Search

This paper presents a comparative study of the extinguishment of liquid heptane and gaseous propane diffusion flames caused by the addition of flame-suppressing agents (N2 and CO2) into a coflowing stream of air. Measurements included mass burning rate, concentrations of suppressants at flame extinction, inflame temperature profiles, and exhaust composition. Adiabatic equilibrium temperatures of the stoichiometric mixtures of the fuels in pure airstream, and in the airstream containing flame suppressants with the concentrations corresponding to flame extinction, were calculated. Although the extinction processes of liquid and gaseous fuel flames are markedly different, even when both are buoyancy dominated, the extinction mechanism of CO2 and N2 in both flames is primarily thermal. The reaction zone temperature in the flame-anchoring region of the heptane flame at extinction agrees well with the adiabatic flame temperature corresponding to the lean flammability limit. In propane flame, where the flame base is lifted off the burner before extinction, the reaction zone temperature at extinction is 100-200 K higher than the adiabatic flame temperature at the lean flammability limit.

A simple laboratory apparatus is described for observing the performance of fire extinguishing agents in total flooding applications. Known compositions of agents mixed with dry air are prepared in a 2 L glass combustion flask, and a flame fed by propane, butane, or other gaseous fuel from a pressure cylinder is inserted into the flask through a standard taper joint. Burning times to extinguishment are measured, both in the absence and presence of added physical or chemical agents. From the known rate of propane burning and initial compositions, the percentages of gas volume in the combustion flask at extinguishment are calculated for each burning time. Agents that are not sufficiently volatile to be investigated with the ordinary cup burner method are readily studied with the new technique.

A pronounced synergism is reported for mixtures of physical and chemical agents used in flame extinguishment. Mixtures containing approximately 90% (by volume) of either carbon dioxide or nitrogen and 10% of CF3Br or CF2ClBr are much more effective in suppressing fires (in cup-burner tests) than would be predicted by an additivity of effects. A method is proposed for delivering solutions of chemical agents dissolved in liquid CO2 to obtain cooperative fire-suppression effects, while simultaneously reducing the danger to humans and the environment. A simple model has been developed to give very good predictions of the extinguishment concentrations of mixtures of physical and chemical agents.

Unconfined uniaxial creep tests were conducted at -8.5 degree C on sand-ice samples with an average wet density of 124 pcf (1,986 kg/m3). The samples were 94% water saturated, on the average, prior to freezing. Thus the data represent the behavior of slightly undersaturated frozen sands. Stress-strain data for different durations of loading were obtained by replotting the creep data. Values of the stress characterizing the change in stress-strain relations are confirmed by creep rate-stress data. It is shown that use of a single continuous stress function to represent stress-strain data in models must be limited to either stresses greater than, or less than, the limiting long-term strength. A single continous stress function cannot represent realistically the entire stress range. /ASCE/

The ever-increasing use of polymeric materials has caused much concern to those involved in fire prevention and control. There is very little information available in the literature specifying the ignition characteristics of the many plastics and rubber compounds now being utilized.

The performance of a conventional hydrocyclone for separation of immiscible liquids was investigated experimentally. The results are particularly useful for evaluation of multi-cyclone process design of same system. The effect of materials of constructions of the hydrocyclone on the separation efficiency was also tested. Significant differences in performance were noted depending on the preferential wettability characteristics of the solid surfaces for the liquids being separated. On a etudie experimentalement l'efficacite d'un hydrocyclone classique pour separer des liquides non-miscibles. Les resultats sont particulierement utiles pour l'evaluation de la conception d'un procede a base de cyclones multiples pour le meme systeme. On a aussi etudie l'effet des materiaux de construction de l'hydrocyclone sur l'efficacite de la separation. On a observe des differences importances dans ladite efficacite selon les degres de mouillage preferentiel des surfaces des matieres solides que presentaient les liquides qu'on separait.

The oscillating quartz crystal viscometer has been used to investigate possible viscoelastic behavior in synthetic lubricating fluids and to obtain viscosity-pressure-temperature data for these fluids at temperatures to 300 F and pressures to 40,000 psig. The effect of pressure and temperature on the density of the test fluids was measured concurrently with the viscosity measurements. Viscoelastic behavior of one fluid, di-(2-ethylhexyl) sebacate, was observed over a range of pressures. These data were used to compute the reduced shear elastic (storage) modulus and reduced loss modulus for this fluid at atmospheric pressure and 100 F as functions of reduced frequency.

