The National Academies Press

Currently Skimming:

Pages 14-71

The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.

From page 14... ... 14 Findings and Application 3.1 Critical Flaw Size Analytical parametric studies were performed in order to establish the critical flaw sizes considered rejectable for typical bridge CJP welds. The results were used to identify the critical flaw size that must be reliably detected and rejected to establish revised acceptance and rejection criteria. Read the entire page →
From page 15... ... 15 for fracture (no length requirement for fracture) and the length limit for fatigue (no height requirement for fatigue) Read the entire page →
From page 16... ... 16 3.1.2.1 Cyclic Stress Range To determine the stress range for evaluation, a reasonable approach is to use the stress range associated with infinite life; for example, using a stress range of 16 ksi for a Category B butt weld detail. However, the stress range associated with the CAFL (constant amplitude fatigue limit) Read the entire page →
From page 17... ... 17 the effects of various flaw types on the performance of CJP welds. While finite element models were developed to evaluate the effects of plasticity for transition welds, the Option 1 failure assessment diagram (FAD) Read the entire page →
From page 18... ... 18 permissible thickness for a certain grade of steel was smaller. The recommended TK term of +25°C was used to account for the scatter in the Charpy correlation. Read the entire page →
From page 19... ... 19 flaws. The second level was equal to two-thirds (66%) Read the entire page →
From page 20... ... 20 range results given in Table 11 matched the target surface and embedded flaws for 4″ thick plates given in Table 9 and Table 10. The 4 ksi stress range results given in Table 11 are larger than the target surface and embedded flaws for 4″ thick plates given in Table 9 and Table 10 since these cases were controlled by fracture rather than fatigue crack growth. Read the entire page →
From page 21... ... 21 flaw length tends to be a bit longer for the ISO acceptance criteria. As the plates become thicker, the ISO acceptance criteria allows larger flaws than were calculated during this project. Read the entire page →
From page 22... ... 22 In order to estimate the effect of the stress concentration on the fatigue resistance of transition welds, a polynomial trendline was fit to the results from the three transitions with the greatest SCF -- 1″ to 1.5″, 1″ to 2″, and 2″ to 4″ transitions -- to obtain the following result, where t is the normalized depth in the plate, with t = 0 on the transition-side face and t = 1 on the opposite face: 1.6587 4.8158 5.0949 3.0826 1.79214 3 2( ) = − + − +SCF t t t t tFlange The SCF estimation was then multiplied by the various nominal stresses shown in Table 6 to obtain a through-thickness fatigue stress profile for each nominal stress level. Read the entire page →
From page 23... ... 23 the SCF below the surface. When all of the thickness transitions were plotted, it was discovered that the thickness transitions with the highest SCF were nearly linear for points between the maximum SCF and the midpoint of the plate. Read the entire page →
From page 24... ... 24 evaluated. Therefore, these results should be valid for typical thickness transitions, regardless of the thickness of the thicker plate. Read the entire page →
From page 25... ... 25 scans (i.e., face and side of weld scanned and corresponding index offset) ; incidence angle range; angular sweep increment; calibration/TCG block details; equipment and transducer make, model, and settings; along with any other information included in AWS D1.5 Table K.2. Read the entire page →
From page 26... ... 26 3.2.2 Flaw Detection and Location The accuracy in reported flaw location was very poor for many of the PAUT and conventional UT technicians. Thus, a "hit" (simply defined as the technician noting that they detected an indication which matched a known flaw) Read the entire page →
From page 27... ... 27 Flaw rejection from TOFD inspections cannot be compared to PAUT or conventional UT inspections since there is no acceptance criteria for this NDT technique in AWS D1.5. For the flaw to be considered rejected, it had to be located correctly as well as meeting any other criteria for rejection included in AWS D1.5. Read the entire page →
