This paper presents theoretical and experimental analyses of the cold heat storage devices. The device includes the following components: compressor, condenser, expansion valve, ice making pipes, and cold heat storage tank. Such devices are employed in storing cold heat by making ice at night and melting the ice in the daytime when air conditioning load is large. The results are: 1) The equations have been established to predict the ice volume and they are confirmed to coincide with the experimental results. These equations contain such parameters as compressor capacity, ice making pipe length and diameter, as well as storage tank volume. 2) The analyses indicate that the length of the ice making pipe has a greater influence upon the volume of ice than the diameter of the pipe does. 3) It is very effective to reduce the superheat zone and to use all pipes as ice making zone. This is achieved by proper selection of the expansion valve and its opening. 4) The rate of ice production has been analysed quantitatively. The analysis takes into account both the growing ice on the pipes and the ice removed from the pipes by a novel means.

View full abstract

An analytical solution is developed for heat and mass transfer in a evaporative cooler, in reference to the theoretical model proposed by Parker and Treybal. The analysis gives two transfer coefficients. The analysis also predicts the variation of fluid temperatures and enthalpy, provided the transfer coefficients are given as functions of mass velocities and physical properties of the fluids. In order to investigate the characteristics of the transfer coefficients, experiments are carried out for various sets of mass velocities with several multi-tubular evaporative coolers. In some experiments, fluid temperatures are measured at several locations along the tubes, being compared with predicted ones.

View full abstract

In the local exhaust systems, only suction flows are widely utilized to control contaminations, and the hoods used such systems are called "Pull Hoods". It is well known that the air velocity in the region in front of the opening under suction decreases with increasing distances from the hood. Accordingly, with increasing distance from the source of contamination to the suction opening, it is necessary to increase the exhaust air volume for the pull hood and to elongate the sizes of hood in order to get satisfactory results with the pull hood. On the other hand, a push flow can be used as the means of separation in the space such as air shutters or conveyors of the impurities, because of the fact that the central velocity of the flow is comparatively kept up to a step forward the push opening. Then, by the combination flow which push and pull flow cooperate each other, it is considered to make efficient use of the both flow characteristics, and the hoods used such objects are called "Push Pull Hoods". It is effective to apply push pull hood to rational equipment such as air shutter, local exhaust and ventilation systems. In practice, there is such as a way of thinking a long time ago, for example, air curtains are put to practical application. Nevertheless, characteristics and economical design method on the push pull hood have not yet been established enough. Therefore, first of all authors, who have noticed the above point, investigate previous reports on the push pull hood, and check up carefully conventional design method and unsolved problem point on the hood. Next, under new method that the flow produced within the hood is composed of one body allied with push flow and pull one, when both flow are handled two dimensionals and assumed to be the potential flow, the relations between the flows and hood flanges have been studied by using combination method of stream lines. Then, from the theoretical and experimental approaches performed during this investigation, the following conclusions can be formed. Generally, it is found that the shapes of the flow within the push pull hoods are considerably affected by the flanges of the hood. The area of push flow being shaped between push opening and suction opening becomes the smallest when suction side has flanges and push side has no flange. This means that this type is best and should adapt one always as far as circumstances permit.

View full abstract

In the first paper, the results of investigation of previous papers and the relationship between the push-pull flow and the hood flanges were reported. In this paper, in order to make clear the rational design method on the two dimensional push-pull hood, "Flow Ratio Method" proposed for the canopy hoods and the lateral hoods is also applied to the push-pull hoods. The interrelation of flow volume on the hood is as follows: Q_3=Q_1+Q_2=Q_1(1+Q_2/Q_1) where Q_1: push flow volume [m^3/min] Q_2: flow volume from surroundings [m^3/min] Q_3: pull flow volume [m^3/min] The aspects of push-pull flow vary according to the values of Q_2/Q_1, and the value of (Q_2/Q_1)_<limit> which shows the limited state in case Q_1 does not leak out around the hood exists always. It can be considered that (Q_2/Q_1)_<limit> is determined by the following relations. K_S=(Q_2/Q_1)_<limit,v_0=0>=f_S(D/E, V/E, W/E, H/E) K_B=(Q_2/Q_1)_<limit,v_0>0>=f_B(v_0/v_1, D/E, V/E, W/E, H/E) where K_S: shutting limit flow ratio [non-dimension] K_B: breaking limit flow ratio [non-dimension] E: width of push opening [m] D: width of pull opening [m] V: push flange length [m] W: pull flange length [m] H: arrival distance between push opening and pull one [m] v_1: push flow velocity [m/s] v_0: side flow velocity [m/s] Then, from the abovementioned point of view, the influences of hood shapes and side flow on the push-pull flow are studied experimentally. The results of experiments are summarized as follows: (1) The pull flange length ratio, W/E, and the push-pull distance ratio, H/E, influence remarkably on the values of K_S. (2) The push flange length ratio, V/E, influences very small on the values of K_S, and it can be considered that there is no effect of the pull opening ratio, D/E, to K_S. (3) If the safety factor for shutting, n, is represented by the ratio of K_B to K_S, the values of n increase with the increase of the side flow velocity ratio, v_0/v_1, and H/E, and are not related to D/E, V/E and W/E. (4) The safety factor for breaking, m, should be selected the values within 1.0≦m≦1.5 for the practical design. Next, the equations for designing two dimensional push-pull hoods are derived from the experimental results, and the calculation charts for the practical design are made.

