PSI - Issue 52

Ben M B Sargeant et al. / Procedia Structural Integrity 52 (2024) 472–479 Ben M B Sargeant , Catrin M Davies and Paul A Hooper / Structural Integrity Procedia 00 (2023) 151-158

475

154

2. Materials and Methods Standard 10 mm square, 55 mm length, V-notch Charpy samples were extracted from the bulk of previously tested 25 mm square SEN(B) samples made of SA508 Grade 3 Steel – Fig. 2. This makes the material identical to this previous work, completed by Rait [5]. Samples were manufactured by wire electrical discharge machining (EDM), in accordance with ASTM E1820 specifications. Rait’s r esults showed equivalent plastic strain at maximum load, penetrated 7.2 mm [5]. This plastic zone is remote from the Charpy samples’ new notch and crack growth path (cut 31.5 mm from the fracture surface) and, therefore, these samples may be considered unaffected by the previous loading.

Fig. 2 Extraction of Charpy-sized samples from fractured SEN(B) samples. A scaled sketch of plastic strain results are overlayed to demonstrate region of possible damage; a) Original SEN(B) sample; b) Cutting plan; c) Extracted Charpy-sized sample

2.1. Sample Preparation Following manufacture, the 10 mm samples were fatigue pre-cracked to create a sharp crack of length 5 mm, giving a crack length to with ratio, a/W of 0.5. This was confirmed by heat tinting the samples by heating to 250 o C in air, for 1.25 hours. This functions to lightly oxidise the surface and distinguish pre-crack from fracture surface after testing with minimal impact on bulk material properties. Finally, samples were set up for potential drop crack monitoring by spot welding threaded studs at the mid-point of both square faced ends for current supply through the cross-section of the sample. Potential drop was measured by welding wire across the crack, following ASTM E1820 standards. 2.2. Experimental Method for Fracture Toughness Assessment Samples were loaded in three-point bending, on an INSTRON 100kN servo-hydraulic machine at a constant displacement rate of 1 mm/min (Quasi-static). Voltage from the in-built 100kN load cell and linear potentiometer was output to a National Instruments (NI) 9219 data logger, with a range set to ±10 V and 24bit resolution. Load and displacement data was then used to determine J-integral by method described in ASTM E1820 [3] and eq.1. To distinguish the fracture surfaces post fracture testing, samples were cooled in liquid nitrogen to bring the material through ductile-to-brittle transition (approx. -62 o C [5]) and fractured through impact. The fracture surface from the room temperature test (predominantly ductile) was sufficiently different in texture to the sub-zero fracture (predominantly brittle) to allow observation of the crack front before and after testing. = ( ) 1.5 , ℎ ( )= 3√ [1.99−( )(1− )(2.15−3.93 +2.7( ) 2 )] 2(1+2 )(1− ) 1.5 0 = 0 + 0 = 2 ′ + ( − 0 ) , { ′ = | ′ = 1− 2 | ሺͳሻ In eq. 1 K is stress intensity factor, a is crack length, W is sample width, P is applied load, S is the sample span, B is sample breadth, ( ) is the sample shape function, J 0 is the J -integral based from initial crack length ( a 0 ), J e0 is the elastic component of J-integral, J p0 is the plastic component of J-integral, E is the material elastic Youngs modulus, ν is Poisson’s ratio, η p is the plastic coefficient (1.9 for SEN(B) samples) and A p is the plastic area under the load-displacement response curve.