From page 28... ... 28 Specimen Details PAUT1 Conventional UT TOFD Flaw ID Flaw Type PAUT1 PAUT2 PAUT3 PAUT4 PAUT Avg UT1 UT2 UT3 UT4 UT5 UT Avg TOFD1 TOFD2 TOFD Avg 1 LOF 1 0 1 1 0.75 0 1 0 1 1 0.6 0 1 0.5 2 LOF 1 1 1 1 1 1 1 1 1 1 1 1 3 LOF 1 1 1 1 1 1 1 1 1 1 1 1 4 Toe Crack 0 0 1 1 0.5 0 0 0 0 0 0 1 0 0.5 5 Crack 1 0 0.5 1 1 1 1 1 1 1 1 1 6 Crack 1 0 0.5 1 1 1 1 1 1 1 1 1 7 Crack 1 0 0.5 1 1 1 0 1 0.8 1 1 1 8 LOF 1 1 0 1 0.75 1 0 1 1 1 0.8 1 1 1 9 LOF 1 0 0 1 0.5 0 1 1 0 0 0.4 1 1 1 10 LOF 1 0 1 1 0.75 1 0 1 1 1 0.8 1 1 1 11 LOF 1 0 1 0 0.5 1 0 1 1 1 0.8 1 1 1 12 Porosity 0 0 1 1 0.5 0 1 1 1 1 0.8 0 1 0.5 13 Slag 0 1 0.5 1 1 1 1 1 1 0 1 0.5 14 Slag 1 0 0.5 1 1 1 1 1 1 0 1 0.5 15 Slag 1 1 1 1 1 1 1 1 1 0 1 0.5 16 Porosity 1 1 1 1 1 0 1 1 0.8 1 1 1 17 Slag 1 1 1 1 1 1 1 1 1 1 0 0.5 18 Slag 0 0 0 0 0 0 0 0 0 0 0 0 1 0.5 19 Slag 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Total Hits 15 2 6 13 36 14 14 15 15 16 74 13 17 30 Hit Rate Avg. 0.79 0.22 0.67 0.68 0.64 0.74 0.74 0.79 0.79 0.84 0.78 0.68 0.89 0.79 1Cells blacked out were not tested by technician Table 18. Read the entire page →
From page 29... ... 29 For instance, flaws with lower classifications according to PAUT as compared to conventional UT include Flaw 1 and Flaws 12–15. These include a small LOF flaw, a group of porosity flaws, and three slag inclusions. Read the entire page →
From page 30... ... 30 Specimen Details PAUT Annex K1 Conventional UT Flaw Type PAUT1 PAUT2 PAUT3 PAUT4 PAUT Avg UT1 UT2 UT3 UT4 UT5 UT Avg 1 LOF 0 0 0 12 0.25 0 1 0 1 1 0.6 2 LOF 1 1 1 1 1 1 1 1 1 3 LOF 1 1 1 1 1 1 1 1 1 4 Toe Crack 0 0 0 0 0 0 0 0 0 0 0 5 Crack 1 0 0.5 1 1 1 1 1 1 6 Crack 1 0 0.5 1 1 1 1 1 1 7 Crack 1 0 0.5 1 1 1 0 1 0.8 8 LOF 1 1 0 1 0.75 1 0 1 1 1 0.8 9 LOF 1 0 0 1 0.5 0 1 1 0 0 0.4 10 LOF 1 0 1 1 0.75 1 0 1 1 1 0.8 11 LOF 1 0 1 0 0.5 1 0 1 1 1 0.8 12 Porosity 0 0 0 1 0.25 0 1 1 1 1 0.8 13 Slag 0 1 0.5 1 1 1 1 1 1 14 Slag 12 0 0.5 1 1 1 1 1 1 15 Slag 0 0 0 1 1 1 1 1 1 16 Porosity 1 1 1 1 1 0 1 1 0.8 17 Slag 1 1 1 1 1 1 1 1 1 18 Slag 0 0 0 0 0 0 0 0 0 0 0 19 Slag 1 1 1 1 1 1 1 1 1 1 1 Total Rejected 13 2 3 11 29 14 14 15 15 16 74 Rejection Rate 0.68 0.22 0.33 0.58 0.52 0.74 0.74 0.79 0.79 0.84 0.78 1Cells blacked out were not tested by technician 2Rejected due to crack classification rather than amplitude and length Flaw ID Table 22. Rejection rate (reject "1"/accept "0") Read the entire page →
From page 31... ... 31 seemed to have groups of sparse porosity intermittently within some of the plates, as shown in the digital RT images in Appendix D Therefore, indications that overlap with these unintended weld flaws should not be indicated as false calls. Read the entire page →
From page 32... ... 32 PAUT and TOFD when length and height sizing were both required to be within one-half to twice the actual dimension, along with being properly located. Due to the large inaccuracies with height sizing and the lack of reported flaw height for TOFD results, no PAUT or TOFD technicians were close to passing the minimum performance requirements. Read the entire page →
From page 33... ... 33 coupled with line scanning using an encoder is prudent. This will help ensure that the small flaws that were accepted in the round robin when using line scanning alone would be rejected in practice. Read the entire page →
From page 34... ... 34 Nine steel specimens were fabricated and tested using conventional UT and PAUT. Table 28 outlines the samples tested and their properties. Read the entire page →
From page 36... ... 36 The variability in ultrasonic inspection of NGI-ESW welds was then assessed following the evaluation of base metal. Unlike the consistent microstructure of base metal, welding produces different zones of varying grain structures. Read the entire page →
From page 37... ... 37 While limiting the probe frequency may address the attenuation issues associated with base metal, the attenuation characteristics of the weld itself, in particular attenuation associated with NGI-ESW, was found to be quite high. In this case, use of a lower frequency, such as 2.25 MHz, still resulted in a large loss of amplitude from sound passing through the weld metal. Read the entire page →
From page 38... ... 38 be made in the far field in order to ensure that the change in amplitude is due to attenuation. API RP 2X requires that the single-V-path-measurement sound path exceeds 4 inches. Read the entire page →
