Determining the BTR by conducting a Trans-Varestraint test using a high-speed camera and two-color pyrometry

Determining the BTR by conducting a Trans-Varestraint test using a high-speed camera and... The Ni-Fe alloy consumable that applies to the welding of 9% nickel steel for LNG storage tanks is highly susceptible to solidification cracks. To evaluate susceptibilities such as the brittleness temperature range (BTR), a Trans-Varestraint test is generally conducted. However, it is difficult to evaluate the minimum strain value for the BTR and the real temperature at both ends of a solidification crack in a conventional Trans-Varestraint test because these values are measured indirectly. In this study, we propose determining the temperature range by conducting in-situ observations during a Trans-Varestraint test using a high- speed camera and two-color pyrometry so that the temperature range can be measured directly from the temperatures at both ends of the crack. Furthermore, we measured the augmented strain from the time elapsed since the initiation of bending. This method allowed us to successfully measure the augmented strain and the temperature range in the Trans-Varestraint test and to determine the BTR more accurately. . . Keywords High-speed camera Two-color pyrometry Trans-Varestraint test 1 Introduction range (BTR). Solidification cracks will occur if the strain curve increases during cooling to the point where it crosses the BTR [1, 2]. Most LNG storage tanks are built using 9% nickel steel, which has a very high degree of strength and toughness. The welding One method that can be used to evaluate a material’sBTR consumable used for 9% nickel steel is generally a high-Ni is a Trans-Varestraint test. In this test, solidification cracks are alloy, such as Ni-Cr alloy, Ni-Mo alloy, or Ni-Fe alloy. These reproduced by forcibly bending and applying strain to the test types of Ni alloys are extremely tough when welded. material with the temperature distribution applied to the eval- However, they are highly susceptible to solidification cracks, uation material by a TIG arc. The BTR is defined by the and a high deposition rate welding technique, such as sub- relationship between the length of the longest solidification merged arc welding (SAW), is adopted for the tank’sfabrica- crack and the cooling rate for TIG welding [3]. tion. Therefore, evaluating solidification crack susceptibility However, the Trans-Varestraint test has some limitations. is of great importance in the fabrication of LNG storage tanks. Although the results of measuring the cooling rate for TIG One of the typical values that indicate susceptibility to so- welding by inserting thermocouples into the molten pool in- lidification cracks is ductility at an elevated temperature. dicate the temperature gradient within the weld direction, Some materials, especially Ni alloys, have a low-ductility some of the cracks reproduced by the Trans-Varestraint test temperature range as defined by the brittleness temperature may beextendedatanangle to the welddirection.Thismay produce a margin of error. In addition, to determine the limit- ing strain for the occurrence of cracks, it is necessary to per- Recommended for publication by Commission IX - Behaviour of Metals form the test many times using different augmented strains. It Subjected to Welding is difficult to accurately determine the limiting strain for the occurrence of cracks if only a limited number of tests are * Daisuke Abe daisuke_abe@ihi.co.jp conducted. In recent years, Osuki et al. reported the strain measure- 1 ment results obtained at crack initiation in a Trans-Varestraint IHI Corporation, 1, Shin-Nakahara-cho, Isogo-ku, test [4]. However, few studies have directly measured the Yokohama-shi, Kanagawa 235-8501, Japan Weld World temperature and/or the strain at crack initiation in a Trans- Varestraint test. Furthermore, Kadoi et al. reported the two- dimensional temperature distribution that they measured using a two-color pyrometry technique that was adopted for a laser Trans-Varestraint test [5, 6]. However, this technique was ap- plied to a laser rather than a TIG arc, which is a general heat source for the Trans-Varestraint test, and its cooling rate is quite different to that of laser welding. It is difficult to measure the temperature distribution during TIG welding because the arc light and bright tungsten electrode may interfere with the measurements. Therefore, the aim of this study is to determine the critical strain at crack initiation and the temperatures at both ends of Fig. 2 Dimensions of the test pieces for the Trans-Varestraint test the longest solidification crack in the Ni-Fe alloy by adopting a high-speed camera and a two-color pyrometry technique in a Trans-Varestraint test. First, a two-color pyrometry system chemical compositions of these two materials are shown in was developed to measure the temperature distribution during Tables 1, 2,and 3. TIG welding while avoiding the effects of the arc light and bright tungsten electrode. Next, the validity of the tempera- tures measured using the two-color pyrometry method was 2.2 Two-color pyrometry technique verified by measuring them at the solid-liquid interface of a pure Ni plate. Finally, a Trans-Varestraint test was carried out 2.2.1 Principles of the two-color pyrometry technique to determine the BTR by using a high-speed camera and a two-color pyrometry technique. As shown in Eq. (1) below, the brightness of the emitted light can be expressed as a function of the wavelength (λ), the temperature (T), and the emissivity (ε). However, since the emissivity varies depending on the object’smaterialand sur- 2 Materials and experimental procedure face conditions (e.g., its roughness and/or its state), it is diffi- cult to accurately measure its temperature from its brightness. 2.1 Materials However, by taking the luminance ratio (I /I ) for the two 1 2 wavelengths as shown in Eq. (2), it is possible to derive the Pure Ni plates were used to verify the validity of the temper- temperature independently of the emissivity (ε). Since the atures measured using the two-color pyrometry method. The two-color temperature measurement method utilizes this prin- test pieces used in the Trans-Varestraint test were made of Ni- ciple, it is necessary to acquire two images at different wave- Fe weld metal that had been welded by conducting submerged lengths for the measurement object [7]. arc welding with A5.14 ERNiMo-8 filler wire. A cross- I ¼ ε⋅ fðÞ λ; T ð1Þ sectional view of the weld metal is shown in Fig. 1.The dimensions were 150 mm × 50 mm × 10 mm (Fig. 2). The I ε⋅ fðÞ λ ; T 1 1 ¼ ð2Þ I ε⋅ fðÞ λ ; T 2 2 2.2.2 Components of the two-color pyrometry system Specimen In this study, an innovative optical system was developed and combined with a high-speed camera for two-color pyrometry. A schematic illustration of the system is shown in Fig. 3. With this optical system, the observed image was split into two images by a splitter. After that, they were passed through band-pass filters with a range of 850 and 950 nm. Finally, the two images were captured by one of the high-speed cam- era’s image sensors. The magnification used for this system Fig. 1 Cross-sectional view of the weld joint and specimen position for the Trans-Varestraint test was × 100. Weld World Table 1 Chemical compositions of the materials used Material Spec No. C Si Mn P S Cr Ni Mo Cu Fe W Pure Ni plate UNS N02201 0.01 0.03 0.09 – <0.001 – 99.71 – 0.04 0.12 – Weld metal (wire) AWS A5.14 ERNiMo-8 0.02 0.02 < 0.1 0.005 0.001 2.1 69.7 19.0 0.02 5.73 3.0 2.2.3 Calibration of the temperature two-color optical system placed