An induced-convection effect upon the peak- boiling heat flux.
Induced convection effect on peak pool-boiling heat flux under variable gravity, pressure and size conditions and various boiled liquids
SEARCH · Search NASA
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.
Induced convection effect on peak pool-boiling heat flux under variable gravity, pressure and size conditions and various boiled liquids
Mathematical correlation of effects of gravity, size, boiled liquid, pressure, and configuration on peak and minimum boiling heat fluxes
An experimental study of the inception of boiling in potassium in forced convection is reported. The boiler consisted of a 0.19-inch inside diameter, niobium-1% zirconium boiler tube approximately six feet long. Heating was accomplished by direct electrical tube wall conduction. Experiments were performed with both all-liquid fill and two-phase fill startup sequences and with a range of flow rates, saturation temperatures, inert gas levels, and fill liquid temperatures. Superheat of the liquid above the equilibrium saturation temperature was observed in all the experiments. Incipient boiling liquid superheat ranged from a few degrees to several hundred. Comparisons of these data with other data and with several analytical treatments are presented.
A photoelectric particle counter was used for the measurement of aerosol boiling points. The operational principle involves raising the temperature of the aerosol by vigorously heating a portion of the intake tube. At or above the boiling point, the particles disintegrate rather quickly, and a noticeable effect on the size distribution and concentration is observed. Stratospheric aerosols appear to have the same volatility as a solution of 75% sulfuric acid. Chemical analysis of the aerosols indicates that there are other substances present, but that the sulfate radical is apparently the major constituent.
Film boiling of a saturated liquid on a vertical surface is analyzed to determine the local heat-transfer rates as a function of height and heater-surface superheat. Experiments show that the laminar-flow model is inadequate. A turbulent-vapor-flow model is used, and the influence of the interfacial oscillations is incorporated on a semiempirical basis. Measurements of local film boiling were obtained with a transient technique using saturated liquid nitrogen.
The conditions at which the void in subcooled boiling starts to undergo a rapid increase were studied experimentally. The experiments were performed in a 12.7 x 9.5 mm rectangular channel. Heating was from a 3.2 mm wide strip embedded in one wall. The pressure ranged from 9.45 to 20.7 bar, mass velocity from 600 to 7000 kg/sq m sec, and subcooling from 16 to 67 C. Photographs were used to determine when detached bubbles first appeared in the bulk flow. Measurements of bubble layer thickness along the wall were also made. Results showed that the point of net vapor generation is close to the occurrence of fully-developed boiling.
The dispersion relation is evaluated numerically for Taylor waves in a viscous unstable interface with surface tension. The solution takes account of transverse curvature and the numerical evaluations apply to horizontal cylindrical, as well as to plane, interfaces. The result is verified with frequency and wavelength data obtained during film boiling on horizontal wires. A very general empirical correlation is given, en passant, for the vapor blanket thickness during film boiling.
The initiation of boiling of water on metal surfaces during pressure transients has been investigated. The data were obtained by a new technique in which light beam fluctuations and a pressure signal were simultaneously recorded on a dual beam oscilloscope. The results obtained agreed with those obtained using high speed photography. It was found that, for water temperatures between 90-150 C, the wall superheat required to initiate boiling during a rapid pressure transient was significantly higher than required when the pressure was slowly reduced. This result is explained by assuming that a finite time is necessary for vapor to fill the cavity at which the bubble originates. Experimental measurements of this time are in reasonably good agreement with calculations based on the proposed theory. The theory includes a new procedure for estimating the coefficient of vaporization.
The quantitative determination of the influence of heater surface orientation and gravity on nucleate pool boiling of liquid nitrogen and liquid hydrogen is described. A transient calorimeter technique, well suited for obtaining pool boiling data under reduced gravity and used earlier by Clark and Merte (1963), was employed after being adapted to flat a surface whose orientation could be varied. The obtained determination results are reviewed.
Review of investigation results on the characteristics of slush (i.e., mixtures of liquid and solid) and boiling methane, and of binary mixtures of methane and other natural gas constituents, intended for use as fuels for high performance aircraft, rocket engines, and motor vehicles. The investigation extends to the characterization of slush prepared from pure methane and binary mixtures of methane with nitrogen, ethane, and propane, and the boiling properties of these mixtures.