From page 39... ... 39 The research team recommends that additional requirements be provided to account for the significant loss of amplitude found when sound propagates through coarse grained NGI- ESW welds. These requirements include verifying the amplitude and location of a 1.5 mm (0.06″) Read the entire page →
From page 40... ... 40 The QT specimens -- specimens 70 and 101 -- have a very low anisotropic ratio. In comparison, it is clear that all of the TMCP specimens demonstrate high anisotropic ratio. Read the entire page →
From page 41... ... 41 for all sound paths with the 70° incidence angle. Therefore, while limiting the incidence angle range to 40°–60° will lessen the impact from the changes in the shear wave velocity, it will not limit the amplitude differences to 2 dB or less over all possible sound paths. Read the entire page →
From page 42... ... 42 orientation of the probe that will be used during scanning. For instance, if the PAUT line scanning will be performed with the probe oriented along the rolled direction, then the measurement of the shear wave velocity or incidence angle shall be performed in the same direction. Read the entire page →
From page 43... ... 43 Figure 13. 0.6" deep SDH in 45ç TMCP plate. Read the entire page →
From page 44... ... 44 to the results of the rolled direction for the 1″ deep hole but was lower than the rolled direction for the 0.6″ hole. Based on these results, the research team recommends additional requirements for acoustically anisotropic materials. Read the entire page →
From page 45... ... 45 surface breaking flaw. CIVA analysis was performed to determine the minimum ligament for the SDH from the surface of the plate in order to use it for skipping off of the backwall. Read the entire page →
From page 46... ... 46 placement of ultrasonic transducers for transfer correction and TCG calibration are shown in Figure 9 and Figure 16, respectively. Finally, the research team recommends that Annex K require the calibration block to be similar in temperature to the test object when calibration is performed. Read the entire page →
From page 47... ... 47 ness or thickness transition) Read the entire page →
From page 48... ... 48 accept an indication that was up to 5 dB above reference [i.e., 5 dB greater amplitude than the 1.5 mm (0.06″) diameter SDH] Read the entire page →
From page 49... ... 49 conservative than AWS D1.5 Annex K since the CIVA results from –6 dB to –13 dB correlate to the conventional UT limits while they are in the acceptable range for Annex K Although it may seem that the CIVA results are overly conservative when compared to Annex K, they correlate well to the conventional UT acceptance criteria, which appears to have provided good historical performance when used for UT inspection. Read the entire page →
From page 50... ... 50 amplitude limits for the various incidence angles in the fixed attenuation tables along with correction for the difference in true attenuation compared to the fixed attenuation model [14–16] Read the entire page →
From page 51... ... 51 the critical volumetric flaw used in the CIVA analysis for the 4 ksi thickness transition and 8 ksi equal thickness cases. This flaw would have been rejectable according to the amplitude limits found during the CIVA analysis for these cases. Read the entire page →
From page 52... ... 52 The maximum amplitude and measured length were used to evaluate each flaw using five different criteria based on the CIVA results: 1. Rejection of flaws with maximum amplitude ≥ –13 dB 2. Read the entire page →
From page 53... ... 53 measurement for flaws use the 6 dB drop method during the manual raster scan. Some PAUT acceptance criteria use a standard amplitude limit for length measurement rather than the 6 dB drop. Read the entire page →
From page 54... ... 54 plate and scanning from both weld faces. A reasonable limit may be limiting the sound path used for full coverage to 12″, since this would still allow for full coverage to be provided at the 70° incidence angles for the 2nd leg in 2″ thick plates (i.e., 4″ deep TCG point) Read the entire page →
From page 55... ... 55 longer within the sound beam coverage, at which point the amplitude drops quickly (0″ index offset) Read the entire page →
From page 56... ... 56 the 0.20″ × 0.20″ embedded flaw with 45° tilt has a maximum amplitude of +9 dB from raster scanning, but a minimum peak amplitude during possible line scan locations of –12 dB even with full coverage from two crossing directions. Therefore, the amplitude of this flaw could be 21 dB below the maximum during the line scanning, even with providing full coverage in two crossing directions. Read the entire page →