on the side opposite the weld direction. To prevent the arc light and the reflected light of the Although the brightness of the emitted light varied widely in electrode from causing an error in the temperature for the two- accordance with the temperature of the target, the dynamic color pyrometry method, a mechanical shutter was mounted range of the high-speed camera was limited. Consequently, on the TIG torch. The shutter covered the arc light and the calibration must be performed to keep the brightness of the electrode at the moment of the observation. A schematic illus- target’s temperature within the dynamic range of the high- tration of the test equipment is shown in Fig. 5. The appear- speed camera. In this study, the measurable temperature range ance of the mechanical shutter is shown in Fig. 6. The obser- of the optical system for two-color pyrometry was adjusted vation parameters were set to a frame rate of 2000 fps and a from 1100 to 1400 °C using a blackbody furnace. The cali- shutter speed for the high-speed camera of 1/2000 s. The strain bration results are shown in Fig. 4. These results indicate that at each point in time during the test was calculated based on the temperature measured using two-color pyrometry matches the relationship between the elapsed time since the initiation the temperature of the blackbody furnace. The error range of bending and the strain measured in advance using a strain between these temperatures was less than 5%. gauge. The temperatures at both tips of the crack for each point in time were calculated based on the two images obtain- ed using the two-color pyrometry technique. 2.3 Verification of temperature results obtained by two-color pyrometry 2.4.2 Conventional method The validity of the temperature results obtained by the two- color pyrometry method was verified by measuring them at For comparison, a Trans-Varestraint test was also conducted the solid-liquid interface of a pure Ni plate. The Ni plate was without a high-speed camera and the two-color pyrometry melted by applying a TIG arc at a current of 200 A for 20 s. method. The heat source conditions were the same as those After the TIG arc had been extinguished, the vicinity of the described above. The augmented strain values were 1.96, molten pool was observed and the temperature distribution 3.23, 4.76, and 5.88%. After the test, the specimen was ob- was measured. The observations were made at a frame rate served, and the BTR was estimated based on the measured of 2000 fps and a shutter speed of 1/2000 s. crack length and the cooling rate. 2.4 BTR results obtained in a Trans-Varestraint test 3 Results 2.4.1 Method using a high-speed camera and a two-color pyrometry technique 3.1 Validation of the temperature measurements using two-color pyrometry ATrans-Varestraint test was carried out to determine the BTR using a high-speed camera and a two-color pyrometry tech- The high-speed camera image and the temperature distribu- nique. The test conditions were as follows: TIG arc current = tion measured by two-color pyrometry in the vicinity of the 200 A; voltage = 15 V; torch traveling speed = 1.6 mm/s; and molten pool of the Ni plate are shown in Fig. 7. Temperatures maximum augmented strain = 5.88%. During the test, the vi- over 1050 °C can be measured using the two-color pyrometry cinity of the molten pool was observed, and the temperature technique. Figure 8 shows the temperature distribution for the distribution was measured using a high-speed camera with a straight line from the molten pool to the base metal indicated Table 2 Parameters for the Trans- Method Current [A] Voltage [V] Travel speed Augmented Ram travel Varestraint test [mm/s] strain [%] speed [mm/s] Conventional 200 14.3 ~ 15.3 1.6 1.96, 3.23, 4.76, 5.88 150 Two-color pyrometry 200 14.3 1.6 5.88 (max.) 150 Weld World Table 3 Parameters for the in-situ observations during the Trans-Varestraint test Method Frame rate [fps] Shutter speed [s] Band-pass filter [nm] Camera angle [°] Two-color pyrometry 2000 1/2000 850, 950 67 by an arrow in Fig. 7. This distribution indicates that the tem- 3.2.2 Temperatures at the tips of the longest crack and time perature decreases continuously from the molten pool to the elapsed since crack initiation base metal. The temperature at the solid-liquid boundary is approximately 1440 °C. Since the liquidus temperature of this In order to evaluate the material’s BTR, the temperatures at material as calculated using Thermo-Calc (database: SSOL4) both tips of the target crack were measured at each specified is 1453 °C, the difference between the two values is 0.9% point in time since the crack initiation. Figure 10 shows (13 °C). high-speed camera images that were passed through an 850-nm band-pass filter at crack initiation (strain, 0.35%), 0.003 s after crack initiation (strain, 0.47%), 0.012 s after 3.2 BTR results obtained in a Trans-Varestraint test crack initiation (strain, 1.63%), and 0.032 s after crack ini- using the two-color pyrometry method tiation (strain, 3.26%). These figures indicate that the cracks increase over time as the strain increases after crack 3.2.1 Observation image and temperature distribution initiation. The length of the target crack is determined by checking the fracture surface of the target crack observed Examples of the observation results for the Trans-Varestraint using an SEM (Fig. 11). The temperature distributions at test are shown in Fig. 9a, b. Figure 9a is the high-speed camera each point in time indicated in Fig. 10 are shown in Fig. 12. image obtained through an 850-nm band-pass filter at a strain The temperature distribution can be measured over the of 3.78%, while Fig. 9b is the image obtained through a 950- crack’s entire length. Figure. 12 shows the results obtained nm band-pass filter at the same time as Fig. 9a. It can be seen by continuously calculating the temperatures at both tips of that two images at different wavelengths were acquired. The the target crack at each specified point in time since crack temperature distribution calculated from these two images initiation. The liquidus temperature of the Ni-Fe alloy is using two-color pyrometry is shown in Fig. 9c. Figure 9d 1427 °C according to Thermo-Calc (database: SSOL4). shows the appearance of the test pieces at the same position The temperature of the crack at the time of crack initiation as that for Fig. 9a, b, c after the Trans-Varestraint test. As these was 1437 °C, and it can be seen that the cracks extend to the figures demonstrate, cracks extending from the molten pool— low temperature side as time passes. Cracking reached its especially the target crack, which is the longest crack in this maximum length during the test at 0.012 s after crack initi- test—can be identified by the high-speed camera image and ation. The corresponding temperature at both ends of the the temperature distribution. However, since the micro fis- cracks (at this time) was 1422 °C on the high-temperature sures (i.e., the micro cracks shown in Fig. 9d) that occur far side and 1157 °C on the low-temperature side. from the molten pool cannot be recognized on the high-speed Subsequently, the temperatures at both the upper end camera images due to the low brightness under the viewing (high-temperature side) and the lower end (low- conditions, these cracks were excluded from the temperature temperature side) of the crack did not change significantly calculations obtained from the two-color pyrometry technique until bending in the Trans-Varestraint test was finished. used in this study. Band-pass Fig. 3 Schematic illustration of filter the two-color pyrometry system Lens system Beam splitter Beam splitter Image sensor Optical system High-speed camera for two-color pyrometry Weld World 3.3 BTR results obtained by a conventional method Figure 15 shows the BTR determined from the Trans- Varestraint test (augmented strain 1.96, 3.23, 4.76, and 5.88%) conducted without using a high-speed camera and two-color pyrometry. The threshold strain for the crack that occurs is approximately 0.7%, and the BTR determined using this method is 330 °C (from 1427 to 1100 °C). 