SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Relative strengths of notched single-layer and multilayer stainless steel pressure vessels at cryogenic temperatures
Stress wave analysis and crack opening displacement to monitor subcritical crack growth for grade B alloy steel pressure vessels
Acoustic emission data were obtained from three point bend fracture toughness specimens of 12-nickel maraging steel, and two pressure vessels of the same material. One of the pressure vessels contained a prefabricated flaw which was extended and sharpened by fatigue cycling. It is shown that the flawed vessel had similar characteristics to the fracture specimens, thereby allowing estimates to be made of its nearness to failure during a proof test. Both the flawed and unflawed pressure vessel survived the proof pressure and 5 cycles to the working pressure, but it was apparent from the acoustic emission response during the proof cycle and the 5 cycles to the working pressure that the flawed vessel was very near failure. The flawed vessel did not survive a second cycle to the proof pressure before failure due to flaw extension through the wall (causing a leak).
The elastic fracture toughness of the three steels is shown to not decrease significantly with decreasing temperature from room temperature to about 244 K (-20 F.). The elastic fracture toughness of the three steels increased with increasing specimen width and thickness. The fatigue-crack-growth data for all three steels fall into relatively narrow scatter bands on plots of rate against stress-intensity range. An equation is shown to predict the upper bounds of the scatter bands reasonably well. Charpy impact energies decreased with decreasing temperature in the nominal temperature range from room temperature to 244 K (-20 F). The nil-ductility temperatures of the steels are discussed.
Manufacture and performance of cryogenically formed stainless steel pressure vessels for liquid gas storage
Two different Kevlar-49 filament-reinforced metal sphere designs were developed, and six vessels of each type were fabricated and subjected to fatigue cycling, sustained loading, and hydrostatic burst. The 61 cm (24 inch) diameter Kevlar-49/cryoformed 301 stainless steel pressure vessels demonstrated the required pressure cycle capability, burst factor of safety, and a maximum pressure times volume divided by weight (pV/W) performance of 210 J/g (834 000 in-lb/lbm) at burst; this represented a 25 to 30% weight saving over the lightest weight comparable, 6A1-4V Ti, homogeneous pressure vessel. Both the Kevlar/stainless steel design and the 97 cm (38 inch) diameter Kevlar-49/2219-T62 aluminum sphere design demonstrated nonfragmentation and controlled failure mode features when pressure cycled to failure at operating pressure. When failure occurred during pressure cycling, the mode was localized leakage and not catastrophic. Kevlar/stainless steel vessels utilized a unique conical boss design, and Kevlar/aluminum vessels incorporated a tie-rod to carry port loads; both styles of polar fittings performed as designed during operational testing of the vessels.
Investigation determined whether irradiated pressure-vessel steels 4340 and 212-B are susceptible to hydrogen embrittlement and to catastrophic failure. Hydrogen-charging conditions which completely embrittled 4340 steel had negligible effect on 212-B steel in tensile and delayed-failure tests.
An experimental study is conducted on 12.5-mm-thick SAE 1020 steel (plain carbon steel) plate to assess hydrogen attack at room temperature after specimen exposure at 525 C to hydrogen and a blend of hydrogen sulfide and hydrogen at a pressure of 3.5 MN/sq m for exposure times up to 240 hr. The results are discussed in terms of tensile properties, fissure formation, and surface scales. It is shown that hydrogen attack from a high-purity hydrogen environment is severe, with the formation of numerous methane fissures and bubbles along with a significant reduction in the room-temperature tensile yield and ultimate strengths. However, no hydrogen attack is observed in the hydrogen/hydrogen sulfide blend environment, i.e. no fissure or bubble formation occurred and the room-temperature tensile properties remained unchanged. It is suggested that the observed porous discontinuous scale of FeS acts as a barrier to hydrogen entry, thus reducing its effective equilibrium solubility in the iron lattice. Therefore, hydrogen attack should not occur in pressure-vessel steels used in many coal gasification processes.
Weld tests of stainless steel pressure vessels fabricated with cryogenic forming techniques
An analytical study was performed to determine a design configuration for a cavity test reactor. Test section criteria were that an average flux of 10 to the 15th power neutrons/sq cm/sec (E less than or equal to 0.12 eV) be supplied to a 61-cm-diameter spherical cavity at 200-atm pressure. Design objectives were to minimize required driver power, to use existing fuel-element technology, and to obtain fuel-element life of 10 to 100 full-power hours. Parameter calculations were made on moderator region size and material, driver fuel arrangement, control system, and structure in order to determine a feasible configuration. Although not optimized, a configuration was selected which would meet design criteria. The driver fuel region was a cylindrical annular region, one element thick, of 33 MTR-type H2O-cooled elements (Al-U fuel plate configuration), each 101 cm long. The region between the spherical test cavity and the cylindrical driver fuel region was Be (10 vol. % H2O coolant) with a midplane dimension of 8 cm. Exterior to the driver fuel, the 25-cm-thick cylindrical and axial reflectors were also Be with 10 vol. % H2O coolant. The entire reactor was contained in a 10-cm-thick steel pressure vessel, and the 200-atm cavity pressure was equalized throughout the driver reactor. Fuel-element life was 50 hr at the required driver power of 200 MW. Reactor control would be achieved with rotating poison drums located in the cylindrical reflector region. A control range of about 18 percent delta k/k was required for reactor operation.
