AU692183B2 - A method and an apparatus for flaw detection - Google Patents

Description

A Method and an Apparatus for Flaw Detection Description of the Invention This invention relates to a method and an apparatus for detecting flaws existing on a member to be inspected such as a steel member.

For detecting if flaws (including cracks) exist on a steel member, there is a known technique wherein the surface region ofsthe member is heated by high frequency induction heating, and the flaws on the surface of the member are detected by measuring the surface temperature of the member with a radiation thermometer. This technique is disclosed in Japanese Provisional Publication No.298846/90 and in corresponding United States Patent No.5069005. As shown more specifically in Fig. 23, the surface region of the member W to be inspected is heated by using a high frequency induction heating coil 101 during transportation. and after that, the temperature distribution on the surface of the member W is measured with a radiation thermometer 102 arranged just next to the coil 101. If a flaw exists on the surface, the temperature at the flaw is different from the sound part (higher or lower than the surroundings, for example) so that the flaw is detected according to this temperature difference.

Although a member covered with scale can be adopted for the flaw detection mentioned above, the member is treated by shot blasting as a pretreatment for removing the scale prior to the flaw detection. The surface of the member W becomes a S: glossy one (shot surface) after the shot blasting and reduces :the emissivity so that the temperature difference AT .detected by the radiation thermometer 102, the level of the flaw signal decreases. Furthermore, handling marks (for *0 example, marks due to roller rubbing, bar applying or wire applying) are easily formed on the surface of the member, and these marks exhibit mirror-like appearance which is much more glossy than the shot surface so that the emissivity decreases significantly thereat. The part of lower emissivity E is detected as a lower temperature part with the radiation thermometer 102. Therefore, it may be difficult to judge whether the low temperature part detected by the radiation thermometer 102 shows a real flaw or a handling mark. This may cause a false detection of regarding a handling mark as a real flaw and may lead to decrease in the detection accuracy.

When the flaw detection is performed on a member covered with scale, the handling marks by rolls, bars, wires, and so on, mentioned above may appear as parts of scale peeling off.

The emissivity is smaller also for such parts than for the surroundings (scale), and the temperature measured with the radiation thermometer becomes lower thereat. This situation is fundamentally the same as that of the handling marks formed on the shot surface.

The object of this invention is to enable a high accuracy of flaw detection even in the case. that the surface emissivity of the member is not uniform because of the handling marks mentioned above. Furthermore, it is also the object of this invention to make the flaw detection more easy and reliable by enhancing the total surface emissivity of the member and increasing the temperature difference AT the flaw signal) on the surface.

For solving the object of this invention, the surface of 25 the member to be inspected is covered with powder for making the surface emissivity uniform, the member is heated in this state, and then the flaws are detected by measuring the temperature distribution on the surface according to the energy emission therefrom.

That is to say, the flaw detection method of th invention comprises the steps of; covering the surface of ember to be inspected with powder deposite reon (for example, by using an attraction 'of static electricity); l~p~ 2a In accordance with one aspect of the present invention, there is provided a method for detecting flaws existing on the surface of a member to be inspected, said method comprising the steps of: partially covering the surface of said member to be inspected with powder particles so that said surface of the member is partly exposed and wherein the thickness of powder coverage on said member is approximately 1/6 1/3 of the particles diameter if the layer of powder deposited is notionally assumed to be levelled and having uniform thickness heating the surface region of said member by high frequency induction heating; measuring the temperature distribution on said surface of the heated member covered with said powder according to the sum of energy emission from particles of sid powder covering said surface and energy reflection from said surface exposed said particles using a radiation thermometer; S 15 detecting the flaws on said member according to the temperature distribution on said surface of the member.

t 4 IN:ALIBE1U592:J4P S3 O heating the surface region of said member by frequency induction heating; 0 measuring the temperature distributi on the surface of the heated member covered with powder according to the energy emission from t surface of said member by using a radiation ther er; and d cting the flaws on said member according to the temperature distribution on the surface of the member.

One example of the members to be inspected are those having surfaces treated by shot blasting (shot surface), but the members are not limited to such ones, and those covered with scale are also adopted for the flaw detection. Anyway, the powder coating of this invention is effective for making the surface emissivity uniform and increasing the total emissivity when the emissivity is distributing on the surface of the member due to a nonuniformity of gloss, and so on, or when the total emissivity is small because the whole surface of the member is glossy as in the case of a shot surface, and so on.