4 Relevance of BTR results obtained by a high-speed camera and two-color pyrometry The BTR is generally characterized by the crack generation limiting strain, the ε minimum, and the upper-end temperature Setting temperature in blackbody furnace ( C) (high-temperature side) and lower-end temperature (low- Fig. 4 Relationship between the temperature in the blackbody furnace temperature side) of the crack. The characteristics of the and the temperature measured by two-color pyrometry BTR determined based on the high-speed camera image and two-color pyrometry were compared with those of the BTR determined based on the results of the conventional Trans- 3.2.3 BTR results obtained by the two-color pyrometry Varestraint test in order to verify the effectiveness of the meth- method od proposed in this study. Based on the relationship between the time elapsed since the initiation of bending and strain, the strain at each specified 4.1 Minimum strain for crack initiation point in time since the crack initiation was calculated from the results shown in Fig. 13. The relationship between the With the conventional method, the minimum strain for crack temperature and the strain of the cracked area of the target initiation in the Trans-Varestraint test is generally estimated by crack is shown in Fig. 14. The strain at the time of crack extrapolating data in high-strain levels since it is difficult to initiation is 0.35%. The corresponding temperature in these test using low-strain levels, which may be the threshold for test results is 1437 °C. Therefore, the ε minimum, which is crack initiation. In this study, the minimum strain for crack the critical strain for solidification crack initiation, is 0.35%. initiation was approximately 0.7% based on the results obtain- From this point, the lower-end temperature of the solidifica- ed with an augmented strain of 1.96% or more. However, with tion crack decreases to 1157 °C at a strain of 1.63% and the method using a high-speed camera and a two-color py- remains nearly constant in the strain region beyond this point. rometry technique, the minimum strain was 0.35%, which was As a result, the BTR that indicates the solidification crack determined by the strain measured using a high-speed camera temperature range seems to be approximately 280 °C (from at the time of crack initiation. Since very few studies have 1157 to 1437 °C). been conducted on the minimum strain for crack initiation in Filter Shutter Weld Lens direction Weld TIG touch High-speed camera with direction Yoke Yoke two-color pyrometry system Specimen Bending block Fig. 5 Schematic illustration of the Trans-Varestraint test using a high-speed camera and two-color pyrometry Temperature measured by two-color pyrometry ( C) Weld World Fig. 6 BTR measured by an innovative method using a high-speed cam- era and two-color pyrometry Fig. 8 Temperature distribution for the line shown in Fig. 6 this type of material, it is difficult to ascertain what the true value is. However, it has been reported that the value for upper-end temperature measured using two-color pyrometry another type of Ni base alloy is approximately 0.1 to 0.4% was 1437 °C at crack initiation. This value is almost the same [8]. In addition, since it is not possible to measure the aug- as the liquidus temperature. Nevertheless, after crack initia- mented strain using the conventional method, the method tion, the temperature was much higher than the liquidus tem- using a high-speed camera and two-color pyrometry is con- perature. One of the reasons for this is that the temperature sidered to be more accurate. measured using two-color pyrometry indicates the tempera- ture of the molten pool that has flowed into the crack opening. 4.2 Upper-end temperature in BTR The upper end of the crack has a large width, and the inside of the crack is filled with a molten pool, as shown in Fig. 12. The upper-end temperature of the BTR roughly corresponds Since the temperature of the molten pool is higher than that of to the nominal liquidus temperature. With the conventional the solid-liquid interface, the temperature measured using method, the upper-end temperature of the BTR is often two-color pyrometry was a higher value. Generally, it is diffi- substituted with the liquidus temperature measured by cult to separate the melt inside the cracks from the crack end performing thermodynamic calculations. In this case, the tem- when using the two-color temperature measurement method, perature for the Ni-Fe alloy used in this study was 1427 °C so it is preferable to substitute the nominal liquidus tempera- according to Thermo-Calc (database: SSOL4). However, the ture for the upper-end temperature of the BTR regardless of the two-color temperature measurement method. 4.3 Lower-end temperature in BTR The lower-end temperature of the BTR corresponds roughly to the solidification completion temperature. However, it did not coincide with the nominal solidus temperature, and it has been reported to correspond to the solidification temperature for the melting point of a liquid film because a liquid film with a low melting point tends to remain due to the solidification segregation of solute elements, such as C, P, S, and B [9]. Consequently, determining the lower-end temperature of the BTR to a high degree of accuracy in the Trans-Varestraint test Image of high- Image of two-color is extremely important for understanding the solidification speed camera pyrometry brittleness characteristics of the material. In this study, the lower-end temperature measured using the two-color pyrom- Fig. 7 High-speed camera image and temperature distribution generated by two-color pyrometry around the molten pool of the pure Ni plate etry technique was approximately 1150 °C, which is almost Weld World Weld direction (a) (b) (c) (d) Micro cracks Target crack Target crack Target crack Target crack 2 mm 2 mm 2 mm 2 mm Fig. 9 Results of the Trans-Varestraint test. a Image from the high-speed camera (band-pass filter: 850 nm. (b) Image from the high-speed camera (band- pass filter, 950 nm). c Temperature distribution measured by two-color pyrometry. d Post-test appearance of the bead 50 °C higher than the temperature of approximately 1100 °C assumed that the temperature remains the same regardless that was measured using the conventional method. of the crack generation direction as long as the distances The reason for this difference is that the target crack was from the molten pool to the lower end of the crack are at an angle of approximately 37° with respect to the weld equal. In reality, however, the shape of the molten pool line. With the conventional method, the lower-end temper- becomes irregular due to turbulence in the arc column ature is calculated based on the relationship between the and fluctuations in the molten pool, and it is conceivable crack length and the cooling rate for TIG arc welding. that turbulence also occurs in the temperature distribution Since the cooling curve was obtained by inserting a ther- around the molten pool. With the conventional method, it mocouple into the molten pool, the temperature history is difficult to address this disturbance in the temperature