1 NACE TM0177 – “Laboratory Testing of Metals for Resistance to Sulfide Stress Cracking and Stress Corrosion Cracking in H2S Environments”
NACE TM0177 is probably one of the most referenced sour testing standards, and it is the source of the original ‘NACE solution,’ which is now NACE TM0177 Solution A. It covers four SSC/SCC test methods, namely A; Tensile, B; Bent-Beam, C; C-ring, and D; Double Cantilever Beam, and it details environments and procedures. It does not give acceptance levels or pass:fail criteria. It covers solution chemistries, solution volume: surface area ratio, test duration, specimen geometry and loading, amongst other testing details for the four different methods, see below.
1.2 Uniaxial Tensile testing
The smooth uniaxial tensile test is Method A in NACE TM0177. Dead weight, proof ring or hydraulic loading may be applied. The specimens are fully machined and, because of the waisted geometry, cannot sample weld surface details or near surface microstructures. There is a standard and a subsize geometry in TM0177 ( 0.25” (6.35mm) diameter, 1” (25.4mm) gauge length and 0.15” (3.81mm) diameter, 0.6” (15mm) gauge length respectively). The standard warns that subsize specimens can result in shorter failure times than full size specimens, but does not suggest that the threshold stress will be lower. The shoulder radius is specified as 15mm minimum, but 20mm minimum is advised for CRAs in EFC 17, to avoid “unwanted preferential cracking at these locations”. Loading has to be by ‘sustained load’ or ‘constant load’ devices, where ‘sustained load’ refers to a spring loaded device such as a proof ring, and ‘constant load’ refers to dead weight or hydraulic loading. The standard warns that slight load relaxation in a sustained load device may mean that the specimen cracks, without failing, and therefore stipulates visual examination of the specimens after test. Although the standard requires load to be ‘applied carefully to avoid exceeding the desired value’, there is no warning against inadvertent torsion or bending (non-axial loading) which can affect results. In order to generate a threshold stress, specimens are tested at a range of loads for up to 720 hours each.
This test can result in SOHIC (stress-oriented hydrogen-induced cracking) in susceptible materials, and thresholds of around 50% yield stress have been recorded in materials which had survived up to around 100% yield stress in a bend test on a welded sample at TWI (Pargeter 1986).
1.3 Bent Beam testing
The bent beam tests are NACE TM0177 Method B. These specimens are very small – 4.5mm wide, 1.5mm thick, with two 0.7mm diameter drilled holes, and the results are difficult to interpret. This test method is not often used.
1.4 C-Ring testing
The C-ring test is NACE TM0177 method C, and there is also guidance given in EFC 16 and EFC 17. It is effectively a bend test, suitable for relatively small tubular material, producing stress in the hoop direction. If the ring is compressed, a tensile stress is generated on the outside but, as pipes usually carry the environment of concern on the inside, it is more usual to expand the ring, as sketched.
1.5 Double Cantilever Beam (DCB) test
The longest standing fixed displacement fracture mechanics test for sour service is the double cantilever beam (DCB) test, which is also NACE TM0177 method D. The specimen geometry is as shown. Load is applied axially via free-rotating pins through the pin holes, and a wedge is inserted to maintain the displacement. The specimen is then exposed to the environment for 14 days, following which it is unloaded, again using the pin holes. A discontinuity in the load:displacement curve indicates the remaining load on the specimen. This, along with the crack dimensions measured on the broken open specimen, allows the K at the end of the test to be calculated. It should be noted that all fixed displacement fracture mechanics tests are crack arrest tests. As the crack grows, the load relaxes, and this wins over the effect of increasing crack length, so that the stress intensity decreases.
If the correct initial load is selected, the DCB test has the potential for giving an answer in one test, which is quite an advantage over a series of fixed load tests to define a threshold. This only works, however, if the correct load is selected – high enough to get a crack moving, but not so high that the crack runs too far, and the test becomes invalid.
Test durations using DCB tests are also reduced compared to other methods because the standard test duration is half that of fixed load tests in TM0177. That is, however, a slight concern as, if the crack has not arrested at 14 days, then the result will be un-conservative.
1.6 Four Point Bend testing
Although it is not uncommon to come across company specifications for four point bend testing ‘according to NACE TM0177’, this method is not in that standard. There has been guidance given in EFC 16 and 17, and also in ISO 7539-2, for some time, and there is now a NACE standard; TM0316. Tests may be on fully machined specimens or with, for example, weld details intact. If this type of test is specified, check which test method the client really intends.
EFC 16: ‘Guidelines on Materials Requirements for Carbon and Low Alloy Steels for H2S-Containing Environments in Oil and Gas Production’ (EFC 16, 3rd Edition) Edited by Svein Eliassen and Liane Smith, Maney Publishing, July 2009, 56pp paperback, ISBN 978 1 90654 033 3
EFC 17: ‘Corrosion Resistant Alloys for Oil and Gas Production: Guidance on General Requirements and Test Methods for H2S Service’, (EFC 17, 2nd Edition), Maney Publishing, May 2002, 96pp, ISBN 978 1 902653 556
ISO 7539-2 ‘Corrosion of metals and alloys - Stress corrosion testing - Part 2: Preparation and use of bent-beam specimens’.
NACE TM0316, 2016: ‘Four-point bend testing of materials for oil and gas applications’.
Pargeter, R J, “Hydrogen-induced stress corrosion cracking and hardness of welded structural and pipeline steels – Final contract report” TWI GSP report 5537/29/86 May 1986 (TWI report archive)