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.
The effects of cavitation flow on piston type, positive displacement, hydraulic pumps are discussed. The operating principles of the pump and the components which are most subject to erosion effects are described. The mechanisms of cavitation phenomena are identified from photographic records. Curves are developed to show the solubility of air in water, oil-water emulsion, and industrial hydraulic oil.
The results of research to identify the cavitation erosion damage mechanisms at the microscopic level for three metals (aluminum, stainless steel, and titanium) representing a range of properties and microstructure are presented. The metals were exposed to cavitation generated in distilled water by a 20-kHz ultrasonic facility operating at a vibration amplitude of 2 mils. Representative properties of the metals and experimental details are summarized. Replicas of the eroded surfaces of the specimens obtained periodically during exposure were examined with a transmission electron microscope to follow progression of the erosion damage and identify dominant erosion mechanisms as a function of exposure time. Eroded surfaces of selected specimens were also examined with a scanning electron microscope to assist in the interpretation.
The phenomenon of cavitation flow damage is discussed. The initial phases of damage and the effect of external stresses on the extent to which damage is incurred are analyzed. Three experimental facilities were used to procure the data required: (1) a water loop with venturi, (2) a mercury loop with venturi, and (3) a vibratory facility (stationary specimen, nonflow system). A description of each system is provided for the clarification of test conditions. Photographs of typical cavitation damage instances are included.
A method for predicting thermodynamic effects of cavitation (changes in cavity pressure relative to stream vapor pressure) is presented. The prediction method accounts for changes in liquid, liquid temperature, flow velocity, and body scale. Both theoretical and experimental studies used in formulating the method are discussed. The prediction method provided good agreement between predicted and experimental results for geometrically scaled venturis handling four different liquids of widely diverse physical properties. Use of the method requires geometric similarity of the body and cavitated region and a known reference cavity-pressure depression at one operating condition.
Four-volume publication contains extensive cavitation data and simplified technique for predicting cavitation performance of pumps.
The evaluation of the transfer function for an axial cavitating inducer in a high performance turbopump is considered. An approach for the evaluation of the compliance and mass flow gain factor can be used as a basis for the determination of the first nonzero terms in the expansions of all the elements in the transfer function. Applications of the described method to turbopumps are discussed and the theoretical results are compared with test observations in the case of the Saturn J2 engine. It is demonstrated that a dynamic transfer function which includes not only a compliance but also a mass flow gain factor is a satisfactory first approximation for a cavitating turbopump.
A semi-empirical entrainment theory was employed to correlate the measured temperature depression, Delta T, in a developed cavity for a venturi. This theory correlates Delta t in terms of the dimensionless numbers of Nusselt, Reynolds, Froude, Weber and Peclet, and dimensionless cavity length, L/D. These correlations are then compared with similar correlations for zero and quarter caliber ogives. In addition, cavitation number data for both limited and developed cavitation in venturis are presented.
Experimental observations of instability of cavitating inducers were made for two different inducers operating at different flow coefficients. In general, instability occurred just before head breakdown. Auto-oscillation and rotating cavitation were observed. Analysis of small-amplitude behavior of the inducer and hydraulic system is carried out, and analytical predictions of stability limits were compared with experiment.
The net positive suction head (NPSH) requirements for a pump are determined by the combined effects of cavitation, fluid properties, pump geometry, and pump operating point. An important part of this determination is the temperature depression (Delta T). Correlations are presented of the temperature depression for various degrees of developed cavitation on venturis and ogives. These correlations, based on a semi-empirical entrainment theory, express Delta T in terms of the dimensionless numbers of Nusselt, Reynolds, Froude, Weber, and Peclet, and dimensionless cavity length (L/D). The Delta T data were obtained in Freon 114, hydrogen and nitrogen for the venturis and in Freon 113 and water for the ogives.
The unsteady dynamics of cavitating cascades and inducer pumps were studied with a view to understanding (and possibly predicting) the dynamic characteristics of these devices. The chronology of the research is summarized as well as the final conculsions for each task. The construction of a dynamic pump test facility and its use in making experimental measurements of the transfer function is described as well as tests conducted using a scale model of the low pressure liquid oxygen turbopump inducer in the shuttle main engine. Auto-oscillation and unsteady inlet flow characteristics are discussed in addition to blade cavity influence and bubbly cavitation.