was recorded only in the weld line direction. If the temper- distribution. With the two-color pyrometry technique, ature distribution around the molten pool is ideal, it can be however, the temperature at each observed position can Fig. 10 High-speed camera images taken during the Trans- Varestraint test Weld World Fig. 11 Fracture surface of the Target crack Weld direction target crack 1 mm Enlarged view A Enlarged view B be calculated directly. If cracks with the same length as that This value is almost the same as the temperature measured of the target crack occur in the weld line direction, the using the conventional method. lower-end temperature of the crack measured using the At least within the scope of this research, the conven- two-color pyrometry method is approximately 1090 °C. tional method was therefore found to include errors since Weld direction Target crack Target crack Target crack Target crack 2 mm 2 mm 2 mm 2 mm Time* : 0.000 s Time* : 0.003 s Time* : 0.012 s Time* : 0.032 s Augmented strain : 0.35% Augmented strain : 0.47% Augmented strain : 1.63% Augmented strain : 3.26% * Time elapsed since crack initiation Fig. 12 Temperature distribution generated by two-color pyrometry during the Trans-Varestraint test Weld World Solid circle : High temperature tip Open circle: Low temperature tip Liquidus temperature: 1,427 C Time elapsed since crack initiation (sec) Fig. 15 BTR measured by a conventional method Fig. 13 Relationship between the time elapsed since crack initiation and the temperature at the target crack tips 5 Conclusion In this study, a Trans-Varestraint test was conducted using a the influence of temperature fluctuations caused by the high-speed camera that incorporates a two-color optical sys- crack generation direction was not taken into consideration. tem to deliver greater accuracy in the measuring of the brittle- In contrast, it is possible to calculate the temperature at the ness temperature range (BTR) for a material. With this test, the crack generation position directly by using the two-color augmented strain at the time of solidification cracking initia- pyrometry technique. Consequently, the BTR results can tion was calculated based on the time elapsed since the initi- be said to be more reliable. According to Kadoi et al. ation of bending. In addition, the temperature range of the [10], the temperature gradient in laser welding in the direc- crack was directly measured using the two-color pyrometry tion of the angle with the weld line differs greatly from the technique. Using this technique, the minimum augmented temperature gradient in the weld direction. Therefore, the strain for crack initiation and the temperatures at both the lower-end temperature of the BTR results obtained using upper end (high temperature side) and lower end (low temper- the conventional method in a Trans-Varestraint test with ature side) of the cracks were measured. laser welding, which has a higher cooling rate than arc As a result, the following conclusions were drawn. welding, will tend to be unreliable, and the deviation from The temperature measurement results obtained using the the results obtained using the two-color temperature mea- two-color pyrometry method at the solid-liquid boundary of surement method will increase. the pure Ni plate melted using the TIG arc was approximately 1440 °C, which is very close to the liquidus temperature of 1453 °C. Therefore, the two-color temperature measurement method used in this study can be said to have obtained rea- sonable results in the high-temperature range. We succeeded in directly measuring the upper-end temper- ature and the lower-end temperature of the crack at each point BTR in time using the high-speed camera image and the two-color temperature measurement method in the Trans-Varestraint test. For the BTR calculated from the obtained data, the min- imum strain for crack initiation was 0.35%, the upper-end temperature of the crack was 1437 °C, the lower-end temper- ature of the crack was 1157 °C, and the temperature range was Min. strain = 0.35% 280 °C. The BTR determined from the Trans-Varestraint test con- Temperature ( C) ducted without using a high-speed camera image and the two- color temperature measurement method was 330 °C. The Fig. 14 BTR measured by an innovative method using a high-speed camera and two-color pyrometry upper-end temperature of the crack was 1428 °C, and the Strain (%) Temperature ( C) Weld World crack susceptibility for weld metals with Trans-Varestraint test. lower-end temperature was approximately 1100 °C. The Trans Jpn Weld Soc 2(2):141–162 lower-end crack temperature measured using the conventional 3. Shinozaki K (2002) Hot cracking in weld zone. J Jpn Weld Soc method was approximately 50 °C lower than that measured 71(6):455–459 using the two-color temperature measurement method. This is 4. Osuki T, Miyabe K, Hirata H, Ogawa K (2008) In-situ observation of solidification cracking during transverse varestraint test. possibly because the cooling curve used in calculating the Preprints of the National Meeting of J.W.S. in Japanese crack temperature did not take into account cracks being at a 5. Wang D, Kadoi K, Shinozaki K, Yamamoto M (2016) Evaluation certain angle to the weld line direction. With the two-color of solidification cracking susceptibility for austenitic stainless steel pyrometry technique, the crack temperature can be measured during laser Trans-Varestraint test using two dimensional tempera- directly regardless of the crack generation direction. ture measurement. ISIJ Int 56(11):2022–2028 6. Yamashita S, Yamamoto M, Shinozaki K, Kadoi K, Mitsui K, Usui Consequently, the lower-end temperature can be considered H (2015) In-situ temperature measurement using a multi-sensor to be more precise than the result obtained using the conven- camera during laser welding. Q J Jpn Weld Soc 33(2):93–97 tional method. 7. Ogawa Y (2011) High speed imaging technique part 1—high speed imaging of arc welding phenomena. Sci Technol Weld Join 16:33– Open Access This article is distributed under the terms of the Creative 8. Yushchenko K, Savchenko V, Chervyakov N, Zvyagintseva A, Commons Attribution 4.0 International License (http:// Guyot E (2011) Comparative hot cracking evaluation of welded creativecommons.org/licenses/by/4.0/), which permits unrestricted use, joints of alloy 690 using filler metals Inconel® 52 and 52 Mss. distribution, and reproduction in any medium, provided you give appro- Weld World 55:28–35 priate credit to the original author(s) and the source, provide a link to the 9. Saida K, Matsushita H, Nishimoto K, Kiuchi K, Nakayama J Creative Commons license, and indicate if changes were made. (2013) Quantitative influence of minor and impurity elements on hot cracking susceptibility of extra high-purity type 310 stainless steel. Yosetsu Gakkai Ronbunshu. Q J Jpn Weld Soc 31(1):56–65 (in Japanese) References 10. Sakoda S, Wang D, Kadoi K, Shinozaki K, Yamamoto M (2014) Investigation of evaluation method for trans-varestraint test with 1. DuPont JN, Lippold JC, Kiser SD (2009) Welding metallurgy and laser welding—investigation of hot cracking susceptibility during weldability of nickel-base alloys. Wiley, Hoboken laser welding (Part I). Preprints of the national meeting of J.W.S. in 2. Senda T, Matsuda F, Takano G, Watanabe K, Kobayashi T, Japanese Matsuzaka T (1971) Fundamental investigations on solidification http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Welding in the World Springer Journals

Determining the BTR by conducting a Trans-Varestraint test using a high-speed camera and two-color pyrometry

Free
10 pages
Loading next page...