In the step of covering the member with powder mentioned S"above, the powder can be deposited on the surface of the member (by using an attractive force of static electricity, and so on) either at a thickness which is sufficient to cover the surface completely or at a very small thickness. In the case of a very small thickness, the surface of the member can be covered with the powder so that the surface is partly "exposed. The degree of such very thin coverage is, for example, t1at the exposed part of the surface is observed in a form of dispersed spots when said surface is microscopically observed.

It is preferable to remove the powder deposited on the oo surface of the member after finishing the measurement of the temperature distributicof on the surface. In this powder removing step, the powder deposited on the member is sucked by a vacuum cleaner or blown away by using high pressure air from a nozzle or from a llower.

When the member consists of a non-magnetic material (a non-magnetic stainless steel, and so on), the depth of the electric current permeation of the high frequency induction heating is set to be larger than that of the flaw to be detected in the method of this invention. The flaw is detected according to that the temperature is lower on the parts where flaws exist than on the parts where no flaw exists.

As shown in Fig. 24, an induction current (eddy current) is generated in the surface region of the member W by the induction heating coil 101. When the permeation depth of the induction current is set to be larger than that of the flaw to be detected the temperature at the flaw, in most of the cases, becomes lower than that of the surroundings since the induction current passes around the flaw thereby causing a reduction of current density thereat, as shown in Fig. 25 (the temperature is conceptually expressed by a leigth of an upward arrow in Fig. 25). The flaw can be detected by measuring the temperature difference AT with the radiation thermometer 102. On the other hand, there may be the case that the temperature at the flaw is higher than the sound part (the part with no flaw) depending upon the morphology of the flaw, for example, in the case of a spot-like flaw or a flaw consisting of spots formed in a row. Particularly, the flaws on the edge portion of a square bar is detected as a high temperature part in most of the cases.

For a magnetic material, the depth of the electric current permeation of high frequency induction heating, 0, is usually smaller than that of the flaw to be detected, d, as shown in Fig. 27. Since the electric current density is larger for the u flaw part than for the sound part, the heat generated by the induction heating is larger for the flaw part, and the temperature measured by the radiation thermometer 102 becomes higher thereat. In this case, the part where the temperature is higher than its surroundings is to be judged as a flaw.

Tne flaw part should be measured as a high temperature part with the radiation thermometer 102 if the flaw detection principle presented in Fig. 27 is employed, but when the flaw part exists, for example, within a handling mark Sl as shown in Fig. 28, the apparent temperature at the flaw part becomes low because of the small emissivity due to the handling mark.

In this case, the flaw detection may result in fail. Such problem, however, can be prevented in this invention since the powder coverage provides a uniform emissivity.

The apparatus of this invention for detecting flaws comprises; 0a transportation line for transporting a membe to be inspected; 0a powder deposition device mounted on sal transportation line for covering the surface of the m er with powder (by using an attraction force of static ectricity, and so on); 0a high frequency inductio eating device mounted on the transportation line for ating the surface region of the member; t a radiation ermometer for measuring the temperature distributio n the surface of the heated member covered with the po r; and 25 a detecting device for detecting flaws on the member according to the temperature distribution.

"On the transportation line of the apparatus) of t i s e.i-t-e a powder removing device can be mounted for removing the powder deposited on the surface of the member.

For enhancing the reliability of flaw data, it is important S8 that the radiation thermometer measures the temperature distribution on the surface of the member heated by the high frequency induction heating device while the temperature drop is as small as possible. Therefore, the radiation thermometer N, ,3 u v^ l) ^j 6 is preferable to measure the temperature just after heating or during heating. The high frequency induction heating device is usually equipped with a heating coil into which the member to be inspected is inserted. The temperature measuring point can be arranged, for example, just next to the heating coil on the transporting line or can be located in the heating coil.

For example, the apparatus can be constituted so that the induction heating device comprises a heating coil wherein a cavity (slit, notch, hole, and so on) formed therethrough from the inside to the outside thereof, and the radiation thermometer receives the energy emitted through the cavity. The apparatus can be constructed also so that the radiation thermometer located on the direction inclined to the heating coil which receives the energy emitted through the gap between the heating coil and the member inserted thereinto.