From page 57... ... 57 incidence angular range was –6 dB and over the 45°–50° incidence angular range was +4 dB. Obviously, the amplitude of this flaw is very sensitive to the incidence angle, as is typical for tilted lack-of-fusion flaws. Read the entire page →
From page 58... ... 58 that the flaw was aligned parallel to the weld axis and, therefore, the probe was perfectly perpendicular to the weld flaw. While this is a valid assumption for raster scanning where the probe will be rotated as well as translated, it may be unconservative for line scanning. Read the entire page →
From page 59... ... 59 -35 -30 -25 -20 -15 -10 -5 0 5 10 0 5 10 15 20 25 Ch an ge fr om R ef er en ce A m pl itu de (d B) Read the entire page →
From page 60... ... 60 for 10° skew. In general, including the 5° skew resulted in a –2 dB to –3 dB decrease in the flaw detection amplitude while the decrease was –9 dB to –12 dB for 10° skew. Read the entire page →
From page 61... ... 61 for any flaw detection limit of –14 dB or less. All of the flaw detection limits for planar flaws determined through the CIVA analysis and presented in Table 39 are below –14 dB, so use of the CIVA results would have resulted in all of the intended flaws to be detected. Read the entire page →
From page 62... ... 62 low incidence angle (40°) , as shown in Figure 31 (right) Read the entire page →
From page 63... ... 63 of harmless indications which require follow-up raster scanning is minimal. According to the CIVA analysis, the –18 dB limit would overestimate the lowest possible amplitude from a critical flaw with skew. Read the entire page →
From page 64... ... 64 Drawing Details Conv. UT Rejection Rate Raster Scan Results Max Amp (dB) Read the entire page →
From page 65... ... 65 welds can be determined using the 6 dB drop method on the encoded line scan results similar to the current Annex K requirements. A marked-up version of Annex K and associated commentary with the recommended changes is given in Appendix G Read the entire page →
From page 66... ... 66 would often consistently report multiple flaws either to the left or right of their actual location. This offset in flaw location was sometimes quite large, resulting in a large number of detected flaws which did not meet the API RP 2X requirements for reported flaw location. Read the entire page →
From page 67... ... 67 large improvement in PAUT inspection quality. In fact, it may actually have the opposite effect as a technician who is not properly performing any given task will become more entrenched in the wrong practice and become more confident that he or she is actually doing it correctly. Read the entire page →
From page 68... ... 68 removal of RT requirements in lieu of in-depth PAUT inspection. The researchers are confident that a meaningful practical examination could be performed in a single day. Read the entire page →
From page 69... ... 69 To be correctly sized, this document recommends that the reported dimensions be within a factor of two of true dimensions (one-half to twice the actual dimension) Read the entire page →
From page 70... ... 70 • Documentation of inspection procedures in writing in accordance with recognized standards and accepted in writing by the engineer • Written procedures that contain at minimum the following information: – Specific operator training requirements – Types of weld joint configurations to be examined – Acceptance criteria – Type of UT equipment (manufacturer and model number) – Type of transducer, including frequency, size, shape, angle, and type of wedge – Scanning surface preparation and couplant requirements – Type of calibration test block(s) Read the entire page →
From page 71... ... 71 metric surface flaw for an 8 ksi stress range in an equal thickness weld. Since these flaws are the same sizes as the critical planar flaws, it is recommended that the limits provided in Table 45 and Table 46 also be used for the maximum size of an individual volumetric flaw (i.e., maximum slag inclusion or pore) Read the entire page →

From page 14...

... 14 Findings and Application 3.1 Critical Flaw Size Analytical parametric studies were performed in order to establish the critical flaw sizes considered rejectable for typical bridge CJP welds. The results were used to identify the critical flaw size that must be reliably detected and rejected to establish revised acceptance and rejection criteria.