 
/lp/springer_journal/determining-the-btr-by-conducting-a-trans-varestraint-test-using-a-gU3lK0kJAM
Publisher
Springer Berlin Heidelberg
Copyright
Copyright © 2018 by The Author(s)
Subject
Materials Science; Metallic Materials; Continuum Mechanics and Mechanics of Materials; Theoretical and Applied Mechanics
ISSN
0043-2288
eISSN
1878-6669
D.O.I.
10.1007/s40194-018-0608-4
Publisher site
See Article on Publisher Site

Abstract

The Ni-Fe alloy consumable that applies to the welding of 9% nickel steel for LNG storage tanks is highly susceptible to solidification cracks. To evaluate susceptibilities such as the brittleness temperature range (BTR), a Trans-Varestraint test is generally conducted. However, it is difficult to evaluate the minimum strain value for the BTR and the real temperature at both ends of a solidification crack in a conventional Trans-Varestraint test because these values are measured indirectly. In this study, we propose determining the temperature range by conducting in-situ observations during a Trans-Varestraint test using a high- speed camera and two-color pyrometry so that the temperature range can be measured directly from the temperatures at both ends of the crack. Furthermore, we measured the augmented strain from the time elapsed since the initiation of bending. This method allowed us to successfully measure the augmented strain and the temperature range in the Trans-Varestraint test and to determine the BTR more accurately. . . Keywords High-speed camera Two-color pyrometry Trans-Varestraint test 1 Introduction range (BTR). Solidification cracks will occur if the strain curve increases during cooling to the point where it crosses the BTR [1, 2]. Most LNG storage tanks are built using 9% nickel steel, which has a very high degree of strength and toughness. The welding One method that can be used to evaluate a material’sBTR consumable used for 9% nickel steel is generally a high-Ni is a Trans-Varestraint test. In this test, solidification cracks are alloy, such as Ni-Cr alloy, Ni-Mo alloy, or Ni-Fe alloy. These reproduced by forcibly bending and applying strain to the test types of Ni alloys are extremely tough when welded. material with the temperature distribution applied to the eval- However, they are highly susceptible to solidification cracks, uation material by a TIG arc. The BTR is defined by the and a high deposition rate welding technique, such as sub- relationship between the length of the longest solidification merged arc welding (SAW), is adopted for the tank’sfabrica- crack and the cooling rate for TIG welding [3]. tion. Therefore, evaluating solidification crack susceptibility However, the Trans-Varestraint test has some limitations. is of great importance in the fabrication of LNG storage tanks. Although the results of measuring the cooling rate for TIG One of the typical values that indicate susceptibility to so- welding by inserting thermocouples into the molten pool in- lidification cracks is ductility at an elevated temperature. dicate the temperature gradient within the weld direction, Some materials, especially Ni alloys, have a low-ductility some of the cracks reproduced by the Trans-Varestraint test temperature range as defined by the brittleness temperature may beextendedatanangle to the welddirection.Thismay produce a margin of error. In addition, to determine the limit- ing strain for the occurrence of cracks, it is necessary to per- Recommended for publication by Commission IX - Behaviour of Metals form the test many times using different augmented strains. It Subjected to Welding is difficult to accurately determine the limiting strain for the occurrence of cracks if only a limited number of tests are * Daisuke Abe daisuke_abe@ihi.co.jp conducted. In recent years, Osuki et al. reported the strain measure- 1 ment results obtained at crack initiation in a Trans-Varestraint IHI Corporation, 1, Shin-Nakahara-cho, Isogo-ku, test [4]. However, few studies have directly measured the Yokohama-shi, Kanagawa 235-8501, Japan Weld World temperature and/or the strain at crack initiation in a Trans- Varestraint test. Furthermore, Kadoi et al. reported the two- dimensional temperature distribution that they measured using a two-color pyrometry technique that was adopted for a laser Trans-Varestraint test [5, 6]. However, this technique was ap- plied to a laser rather than a TIG arc, which is a general heat source for the Trans-Varestraint test, and its cooling rate is quite different to that of laser welding. It is difficult to measure the temperature distribution during TIG welding because the arc light and bright tungsten electrode may interfere with the measurements. Therefore, the aim of this study is to determine the critical strain at crack initiation and the temperatures at both ends of Fig. 2 Dimensions of the test pieces for the Trans-Varestraint test the longest solidification crack in the Ni-Fe alloy by adopting a high-speed camera and a two-color pyrometry technique in a Trans-Varestraint test. First, a two-color pyrometry system chemical compositions of these two materials are shown in was developed to measure the temperature distribution during Tables 1, 2,and 3. TIG welding while avoiding the effects of the arc light and bright tungsten electrode. Next, the validity of the tempera- tures measured using the two-color pyrometry method was 2.2 Two-color pyrometry technique verified by measuring them at the solid-liquid interface of a pure Ni plate. Finally, a Trans-Varestraint test was carried out 2.2.1 Principles of the two-color pyrometry technique to determine the BTR by using a high-speed camera and a two-color pyrometry technique. As shown in Eq. (1) below, the brightness of the emitted light can be expressed as a function of the wavelength (λ), the temperature (T), and the emissivity (ε). However, since the emissivity varies depending on the object’smaterialand sur- 2 Materials and experimental procedure face conditions (e.g., its roughness and/or its state), it is diffi- cult to accurately measure its temperature from its brightness. 2.1 Materials However, by taking the luminance ratio (I /I ) for the two 1 2 wavelengths as shown in Eq. (2), it is possible to derive the Pure Ni plates were used to verify the validity of the temper- temperature independently of the emissivity (ε). Since the atures measured using the two-color pyrometry method. The two-color temperature measurement method utilizes this prin- test pieces used in the Trans-Varestraint test were made of Ni- ciple, it is necessary to acquire two images at different wave- Fe weld metal that had been welded by conducting submerged lengths for the measurement object [7]. arc welding with A5.14 ERNiMo-8 filler wire. A cross- I ¼ ε⋅ fðÞ λ; T ð1Þ sectional view of the weld metal is shown in Fig. 1.The dimensions were 150 mm × 50 mm × 10 mm (Fig. 2). The I ε⋅ fðÞ λ ; T 1 1 ¼ ð2Þ I ε⋅ fðÞ λ ; T 2 2 2.2.2 Components of the two-color pyrometry system Specimen In this study, an innovative optical system was developed and combined with a high-speed camera for two-color pyrometry. A schematic illustration of the system is shown in Fig. 3. With this optical system, the observed image was split into two images by a splitter. After that, they were passed through band-pass filters with a range of 850 and 950 nm. Finally, the two images were captured by one of the high-speed cam- era’s image sensors. The magnification used for this system Fig. 1 Cross-sectional view of the weld joint and specimen position for the Trans-Varestraint test was × 100. Weld World Table 1 Chemical compositions of the materials used Material Spec No. C Si Mn P S Cr Ni Mo Cu Fe W Pure Ni plate UNS N02201 0.01 