Particularly for non-magnetic steels, the absorption of electric power is smaller than for magnetic steels so that the flaw detection is needed to be performed at a low transportation speed. In this case, the temperature measurement is performed preferably in the heating coil for avoiding the temperature drop due to heat conduction, and so on, as small as possible.

erobM~e.

i' In this i nvention, the surface of the member to be inspected is covered with powder deposited by using static 0 25 electricity, and so on. Therefore, even if low emissivity parts (such as handling marks) exist, the surface emissivity becomes uniform due to the powder coverage, and the flaw signals are detected in this state so that the probability of S.false detections can be reduced. Since the shot surface, which has a low emissivity, is enhanced its total omissivity by the powder coating, the temperature difference AT can be •increased. Furthermore, even if handling marks having lower emissivities than a shot surface exist, the emissivity of the surface becomes uniform due to the powder coating. That is to say, the level of emissivity can be improved with a simultaneous achievement of a uniformity of the emissivity for a shot surface, and so on, so that strong flaw signals and precise results of flaw detection can be obtained.

In the case of a large thickness of the powder deposition on the surface of the member, the effect of making the emissivity uniform is enhanced. On the other hand, when the powder coverage is very thin (for example, in a degree of the surface partly exposed), of course an economical advantage is attained, and besides this, the time and energy for heating the member can be almost the same as in a conventional method since only smP.ll heat is needed for heating the powder deposited on the member. Furthermore, such very thin powder coverage is preferable to improve the detection sensitivity because sharp profiles of flaw signals can be obtained, while the profiles of the flaw signals broaden in the case of a thick coverage due to the heat conduction in the powder layer, and so on.

This powder coverage is performed usually with charging the powder and the member at negative and positive polarities, respectively, according to the same principle as a conventional electrostatic powder coating. The powder is required only to cover the surface of the member, and baking as in a conventional powder coating is not necessary. After the powder coating, the high frequency induction heating and the subsequent temperature measurement with a radiation thermometer are performed, and then flaws are detected according to the temperature distribution on the surface as well as in the ".-method of the prior art. The temperature at the flaw to be detected becomes higher or lower than at the sound part S: depending upon whether the depth of the permeation of the induction current for heating is larger or smaller than the depth of the flaw to be detected, or depending upon the morphology of the flaw.

8 In the Ainvcntien, comprising the step or the device for removing the powder, the powder deposited on the member is removed by using a vacuum cleaner, and so on, after finishing the temperature measurement with the radiation thermometer.

Therefore, no powder remains on the surface of the member, and bad influence due to the residual powder can be prevented in the subsequent steps. In the Anv4zin of measuring the temperature distribution on the surface of the member in the heating coil, a high level of flaw signal is obtained since no temperature drop occurs.

In the accompanying drawings: Fig 1. is a figure conceptually presenting the flaw detection method and the apparatus of one embodiment of this invention; Fig. 2 is a perspective view specifying the concept of the apparatus presented in Fig. 1; Fig. 3 is a partial enlarged view of Fig. 2; Fig. 4 is a figure showing the arrangement of the radiation thermometer in Fig. 2; Fig. 5 is a figure presenting the principle of electrostatic powder coating; Fig. 6 is a figure showing the detail of Fig. Fig. 7 is a figure explaining an example of powder coating; Fig. 8 is a figure explaining the effect when the powder is deposited at a very small thickness so that the surface of the member is partly exposed; Fig. 9lis a figure explaining the state that the powder is deposited as thick as the surface is covered completely; Fig. 10 is a cross-sectional view showing the flaw etdetection process for the member; Fig. 11 is a figure showing the relationship between the flaw existing on the member and the temperature measured by the radiation thermometer; K-iPr 770 9 Fig. 12 is a figure showing the case of using a coil wherein a slit is formed in the center portion; Fig. 13 is a front view of Fig. 12; Fig. 14 is a side view of Fig. 12; Fig. 15 is a figure explaining the first example of the case that the temperature measurement for the member is performed in the heating coil; Fig. 16 is a figure explaining the second example of the same case; Fig. 17 is a figure explaining the first graph of the case that the temperature measurement is performed for the member in the heating coil.