During an investigation of the time for ignition of polymeric materials under the influence of radiant heating, it was found that the polymer surface reflectance-absorptance characteristics were a major factor in the variance of the ignition times. A subsequent research study was made of the reflectance-absorptance characteristics of those polymers used in the ignition testing. Reflectance values were obtained over the wavelength of 0.3 to 2.5 microns using a double-beam Cary model 14 spectrophotometer with an integrating sphere reflectometer and over the wavelengths of 1.0 to 10.0 microns using a Gier-Dunkle Hohlraum with a Perkin-Elmer spectrophotometer. Absorptance values were obtained by means of Kirchoff's Law, Drawings and graphs are included which illustrate the test apparatus and type of data collected. A table of average absorptances of several polymers are given and listed according to the particular type of heat source used. Average absorptances of the polymers over the monochromatic wavelength span of the heat sources were calculated using the equation Mathematical analyses were developed and are presented for both the integrating sphere reflectometer and Gier-Dunkle Hohlraum unit. Drawings and graphs are included which illustrate the test apparatus and type of data collected. A table of average absorptances of several polymers are given and listed according to the particular type of heat source used.

Unlike coal, for which solvent liquefaction dates back to the beginning of this century, treatment of oil shale using various solvents has received attention only in recent years as an alternative process to producing liquid fuels by retorting. Solvent extraction offers the promise of overcoming the relatively low yields of desirable products obtained and the high energy expenditures required in pyrolysis. The present study was undertaken to investigate the solubilization of a Moroccan oil shale at 300{degree}C and 3000 psig in toluene. In this case the extent of solubilization was measured in terms of tetrahydrofuran extractables at refluxing conditions. Analyses of the soluble material produced were obtained by gel permeation chromatography and proton NMR since these provide valuable information for mechanistic interpretations.

The objective of this experimental investigation was to obtain information on the reaction kinetics for the hydrogen reduction of copper from copper sulfate solutions in a continuous, countercurrent tubular reactor. Aqueous copper sulfate solution was introduced at the top of the reactor while hydrogen was sparged into the bottom. Separations of the bottom (liquid-solid) and overflow (gas-liquid) products were accomplished in individual high pressure separators. The rate of reaction increases with temperature up to about 180 to 220 °C depending on the residence time and thereafter levels off or decreases slightly. Feed solutions having a pH greater than 1.8 yielded a product which was contaminated with cuprus oxide. Over the range of this study, the operating pressure, and consequently the solubility of hydrogen in the aqueous reactant demonstrated negligible effect on the depletions. Likewise, very little change in depletions was observed beyond residence times of 10 minutes. Acid addition to the feed retarded the rate of reaction, whereas addition of sulfate ion increased the rate; these two observations agree with the results of previous investigators.

Previous experimental studies on the ignition of dry wood by thermal radiation show poor agreement because all of the primary variables have not been monitored, and consequently they are not included in the resulting correlations. Although the density of the wood, the thickness of the specimen, and the magnitude of the incident irradiance have generally been recognized as principal parameters, there is little evidence in the literature that the spectral absorptance of the wood relative to the spectral quality of the incident radiation has been appreciated as a dominant variable. Therefore, the purpose of this investigation was to examine the significance of all of these variables. Piloted ignition data were obtained for thirteen dry wood species having thicknesses from less than 0.2 cm to more than 2.5 cm and densities from less than 0.1 g/cm 3 to more than 1.0 g/cm 3 . Diffusion flames and high-temperature tungsten lamps were used as sources of thermal radiation at incident irradiances from about 0.6 cal/cm 2 -sec to about 3.5 cal/cm 2 -sec. The spectral absorptances of the wood samples were measured in order to calculate an average absorptance over the range of spectral quality of the incident radiation. The ignition data were correlated on the basis of relationships suggested by considering the wood specimens as infinite slabs of inert, opaque material. It was found that, when the density, thickness, and absorbed irradiance were included in the correlation of the experimental data, a unique ignition correlation could be defined.

In the past, the authors and their colleagues have studied the effects of wind-blown flames involving burning gases and liquid pool fires. It would seem logical to conduct a similar study with solid fuels. This the authors have done, and they report their observations here.

A simplified and improved correlation for the drag coefficient of windblown natural gas flames is given. Experimental results leading to the correlation were obtained in a low-speed wind tunnel specifically designed for such studies at the University of Oklahoma North Campus.

