Search

The paper reports extensive connective heat transfer data for turbulent flow of air around a U-bend with a ratio of bend radius:pipe diameter of 3.375:1. Experiments cover Reynolds numbers from 2 × 104 to 1.1 × 105. Measurements of local heat transfer coefficient are made at six stations and at five circumferential positions at each station. At Re = 6 × 104 a detailed mapping of the temperature field within the air is made at the same stations. The experiment duplicates the flow configuration for which Azzola and Humphrey [3] have recently reported laser-Doppler measurements of the mean and turbulent velocity field. The measurements show a strong augmentation of heat transfer coefficient on the outside of the bend and relatively low levels on the inside associated with the combined effects of secondary flow and the amplification/suppression of turbulent mixing by streamline curvature. The peak level of Nu occurs halfway around the bend at which position the heat transfer coefficient on the outside is about three times that on the inside. Another feature of interest is that a strongly nonuniform Nu persists six diameters downstream of the bend even though secondary flow and streamline curvature are negligible there. At the entry to the bend there are signs of partial laminarization on the inside of the bend, an effect that is more pronounced at lower Reynolds numbers.

The heat transfer downstream of an axisymmetric abrupt expansion in a pipe in the transition Reynolds number regime has been investigated experimentally. In these experiments the wall of the downstream pipe was heated to give a constant heat flux into the air flow. The ratio of the upstream to downstream pipe diameters was 0.8 and the downstream Reynolds number ranged from 1420 to 8060. Within a narrow range of Reynolds numbers, around 5000, the position of the maximum Nusselt number was found to shift considerably. This interesting behavior may be associated with the flow instabilities in sudden expansions which have been observed by others.

The thermal behavior of a horizontal storage tank has been investigated numerically and experimentally. The purpose of the present work is to compare a simple, one-dimensional model to the measured temperature stratification in this horizontal tank. For the test conditions used in these experiments, axial temperature gradients were found to be negligible compared to vertical temperature gradients. It was observed that significant mixing of the tank occurred when water was removed unless a diffuser manifold was placed on the makeup water inlet. The numerical model used was one-dimensional (vertical) with some degree of mixing at the inlet and outlet boundaries. The temperatures at the top of the tank were predicted fairly well, but the temperatures at the bottom deviated somewhat from the predictions. It appears that some additional interior mixing, not included in the present model, occurs.

A computer program has been developed to analyze the thermodynamic performance of a cogeneration plant. Results of parametric studies using this program include partial load operation, inlet air cooling, and supercharging. Fuel chargeable to power, a previously used figure-of-merit, is used to evaluate the performance for partial-load operation. Fuel chargeable to incremental power and differential steam efficiency are two new figures-of-merit for evaluating other design alternatives: the former for incremental increases in electric power output, the latter for incremental increases in steam output at the expense of electrical power output.

Measurements of the local heat transfer coefficients on a pin fin (i.e., a short cylinder in crossflow) in a duct have been made using two methods, both of which employ liquid crystals to map an isotherm on the surface. The transient method uses the liquid crystal to determine the transient response of the surface temperature to a change in the fluid temperature. The local heat transfer coefficient is determined from the surface response time and the thermal properties of the substrate. The heated-coating method uses an electrically heated coating (vacuum-deposited gold in this case) to provide a uniform heat flux, while the liquid crystal is used to locate an isotherm on the surface. The two methods compare well, especially the value obtained near the center stagnation point of the pin fin where the difference in the thermal boundary condition of the two methods has little effect. They are close but differ somewhat in other regions.

Detailed heat transfer, temperature, and velocity data are reported for the turbulent flow downstream of an abrupt increase in tube diameter (2.5:1) in which the downstream tube is maintained at a uniform elevated temperature. The heat transfer experiments cover downstream Reynolds numbers ranging from 4300 to 44,500, the flow being fully developed at the exit of the small tube (i.e., the abrupt expansion step). Maximum local heat transfer coefficients are proportional to the upstream Reynolds number to the power 2/3 with the location of the maximum moving upstream slightly as the Reynolds number is increased. Heat transfer data at a Reynolds number of 17,300 are supplemented by velocity and temperature profiles, which are especially informative. They bring out clearly that the viscous sublayer, despite the thinning that is believed to occur in the vicinity of the reattachment point, still provides the major resistance to heat transfer. The correct prediction of the Nusselt–Reynolds number relation in reattaching flows is thus crucially dependent on the variation of turbulent diffusivity in the “buffer” region of the flow.

Experiments and computations have been performed for turbulent air flow in a circular duct in which uniform heating is applied over a 60 deg arc of the circumference. The results are compared to those for both a uniform heat flux around the full circumference of the duct and for a uniform heat flux around half the circumference. The flow was hydrodynamically fully developed at the start of heating and there were negligible buoyancy effects. Both the experimental results and the computations show that the local heat transfer coefficients for the narrow strip (60 deg wide) are less than the values for uniform heating around the full circumference but greater than the values for uniform heating around half the circumference (180 deg). This rather surprising result is a consequence of the temperature distribution in the fluid and the definition of the local heat transfer coefficient, which is based on the bulk temperature.

An experimental technique for obtaining a uniform wall heat flux has been used to determine the local heat transfer coefficients around a cylinder. Data are presented for a single cylinder, for cylinders in tandem, and cylinders located in the entrance of a tube bank. Results are compared to those of other studies with uniform wall heat flux. For the single cylinder, these are found to depend upon blockage, aspect ratio, and free-stream turbulence. For both inline and staggered tube arrangements, the heat transfer coefficient distribution depends on row location but appears to be nearly established by the third row.