View full abstract

To control particulate air pollutant emissions, there have been many attempts to include the effect of electrostatic force in conventional scrubbers. They can be classified into the following three types. 1) Dust particles are charged electrically to improve collection efficiencies of conventional scrubbers. 2) Charged water droplets are sprayed electrohydrodynamically and dust particles are collected by impact scrubbing. 3) Dust particles and water droplets are charged electrically and dust particles are collected by electrostatic force. Type 3) can be also classified as follows; 3-1) Dust particles and water droplets are oppositely charged. The dust particles are deposited to the water droplet surface by Coulomb force, and the droplets are collected to the inside wall of the precipitator by electrostatic diffusion. 3-2) Dust particles and water droplets are charged to the same polarity. The dust particles and water droplets are collected to the inside wall of the precipitator by electrostatic diffusion. The collection efficiencies of the types 3-1) and 3-2) are similar, so, if particles and droplets can be charged at a time by a unit charge device, the type 3-2) will be simpler in the structure. In the type of 3) water droplets must be small enough to keep long life in the precipitator. In this paper, we considered the Vapor Inject & Charge Method belongs to 3-2), in which saturated vapor is injected into the dust laden gas, condensed water droplets are charged by corona discharge and precipitated to the inside wall of the precipitator duct by the effect of electrostatic diffusion. In comparison with the conventional electrostatic precipitators, which utilize electric field made by the space charge of continuously supplied corona, our method utilizes the electric field made by the space charge of water droplets which must be gradually collected. Therefore, it cannot outperform electrostatic precipitators in collection efficiency, but the structure of dust precipitating zone is much simpler. At first, we deduced, for monodisperse dust particles, theoretical equations to predict collection efficiencies based on a few assumptions. The equations showed that the factors which have direct effects upon collection efficiencies η are: inlet air velocity, V_a, temperature and humidity of inlet air, conditions of injected vapor, mass rate of injected vapor G_v, dimensions of the corona charger, current in the corona charger i_c and length of the precipitator duct L. Another factors which will have indirect effects upon collection efficiencies are: diameter of dust particles δ_p, qualities of dust particles, number concentration of dust particles n_p, diameter of the precipitator duct and the shape of vapor injection nozzle. Next, we made experiments on the effects of six factors of V_a, G_v, i_c, L, δ_p, n_p on collection efficiencies. Inlet air conditions were 20℃ dry bulb temperature and 80% relative humidity. The corona charger was made of co-cylindrical copper electrodes, and the precipitator duct was made of grounded aluminium pipe with diameter 32mm and thermally insulated from room air temperature. The uniform diameter particles of polystylene latex were fed to the inlet air as test aerosol, and the particle number cocentration in outlet air was counted by B & L Dust Counter 40-1 after diluting and drying outlet air with large amount of dry clean air. Experimental results were in approximate agreement with theoretical calculations except on the case of lower G_v. An example of experimental results was η=97.5% at V_a=1m/s,G_v=1/10, i_c=8μA/cm, L=10m, δ_p=0.312μm, n_p=6×10^1_0 particles/m^3. So we concluded that the Vapor Inject & Charge Method has enough collection efficiencies to control submicron particulate air pollutant emission.

