Laminar free-convection heat-transfer was treated here theoretically for the simplest case where it was assumed that with respect to a flat surface in infinite extended fluid, the surface and the fluid temperatures were uniform and its physical constants as well did not vary due to temperature. For the first was found the basis of the Pohihausen's transformation, then a similar transformation was introduced to the momentum-and energy-equation to solve them approximately, and was found a solution with arbitrary velocity-and temperature-profiles of the boundary layer. In acordance with this result, the approximate solutions of Squire, Yamagata and Sugawara-Michiyoshi were compared with each and with the Ostrach's exact solutions. Two approximate solutions are presented in this paper where approximate velocity- and temperature-profiles of the boundary layer are represented by the following equations respectively ; [table] where s is a function of the Prandtl number P_r. The solution (1) for about P_r<0.1 and the solution (2) for all P_r values give closer solutions than any conventional approximate solutions as to velocity profile, temperature profile and heat-transfer coefficient.

View full abstract

It was studied that physical constants for what temperature should be taken in arranging the experimental data on heat-transfer coefficients. From the numerical calculations made on water and ethylene-glycol, were given the following conclusions. Prandtl number, kinematic viscosity and thermal conductivity ;(the surface temperature)-1/4 (temperature difference between the surface and the fluid). Coefficient of cubical expansion ; the mean coefficient of cubical expansion to be made up to the arithmetic mean of the surface temperature and the fluid temperature with the fluid temperature as basis. For the second problem we derived from the momentum and energy equations an ordinary differential equation, which would give a solution for any Prandtl number and an arbitrary temperature difference distribution. The equation is soluble elementarily in the case where P_r→∞. From the several examples of numerical calculations were derived the following conclusions.(1) There is no effect of P_r upon the correlation between surface temperature distribution and distribution of heat transferred.(2) When the temperature at an approximately half height is taken as a typical temperature, the total heat transferred is generally equal to the value in the case of uniform surface temperature.

View full abstract

Generation of a vortex street layer similar to Karman's vortex street was found in the cource of transition from laminar flow to turbulent flow in observation made on a free-convection boundary layer by Schlieren method. When a laminar boundary layer becomes unstable, it changes itself to a vortex street layer. A vortex line forming the vortex street, producing a definite number of projections in its flow direction, takes a form of a horse shose line. Then, according to its nature ensues turbulent state. The point of transition from a laminar layer to a vortex street layer was G_rP_r≒4×10⁹. The point of transition from a vortex street layer to a turbulent layer was G_rP_r≒10¹⁰. The number of oscillation and the wave lengths of Schlieren photos, that is the number of vortex lines generated per unit time and their mutual distances in the vortex street, hold constant until the flow becomes turbulent.

View full abstract

Boiling heat transfer is greatly affected by ease of bubble formation conditioned by surface quality and contamination of the heating surface. This tendency is represented quantitatively as the proportionality constant ζ=αR/Δθ² where α=coefficient of heat transfer, R=radius of heating surface and Δθ=temperature difference between heating surface and boiling liquid. ζ decreases with time from the initial value ζ₀ to the final value ζ_∞ corresponding to the steady state of boiling. This process can be written ζ=ζ₀+(ζ₀-ζ_∞) exp (-kt), k=1.5 1/h It is shown that k is independent of the applied heat flux, surface quality of the heating surface and presence of solute, unless it deposits scale on the heating surface. The cause of the variation of ζ is desorption of adsorbed gas during exposure of the freshly polished heating surface to the atmosphere.

