Low Frequency-saft Inspection Methodology for Coarse-grained Steel Rail Components (manganese Steel Frogs)

In the rail industry, sections of high strength Manganese steel are employed at critical locations in railroad networks. Ultrasonic inspections of Manganese steel microstructures are difficult to inspect with conventional means, as the propagation medium is highly attenuative, coarse-grained, anisotropic and nonhomogeneous in nature. Current in-service inspection methods are ineffective while pre-service X-ray methods (used for full-volumetric examinations of components prior to shipment) are time-consuming, costly, require special facilities and highly trained personnel for safe operations, and preclude manufacturers from inspecting statistically meaningful numbers of frogs for effective quality assurance. In-service examinations consist of visual inspections only and by the time a defect or flaw is visually detected, the structural integrity of the component may already be compromised, and immediate repair or replacement is required. A novel ultrasonic inspection technique utilizing low frequency ultrasound (100 to 500 kHz) combined with a synthetic aperture focusing technique (SAFT) for effective reduction of signal clutter and noise, and extraction of important features in the data, has proven to be effective for these coarse grained steel components. Results from proof-of-principal tests in the laboratory demonstrate an effective means to detect and localize reflectors introduced as a function of size and depth from the top of the frog rail. Using non-optimal, commercially available transducers coupled with the low-frequency/SAFT approach, preliminary evaluations were conducted to study the effects of the material microstructure on ultrasonic propagation, sensitivity and resolution in thick section frog components with machined side-drilled holes. Results from this study will be presented and discussed. Introduction: Ultrasonic nondestructive testing (NDT) has a long and successful history of application to the rail networks used by freight and high-speed rail services. This form of inspection is now routinely performed using special test “cars”, such as those developed and operated by Sperry Rail Services and several other vendors. Such systems can work well for the inspection of normal rails. However, problems are encountered when a car passes over the short lengths of Manganese steel rail (known as “frogs” in railroad acumen) used at critical locations in the railroad network. The need for an effective and reliable means for pre-service examination of coarse-grained, thick-section, Manganese (Mn) steel frog components in both the freight and high-speed rail industries is well established. In-service examinations consist of visual inspections only, as conventional inspection methods are ineffective. Typically, by the time a defect or flaw is visually detected, the structural integrity of the component may already be compromised and immediate repair or replacement is required. Periodically, frogs are received inherently flawed from a manufacturer, and put into service, as most rail road operators do not have a means to conduct pre-service examinations once these components are received. The problem generates a more significant impact when the cost of labor for repair and/or replacement is then added to the equation. In some cases, warranty claims cannot be made due to the lack of part identification and uncertainty in the root cause of the failure. Accordingly, there is a need for a pre-service inspection methodology that can provide a rapid, cost-effective and non-intrusive inspection capability for detection of defects, flaws, and other anomalies in frog components that eventually lead to premature initiation of cracks or failures of these components during service. At present, X-ray methods are used for full-volumetric examinations of small numbers of frog components prior to shipment from manufacturing plants. This method is time-consuming, costly, and requires special facilities and highly trained personnel for safe operations, precluding manufacturers from inspecting statistically meaningful numbers of frogs for effective quality assurance. This leads to an unacceptable rate of failure in the field, often within the first 6 months of service. The work described here illustrates an effective ultrasonic solution to this inspection problem by adapting technology developed for the U.S. Nuclear Regulatory Commission (NRC) for inspecting thick-section, coarse-grained steel, nuclear reactor piping components. The proof-of-concept effort detailed here employed an innovative approach combining and adapting existing capabilities in ultrasonic methodologies and advanced signal processing techniques to test the effectiveness of an ultrasonic pre-service inspection method using low frequency-SAFT Ultrasonic Testing (UT) inspection techniques for examination of Manganese steel frog components. From the standpoint of ultrasonic inspections, cast stainless steel and Manganese steel microstructures are quite similar, as the propagation medium