View full abstract

The present paper describes comparison of data measured for an actual solar house with computer simulation results. The measured data is based on data obtained for Yazaki Experimental Solar House No.1 (See Fig.1) for cooling period of Aug. 10 through 17, 1975 and for heating period of Jan. 16 through 22, 1976. The simulation was conducted not by using a program developed for the present comparative studies, but by utilizing the program already used for other case studies. The program comprises simpler mathematical models giving attention to shortening of computing time so that yearly simulation may be effected. In order to conduct the simulation on the same conditions as those of the actual system, hourly measured data was used as input data of simulations, such as solar radiation, outdoor temperature, cooling and heating load, hot water consumption and condensing water temperature. Results of simulations, such as solar heat collected, auxiliary heat, heat consumption by chiller and water temperatures in thermal storage, were compared with the measured data. Parameters which had influence on performance of components of the system were used mainly according to the design specification for the system. For some of the parameters, however, simulations were repeated and then parameters most comparable with actual measurement were adopted. Such repetitions were necessitated by actual conditions of the system rather than by problems existing in the simulation model. Examples of the said conditions are such as damping of thermal insulation in the thermal storage, smearing of glass surface of the collector, and lowering of efficiency due to repetitive on-off operation of the absorption chiller; all these cannot be expected from the specification or, even if so, it is difficult to anticipate significant values. Thus, provided that appropriate parameters were established, coincidence was found between the simulation results and the measured data, as shown in Case (3) of Table 2 and Figs. 5, 6, 7 and 8. Model for the collectors was represented by simple mathematical models which consisted of equivalent outdoor temperature and equivalent heat transmission coefficient, keeping satisfactory accuracy of computation. Although influence of outdoor wind velocity and convection and emittance in the collector were fixed by parameters, it caused no lowering of computation accuracy. The stratified type thermal storage was modelled by a full-mixing type thermal storage composed of three continuous portions. In spite of such a simple model, simulation results obtained revealed values similar to the data measured for momently variance of water temperature in the storage and for temperature difference in high temperature, medium temperature and low temperature protions (See Figs. 7 and 8). In model for the absorption chiller, property diagrams made by the manufacturer were used correcting the lowering of efficiency due to on-off operation. Simulation results obtained by the model showed substantial similarity to the measurement. Simulation of auxiliary heat, whose discrepancy of 15 to 20% was the most inaccurate of all the comparison items, was thoroughly influenced by accumulated discrepancies of other mathematical models and inaccuracy of the measured auxiliary heat which was in fact obtained from calculation of the measured consumption of kerosene for auxiliary boiler and the assumed boiler efficiency factor of 80%. Although there were found defects that the measuring period was short and that the simulation models did not reflect all operation conditions, this simulation program was proved to have satisfactory accuracy in comparison with the measurement and to withstand practical use.

View full abstract

The T-shaped branch pipe is one of the most basic elements in pipe line systems. Because the evaluations on the energy losses in this pipe are very important for practical use, many investigations have been made. In this branch pipe, when the pressure becomes lower and/or the velocity in main pipe increases, the cavitation is induced in the neighborhood of the junction of main and lateral pipes. In the past, as far as the authors know, these cavitation characteristics have not been published except in that of the T-shaped branch pipe with a single lateral pipe. In this paper, therefore, the cavitation characteristics of the T-shaped branch pipe consisting of main pipe and two symmetrical lateral pipes were experimentally investigated. First, for the T-shaped branch pipe with equal square cross sections of main and lateral pipes, called the fundamental type branch pipe, the aspects of cavitation occurrence were observed in detail and its critical condition was pursued experimentally. Then, for the T-shaped branch pipe with different cross sections of main and lateral pipes and the one with different shapes of branch part from the fundamental type, the experiments were made in the same manner as above described. The shapes of branch part used in this experiment were: one having the radius of curvature in branch corners, one having the extension which represents the continuation of main pipe, and one placing the dividing plate at the front wall of lateral pipe. Also, under the different conditions of flowing-in into the junction the cavitation characteristics of this branch pipe were investigated. It was recognized that the cavitation appeared in separated flow from branch corners and in a pair of vortices formed by secondary flow at branch part, the critical condition of cavitation occurrence depended on shape of the neighborhood of junction and ratio of flow rate, and the cavitation characteristics were best in branch pipe with rounded corner at the junction of main and lateral pipes.