Heat transfer and fiuid fiow were studied in a new type of high pressure tubular reactor, which utilizes a machined helix between the liner and body for passage of a heating or cooling fiuid. Heat transfer rates for the reactor are very high. Over-all heat transfer coefficients based on the i-nside area of the reactor were as high as 540 Wu/(hr) (sq. ft.)( deg F) for condensation of steam by water. The friction loss for flow through the helix is higher than for straight pipe, because of secondary flow. Pressure drop data indicate that in laminar flow, the friction factor for the coil can be calculated by Prandtl's equation for coiled pipe. (auth)

The generally accepted models on the ignition of wood are not valid. So say the authors, who present an improved model, which correlates data on five different woods.

The bending of a flame by wind influences the amount of heat transferred by radiation and convection, the fuel burning rate, and the flame spread rate. To what extent will a flame be bent by wind? The author presents correlations of data taken from liquid pool fires, which enable us to predict flame bending and trailing for large fires.

In the first of this two-part series, the authors discussed the effects of wind on flame bending and trailing. This paper describes investigations of liquid fuel burning rates, flame lengths, wake gas temperatures, and flame radiation. Data are taken from the liquid pool fires described in the earlier article.

In spite of advice to the contrary, the authors spent two years developing a flame ignition cabinet, which throws a new light on ignition criteria. That the cabinet gives good results has been established by comparison with data obtained by others on nonflame ignitions.

What are the effects of flame behavior of a number of fires burning in close proximity to one another? The results of measurements of burning rates, heat feedback, flame height, and flame trailing are reported for fires involving liquid pools.

The present investigation is concerned with the dynamic characteristics of a 12-in. diameter continuous agitated-tank reactor vessel. Response of the vessel effiuent temperature to a change in coolant flow rate through an internal cooling surface is the subject of theoretical and experimental study. Experimental data were obtained through the use of frequency and transient response techniques. Studies were made for the passage of vessel-charge fluids with widely differing physical properties. Also data were taken for various conditions of fluid turbulence both inside and outside the internal heat transfer coils. Experimental and theoretical results are graphically compared. Recommendations are presented for the development of theoretical dynamic relationships.

Use of a torsionally oscillating quartz rod as a high‐pressure viscometer requires knowledge of the effect of pressure on the resonant frequency of the rod. Measurements were made to determine the change in resonant frequency with pressure to 8000 atm at ambient temperatures (23°–26°C) for crystals with resonant frequencies in air at atmospheric pressure of about 20 and 60 kHz. Smoothed data of the change in resonant frequency with pressure for both crystals are correlated to within ±2 Hz to 6000 atm by Δf0*(P)/f0(1 atm)=−6.897×10−7+1.122×10−4P+2.047×10−5P2, where P is in 1000 atm, and f0*(P) and f0 are in kHz. Above 6000 atm the correlation is less precise because of lower reliability of the 60 kHz data.

A numerical technique is presented for predicting the two-dimensional transient temperature distribution in freezing or thawing wet soil. The technique is applied in estimating the influence of seepage stream velocity on an artificial free-draining wet soil by a row of freeze pipes. It is shown that small ground-water flows perpendicular to the vertical axis of a freeze pipe have no significant effect on the rate of growth of the ice boundary in the horizontal direction. In fact, a flow on 0.01 cu ft per hr per sq ft to possibly 0.02 cu ft per hr per sq ft may even enhance the rate of propagation of the ice boundary. Increases in flow above those limits diminishes the rate of growth of the ice boundary quite markedly. At a flow rate of 0.50 cu ft per hr per sq ft, closure of the ice boundary between a row of 6-in freeze pipes with a center-to-center spacing of 8.6 ft will never occur.

During ignition testing of several polymeric materials, it was found that the surface reflectance/absorptance characteristics had a definite influence on the time of ignition. Since several of the polymers were either transparent or translucent, the monochromatic surface reflectance/absorptance had to be adjusted to account for energy transmittance. As a result, reflectance and transmittance measurements were made over the monochromatic wavelengths of 0.3-10.0 microns. Average values of reflectivity and transmissivity were calculated by the equation: \[{{(f)}_{av}}=\frac{\int {{f}_{\lambda }}{{e}_{\lambda }}d\lambda }{\int {{e}_{\lambda }}d\lambda !}\] where (f)av is either the average reflectivity, ray, or average transmissivity Tay, where fx is either the monochromatic reflectance rX or the monochromatic transmittance TX. A is the wavelength and a is the monochro-matic emissive power of the heat sourcA. Average values of reflectivity and transmissivity were calculated for several heat sources; radiant panel, carbon arcs (2900°K, 3900°, and 6000°K), solar radiation, benzene flame, and blackbody radiation. Plots of average reflectivity and transmissivity are illustrated, and charts of transmittance vs. wave length, and reflectance vs. wave length are included.