These Guidelines integrate state-of-the-art Elastic-Plastic Fracture Mechanics (EPFM) and proof test implementation issues into a comprehensive proof test analysis procedure in the form of a Road Map which identifies the types of data, fracture mechanics based parameters, and calculations needed to perform flaw screening and minimum proof load analyses of fracture critical components. Worked examples are presented to illustrate the application of the Road Map to proof test analysis. The state-of-the-art fracture technology employed in these Guidelines is based on the EPFM parameter, J, and a pictorial representation of a J fracture analysis, called the Failure Assessment Diagram (FAD) approach. The recommended fracture technology is validated using finite element J results, and laboratory and hardware fracture test results on the nickel-based superalloy IN-718, the aluminum alloy 2024-T351 1, and ferritic pressure vessel steels. In all cases the laboratory specimens and hardware failed by ductile mechanisms. Advanced proof test analyses involving probability analysis and Multiple Cycle Proof Testing (MCPT) are addressed. Finally, recommendations are provided on to how to account for the effects of the proof test overload on subsequent service fatigue and fracture behaviors.
These guidelines integrate state-of-the-art elastic-plastic fracture mechanics (EPFM) and proof test implementation issues into a comprehensive proof test analysis procedure in the form of a road map which identifies the types of data, fracture mechanics based parameters, and calculations needed to perform flaw screening and minimum proof load analyses of fracture critical components. Worked examples are presented to illustrate the application of the road map to proof test analysis. The state-of-the art fracture technology employed in these guidelines is based on the EPFM parameter, J, and a pictorial representation of a J fracture analysis, called the failure assessment diagram (FAD) approach. The recommended fracture technology is validated using finite element J results, and laboratory and hardware fracture test results on the nickel-based superalloy Inconel 718, the aluminum alloy 2024-T3511, and ferritic pressure vessel steels. In all cases the laboratory specimens and hardware failed by ductile mechanisms. Advanced proof test analyses involving probability analysis and multiple-cycle proof testing (MCPT) are addressed. Finally, recommendations are provided on how to account for the effects of the proof test overload on subsequent service fatigue and fracture behaviors.
Progressive damage and fracture of large shell structures is investigated. A computer model is used for the assessment of structural response, progressive fracture resistance, and defect/damage tolerance characteristics. Critical locations of a stiffened conical shell segment are identified. Defective and defect-free computer models are simulated to evaluate structural damage/defect tolerance. Safe pressurization levels are assessed for the retention of structural integrity at the presence of damage/ defects. Damage initiation, growth, accumulation, and propagation to fracture are included in the simulations. Damage propagation and burst pressures for defective and defect-free shells are compared to evaluate damage tolerance. Design implications with regard to defect and damage tolerance of a large steel pressure vessel are examined.
Numerical simulations are used to predict the cosmic-ray induced background in a passively shielded gas scintillation proportional counter to be used as the focal plane instrument of a balloon-borne hard x-ray telescope system. The predicted background in the 20-75 keV operating range of the detector shows a spectral shape characteristic of a power law photoelectrically absorbed by the stainless steel pressure vessel walls. It is found that the dominant background radiation component is simply cosmic diffuse and atmospheric gamma-ray leakage though there is a contribution from nuclear interactions of energetic particles in the walls and nearby payload structures. The correlation between shielding properties and background flux is examined parametrically using one-dimensional slab models, within material and weight constraints, in order to optimize predicted sensitivity. The addition of a few mm thick Sn passive shielding reduces the primary photon contribution by up to 50% at the spectral peak, E 150 keV, and adds only negligibly to the flux due to photon production within the high-Z shields at lower energies.