View full abstract

We have been studying experimentally the heat transmission characteristic of our tubes of new type as the case of air flowing normal to a bank of staggered tubes, and have already reported a portion of the results twice. Now we will report about the characteristic of the bundles of No.I tubes (each has two fins 10 mm height, stretching in axial direction on its front surface of circular tube holding the angle 2θ=40°between the two fins), of No.III tubes (each has two fins of 14 mm height, on the front (θ=0°) and the back (θ=180°) of the circular tube respetively) and of the ordinary circular tubes named as "simple circular tubes" to make a distinction from the two above. Here all the tubes have the same outer diameter of 38 mm. The main results of our researches are as follows : (1) No.I tubes as bundle have the characteristic of higher coefficient of heat transfer (as for the total area of its outer surface) and less pressure loss, comparing with the case of simple circular tubes. Especially this characteristic is most striking when the tubes are arranged in staggered type holding the same spacings of 80 mm to the transverse and longitudinal directions respectively. (2) We know that No.III tube has inferior characteristic when it is used as single tube (1st reprot). Now we see that the bundle of No.III tubes in staggered arrangement with equal spacings of 80 mm has in appearance the characteristic approximate to that of the case of No.I tubes, if we do not consider the increase of tube weight. But according to the order of magnitude of coefficients of heat transfer, we can say that the largest is No.I tubes', the second is the simple circular tubes' and the third is No.III tubes'.

View full abstract

This report deals with temperature measurement of the piston surface. The test engine used is a four-stroke spark-ignition engine with a water cooled single cylinder, bore×stroke being 114.3×140 mm and its output is 6 HP at 850 rpm. The iron-constantan thermocouple is used and its hot junction is located in the surface centre of the aluminium-alloy piston head. In order to measure accurately the periodic temperature change, the hot junction is constructed with the metallic film. Two wires-iron and constantan-are electricaly insulated and led to the piston surface, and there they are shorted by the metallic film. The film is constructed successfully with the aluminium metallikon. In connecting thermocouple to the amplifier, the input transformer is used. By these methods, the temperature change of the aluminium-alloy piston has been measured without being disturbed by the spark plug.

View full abstract

The preceding papers have set out the calculations of the performance of turbocharged two-cycle Diesel engines on the basis of given values of the variable parameters. These are the kind of calculations which must be carried out in order to estimate the performance at the design point of an engine. In gas turbines, the part load performance can be estimated by linking together in correct manner the performance of several components, most of whose characteristics cannot be expressed mathematically. In this paper, we attempt to estimate the part-load performance of turbocharged two-cycle Diesel engines by the same process, replacing the combustor of a simple gas turbine with the Diesel cylinders. At the same sime, the influences of changes in the assumed component performance features on the part-load performance of an engine are found out by this method.

View full abstract

The operating characteristics of turbocharged four-cycle Diesel engines are estimated with the equilibrium operating lines, which are drawn by linking together the characteristic curves of the components of the given engine system. In matching the engine output to the equilibrium operating line, two methods are applied. The one is the theoretical method as shown on two-cycle engines in the previous paper, and the other is the experimental one with the practical engine which is mechanically supercharged.

View full abstract

When two exhaust pipes are connected with a short plain pipe or a diffuser at their open ends, it is possible to promote the scavenging of cylinders by the rarefaction waves produced in the exhaust pipes with the blow down in either of them. By measuring the pressure variations in the exhaust pipes of an air model, the effect of the short pipe or the diffuser on the energy of rarefaction waves has been investigated. Further the optimum area-ratio of the short pipe or the diffuser to the exhaust pipe to produce the maximum energy of the rarefaction wave has been theoretically calculated on the assumption that the relation of steady flow can be applied to the short pipe or the diffuser. The results obtained agree fairly well with those of the experiment.

View full abstract

It is a wellknown fact that the scavenging of an internal combustion engine by utilizing the wave effect in the exhaust pipe is promoted with a stepped or conical open end. However, most of the works along this line hitherto performed are confined to the qualitative ones. The present authors have theoretically treated the wave phenomena in a pipe with a stepped or conical open end, and determined the optimum area-ratio of the step or the diffuser to produce the maximum energy of rarefaction wave. The results obtained are experimentally confirmed with an air model by measuring the energy of rarefaction waves.

View full abstract

In the previous paper, the authors already reported the effects of engine speed and pressure difference on the ineffective lift in an internal combustion engine, using a model (A) of piston head. Successively, they have carried out some experiments with another model (B) of piston head in order to examine the effect of tappet clearance on the ineffective lift, by means of two methods, that is, measuring air flow and recording the pressure flactuation in the cylinder. Further, using the regular piston head, they have determined the ineffective lift of suction and exhaust valves by the pressure diagram. Some results obtained in such experiments are summarized as follows : a) If the ineffective lift be used instead of the ineffective angle, it is not necessary at all to consider the values of tappet clearance and the kinds of valve lift curve. b) It is possible to estimate the ineffective lift not only by the measurements of air flow, but also by the analyses of the pressure diagrams. c) In all cases of using the piston head models (A) and (B) and the regular piston head, the ineffective lift increases in proportion to the engine speed and decreases with the increase of the pressure difference between suction and exhaust sides. d) In all cases, the minimum value of ineffective lift remains constant in spite in spite of engine speed and pressure difference as shown already in the previous paper.