Fig. 18 is a figure explaining the second graph of the same case; Fig. 19 is a figure explaining the surface state of the member corresponding to the comparative example and example 1; Fig. 20 is a graph showing the result of the flaw detection in the comparative example; Fig. 21 is a graph showing che result of the flaw detection in example 1; Fig. 22 is a sketch of the micro photograph for the :surface of the member in example 1; Fig. 23 is a perspective view presenting an example of a conventional flaw detection method; ,0 25 Fig. 24 is a cross-sectional view of the coil for high frequency induction heating in Fig. 23.

Fig. 25 is a figure explaining the relationship between a othe positjon of the flaw and the path of the induction i current; Fig. 26 is a figure explaining the energy emission in the case that a handling mark exists on the member; Fig. 27 is a figure explaining the case that the depth of the electric current permeation in the surface region of the member is smaller than the depth of the flaw; Fig. 28 is a figure showing the case that a flaw exists within a handling mark.

Several embodiments of this invention will now be described with reference to accompanied drawings.

In Fig. i, the member to be inspected (simply as 'the member', hereinafter) W consists of a non-magnetic material formed in a longitudinal shape having a square cross-section, and its outer surface presents, for example, a shot one as a result of removing the scale thorough shot blasting, and so on. This member W is transported by the transportation line comprising a bobbin-shaped roll 1 for transportation, and so on.

A powder deposition device 2 is mounted on the transportation line. This device deposits designated powder on the surface of the member W by using static electricity, and the principle of the deposition is the same as that of a conventional electrostatic powder coating. Fig. 5 and 6 conceptually show the principle of the powder coating. The designated powder is fed to the a powder spray nozzle 3 from a powder tank 4. A high voltage power generator 5 applies an voltage between the powder spray nozzle 3 and the member W at negative polarity for the spray nozzle 3 and at positive polarity for the member W. The member W is grounded.

As shown in Fig. 6, each powder particle P, for example, has a core consisting of C (carbon) or Ti (titanium) and a shell consisting of resin (for example, acrylic resin, and so on) The average particle diameter is, for example, approximately in the range of 10 100 gm. The average diameter of particles is preferably 20 50 gm and more preferably 40 pm because of an economical reason. The resin shell of the particles P are negatively charged under an oee, "ionization atmosphere formed by the voltage mentioned before (for example, 100 kV with up to 50 AA of current), attracted to the W in the anode side by the Coulomb's force and r/ O 1c,"' t C: o cover its surface depositing thereto. The thickness of the powder coverage or deposition can be set so that the surface of the member W is exposed in a form of dispersed spots. More specifically, the thickness of the powder coverage in this case has a very small thickness of approximately 1/3 1/6 of the particle diameter if the layer of the powder deposited is assumed to be leveled as having a uniform thickness.

As shown in Fig. 7, for example, the powder spray nozzle 3 can be constituted to be able to cover each surface of the member W with the powder individually along the direction of the transportation. In this case, all surfaces of the member W can be coated simultaneously with the powder also by using four such powder spraying nozzles 3 arranged around the merober W. The member W can be coated with the powder by using only one spray nozzle 3 in such way that one half side of the member W is coated by the spray nozzle 3, then the member W is turned and taken back to the upper stream in a reciprocal motion, and finally the other half side is coated by the same nozzle 3.

As shown in Fig. 10 for example, .f handling marks S1 ar- existing on the surface S of the member W, the surface emissivity of the member W becomes almost uniform since the whole surface S of the member W, having the handling marks S1 thereon, is coated with the powder particles P (Fig. 10 There is no limitation upon the color of the powder particles.

The powder particles P deposited on the surface S of the member W are bad conductors and maintain negative charge so .that the particles never drop from the surface S within a 0:certain period even if the member W is cut from the ground. In Fig. 10, the deposition thickness of the powder particles P is intentionally enlarged because of the convenience for drawing.

o• When the member W is treated by shot blasting, the surface S becomes a shot surface, and the handling marks S1 formed by rolls, bars, wires, and so on, have mirror-like appearances l which are more glossy than the shot surface. On the other hand, when the member W is the one covered with scale before shot blasting, the surface S and the handling marks Sl correspond to the scale and the parts of scale peeling off, respectively. The surface S is, however, a shot one in many cases.