This paper describes an experimental technique for measuring the local heat transfer coefficient in complex flows. The technique employs a gold-coated plastic sheet to generate a nearly uniform wall heat flux. The sheet is mounted on an insulated substrate of the desired shape and heating is achieved by passing an electrical current through the gold coating. Three applications of this technique are described: heat transfer downstream of an abrupt expansion, heat transfer from a cylinder in crossflow, and nonuniform circumferential heating in a pipe. For each of these applications the effect of the wall conduction on the surface heat flux is evaluated and found to be small. The effect of nonuniformities in the gold coating, and of the resistance temperature coefficient, are also evaluated and likewise are found to be quite small. Methods of correcting for each of these small effects are presented for the above applications.

Measurements have been made of the local heat transfer coefficients to an air flow downstream of an axisymmetric abrupt expansion in a circular pipe with a constant wall heat flux. The experimental technique uses an electrically heated thin gold film on a plastic substrate. The flow upstream of the expansion was unheated and fully developed. Runs were made with small diameter to large diameter ratios of 0.267 to 0.800 and over the Reynolds numbers range of 5,300 to 87,000 (based on downstream diameter). The results include measurements near the expansion corner region where no previous measurements have been reported. These provide clear evidence of a secondary recirculation. Comparisons are also made with previous experimental results in the region of reattachment.

Experiments have been performed with turbulent air flow in a tube with uniform and with top-half heating; the flow was hydrodynamically fully developed at the start of the heated section, and there were negligible buoyancy effects. Numerical solutions of the mean momentum and energy equations which use the suggestion [1] that the ratio of radial-to-circumferential turbulent diffusivities is equal to the corresponding ratio of mean square velocity fluctuations in these directions agree well with the experimental results.

The thermodynamic performance of a gas turbine cogeneration system is predicted using a computer model. The predicted performance is compared to the actual performance, determined by measurements, in terms of various thermodynamic performance parameters which are defined and discussed in this paper. These parameters include the electric power output, fuel flow rate, steam production, electrical efficiency, steam efficiency, and total plant efficiency. Other derived parameters are the net heat rate, the power-to-heat ratio, and the fuel savings rate. This paper describes the cogeneration plant, the computer model, and the measurement techniques used to determine each of the necessary measurands. The predicted and the measured electric power compare well. The predicted fuel flow and steam production are less than measured. The results demonstrate that this type of comparison is needed if computer models are to be used successfully in the design and selection of cogeneration systems.

Computer models have been used to analyze the thermodynamic performance of a gas turbine (GT) cogeneration system and an internal combustion engine (IC) cogeneration system. The purpose of this study was to determine the effect of thermal matching of the load (i.e., required thermal energy) and the output steam fraction (fraction of the thermal output, steam and hot water, which is steam) on the thermodynamic performance of typical cogeneration systems at both full and partial output. The thermodynamic parameters considered were: the net heat rate (NHR), the power-to-heat ratio (PHR), and the fuel savings rate (FSR). With direct use (the steam fractions being different), the NHR of these two systems is similar at full output, the NHR of the IC systems is lower at partial output, and the PHR and the FSR of the GT systems are lower than those of the IC systems over the full range of operating conditions. With thermal matching (to produce a given steam fraction) the most favorable NHR, PHR, and FSR depend on the method of matching the load to the thermal output.

Leaf water potential was measured on the same leaf with an in situdewpoint hygrometer of the Neumann‐Thurtell design and with a pressure chamber, for potato (Solarium tuberosum‘Russet Burbank’), sunflower (Helianthus annuus‘Russian Mammoth’), pepper (Capsicum annuum‘California Wonder’), soybean (Glycine max‘Wells’), and oat (Avena sativa‘Lodi’). On the average the pressure chamber measurement was within 1.5 bars of the in situpotential, although the scatter in the data was large. The comparisons showed a trend in which the pressure chamber gave lower (drier) leaf water potentials than the hygrometer in the high (wet) potential range, and higher potentials than the hygrometer in the dry potential range. This trend was most distinct in sunflowers, soybeans, and oats. Serious error in the pressure chamber measurement can result if evaporative loss of leaf water is not prevented before and during pressurization or if leaves are stored for an extended period between excision and pressurization.

Dewpoint hygrometers were used to compare the in situwater potential of a leaf with the potential of a tissue piece excised from the same leaf and measured either by a hygrometer or psychrometer. Excised tissue potential was found to average 1.0 to 3.0 bars lower (drier), depending on species and water potential range, than the in situpotential for potato (Solanum tuberosumL.), sunflower (Helianthus annuusL.), pepper (Capsicum annuumL.), and soybean (Glycine maxL.). With oat (Avena sativaL.) the potential increased after the tissue was excised. Internal tissue changes following excision contributed significantly to the potential changes in oat and soybean. Internal changes could not be separated from tissue drying for the other three species. Loss of tissue water during sample preparation and confinement in sample holder was superimposed on the internal shifts; drying losses appeared greatest for the psychrometer, on the average. All our data were obtained on glasshouse‐grown plants, so there is no immediate extrapolation of our data to field‐grown plants, nor to other species. It is clear, however, that water potential methods requiring excised tissue are liable to incur significant errors from the excision, especially in the wet range where such errors can be relatively large.