View full abstract

A scale model test of a heat storage tank has been conducted to elucidate the similarity law and to obtain fundamental performance data. The model used in this experiment was a 1/10 scale tank of a typical one of the 18 serial tanks constructed at Toyama Medical and Pharmaceutical University in 1977. The size of the model was 470mm wide, 1,100mm long, 680mm deep with the capacity of 0.33m^3. The model tank was previously filled with warm water and made uniform in temperature, and then supplied stepwise with cold water. The variation of water temperature was measured at the inlet and the outlet, and the inner flow pattern was observed by visualizing cold water with methylene-blue. The cold water flowed in horizontally through the inlet opening (370mm width by 70mm height, located at the bottom of a wall) stagnated to a certain extent under the warm water layer because of density difference. The warm water was then pushed out consequently through the outlet over submerged weir of 370mm width by 70mm water depth located at the top of the opposite side wall. Based on a few assumptions, and using dimensional analysis of Navier-Stokes' equation of motion, it became clear that the flow characteristics are subject principally to Archimedes number Ar, and Reynolds number Re. Ar is a non-dimensional parameter defined by Ar=LgβΔT/U^2, and in this paper, U stands for mean velocity at inlet opening and L for the water depth. Therefore, the test was carried out under various Re for each Ar of 1.0, 1.5, 2.4, 4.0, 6.0 and 10.0. The measured data indicate that he temperature response is directly dependent on Ar and the effect of Re is negligible within the measured range. Thus, a model test on a heat storage tank can be scaled up directly, merely by adjusting Ar coincident with the expected prototype. Obtained results show also that the heat storage efficiency η decreases with decreasing Ar. This loss is mainly due to the overflow of cold water caused by predominant inertia in the early stage of operation. On the other hand, even though Ar is extremely increased, η appears to have an upper limit distinctly lower than 100%, and the effect of the depth of the outlet weir which is as large as 10% of total depth is considered to be the main reason of efficiency reduction. In the case of the prescribed prototype, evaluated Ar is 5, and then the unit tank efficiency was estimated at η=0.88 using the obtained data. This efficiency is considerably higher than the theoretical value of perfectly mixed tank (η=0.632), therefore it is quite clear that the adoption of a heat storage tank of this type is advantageous.

View full abstract

In this report we calculated annual cooling and heating coil loads, thermal efficiency, mixing loss and energy saving index of VAV systems and compared them with those of 4-pipe induction unit system in perimeter zones of model building. The hourly heating and cooling loads for 365 days of year were calculated with aids of HASP/ACLD 7101 and the standard meteorological data in Tokyo by SHASE. For the minimum flow rate of outlet air of VAV systems, two values, that is, 50% of the maximum flow rate and the greater one of 25% of the maximum and the minimum outside ventilation air (6m^3/m^2・h), were set and computed respectively. For all systems, the load was calculated in following three cases: employing neither rotary type air to air heat exchanger nor free cooling (Case 1); employing the exchanger with 70% efficiency, but not free cooling (Case 2); and employing both the exchanger with 70% efficiency and free cooling whenever possible. The cooling energy saving index is defined as follows: CESI=(q_C)/(Q_C), CESI'=(q'_C)/(Q_C), CESI"=(q"_C)/(Q_C) and the heating energy saving index-as follows: HESI=(q_H)/(Q_H), HESI'=(q'_H)/(Q_H), HESI"=(q"_H)/(Q_H) where q_C: annual cooling coil load of one zone q_H: annual heating coil load of one zone Q_C: basic annual cooling coil load Q_H: basic annual heating coil load (') and (") show the value of Case 2 and Case 3, respectively. As the basic coil loads, in this report, we adopted the mean cooling and heating coil loads of Case 1 of 4-pipe induction unit system in the perimeter zone. As a result, VAV systems of Case 3 obtained more energy saving than 4-pipe induction unit system, when the minimum flow rate of supply air was set at 25% of the maximum air and speed control was adopted at supply fan.

View full abstract

This paper describes the estimation technique and its application to determine thermal response factors from experimental room air temperature and surface heat flux data. This technique can be applied to multi-layer walls and neither homogeneous nor isotropic walls. Response factors are estimated by the estimation technique so that they may satisfy physical constraints. Least squares method is used in the technique. Physical constraints are the reciprocity of response factors and the relation between response factors and heat transmission coefficients. Several digital simulation studies have been carried out. Every value of estimated response factors agrees with theoretical values, when response factors are estimated from the data generated by numerical analysis. The applicability of the technique is assured to actual walls. In the application to experimental data, every term of estimated response factors, except for the initial term of the heat transmission response factors, agrees with theoretical value which is known about a test wall.

View full abstract