As shown in Fig. 1, a high frequency induction heating device (induction heating device, hereinafter) 8 is mounted on the lower stream of the transportation line. This device is a conventional one for generating an induction current (eddy current) in the surface region of the member W, and formed in an annular shape surrounding the member W as shown in Fig.

24. For the member W consisting of a non--magnetic material, the induction frequency is set so that the depth of current pei.;,ation, 6 (Fig. 25), is larger than the depth of the flaw to be detected, d.

As shown in Fig. 10 the surface region of the member W is heated by the induction current (eddy current) generated therein. Except for the cases of flaws or cracks on the edge portion of a square bar, and so on, which exhibit higher temperature than in the sound part, the induction current passes around the flaw so that the temperature becomes lower at a flaw than at a sound part in most of the cases. The temperature of the member W increases 20 30 "C from ambient temperature by heating as mentioned above, the member W is heated to 40 50 'C when the ambient temperature is 20 *C.

In Fig. i, a radiation thermometer 9 is arranged just next to the induction heating device 8 and measures the surface temperature of the member W whose surface region is heated.

The radiation thermometer 9 is a conventional one which transforms the energy emission from the surface of the member *o W into an electric output by using sensors and displays this output. A plurality of such radiation thermometers 9 are arranged in the apparatus of Fig. 1, and each thermometer 9

A/T

Z

corresponds to one face or one edge of the member W.

As shown in Figs. 12 14, a fork type of heating coil 8, in other words, a coil 8 having a cavity (such as slit 8a) formed therethrough from the inside to the outside thereof can be used. The member W is inserted into the coil 8 and heated therein. The surface temperature of the part heated is measured in the coil 8 by using appropriate number of radiation thermometers 8 through the slit 8a as shown in Fig.

Such temperature measurement in the heating coil 8 provides a high accuracy of measurement since almost no temperature drop occurs.

The same effect can be obtained also by using the device presented in Fig. 16. In this case, the radiation thermometer 9 is arranged in the position and direction so that it is directed to the member ThT in the heating coil 8 through the gap between the heating coil 8 and the member W inserted therein.

Fig. 17 is a graph presen:ing the sensitivity (intensity) of the flaw signal AT measured along the direction parallel to the center axis of the heating coil 8 which is defined as the origin. The electric power supplied to the heating coil 8 was .fixed to 130 kVA, and three transportation speed of the member ,2 W (relative to the coil 8) was used, 10 m/min, 20 m/min and 40 mimin. According to this result, regardless of the transportation speed, it is preferable to measure the temperature at around the region from the center to the exit of the heating coil 8 for enhancing the level of the flaw signal AT.

18 shows the ratio of the flaw signal AT to the temperature raisek-t the surface of the member W the degree of how strong flaw signal is ctained against the temperature raise). Good results are obtained within the width of the heating coil 8.

eel° 0If a flaw exists on the surface of the member W as shown in Fig. 11 and for example, the temperature distribution measured by the radiation thermometer 9 becomes as 14 shown in Fig. 11 Even if a handling mark S1 exists as shown in Fig. 10 the mark Si is covered with the powder P so that the surface emissivity of the member W becomes almost uniform, and the temperature measured by the radiation thermometer 9 does not exhibit a lower value at the handling mark S1. Therefore, the handling mark S1 is never detected talsely as a flaw.

Although the flaw existence can be judged through a visual observation of the temperature distribution according to the values measured by the radiation thermometer 9, the apparatus can be constituted so that the radiation thermometer 9 is connected with a judging means 10 comprising a compater, and so on, and the apparatus makes a warning signal when the judging means 10 detects a temperature abnormality, with showing the data of the flaw detected such as the location, depth, and so on, on a displaying means as conceptually shown in Fig. 1. The apparatus can comprise a marking device which makes a mark indicating the location of the flaw when receiving a signal from the judging means 10. Furthermore, the apparatus can comprise a grinder 16 vhich removes detected flaws automatically receiving a signal of flaw location from the judging means On the lower stream side of the radiation thermometer 9, there are a powder removing device 11 equipped with a suction portion 12 mounted as surrounding the member W, a suction pump 13 connected witli the suction portion 12 and a powder collection portion 14. This device removes the powder deposited on the membei W by sucking and collects the powder after the atemperature measurement with the radiation thermometer 9. The suction portion 12 of the powder removing device 11 can be mounted movably along the longitudinal direction of the member

W.