Dynamic transfer functions for two cavitating inducers of the same geometry but different size are presented, compared and discussed. The transfer functions for each inducer indicate similar trends as the cavitation number is decreased; only minor changes are noted with changes in the flow coefficient, the uniformity of the inlet flow or the temperature of the water (21-74 C). The non-dimensional results for the two sizes are compared with themselves and with theoretical calculations based on the bubbly flow model. All three sets of results compare well and lend further credence to the theoretical model. The best values of the two parameters in the model are evaluated and recommended for use in applications.
Cavitation erosion experimental data was analyzed by using normalization and curve-fitting techniques. Data were taken from experiments on several materials tested in both a rotating disk device and a magnetostriction apparatus. Cumulative average volume loss rate and time data were normalized relative to the peak erosion rate and the time to peak erosion rate, respectively. From this process a universal approach was derived that can include data on specific materials from different test devices for liquid impingement and cavitation erosion studies.
The authors are currently carrying out a study of the cavitation erosion of different bearing metals and alloys in mineral oils were studied. The variations of weight loss, the pit diameter and depth due to cavitation erosion on Al 6061-T6 in mineral oil and water are presented.
Cavitation-erosion experimental data previously covering several materials tested in a rotating disk device and a magnetostriction apparatus were analyzed using new normalization and curve-fitting techniques. From this process a universal approach is derived which can include data from cavitation and liquid impingement studies for specific materials from different test devices.
The variations of the mean depth of penetration, the mean depth rate of penetration, MDRP, the pit diameter 2a and depth h due to cavitation attack on Al 6061-T6, Cu, brass of composition Cu-35Zn-3Pb and Ti-5A1-2.5Sn are presented. The experiments are conducted in a mineral oil of viscosity 110 CS using a magnetostrictive oscillator of 20 kHz frequency. Based on MDRP on the materials, it is found that Ti-5Al-2.5Sn exhibits cavitation erosion resistance which is two orders of magnitude higher than the other three materials. The values of h/a are the largest for copper and decreased with brass, titanium, and aluminum. Scanning electron microscope studies show that extensive slip and cross slip occurred on the surface prior to pitting and erosion. Twinning is also observed on copper and brass.
Cavitation erosion studies of aluminum 6061-T6 in mineral oil and in ordinary tap water are presented. The maximum erosion rate (MDPR, or mean depth of penetration rate) in mineral oil was about four times that in water. The MDPR in mineral oil decreased continuously with time, but the MDPR in water remained approximately constant. The cavitation pits in mineral oil were of smaller diameter and depth than the pits in water. Treating the pits as spherical segments, we computed the radius r of the sphere. The logarithm of h/a, where h is the pit depth and 2a is the top width of the pit, was linear when plotted against the logarithm of 2r/h - 1.
Size scaling in cavitation erosion is a major problem confronting the design engineers of modern high speed machinery. An overview and erosion data analysis presented in this paper indicate that the size scale exponent n in the erosion rate relationship as a function of the size or diameter can vary from 1.7 to 4.9 depending on the type of device used. There is, however, a general agreement as to the values of n if the correlations are made with constant cavitation number.
Experimental investigations have been conducted to determine the early stages of cavitation attack on 6061-T6 aluminum alloy, electrolytic tough pitch copper, brass, and bronze, all having polycrystalline fcc matrices. The surface profiles and scanning electron micrographs show that the pits are initially formed at the grain boundaries, while the grain surfaces are progressively roughened by multiple slip and twinning. The initial erosion is noted to have occurred from the material in the grain boundaries, as well as by fragmentation of part of the grains. Further erosion occurred by shearing and necking of the surface undulations caused by plastic deformation. The mean penetration depth, computed on the basis of mass loss, was lowest on the bronze and greatest on the copper. Attention is given to the relation of cavitation attack to grain size, glide stress and stacking fault energy.