OR WAIT null SECS
CAN TURBOMACHINERY PARTS BUILT USING ADDITIVE
MANUFACTURING — THE INDUSTRIAL VERSION OF 3D PRINTING —
MEET THE CHALLENGE OF SOUR ENVIRONMENTS?
BY BYRON MOHR AND WILLIAM KOVACS
Additive manufacturing (AM) is one of many emerging technologies positively impacting the turbomachinery industry. AM offers several advantages compared to traditional production methods, including quick prototyping, shorter lead times and reduced waste. However, it also presents challenges, such as variable quality and anisotropy (exhibiting different properties in different directions). This could pose problems for equipment that is operating in sour environments.
There are several industry standards for sour service capability of wrought and cast materials. This includes NACE MR0175/ ISO15156 Petroleum and Natural Gas Industries— Materials for Use in H2S-containing Environments in Oil and Gas Production and MR0103/ISO17945 Materials Resistant to Sulfide Stress Cracking in Corrosive Petroleum Refining Environments. However, they do not directly apply to AM production methods. As a result, additional qualification and material testing are needed to determine the suitability of AM components for sour service. Testing for sulfide stress cracking (SSC) can be done, for example, on 17-4 stainless steel (UNS S17400) produced by applying the AM powder bed fusion process.
Results can be compared with traditional welded and wrought testing of the same alloy. This testing helps determine if a set of AM specimens can comply with the testing requirements for wrought and welded materials. Various techniques were used to evaluate the material properties of the wrought, welded and AM specimens. Rockwell hardness testing reading in the “C” scale were taken from AM, age-hardened rod, and welded specimens. The Rockwell scale hardness test is based on indentation hardness of a material by determining the depth of penetration of an indenter under a large load compared to the penetration made by a preload.
There are different scales, denoted by a single letter (e.g., HRB, HRC, and so on), that use different loads or indenters. All material had been heat-treated. Readings of the AM specimens were distributed onto different XY- and Z- planes. All specimens met the maximum hardness specification, but AM components exhibited slightly lower readings. AM specimens had inconsistent readings. One specimen displayed a range (min to max) of 7.9 HRC. Ranges for the rod and welded specimens, on the other hand, were 1.2 and 2.4 HRC, respectively. Yield Strength Research has shown that yield strength strongly correlates with relative cracking susceptibility of UNS S17400. To that end, duplicate specimens were turned and tested from AM blanks (Z-direction).
The tensile properties of the rod and welded specimens were also recorded. Testing revealed an average yield strength just above 130 kilopounds per square inch (ksi) for the AM specimen. This was higher than other materials with the same heat treatment condition. As there are some correlations between hardness and yield strength or tensile strength, the use of yield strength alone as a discriminator for AM parts is not encouraged.
AM parts manufacturers could, however, incorporate tension specimens for quality control in each of the XY, and Z directions with every build that is carried out. Previous research has demonstrated that the most likely direction to experience reduced properties is in the Z-build direction. This is a result of the solidification direction and segregation of impurities. Uniaxial tension is also useful in evaluating the performance of various fabrication routes (AM, wrought, and cast).
Testing should be done in two environments: Low H2S and high H2S. Results indicate capability in condensed water conditions beyond the NACE MR0175 listed conditions for plate and welded materials. Relevant NACE MR0175 limits, regarding stress level and partial pressure of H2S are shown in the figure. It also shows the results of tested materials and additional values obtained from literature for SSC testing of UNS S17400 stainless steel. As can be seen in the figure, results of SSC testing of AM parts with the longitudinal direction of the specimen in the Zaxis indicate a significant performance deficit of the AM material tested relative to reference tests on wrought and welded types. When exposed to either the “high H2S” or “low H2S” conditions, each set of three AM specimens failed in less than 24 hours. Independent crack initiation sites were observed on the fracture face.
Energy Dispersive x-ray Spectroscopy (EDS) element maps of these sites showed an elevation in the sulfur content. The compositional analysis also showed elevated levels of niobium, manganese, silicon and also in oxygen levels. Multiple factors may have resulted in the AM specimens’ poor performance in the sour environment. This includes inclusions, precipitates, intra-layer defects, porosity, material condition and heat treatment, and microstructure. There were discontinuities seen in the polished AM surfaces (XZ-plane), along with inferior general and localized corrosion resistance in this direction. This is likely a result of inhomogeneities that occurred during fabrication.
All AM specimens met the compositional, heat treatment and maximum hardness limits specified in NACE MR0175. However, the uniaxial tension test did not achieve the performance in a sour environment compared to wrought and welded counterparts. AM parts could be distinguished from wrought and welded counterparts by their variability in (macro) hardness, higher yield strength to tensile strength (YS/TS) ratio, reduction of area, and lower ductility in the Z-direction. These test methods and criteria may be useful tools for AM batch process quality control.
Overall, additional testing is needed to more accurately determine the suitability of AM components for turbomachinery in sour environments. Improvements in the AM microstructural control, porosity and mechanical properties would be beneficial to minimize the sources of variability in sour qualification testing.
Byron Mohr is Manager of Materials, Welding, Solid Mechanics, and Development Engineer Departments for the Dresser-Rand business, part of Siemens Power and Gas. For more information visit www.dresser-rand.com William Kovacs III is Senior Engineer, Materials Compatibility Group at DNV GL. This article is based on the paper, “Additive manufacturing for sour service, an experimental investigation” presented at the 2017 NACE Corrosion Conference and Expo. As well as Byron Mohr and W. Kovacs III, additional authors from the DNV GL Pipeline Services Department were L. Cao, K. Evans, C. Taylor, S.A. Waters, Z. Berg and J. Silva.