0.03 0.09 – <0.001 – 99.71 – 0.04 0.12 – Weld metal (wire) AWS A5.14 ERNiMo-8 0.02 0.02 < 0.1 0.005 0.001 2.1 69.7 19.0 0.02 5.73 3.0 2.2.3 Calibration of the temperature two-color optical system placed on the side opposite the weld direction. To prevent the arc light and the reflected light of the Although the brightness of the emitted light varied widely in electrode from causing an error in the temperature for the two- accordance with the temperature of the target, the dynamic color pyrometry method, a mechanical shutter was mounted range of the high-speed camera was limited. Consequently, on the TIG torch. The shutter covered the arc light and the calibration must be performed to keep the brightness of the electrode at the moment of the observation. A schematic illus- target’s temperature within the dynamic range of the high- tration of the test equipment is shown in Fig. 5. The appear- speed camera. In this study, the measurable temperature range ance of the mechanical shutter is shown in Fig. 6. The obser- of the optical system for two-color pyrometry was adjusted vation parameters were set to a frame rate of 2000 fps and a from 1100 to 1400 °C using a blackbody furnace. The cali- shutter speed for the high-speed camera of 1/2000 s. The strain bration results are shown in Fig. 4. These results indicate that at each point in time during the test was calculated based on the temperature measured using two-color pyrometry matches the relationship between the elapsed time since the initiation the temperature of the blackbody furnace. The error range of bending and the strain measured in advance using a strain between these temperatures was less than 5%. gauge. The temperatures at both tips of the crack for each point in time were calculated based on the two images obtain- ed using the two-color pyrometry technique. 2.3 Verification of temperature results obtained by two-color pyrometry 2.4.2 Conventional method The validity of the temperature results obtained by the two- color pyrometry method was verified by measuring them at For comparison, a Trans-Varestraint test was also conducted the solid-liquid interface of a pure Ni plate. The Ni plate was without a high-speed camera and the two-color pyrometry melted by applying a TIG arc at a current of 200 A for 20 s. method. The heat source conditions were the same as those After the TIG arc had been extinguished, the vicinity of the described above. The augmented strain values were 1.96, molten pool was observed and the temperature distribution 3.23, 4.76, and 5.88%. After the test, the specimen was ob- was measured. The observations were made at a frame rate served, and the BTR was estimated based on the measured of 2000 fps and a shutter speed of 1/2000 s. crack length and the cooling rate. 2.4 BTR results obtained in a Trans-Varestraint test 3 Results 2.4.1 Method using a high-speed camera and a two-color pyrometry technique 3.1 Validation of the temperature measurements using two-color pyrometry ATrans-Varestraint test was carried out to determine the BTR using a high-speed camera and a two-color pyrometry tech- The high-speed camera image and the temperature distribu- nique. The test conditions were as follows: TIG arc current = tion measured by two-color pyrometry in the vicinity of the 200 A; voltage = 15 V; torch traveling speed = 1.6 mm/s; and molten pool of the Ni plate are shown in Fig. 7. Temperatures maximum augmented strain = 5.88%. During the test, the vi- over 1050 °C can be measured using the two-color pyrometry cinity of the molten pool was observed, and the temperature technique. Figure 8 shows the temperature distribution for the distribution was measured using a high-speed camera with a straight line from the molten pool to the base metal indicated Table 2 Parameters for the Trans- Method Current [A] Voltage [V] Travel speed Augmented Ram travel Varestraint test [mm/s] strain [%] speed [mm/s] Conventional 200 14.3 ~ 15.3 1.6 1.96, 3.23, 4.76, 5.88 150 Two-color pyrometry 200 14.3 1.6 5.88 (max.) 150 Weld World Table 3 Parameters for the in-situ observations during the Trans-Varestraint test Method Frame rate [fps] Shutter speed [s] Band-pass filter [nm] Camera angle [°] Two-color pyrometry 2000 1/2000 850, 950 67 by an arrow in Fig. 7. This distribution indicates that the tem- 3.2.2 Temperatures at the tips of the longest crack and time perature decreases continuously from the molten pool to the elapsed since crack initiation base metal. The temperature at the solid-liquid boundary is approximately 1440 °C. Since the liquidus temperature of this In order to evaluate the material’s BTR, the temperatures at material as calculated using Thermo-Calc (database: SSOL4) both tips of the target crack were measured at each specified is 1453 °C, the difference between the two values is 0.9% point in time since the crack initiation. Figure 10 shows (13 °C). high-speed camera images that were passed through an 850-nm band-pass filter at crack initiation (strain, 0.35%), 0.003 s after crack initiation (strain, 0.47%), 0.012 s after 3.2 BTR results obtained in a Trans-Varestraint test crack initiation (strain, 1.63%), and 0.032 s after crack ini- using the two-color pyrometry method tiation (strain, 3.26%). These figures indicate that the cracks increase over time as the strain increases after crack 3.2.1 Observation image and temperature distribution initiation. The length of the target crack is determined by checking the fracture surface of the target crack observed Examples of the observation results for the Trans-Varestraint using an SEM (Fig. 11). The temperature distributions at test are shown in Fig. 9a, b. Figure 9a is the high-speed camera each point in time indicated in Fig. 10 are shown in Fig. 12. image obtained through an 850-nm band-pass filter at a strain The temperature distribution can be measured over the of 3.78%, while Fig. 9b is the image obtained through a 950- crack’s entire length. Figure. 12 shows the results obtained nm band-pass filter at the same time as Fig. 9a. It can be seen by continuously calculating the temperatures at both tips of that two images at different wavelengths were acquired. The the target crack at each specified point in time since crack temperature distribution calculated from these two images initiation. The liquidus temperature of the Ni-Fe alloy is using two-color pyrometry is shown in Fig. 9c. Figure 9d 1427 °C according to Thermo-Calc (database: SSOL4). shows the appearance of the test pieces at the same position The temperature of the crack at the time of crack initiation as that for Fig. 9a, b, c after the Trans-Varestraint test. As these was 1437 °C, and it can be seen that the cracks extend to the figures demonstrate, cracks extending from the molten pool— low temperature side as time passes. Cracking reached its especially the target crack, which is the longest crack in this maximum length during the test at 0.012 s after crack initi- test—can be identified by the high-speed camera image and ation. The corresponding temperature at both ends of the the temperature distribution. However, since the micro fis- cracks (at this time) was 1422 °C on the high-temperature sures (i.e., the micro cracks shown in Fig. 9d) that occur far side and 1157 °C on the low-temperature side. from the molten pool cannot be recognized on the high-speed Subsequently, the temperatures at both the upper end camera images due to the low brightness under the viewing (high-temperature side) and the lower end (low- conditions, these cracks were excluded from the temperature temperature side) of the crack did not change significantly calculations obtained from the two-color pyrometry technique until bending in the Trans-Varestraint test was finished. used in this study. Band-pass Fig. 3 Schematic illustration of filter the two-color pyrometry system Lens system Beam splitter Beam splitter Image sensor Optical system High-speed camera for two-color pyrometry Weld World 3.3 BTR results obtained by a conventional method Figure 15 shows the BTR determined from the Trans- Varestraint test (augmented strain 1.96, 3.23, 4.76, and 5.88%) conducted without using a high-speed camera and two-color pyrometry. The threshold strain for the crack that occurs is approximately 0.7%, and the BTR determined using this method is 330 °C (from 1427 to 1100 °C). 