Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
72 Conclusions and Suggested Research 4.1 Conclusions NCHRP Project 14-35: Acceptance Criteria of Com- plete Joint Penetration Steel Bridge Welds Evaluated Using Enhanced Ultrasonic Methods had the objectives of develop- ing guidelines to evaluate CJP welds in steel bridges based on updated acceptance criteria and to develop proposed modi- fications to AWS D1.5. While AWS D1.5 currently includes PAUT inspection procedures in Annex K, these acceptance criteria were workmanship-based and were carried over from previous D1.1 conventional UT methods. AWS D1.5 did not provide means for alternative methods such as FMC/TFM PAUT or TOFD, which are suited for evaluation of flaw criti- cality based on measurements of flaw size rather than ampli- tude responses. The research has shown that (1) the critical flaw size of bridge welds could be developed using FFS, (2) a large amount of variability was possible when weld inspections were performed using current AWS D1.5 conventional UT and PAUT scanning procedures and the current workforce, (3) computer modeling could be used to evaluate ultrasonic responses of weld flaws and reference reflectors, (4) acoustic properties of bridge steels may vary widely and may not be isotropic, (5) revised acceptance criteria for Annex K could be developed to detect and reject critical weld flaws utilizing amplitude-based criteria, and (6) additional technician per- formance requirements including independent practical examination were necessary. Major findings of this research were that the current scan- ning procedures and acceptance criteria in Annex K did not correlate to traditional limits used in conventional UT per Clause 6 and were not adequate for rejection of critical weld flaws according to FFS. This research also found that differ- ences in acoustic properties between the calibration block and test object could result in significant error in reference sensi- tivity for frequencies that were allowed in Annex K. The cur- rent version of Annex K does not include any requirements on calibration block acoustic properties. Variations in shear wave velocity were also found to be significant for common grades of bridge steels. These variations resulted in significant error in beam refraction angle, which could result in inaccurate flaw location and significant loss of amplitude. These variations affect both conventional UT and PAUT. Proposed revisions to Annex K were provided in this report that would account for these differences, but similar revisions should be included in Clause 6 for conventional UT. The final product of NCHRP Project 14-35 was proposed revisions to AWS D1.5 Annex K for improved flaw detection and rejection. These revisions include minimum requirements for technician qualification, requirements on the acoustic properties of calibration blocks in order to represent the con- ditions found in the test object, requirements on the scanning procedure and sound coverage, and requirements on accep- tance criteria to detect and reject critical weld flaws. Recommendations were also provided which would allow alternate UT methods to be used in lieu of PAUT or conven- tional UT. Use of these methods relies upon written agree- ment by the engineer and contractor along with procedure development and demonstration on weld flaw specimens. Limits on acceptable flaw sizes that could be incorporated into an acceptance criteria based on flaw size measurements were also provided. 4.2 Suggested Research While NCHRP Project 14-35 has resulted in important find- ings and recommendations for modifications for AWS D1.5 regarding application of PAUT for the inspection of bridge welds, additional research is suggested which could aid in application of these recommendations. Five topics of sug- gested research are proposed: (1) performing a round robin testing program to compare inspection results using RT to inspection results using PAUT in accordance with the revised version of Annex K, including the proposed modifications; C H A P T E R 4
73 (2) developing a performance-based qualification program for PAUT technicians and verifying improvement in inspec- tion results; (3) developing specific scan plan recommen- dations for probe selection and line scan index offset for typical weld geometries; (4) collecting additional data on variability of acoustic properties for steel bridge base metals; and (5) developing automated scanning and data processing techniques to minimize variability due to human factors. The first research topic will compare round robin inspec- tion results of RT and PAUT using the revised version of Annex K. While this research project used computer model- ing and experimental test results to develop and verify the revisions to Annex K, the round robin results would be able to collect information on human factors and scatter in results. This study would aid in adoption of the revised Annex K inspection procedure to be used in lieu of RT by providing a direct comparison of inspection results using both methods. The second research topic will involve development of PAUT technician qualification requirements, which will ade- quately demonstrate use of the revised Annex K procedure. This research would involve the establishment of the number, size, and type of flaws needed for a meaningful practical examination. This research will also provide verification whether the PAUT inspection quality will meet or exceed RT after requiring proper training and performance testing of PAUT technicians along with the revisions to Annex K. The third research topic will develop specific scan plan recommendations for typical weld geometries. Currently, there is no ANST Level III for PAUT; therefore, the ASNT Level II PAUT technicians are responsible for developing the scan plan. This includes probe selection, index offset loca- tion for line scans, and focal law configurations. As shown in NCHRP Project 14-35, the probe frequency and aperture should be properly selected in order to result in optimal inspection results. For instance, inspections of thick plates with long sound paths may require use of lower frequency and larger apertures in order to account for loss of amplitude due to attenuation and beam spread. This research has also shown that the amplitude response of flaws will be greatly influenced by probe location. Since typical bridge welds utilize very similar geometries based on the plate thickness, recommendations to probe parameters and index positions could be tabulated for typical welds. This would cut down on the effort for scan plan development and would result in more consistent inspections from technician to technician within the QA/QC process. The fourth research topic will collect additional data on the variability of acoustic properties for steel bridge base metals. This research topic would involve collecting data on the variability of acoustic properties of applicable steel bridge grades along with the variability of acoustic properties at dif- ferent locations within the same heat of steel. This research topic will aid in determining the overall scatter of typical steels and in developing refined calibration standards. This project should include combinations of possible heat treat- ment within each grade, including A709-50CR (i.e., A1010) since the acoustic properties of this mild stainless steel were not investigated during NCHRP Project 14-35. The fifth research topic will develop automated scanning and data processing techniques to minimize variability due to human factors. This research topic would involve evalu- ating automated scanners, which would result in consistent data collection and probe manipulation. It would also eval- uate techniques for automatic data processing in order to consistently evaluate scan results according to the proposed acceptance criteria.