During the first year (Phase I) of this research program, the following elements of research have been performed: (1) the collection and processing of data for pure components, (2) the evaluation of the generalized MBWR equation of state for halocarbon saturated thermodynamic properties, (3) the determination of pure component parameters for the MBWR equation for ten halocarbons, isobutane and, over a limited range, for ammonia and water, (4) the investigation of modifications of the MBWR equation for improved prediction of properties of hydrogen bonding and polar fluids. With the results of the present work, MBWR parameters have now been determined for twenty-seven pure fluids, making the use of the MBWR equation in geothermal cycle calculations feasible for most candidate working fluids. From this research, it has been concluded that the MBWR equation can be used to correlate the thermodynamic properties of many of the hydrocarbons and halocarbons which are presently potential candidate working fluids for geothermal binary cycles. However, for ammonia and water, the MBWR equation cannot accurately describe behavior in the complete range of fluid states. In anticipation of future needs to describe complex fluids (such as fluorinol-water mixtures) which have been proposed as working fluids, efforts have beenmore » initiated to develop an equation of state capable of describing the behavior of hydrogen bonding and polar as well as nonpolar fluids.« less

Boiling heat transfer to cryogenic liquids has an obvious immediate application in the cooling of nuclear reactors, regenerative cooling of rocket engines, and industrial processes. In many instances these systems involve simultaneous, localized nucleate and film boiling as well as convective heat transfer at elevated pressures. Gross design data can be obtained for specific applications with all of these heat transfer modes occurring, but the key to eventual understanding lies in obtaining information on each process separately.

The strength and density of powder metallurgy (P/M) parts are influenced by powder characteristics, consolidation technique and sintering conditions. In the case of an alloy the properties are further improved by heat treatment. However, the properties of such P/M parts, which are not subsequently subjected t o a forging operation, are always lower than those obtained by conventional techniques for wrought metals and alloys. The objective of this study is t o investigate the effect of one of the processing variables, heat treatment, on the tensile strength and hardness of an isostatically compacted aluminum alloy corresponding t o 6061 (1.0 pct Mg, 0.6 pct Si, 0.25 pct Cu and 0.25 pct Cr). Although aluminum powder metallurgy is now practiced quite extensively, relatively few r e s e a r c h papers on this subject have been published.1"4 Aluminum-copper alloys (2xxxs e r i e s and A1-Mg-Si alloys (6xxx s e r i e s ) are usually solution heat t r e a t e d for dissolving maximum amounts of solute elements such as copper, magnesium and silicon in the aluminum solid solution at the highest possible temperature below the eutectic melting point . Homogenization of the solid solution thereby achieved increases the strength and hardness properties of the component with increasing solution heat treating temperature. For heat treatable alloys in the 2xxx and 6xxx s e r i e s , this treatment is conventionally done at temperatures in the range of 940 t o 970°F (777 to 794 K) which is close t o the eutectic melting temperature for most alloys in the s e r i e s except the alloy 6061. This aluminum-magnesium-silicon alloy contains 1.0 pct magnesium and 0.6 pct silicon, whichforms 1.5 pct magnesium silicide. Since some of the magnesium silicide remains undissolved when heat treating the alloy at the conventional temperature of 970°F (794 K) which is below the complete dissolution temperature of 1025°F (824 K), the strength of the alloy should be improved further by heat treating at higher temperatures.%° Therefore, the purpose of this study was to investigate the effect of solution heat treatment temperature on the tensile strength and hardness of 6061 components made from isostatically compacted powder of this blend. The commercial grade powderblend, 601AC, containing 1.0 pct magnesium, 0.6 pct silicon and 0.25 pet copper and which is similar in composition to commercial 6061 alloy, was processed in the following manner. The blend was isostatically compacted at a pressure of 40,000 ps i (276 MN/m2) in a