth probe
Height Pg of echoes from collective fine reflectors in the path of ultrasound is
represented as equation (1), and is proportional to the square root of pulse duration of
ultrasound,  .
Where, x is a distance from the probe to the reflector, a0 is the attenuation
From equation (1), it is understood that shortening the pulse duration of ultrasound is
effective to make Pg small. As reflectivity of ultrasound at a flaw is not dependent on
the pulse duration of ultrasound, signal-to-noise ratio of a flaw echo can be raised by
use of a broad bandwidth probe. Consequently, detectability can be raised by use of a
broad bandwidth probe. As the conventional probe sends out the oscillation of
piezoelectric element without damping to get high sensitivity, of ultrasound pulse is
several times longer than its wavelength. Therefore, Pg has a large value in the testing
with the conventional narrow bandwidth probe.
3.2 Development of a broad bandwidth surface wave probe with high sensitivity
The oscillation of a piezoelectric element is damped by a damping block to get short
pulse in case of a general type of broad bandwidth probe. Consequently, both
transmitting and receiving sensitivity of a broad bandwidth probe are much lower than
those of the conventional probe. Gain of a receiving amplifier needs to be set a high
value to compensate the decline in sensitivity, so miss-detection of electrical noises
coming over the air is liable to happen. The developed probe is improved in the
structure and in its material to get high sensitivity to prevent miss-detection of electrical
noises coming over the air.
Selection of materials (piezoelectric element, resin wedge)
To get high sensitivity and to lower the amplitude of spurious echo, a piezoelectric
element with high electro-mechanic coupling factor in the thickness direction and low
electro-mechanic coupling factor in the radius direction is used as the oscillator of
ultrasound. The resin with low ultrasound attenuation and high acoustic impedance is
used as the wedge material.
Modification of incident angle
Incident angle qi from the wedge to roll surface is
determined on the basis of surface wave velocity in
high-speed tool steel to satisfy the Snell's law defined
as equation (2).
Where, CW is the ultrasound velocity in the resin
wedge, CR is the surface wave velocity in the test body,
and qi is the refraction angle (90° .in the case of the
surface wave) .
3.3 Performance and usefulness of the developed
Photo.1 shows the waveform of surface wave
transmitted by the developed probe as contrasted with
the waveform of surface wave transmitted by a
conventional surface wave probe. Sensitivity
of the developed probe with the center
frequency of 2MHz and the pulse duration of
1.5 wavelength is equal to that of the
Photo.2 shows some test results using an
artificial surface flaw with 0.2mm in depth
and 0.5mm in width. The height of structural
noises observed by use of the developed probe
is lower by approximately 6dB than that
observed by the use of the conventional probe.
As a result, surface flaws in rolling rolls can
be detected by use of the developed probe
with high signal-to-noise ratio. Sallow small
surface flaws can be also detected with high
signal-to-noise ratio. Test results using
artificial surface flaws with the developed
probe have proved that surface flaws with a
cross section above 0.06mm2 can be detected
with signal-to-noise ratio beyond 10dB. As for
the conventional probe, it is said that the
artificial flaw with a cross section 0.2mm2 is
critical for the detection. The flaw, which has
a cross section 1/3 times smaller than the conventional critical cross section, can be
detected by use of the developed probe.
(a) developed probe
(b) conventional probe
|Photo 1: Waveforms of echoes
from side wall
Fig.4 shows the test results of a work roll with a crack formed during rolling by use of
the developed probe and the conventional probe installed in an automatic roll surface
flaw detector. Gain of ultrasonic receiver was set the same value in both tests. While
height of the crack echo with the developed probe is as high as with the conventional
probe, height of the noise with the developed probe is much lower than that with the
conventional probe. As a result, cracks can be detected with high signal-to-noise ratio
by use of the developed probe.
MB:Main Bang F:Flaw Echo B:Edge Echo|
SE:Spurious Echo in Wedge
ME:Mode Converted Echo
Photo 2: Test results of artificial surface
Fig.5 shows the test results of a ground work roll by use of the developed probe and
the conventional probe installed in an automatic roll surface flaw detector. High
indications are observed with the conventional probe. These are falsely detected
collective minute harmless fire cracks remaining in ground surface. On the contrary, no
false indication can be observed with the developed probe. The developed probe is
effective to prevention of the false detection of collective minute harmless fire cracks.
As a result, excessive grinding of roll surface owing to false detection of collective
minute harmless fire cracks could be prevented, and roll consumption could be reduced
more than 20% while actual operations of roll grinding in Kawasaki Steel Corporation.
Fig 4: Comparison of the test results of roll with an actual surface flaw
Fig 5: Comparison of the test results of roll with harmless small fire cracks in its surface
A broad bandwidth surface wave probe with high sensitivity has been developed.
Detectability of surface flaws can be raised and excessive grinding of roll surface owing
to false detection of collective minute harmless fire cracks can be prevented by use of
this probe. Consequently, roll consumption could be reduced more than 20% while
actual operations of roll grinding in Kawasaki Steel Corporation.
The developed probe can be applied to the surface testing of work rolls for cold
rolling and backup rolls as well.
- Y.Ando, H.Sorano, Y.Sano: Surface Crack Detector for Rolling Mill Rolls by Surface
Ultrasonic Wave, CAMP-ISIJ, Vol.5, No.2(1992), 515(in Japanese)
- Y.Tsuchiya, Y.Ohnishi, A.Koido, K.Shiraishi, K.Masuda, K.Okimoto: Application of
Surface Crack Detector for Work Rolls of Hot Strip Mill, CAMP-ISIJ, Vol.6,
No.2(1993), 505(in Japanese)
- Japanese Unexamined Patent Publication No. 7-294493
- I .N. Ermolov : NDT International DECEMBER(1976), 275-280