4 Relevance of BTR results obtained by a high-speed camera and two-color pyrometry The BTR is generally characterized by the crack generation limiting strain, the ε minimum, and the upper-end temperature Setting temperature in blackbody furnace ( C) (high-temperature side) and lower-end temperature (low- Fig. 4 Relationship between the temperature in the blackbody furnace temperature side) of the crack. The characteristics of the and the temperature measured by two-color pyrometry BTR determined based on the high-speed camera image and two-color pyrometry were compared with those of the BTR determined based on the results of the conventional Trans- 3.2.3 BTR results obtained by the two-color pyrometry Varestraint test in order to verify the effectiveness of the meth- method od proposed in this study. Based on the relationship between the time elapsed since the initiation of bending and strain, the strain at each specified 4.1 Minimum strain for crack initiation point in time since the crack initiation was calculated from the results shown in Fig. 13. The relationship between the With the conventional method, the minimum strain for crack temperature and the strain of the cracked area of the target initiation in the Trans-Varestraint test is generally estimated by crack is shown in Fig. 14. The strain at the time of crack extrapolating data in high-strain levels since it is difficult to initiation is 0.35%. The corresponding temperature in these test using low-strain levels, which may be the threshold for test results is 1437 °C. Therefore, the ε minimum, which is crack initiation. In this study, the minimum strain for crack the critical strain for solidification crack initiation, is 0.35%. initiation was approximately 0.7% based on the results obtain- From this point, the lower-end temperature of the solidifica- ed with an augmented strain of 1.96% or more. However, with tion crack decreases to 1157 °C at a strain of 1.63% and the method using a high-speed camera and a two-color py- remains nearly constant in the strain region beyond this point. rometry technique, the minimum strain was 0.35%, which was As a result, the BTR that indicates the solidification crack determined by the strain measured using a high-speed camera temperature range seems to be approximately 280 °C (from at the time of crack initiation. Since very few studies have 1157 to 1437 °C). been conducted on the minimum strain for crack initiation in Filter Shutter Weld Lens direction Weld TIG touch High-speed camera with direction Yoke Yoke two-color pyrometry system Specimen Bending block Fig. 5 Schematic illustration of the Trans-Varestraint test using a high-speed camera and two-color pyrometry Temperature measured by two-color pyrometry ( C) Weld World Fig. 6 BTR measured by an innovative method using a high-speed cam- era and two-color pyrometry Fig. 8 Temperature distribution for the line shown in Fig. 6 this type of material, it is difficult to ascertain what the true value is. However, it has been reported that the value for upper-end temperature measured using two-color pyrometry another type of Ni base alloy is approximately 0.1 to 0.4% was 1437 °C at crack initiation. This value is almost the same [8]. In addition, since it is not possible to measure the aug- as the liquidus temperature. Nevertheless, after crack initia- mented strain using the conventional method, the method tion, the temperature was much higher than the liquidus tem- using a high-speed camera and two-color pyrometry is con- perature. One of the reasons for this is that the temperature sidered to be more accurate. measured using two-color pyrometry indicates the tempera- ture of the molten pool that has flowed into the crack opening. 4.2 Upper-end temperature in BTR The upper end of the crack has a large width, and the inside of the crack is filled with a molten pool, as shown in Fig. 12. The upper-end temperature of the BTR roughly corresponds Since the temperature of the molten pool is higher than that of to the nominal liquidus temperature. With the conventional the solid-liquid interface, the temperature measured using method, the upper-end temperature of the BTR is often two-color pyrometry was a higher value. Generally, it is diffi- substituted with the liquidus temperature measured by cult to separate the melt inside the cracks from the crack end performing thermodynamic calculations. In this case, the tem- when using the two-color temperature measurement method, perature for the Ni-Fe alloy used in this study was 1427 °C so it is preferable to substitute the nominal liquidus tempera- according to Thermo-Calc (database: SSOL4). However, the ture for the upper-end temperature of the BTR regardless of the two-color temperature measurement method. 4.3 Lower-end temperature in BTR The lower-end temperature of the BTR corresponds roughly to the solidification completion temperature. However, it did not coincide with the nominal solidus temperature, and it has been reported to correspond to the solidification temperature for the melting point of a liquid film because a liquid film with a low melting point tends to remain due to the solidification segregation of solute elements, such as C, P, S, and B [9]. Consequently, determining the lower-end temperature of the BTR to a high degree of accuracy in the Trans-Varestraint test Image of high- Image of two-color is extremely important for understanding the solidification speed camera pyrometry brittleness characteristics of the material. In this study, the lower-end temperature measured using the two-color pyrom- Fig. 7 High-speed camera image and temperature distribution generated by two-color pyrometry around the molten pool of the pure Ni plate etry technique was approximately 1150 °C, which is almost Weld World Weld direction (a) (b) (c) (d) Micro cracks Target crack Target crack Target crack Target crack 2 mm 2 mm 2 mm 2 mm Fig. 9 Results of the Trans-Varestraint test. a Image from the high-speed camera (band-pass filter: 850 nm. (b) Image from the high-speed camera (band- pass filter, 950 nm). c Temperature distribution measured by two-color pyrometry. d Post-test appearance of the bead 50 °C higher than the temperature of approximately 1100 °C assumed that the temperature remains the same regardless that was measured using the conventional method. of the crack generation direction as long as the distances The reason for this difference is that the target crack was from the molten pool to the lower end of the crack are at an angle of approximately 37° with respect to the weld equal. In reality, however, the shape of the molten pool line. With the conventional method, the lower-end temper- becomes irregular due to turbulence in the arc column ature is calculated based on the relationship between the and fluctuations in the molten pool, and it is conceivable crack length and the cooling rate for TIG arc welding. that turbulence also occurs in the temperature distribution Since the cooling curve was obtained by inserting a ther- around the molten