Subscribe to our newsletter to receive the latest news and events from TWI:

Subscribe >

Lab-Scale Full-Thickness Compact Crack Arrest Testing

Introduction

Recent TWI/NSIRC/LRF PhD research by Jessica Taylor compared crack initiation and crack arrest behaviour in six different structural steels (such as those used for pipelines, pressure vessels, offshore structures or wind turbine foundations), based on predictions from a range of standard small-scale mechanical tests. The work, summarised in a recent open-access journal paper [1], concluded that measures of toughness such as Charpy impact energy or CTOD, particularly on the upper shelf, are not able to predict whether modern structural steels could nonetheless exhibit poor crack arrest behaviour. In such structures, a brittle fracture could initiate locally due to a dynamic impact event, or from a local brittle zone in a weld joint. In these cases, having sufficient crack arrest toughness in the steel can limit the damage and avoid a catastrophic failure. However, many design approaches are based only on Charpy or CTOD requirements, without explicit consideration of crack arrest behaviour.

Of the small-scale test results reported in the paper [1], the nil-ductility transition temperature (NDTT), as determined from a set of Pellini tests, was considered to best represent crack arrest behaviour in the steels. Instrumented Charpy tests allow the prediction of the temperature at which the arrest force after fracture is 4kN (T4kN), and this parameter also correlates with the NDTT. For both of these parameters, empirical formulae exist in the literature to predict the brittle arrest toughness (KIa) behaviour with temperature [2, 3], with the prediction based on NDTT being consistently the lower of the two.

Work Programme

Jessica completed her PhD work by carrying out larger scale compact crack arrest tests for one of the steels (the EH 47 shipbuilding steel, which showed the largest difference between the two KIa curves), in order to measure the crack arrest toughness directly from the tests, according to methods given in ASTM E1221, and compare the results against the empirical predictions. Each compact crack arrest (CCA) specimen was over 300mm square and 80mm thick. The initiation of a brittle crack was done via the application of a wedge load into a hole at one end of the specimen, which loads a slit cut into a low toughness weld bead at the base of the specimen starter notch. The wedge load is applied and then removed through several increasing load cycles until a brittle crack occurs. Side-grooves are machined to direct the brittle crack in a planar direction through the specimen. Each test was done at a given temperature, cooled using a liquid-nitrogen cooled stage and insulated throughout the test. The experimental set-up is shown in Figures 1 and 2.

After the test, the specimen was heat-tinted and then broken open to reveal the length of the arrested crack, from which the arrest toughness KIa could be determined. An example is shown in Figure 3.

Larger scale crack arrest tests requiring a running brittle crack to arrest within the ligament of the test specimen are notoriously challenging, subject to scatter, and sometimes unpredictable. However, of the eight CCA tests carried out on this steel, all successfully managed to arrest a brittle crack, and allowed a value of KIa to be determined, for test temperatures between -10°C and -90°C, Figure 4. The values of KIa calculated using the ASTM E1221 method were plotted against temperature, and compared to the empirical predictions based on NDTT [2] and T4kN [3], shown in Figure 5. The results were consistently close to, but above the curve based on NDTT predictions, but the results were over-estimated by the T4kN prediction.

Figure 1. Setting up the compact crack arrest specimens
Figure 1. Setting up the compact crack arrest specimens
Figure 2. The CCA specimen cooled and under test
Figure 2. The CCA specimen cooled and under test
Figure 3. Fracture appearance of one of the CCA specimens after test and heat-tinting of the arrested brittle crack
Figure 3. Fracture appearance of one of the CCA specimens after test and heat-tinting of the arrested brittle crack
Figure 4. Arrested crack length plotted against CCA test temperature
Figure 4. Arrested crack length plotted against CCA test temperature
Figure 5. Comparison of the CCA KIa results, with predictions of arrest toughness based on small-scale tests
Figure 5. Comparison of the CCA KIa results, with predictions of arrest toughness based on small-scale tests

Concluding Remarks

The work here confirms that lab-scale crack arrest testing to determine the arrest toughness, KIa, can be done, as shown by the successful tests performed at TWI. For applications where small-scale testing is used to estimate crack arrest behaviour, Pellini testing to determine NDTT, and estimations of KIa based on that, appear to provide a conservative prediction of the behaviour expected from compact crack arrest testing.

Get in touch with TWI to discuss your fracture testing requirements, including Charpy, Pellini, instrumented-Charpy, fracture toughness, or crack arrest testing.

References:

[1] Jessica Taylor, Ali Mehmanparast, Rob Kulka, Philippa Moore, Li Xu, and Gholam Hossein Farrahi, 2020; ‘Experimental study of the relationship between fracture initiation toughness and brittle crack arrest toughness predicted from small-scale testing’, Theoretical and Applied Fracture Mechanics 110 (2020) 102799. https://doi.org/10.1016/j.tafmec.2020.102799

[2] C.E. PUGH, W.R. CORWIN, R.H. BRYAN and B.R. BASS, 1986; ‘Some advances in fracture studies under the heavy section steel technology program’, Nuclear Engineering and Design 96 (1986) 297-312 297.

[3] Wallin K, Karjalainen-roikonen P, 2016; ‘Crack Arrest Toughness Estimation for High Strength Steels from Sub-Sized Instrumented Charpy-V Tests’, Proceedings of the Twenty-sixth International Ocean and Polar Engineering Conference ISOPE-2016, Rhodes, Greece, June 26-July 1, 2016.

Avatar Philippa Moore Metallurgical Integrity Engineer - Materials & Structural Integrity Group

Philippa has over 15 years’ experience in project management and research related to fracture toughness testing, welding engineering, steel metallurgy, failure investigation, engineering critical assessment, and structural integrity.

For nearly a decade, she has represented the UK on standards work related to fracture toughness testing for BSI and ISO, and was leading the development of Standard BS 8571 on SENT testing. She has published more than 45 papers in various international journals and conferences.

Media /

Related Insights

View all insights
}