Experimental validation of the TRANGLE and related NDT codes for modelling the ultrasonic inspection of rough cracks

被引:8
作者
Burch, SF [1 ]
Collett, N
Chapman, RK
Toft, MW
机构
[1] AEA Technol Plc, El Culham Sci Ctr, Abingdon OX14 3ED, Oxon, England
[2] British Energy Generat Ltd, Gloucester GL4 3RS, England
[3] BNFL Magnox Generat, Berkeley GL13 9PB, Glos, England
关键词
D O I
10.1784/insi.46.2.74.55550
中图分类号
TH7 [仪器、仪表];
学科分类号
0804 ; 080401 ; 081102 ;
摘要
British Energy and BNFL Magnox Generation maintain a suite of computer codes for modelling ultrasonic inspection. One of these, TRANGLE, models the inspection of embedded rough planar cracks in ferritic steel, detected by back-scatter of the ultrasound from the rough defect faces. TRANGLE represents rough defect faces as a fine mosaic of interlocking triangular facets. Representations of rough surfaces can either be 'direct' (based on heights obtained by mapping of a particular surface using a roughness measuring machine) or 'statistical' using an ensemble of theoretically generated random surfaces conforming to a particular statistical description. Ultrasound scattering from these two types of surfaces is calculated as the sum of the diffracted fields from individual facets, using the Kirchhoff approximation. This paper describes recent work on the experimental validation of TRANGLE. Three specimens containing rough defects of different sizes were manufactured using a novel method which allowed each defect to be embedded into a block without any welds in the line of sight between the ultrasonic probes and the defects. Pulse-echo back-scatter measurements were then made and compared with both the 'direct' and 'statistical' predictions from TRANGLE.
引用
收藏
页码:74 / 76
页数:3
相关论文
共 1 条
[1]  
Chapman R. K., 1990, Journal of Nondestructive Evaluation, V9, P197, DOI 10.1007/BF00566394