is highly attenuative, coarsegrained, anisotropic and nonhomogeneous in nature. Current in-service rail inspection technology is ineffective and unable to adequately inspect these sections of Manganese steel rail. Manganese steel rail components are fabricated from material that has a large grain size, and it is this factor, which most severely limits current ultrasonic inspection techniques. Previous experience at Battelle with work performed for the U.S. NRC has shown that novel implementations of low frequency ultrasound, combined with an advanced signal processing methodology, can successfully inspect coarse-grained steels. This work has demonstrated high levels of sensitivity and resolution for detection and localization of thermal and mechanical fatigue cracks that compromise the structural integrity of these components. Recent proof-ofprinciple work conducted at Battelle on Manganese steel frog components provided by Burlington Northern-Santa Fe Railroad (BNSF) and Rail Products and Fabrications, have yielded very favorable results that indicate a solution to the inspection problem using ultrasonic technology. This effort has successfully demonstrated a low frequency ultrasonic inspection protocol that implements the synthetic aperture focusing technique (SAFT), on samples of Manganese steel frog components with various geometric reflectors, side-drill holes and even a laminar discontinuity. The fundamental problem encountered, when seeking to use ultrasound to inspect sections of cast Manganese steel rail, arises due to the scattering of the ultrasonic energy by the large grains. Similar problems have been previously encountered for thick section cast steel components used in the primary pressure boundary of U.S. nuclear reactors. The approach reported here is based upon ongoing work being conducted for the U.S. NRC as part of a study designed to evaluate and develop ultrasonic in-service inspection techniques for detection of thermal and mechanical fatigue cracks in centrifugally and statically cast stainless steel piping components, which are used within the primary pressure boundary of commercial light-water reactors. Figure 1 shows the coarse microstructure of these thick-section reactor piping components. Manganese steel (Hadfield Steel -12% Mn, 1% C) is an anisotropic and nonhomogeneous material. The manufacturing process can result in the formation of a long columnar grain structure, oriented approximately normal to the surface with grain growth oriented along the direction of heat dissipation, forming dendritic features often several centimeters in length. During the solidification of the material, columnar, equiaxed (randomly speckled microstructure), or a mixed structure can result that is dependent upon chemical content and control of the cooling process. Figure 1. Circumferential and Axial Cross Sections of Thick-Section, Centrifugally Cast Stainless Steel Pipe, Illustrating a Mixed Columnar-Equiaxed Macrostructure with Multiple Layers and Nonuniform Mixing of Grains The large size of the Mn steel anisotropic grains (typically 1⁄4” diameter), relative to the acoustic pulse wavelength (typically 1⁄4” diameter at 1 MHz), strongly affects the propagation of ultrasound by causing severe attenuation, changes in velocity, scattering of ultrasonic energy, and refraction and reflection of the sound beam at the grain boundaries. These phenomena can result in defects being missed, incorrectly reported, specific volumes of material not being examined, or all of these conditions. When coherent reflection and scattering of the sound beam occur at grain boundaries, ultrasonic indications occur which are difficult to distinguish from flaw signals. When inspecting coarse-grained materials, where the signal-to-noise ratio is relatively low, ultrasonic examinations can be confusing, unpredictable, and unreliable. Hence, novel approaches are required. Results: There is a significant degree of variability in these microstructures from lot to lot and more importantly, between manufacturers of these components. Of primary concern to roadmasters and track welders are the QA/QC processes that currently exist for Manganese steel frogs at various manufacturing plants. Quite often, newly purchased frogs will exhibit significant internal sand inclusions that act as crack initiation points for components under impact stresses. A novel ultrasonic inspection technique that implements low frequency ultrasound (250 to 450 kHz) combined with a multi-angle, custom-designed transducer head and coupled with an inspection protocol that incorporates advanced signal processing methods for extracting features in the data, has proven to be effective for these coarse grained steel components. The proposed technique uses a zone-focused, low frequency (250-450 kHz) broadband transducer and inspection scanner coupled with an advanced signal processing protocol for reduction of signal clutter and noise and enhancement of signals from defects and anomalies. Figure 2 illustrates the raw ultrasonic rf waveform (time-series) from a sample frog specimen used in this study.