Fig. 2 4 shows an apparatus which specifies the conceptual image of the apparatus presented in Fig. i. In Fig. 2, a i9'" <K'p steel member W is transported by a plurality of bobbin-shaped rolls 1 in its own longitudinal direction. When the steel member W passes through the powder depositing box 20, the powder is electro-statically deposited on the surface of the member W. In the box 20, there are, for example, two powder spraying guns (not drawn in the figures) each of which corresponds to one face or one edge of the steel member W. On the lower stream side of the box 20, a heating coil 8 as shown in Fig. 3 is mounted, and a high frequency current is supplied thereto from a power source 21 via a coil transformer 22.

During the step of passing through the heating coil 8, the steel member W is heated, and the powder deposited thereon is heated by the heat transfer from the member W. On the lower stream side thereof, a plurality of radiation thermometer 9 are mounted as shown in Fig. 2. These radiation thermometer 9 are arranged, for example, with two for the faces 9a and two for the edges 9b four in total), and measure the temperatures on a half side of the member in one measurement.

On the further lower stream, there is a powder collecting box 23 as shown in Fig. 2, which collects the powder from the surface of the steel member W. 24 is a powder collecting device. The flaw detection is performed to one side of the steel member W in one cycle of measurement, and then performed for the other half side in the next cycle after returning the ee 25 member W to the upper stream or transportinc the member W to the further lower stream.

Now, the surface of the member W is assumed to be covered :with a pliu[ality of the layers of powder particle P as shown in Fig. 9. The surface is completely covered with the powder layers. In this case, the emissivity of the surface is dominated by that of the powder particle P so that the object to 0e making the emissivity uniform is achieved. A thick powder coverage, however, takes long heating time for sufficiently heating the member along with the powder deposited thereon so that a slow transportation of the member W or an increase in the heat supply per a unit time is demanded when the heating time is held to be constant. Furthermore, a temperature gradient corresponding to a flaw on the member W, such as a trough, diffuses due to the heat conduction through the thick layer of powder P, and the sensitivity may be reduced because of broadening of the flaw signal.

Although this invention never excludes the embodiment of forming a powder coverage thick enough to cover the surface of the member W perfectly, the thickness of coverage is preferable to be as thin as possible for avoiding the harmful effect mentioned above. If the surface were covered with only one layer of powder particle P, the surface could be covered almost perfectly with the least amount of powder, and the heat conduction could proceed efficiently. Such coverage with only one layer is, however, difficult to be achieved, and tLhe surface may be partly exposed if an ultimately small thickness is pursued. Such state appears to be undesirable, but this partial exposure of the surface rather provides a good result.

In Fig. 8 and so on, the surface (shot surface, for example) of the member W is assumed to be covered with the powder P at a very small thickness so that the surface is iJ partly exposed. Fig. 8, however, presents only a concept and does not provide a precise expression of the real state of the 25 powder deposition. The average distance between the particles P is not large enough to include whole part of a flaw to be detected (or a handling mark) so that flaws, and so on, exist being hidden under many particles P as shown in Fig. 8 (b) When the member W and the powder particles P are heated to the temperature T as shown in Fig. 8 and when the powder particles P are regarded as black bodies, the energy emission Q is proportional tu T 4 (Stefan-Boltzmann's low). As shown in Fig. 8 the energy emission from the powder particles P is Q (where the emissivity is unity), and that from the

I,!