Representative boiling heat transfer and pressure drop information obtained primarily from past NASA and AEC programs is presented which is applicable to forced flow and once-through boiler systems. The forced convection boiler has a number of advantages: little possibility of flow mal-distribution; heat transfer characteristics are usually consistent; and conductances are predictable, so that higher heat fluxes may be employed with safety (which leads to more compact, lighter weight equipment). It was found that in gas-fired systems particularly, the controlling heat transfer resistance may be on the hot side, so that increased fluxes would require extended surfaces. If in a power generation system the working fluid is very expensive, a forced flow boiler can be designed especially for small holdup volume. If the fluid is temperature sensitive, the boiling side wall temperatures can be tailored to maintain maximum heat transfer rates without overheating the fluid. The forced flow and once-through configurations may be the only type which can satisfy a specific need (such as the automotive Rankine cycle power plant design having a very short time-response boiler).
Previously reported vaporization time data of liquid nitrogen drops in film boiling on a flat plate are about 30 percent shorter than predicted from standard laminar film boiling theory. This theory, however, had been found to successfully correlate the data for conventional fluids such as water, ethanol, benzene, or carbon tetrachloride. Experimental evidence that some of the discrepancy for cryogenic fluids results from ice contamination due to condensation is presented. The data indicate a fairly linear decrease in droplet evaporation time with the diameter of the ice crystal residue. After correcting the raw data for ice contamination along with convection, a comparison of theory with experiment shows good agreement.
Previously reported vaporization time data of liquid nitrogen drops in film boiling on a flat plate are about 30 percent shorter than predicted from standard laminar film boiling theory. This theory, however, had been found to successfully correlate the data for conventional fluids such as water, ethanol, benzene, or carbon tetrachloride. This paper presents experimental evidence that some of the discrepancy for cryogenic fluids results from ice contamination due to condensation. The data indicate a fairly linear decrease in droplet evaporation time with the diameter of the ice crystal residue. After correcting the raw data for ice contamination along with convection, a comparison of theory with experiment shows good agreement.
A propulsion system and method are disclosed for controlling the attitude and drag of a space vehicle. A helium dewar contains liquid helium which cools an experiment package. The helium is heated or vented to keep the temperature between 1.5 and 1.7 degrees K to maintain adequate helium boil-off gas as a propellant without adversely affecting the experiment package which is contained in the helium dewar for protection from solar heating. The apparatus includes auxiliary heater and temperature sensor for controlling the temperature of the helium. The boil-off gas propellant is delivered to thruster modules to control vehicle attutude and compensate for drag.
A computer model is developed and verified which is capable of determining the cross-sectional temperature distribution within a heat pipe with an attached radiator fin; such heat pipes would be plugged into contact heat exchangers designed to carry heat from a space station habitation module to the radiator elements through a centralized fluid loop. The model can furnish information for determining the susceptibility of the monogroove heat pipe to boiling, as well as the location and magnitude of that boiling.
An experimental study was made of subcooled flow boiling with LOX; data were obtained with thermocouples attached to the walls of an electrically heated tube through which the LOX flowed. Pressure levels in the tube ranged from 200 psia to supercritical values, and the cooling level varied from the saturated condition to 50 F. Reynolds numbers ranged from 1.5 x 10 to 3.5 x 10 to the 5th. Results show that an increase in either pressure, subcooling level, or Reynolds number produces an increase in the film boiling heat transfer coefficient. Data for this range of parameters were found to correlate in terms of the Stanton number vs (T sub w - T sub s)/(T sub s - T sub b), where T is temperature, w is the wall, s is the saturated condition, and b is the bulk.
The paper presents results obtained during pool boiling from flat surfaces over body force levels of a/g = -1 to 0 to 20 for LN2 and R-113 under steady and transient conditions. Pyrex and metal surfaces are used. Particular attention is given to the case where the effective gravity is perpendicular to the heating surface. At the lower levels of heat flux with nucleate boiling, the heater surface superheat first increases and then decreases as the body force increases above a/g = 1. This is due to the greater contribution of nonboiling convection.
Results are presented for a preliminary experiment involving saturated pool boiling at 1 atm from rotating 2 and 3 in. diameter spheres which were immersed in liquid nitrogen (LN2). Additional results are presented for a stationary, 2 inch diameter sphere, quenched in LN2, which were obtained utilizing a more versatile and complete experimental apparatus that will eventually be used for additional rotating sphere experiments. The speed for the rotational tests was varied from 0 to 10,000 rpm. The stationary experiments parametrically varied pressure and subcooling levels from 0 to 600 psig and from 0 to 50 F, respectively. During the rotational tests, a high speed photographic analysis was undertaken to measure the thickness of the vapor film surrounding the sphere. The average Nusselt number over the cooling period was plotted against the rotational Reynolds number. Stationary sphere results included local boiling heat transfer coefficients at different latitudinal locations, for various pressure and subcooling levels.