View full abstract

Near the T.D.C. of the expansion stroke of the engine, the piston-ring speed is 0 and its load and temperature reach a maximum. If it were based ob the steady-flowhydrodynamics, the oil film between the ring and the cylinder would have to be broken down. In practice, however, the fluid-film lubrication is achieved in the high speed region. These phenomena can be explained by a dynamical theory of the hydrodynamics, the piston-ring speed and load vary so extremely that the oil-film thickness changes rapidly, and this rate of change in the oil-film thickness generates a load carrying capacity by the squeeze action. This paper introduces the equations of the above-mentioned theory, and some calculations are carried out for a ring, the surface profile of which consists of the quadratic and the parallel line.

View full abstract

Our object is to perform an exact experiment, in accordance with my own similarity, on oil-cooling of laminar flow in the circular pipe with constant temperature. Here, the similarity is that the mean temperature θ_m of passed flow oil depends not only on G_z=Wc/kL but also on s'=(t_i-t_w)/(t_i-t_∞) and K=ω/R (t_i-t_∞), and these are dimensionless quantities. We have carried out a series of experement on θ_m to G_z under the condition that s' and K have been kept at constant values. And we have varied the values of s' and K at each series of experiment by changing oil or temperature ranges. As the result of our experiments we have made sure that curves of θ_m to G_z are distinguished from the Nusselt's curve and separates with one another by the values of s' or K, and so it can be recognized that my similarity is right.

View full abstract

This paper deals with a mathematical solution on the heat transfer of cooling pipe with constant temperature and laminar flow oil in it. The partial differential equation of temperature distribution and equations of velocity and of pressure gradient were derived, as the viscosity of oil varies only with temperature and other properties of oil remain unchanged. I have attempted to solve these simultaneous equations. As the process I have taken in a conception of thermal boundary layer, and introduced a thermal equation corresponding to the Karman's momentum equation of boundary layer problems. The calculations of mean temperature θ_m were compared with my experimental results, although sufficient coincidence is difficult to obtain, for these were thermal experiments, but I have recognized that this theory and the experiments agree with each other. Moreover, I am of the opinion that we could ascertain the effect of s' and K following my similarity.

View full abstract

In this paper, the author has experimentally investigated the influences of the roughness of the inner surface of pipe upon the pressure drop due to friction in two-phase flow of the mixture of air and water in inclined and vertical pipes. The results of investigation may be summarized as follows : (1) For the wide range of the roughness, the ratio of pressure drop is written as follows : [numerical formula] where ΔP_TP pressure drop due to friction in two-phase flowing, ΔP_w that in water alone flowing, φ_d dryness fraction, Z value decided by the roughness and inclination angle of pipe. (2) The variations of the values of Z due to the roughness of inner surface and the inclination angle of pipe been clarified. (3) The influences of roughness and inclination angle upon the ratio of the coefficient of friction are presented.

View full abstract

For the cross stays provided in the neighbourhood of the line where the semi-circular crown plate of a loco-type boiler and its flat side plates are joined, no designing formula has been specified in any Rules or Regulations. Therefore, the authors, carring out the theoretical analyses and experiments, could establish the designing formulae and moreover make the following matters clear. (1) The semi-circular crown plate itself, if not connected to the inner firebox by crown stays, is in equilibrium with the internal pressure. Consequently it is sufficient for cross stays to be designed so as to support only the flat portion below the stays. (2) In the case of a boiler having crown stays, the design of cross stays should be carried out considering the additional force due to the downward pull of the crown stays acting on the crown plate. (3) The lower part of the fireboxes must be carefully designed to prevent the early substantial destruction.

View full abstract