surface exposed between the particles P is PQ since the surface is a glossy shot one which has an low emissivity Although the temperature at the exposed surface is expected to be lower than the surroundings in the discussion above, the energy emitted from the particles P is partly reflected on the exposed surface having a high reflectance because it is a shot one, and this reflected energy (which can be nearly regarded as 7Q where y is the reflectance of the exposed surface) is added to the energy emission SQ mentioned above so that the radiation thermometer 9 detects the sum of these energy, SQ yQ as shown only a simple concept in Fig. 8 That is to say, what the surface of a large reflectance is partly exposed means that the reflection of the energy emitted from the particles P becomes considerably large on the exposed surface although the energy directly emitted from the particles P becomes small. In a simple concept for an easy interpretation, the sum of the energy is regarded to be SQ yQ Q because E 7 i. This provides a result almost the same as when the surface is completely S.covered with the powder particles P as shown in Fig. 8 and the particles P are heated instantaneously to the same temperature as the steel member W. The level of the detected temperature for the shot surface is generally increased while 25 the temperature at the flaw part does not changes significantly so that the trough-like flaw signal becomes deeper resulting an enhancement of the sensitivity as shown in Fig. 8 As is described above, the partial exposure of the surface provides not only an result comparable to the surface completely covered with the powder P, without provoking no inconvenience, but also a new effect of the addition of energy due to the reflection. As long as almost the same effect can be obtained, the powder P is used preferably in an amount as t ~t small as possible from view points of time of heating the member W and the intensity of the flaw signals. Therefore, it is recommended to deposit the powder P on the member W so that a part of the surface is exposed intentionally. This inven? -on, however, does not exclude the embodiment wherein the powder P is deposited thick enough to cover the surface without exposure as mentioned before.

Now, it will be described about the case that the induction heating flaw detection method of this invention is applied to a member having a surface condition as shown in Fig. 19.

(comparative example) The member shown in Fig. 19 is subjected to a flaw detection by means of a conventional flaw detection method, and the result is presented in Fig. 20. As it '~ear from Fig. 20, the parts of rubbed marks (handling exhibit low temperature as well as the flaw part so tnat it is difficult to distinguish the flaw and the rubbed marks.

(example 1) The surface of the member shown in Fig. 19 was covered with powder particles with 40 gm of average diameter. The rate of powder spraying was 50 m/min. The result of the flaw detection for this member is presented in Fig. 21. Fig. 21 shows that only the flaw part is detected as a low temperature part.

Fig. 22 is a sketch of the micro photograph taken for the surface of the member in example 1. The surface is observed to be partly exposed in Fig. 22. The exposed part (whice part in the figure) of the surface exhibits a distribution in a form of dispersed spots.

Claims (4)

1. A method for detecting flaws existing on the surface of a member to be inspected, said method comprising the steps of: partially covering the surface of said member to be inspected with powder particles so that said surface of the member is partly exposed andfwherein the thickness of powder coverage on said member is approximately 1/6 1/3 of the particles diameter if the layer of powder deposited is notionally assumed to be levelled and having uniform thickness; heating the surface region of said member by high frequency induction heating; measuring the temperature distribution on said surface of the heated member covered with said powder according to the sum of energy emission from particles of said powder covering said surface and energy reflection from said surface exposed between said particles using a radiation thermometer; detecting the flaws on said member according to the temperature distribution on said surface of the member.

2. The method according to claim 1 wherein said temperature distribution on said surface is measured with said radiation thermometer at the place where said high frequency induction heating is performed.

3. The method according to claim 1 comprising the step of removing said powder deposited on the surface of said member after completing the measurement of temperature distribution on said surface of the member. IN:\LIEIO1592:JJP

4. The method according to claim 1 wherein said member to be inspected consists of a non-magnetic material, wherein the depth of the electric current permeation of said high frequency induction heating is larger than the depth of the flaw to be detected, and wherein the flaw is detected according to that the temperature is lower or higher on the part where the flaw exists than on the part where no flaw exists. A method for detecting flaws, substantially as herein described in relation to any embodiment with reference to the drawings. DATED this Fifth Day of January 1998 Daido Tokushuko Kabushiki Kaisha Patent Attorneys for the Applicant SPRUSON FERGUSON (n:\libEl01461:SEC A Method and an Apparatus for Flaw Detection Abstract The surface of the member. to be inspected is covered with powder deposited by using, for example, staic electricity prior to the flaw detection. The surface of the member can be covered with the powder completely or at a very small thickness so -that the surface is partly exposed. Subsequently, the surface region of the member is heated by high frequency induction heating and then the temperature distribution on the surface is measured with a radiation thermometer The part for which the temperature measured is different from the surroundings is judged as a flaw. The surface of the member is covered with the powder so that the surface emissivity becomes almost uniform, and the resulting temperature distribution measured with the radiation thermometer becomes almost equal to the real one. Therefore, flaws existing oi. the surface of the member are precisely detected according to the measurement with the radiation thermometer (9) even in the case that there are parts of low emissivity, such as handling marks, on the surface. Figure 1. 6331W/GMM I II I