pool. With the conventional method, it mocouple into the molten pool, the temperature history is difficult to address this disturbance in the temperature was recorded only in the weld line direction. If the temper- distribution. With the two-color pyrometry technique, ature distribution around the molten pool is ideal, it can be however, the temperature at each observed position can Fig. 10 High-speed camera images taken during the Trans- Varestraint test Weld World Fig. 11 Fracture surface of the Target crack Weld direction target crack 1 mm Enlarged view A Enlarged view B be calculated directly. If cracks with the same length as that This value is almost the same as the temperature measured of the target crack occur in the weld line direction, the using the conventional method. lower-end temperature of the crack measured using the At least within the scope of this research, the conven- two-color pyrometry method is approximately 1090 °C. tional method was therefore found to include errors since Weld direction Target crack Target crack Target crack Target crack 2 mm 2 mm 2 mm 2 mm Time* : 0.000 s Time* : 0.003 s Time* : 0.012 s Time* : 0.032 s Augmented strain : 0.35% Augmented strain : 0.47% Augmented strain : 1.63% Augmented strain : 3.26% * Time elapsed since crack initiation Fig. 12 Temperature distribution generated by two-color pyrometry during the Trans-Varestraint test Weld World Solid circle : High temperature tip Open circle: Low temperature tip Liquidus temperature: 1,427 C Time elapsed since crack initiation (sec) Fig. 15 BTR measured by a conventional method Fig. 13 Relationship between the time elapsed since crack initiation and the temperature at the target crack tips 5 Conclusion In this study, a Trans-Varestraint test was conducted using a the influence of temperature fluctuations caused by the high-speed camera that incorporates a two-color optical sys- crack generation direction was not taken into consideration. tem to deliver greater accuracy in the measuring of the brittle- In contrast, it is possible to calculate the temperature at the ness temperature range (BTR) for a material. With this test, the crack generation position directly by using the two-color augmented strain at the time of solidification cracking initia- pyrometry technique. Consequently, the BTR results can tion was calculated based on the time elapsed since the initi- be said to be more reliable. According to Kadoi et al. ation of bending. In addition, the temperature range of the [10], the temperature gradient in laser welding in the direc- crack was directly measured using the two-color pyrometry tion of the angle with the weld line differs greatly from the technique. Using this technique, the minimum augmented temperature gradient in the weld direction. Therefore, the strain for crack initiation and the temperatures at both the lower-end temperature of the BTR results obtained using upper end (high temperature side) and lower end (low temper- the conventional method in a Trans-Varestraint test with ature side) of the cracks were measured. laser welding, which has a higher cooling rate than arc As a result, the following conclusions were drawn. welding, will tend to be unreliable, and the deviation from The temperature measurement results obtained using the the results obtained using the two-color temperature mea- two-color pyrometry method at the solid-liquid boundary of surement method will increase. the pure Ni plate melted using the TIG arc was approximately 1440 °C, which is very close to the liquidus temperature of 1453 °C. Therefore, the two-color temperature measurement method used in this study can be said to have obtained rea- sonable results in the high-temperature range. We succeeded in directly measuring the upper-end temper- ature and the lower-end temperature of the crack at each point BTR in time using the high-speed camera image and the two-color temperature measurement method in the Trans-Varestraint test. For the BTR calculated from the obtained data, the min- imum strain for crack initiation was 0.35%, the upper-end temperature of the crack was 1437 °C, the lower-end temper- ature of the crack was 1157 °C, and the temperature range was Min. strain = 0.35% 280 °C. The BTR determined from the Trans-Varestraint test con- Temperature ( C) ducted without using a high-speed camera image and the two- color temperature measurement method was 330 °C. The Fig. 14 BTR measured by an innovative method using a high-speed camera and two-color pyrometry upper-end temperature of the crack was 1428 °C, and the Strain (%) Temperature ( C) Weld World crack susceptibility for weld metals with Trans-Varestraint test. lower-end temperature was approximately 1100 °C. The Trans Jpn Weld Soc 2(2):141–162 lower-end crack temperature measured using the conventional 3. Shinozaki K (2002) Hot cracking in weld zone. J Jpn Weld Soc method was approximately 50 °C lower than that measured 71(6):455–459 using the two-color temperature measurement method. This is 4. Osuki T, Miyabe K, Hirata H, Ogawa K (2008) In-situ observation of solidification cracking during transverse varestraint test. possibly because the cooling curve used in calculating the Preprints of the National Meeting of J.W.S. in Japanese crack temperature did not take into account cracks being at a 5. Wang D, Kadoi K, Shinozaki K, Yamamoto M (2016) Evaluation certain angle to the weld line direction. With the two-color of solidification cracking susceptibility for austenitic stainless steel pyrometry technique, the crack temperature can be measured during laser Trans-Varestraint test using two dimensional tempera- directly regardless of the crack generation direction. ture measurement. ISIJ Int 56(11):2022–2028 6. Yamashita S, Yamamoto M, Shinozaki K, Kadoi K, Mitsui K, Usui Consequently, the lower-end temperature can be considered H (2015) In-situ temperature measurement using a multi-sensor to be more precise than the result obtained using the conven- camera during laser welding. Q J Jpn Weld Soc 33(2):93–97 tional method. 7. Ogawa Y (2011) High speed imaging technique part 1—high speed imaging of arc welding phenomena. Sci Technol Weld Join 16:33– Open Access This article is distributed under the terms of the Creative 8. Yushchenko K, Savchenko V, Chervyakov N, Zvyagintseva A, Commons Attribution 4.0 International License (http:// Guyot E (2011) Comparative hot cracking evaluation of welded creativecommons.org/licenses/by/4.0/), which permits unrestricted use, joints of alloy 690 using filler metals Inconel® 52 and 52 Mss. distribution, and reproduction in any medium, provided you give appro- Weld World 55:28–35 priate credit to the original author(s) and the source, provide a link to the 9. Saida K, Matsushita H, Nishimoto K, Kiuchi K, Nakayama J Creative Commons license, and indicate if changes were made. (2013) Quantitative influence of minor and impurity elements on hot cracking susceptibility of extra high-purity type 310 stainless steel. Yosetsu Gakkai Ronbunshu. Q J Jpn Weld Soc 31(1):56–65 (in Japanese) References 10. Sakoda S, Wang D, Kadoi K, Shinozaki K, Yamamoto M (2014) Investigation of evaluation method for trans-varestraint test with 1. DuPont JN, Lippold JC, Kiser SD (2009) Welding metallurgy and laser welding—investigation of hot cracking susceptibility during weldability of nickel-base alloys. Wiley, Hoboken laser welding (Part I). Preprints of the national meeting of J.W.S. in 2. Senda T, Matsuda F, Takano G, Watanabe K, Kobayashi T, Japanese Matsuzaka T (1971) Fundamental investigations on solidification

Journal

Welding in the WorldSpringer Journals

Published: Jun 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off