The use of small laboratory specimens to predict the integrity of large, complex structures relies on the validity of single parameter fracture mechanics. Unfortunately, the constraint loss associated with large scale yielding, whether in a laboratory specimen because of its small size or in a structure because it contains shallow flaws loaded in tension, can cause the breakdown of classical fracture mechanics and the loss of transferability of critical, global fracture parameters. Although the issue of constraint loss can be eliminated by testing actual structural configurations, such an approach can be prohibitively costly. Hence, a methodology that can correct global fracture parameters for constraint effects is desirable. This research uses micromechanical analyses to define the relationship between global, ductile fracture initiation parameters and constraint in two specimen geometries (SECT and SECB with varying a/w ratios) and one structural geometry (circumferentially cracked pipe). Two local fracture criteria corresponding to ductile fracture micromechanisms are evaluated: a constraint-modified, critical strain criterion for void coalescence proposed by Hancock and Cowling and a critical void ratio criterion for void growth based on the Rice and Tracey model. Crack initiation is assumed to occur when the critical value in each case is reached over some critical length. The primary material of interest is A516-70, a high-hardening pressure vessel steel sensitive to constraint; however, a low-hardening structural steel that is less sensitive to constraint is also being studied. Critical values of local fracture parameters are obtained by numerical analysis and experimental testing of circumferentially notched tensile specimens of varying constraint (e.g., notch radius). These parameters are then used in conjunction with large strain, large deformation, two- and three-dimensional finite element analyses of the geometries listed above to predict crack initiation loads and to calculate the associated (critical) global fracture parameters. The loads are verified experimentally, and microscopy is used to measure pre-crack length, crack tip opening displacement (CTOD), and the amount of stable crack growth. Results for A516-70 steel indicate that the constraint-modified, critical strain criterion with a critical length approximately equal to the grain size (0.0025 inch) provides accurate predictions of crack initiation. The critical void growth criterion is shown to considerably underpredict crack initiation loads with the same critical length. The relationship between the critical value of the J-integral for ductile crack initiation and crack depth for SECT and SECB specimens has been determined using the constraint-modified, critical strain criterion, demonstrating that this micromechanical model can be used to correct in-plane constraint effects due to crack depth and bending vs. tension loading. Finally, the relationship developed for the SECT specimens is used to predict the behavior of circumferentially cracked pipe specimens.
A damage tolerance assessment was conducted of an 8,000 gallon pressurized Liquid Oxygen (LOX) tank. The LOX tank is constructed of a stainless steel pressure vessel enclosed by a thermal-insulating vacuum jacket. The vessel is pressurized to 2,250 psi with gaseous nitrogen resulting in both thermal and pressure stresses on the tank wall. Finite element analyses were performed on the tank to characterize the stresses from operation. Engineering material data was found from both the construction of the tank and the technical literature. An initial damage state was assumed based on records of a nondestructive inspection performed on the tank. The damage tolerance analyses were conducted using the NASGRO computer code. This paper contains the assumptions, and justifications, made for the input parameters to the damage tolerance analyses and the results of the damage tolerance analyses with a discussion on the operational safety of the LOX tank.
The multimodal friction ignition tester (MFIT) is a testbed for experiments on the thermal and mechanical effects of friction on material specimens in pressurized, oxygen-rich atmospheres. In simplest terms, a test involves recording sensory data while rubbing two specimens against each other at a controlled normal force, with either a random stroke or a sinusoidal stroke having controlled amplitude and frequency. The term multimodal in the full name of the apparatus refers to a capability for imposing any combination of widely ranging values of the atmospheric pressure, atmospheric oxygen content, stroke length, stroke frequency, and normal force. The MFIT was designed especially for studying the tendency toward heating and combustion of nonmetallic composite materials and the fretting of metals subjected to dynamic (vibrational) friction forces in the presence of liquid oxygen or pressurized gaseous oxygen test conditions approximating conditions expected to be encountered in proposed composite material oxygen tanks aboard aircraft and spacecraft in flight. The MFIT includes a stainless-steel pressure vessel capable of retaining the required test atmosphere. Mounted atop the vessel is a pneumatic cylinder containing a piston for exerting the specified normal force between the two specimens. Through a shaft seal, the piston shaft extends downward into the vessel. One of the specimens is mounted on a block, denoted the pressure block, at the lower end of the piston shaft. This specimen is pressed down against the other specimen, which is mounted in a recess in another block, denoted the slip block, that can be moved horizontally but not vertically. The slip block is driven in reciprocating horizontal motion by an electrodynamic vibration exciter outside the pressure vessel. The armature of the electrodynamic exciter is connected to the slip block via a horizontal shaft that extends into the pressure vessel via a second shaft seal. The reciprocating horizontal motion can be chosen to be random with a flat spectrum over the frequency range of 10 Hz to 1 kHz, or to be sinusoidal at any peak-to-peak amplitude up to 0.8 in. (.2 cm) and fixed or varying frequency up to 1 kHz. The temperatures of the specimen and of the vessel are measured by thermocouples. A digital video camera mounted outside the pressure vessel is aimed into the vessel through a sapphire window, with its focus fixed on the interface between the two specimens. A position transducer monitors the displacement of the pneumatic-cylinder shaft. The pressure in the vessel is also monitored. During a test, the output of the video camera, the temperatures, and the pneumatic-shaft displacement are monitored and recorded. The test is continued for a predetermined amount of time (typically, 10 minutes) or until either (1) the output of the position transducer shows a sudden change indicative of degradation of either or both specimens, (2) ignition or another significant reaction is observed, or (3) pressure in the vessel increases beyond a pre-set level that triggers an automatic shutdown.
Tool, maraging, and stainless steels fabricated into biaxial pressure vessels