1、I NASA N w- 0 Y n z c Q v9 Q z TECHNICAL NOTE SUBCOOLED- AND NET-BOILING HEAT TRANSFER TO LOW-PRESSURE WATER IN ELECTRICALLY HEATED TUBES by James R. Stone Lewis Research Center Cleveland, Ohio 44135 NATIONAL AERONAUTICS AND SPACE ADMINISTRATION WASHINGTON, D. c. JULY 19ii Provided by IHSNot for Res
2、aleNo reproduction or networking permitted without license from IHS-,-,-TECH LIBRARY KAFB, NM - 1. Report No. NASA TN D-6402 4. Title and Subtitle I 2. Government Accession No. I 3. Recipients Catalog No. I 5. Report Date SUBCOOLED- AND NET-BOILING HEAT TRANSFER TO LOW- j July 1971 PRESSURE WATER .I
3、N ELECTRICALLY HEATED TUBES I 6. Performing Organization Code 7. Author(s) James R. Stone 8. Performing Organization Report No. I E-6207 - 10. Work Unit No. 9. Performing Organization Name and Address 120-27 Lewis Research Center National Aeronautics and Space Administration Cleveland, Ohio 44135 13
4、. Type of Report and Period Covered National Aeronautics and Space Administration Washington, D. C. 20546 11. Contract or Grant No. 2. Sponsoring Agency Name and Address Technical Note 14. Sponsoring Agency Code 5. Supplementary Notes “ 6. Abstract - Experimental data are presented on subcooled and
5、net-quality boiling heat transfer to water flowing vertically upward in tubes with uniform heat flux. Axial inner-wall-temperature dis- tributions are tabulated for mass velocities from 0.67 to 141 kg/(sec)(m 2 ), heat fluxes from 43.8 to 11 400 kW/m2, exit pressures from 24 to 690 kN/m 2 abs, exit
6、qualities up to 0.65, and liquid subcoolings as high as 151 K. Since no satisfactory correlations are available for the full range of test conditions, these experimental data over a wide range of test conditions should be useful to the designer. It appears that, for low-quality and subcooled boiling
7、, non- equilibrium effects must be taken into account, and no presently available model appears to be valid over the full range of subcooling. 7. Key Words (Suggested by Authorls) I - Boiler Net-quality boiling Heat transfer Experimental Nonequilibrium Correlations Subcooled boiling 18. Distribution
8、 Statement Unclassified - unlimited 19. Security Classif. (of this report) 22. PriceX 20. Security Classif. (of this page) 21. No. of Pages Unclassified $3.00 Unclassified For sale by the National Technical Information Service, Springfield, Virginia 22151 II I Provided by IHSNot for ResaleNo reprodu
9、ction or networking permitted without license from IHS-,-,-I SUBCOOLED- AND NET-BOILING HEAT TRANSFER TO LOW-PRESSURE WATER IN ELECTRICALLY HEATED TUBES by James R. Stone Lewis Research Center SUMMARY Experimental data are presented on subcooled and net-quality boiling heat transfer to water flowing
10、 vertically upward in 0. 584- and 1.219-centimeter inside diameter tubes with uniform heat flux. Axial inner-wall-temperature distributions are tabulated for mass velocities from 0.67 to 141 kg/(sec)(m ), heat fluxes from 43.8 to 11 400 kW/m , 2 2 exit pressures from 24 to 690 kN/m abs, exit qualiti
11、es up to 0.65, and liquid subcool- ings as high as 151 K. Since no satisfactory correlations are available for the full ranye of test conditions, these experimental data over a wide range of test conditions should be useful to the designer. 2 The subcooled-boiling data are compared with some existin
12、g correlations. These correlations give the heat flux as a function of wall temperature minus saturation tem- perature. The data show only approximate agreement with the correlations, due prob- ably to the effects of mass velocity, local subcooling, and distance from the inception of boiling. But th
13、e correlations are in the range of the data and may be useful for some ap- plications. Plots of net-boiling heat-transfer coefficients against quality do not agree with any existing correlation; the data show that, for constant heat flux, mass velocity, and qual- ity, the local heat-transfer coeffic
14、ient increases as pressure increases, whereas the correlations predict the opposite. Heat-transfer coefficients, based on a wall-to-liquid temperature difference correc- ted for nonequilibrium, vary less with quality and show more consistent trends than do heat-transfer coefficients based on either
15、wall-to-bulk or wall-to-saturation temperature differences. But no presently available nonequilibrium model appears to be valid over the entire range of liquid subcoolings of this study. Thus, a general heat-transfer cor- relation must be preceded by a nonequilibrium model valid for all subcoolings
16、encoun- tered. “ Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-INTRODUCTION An understanding of forced-flow boiling phenomena is necessary for the rational de- sign of Rankine-cycle power systems, especially those for use in space, where compact- n
17、ess is important. Boiling, with its high heat-transfer coefficients, is also applicable to cooling problems where high heat fluxes are involved, such as in rocket-nozzle cool- ing. To obtain efficient, compact space power systems, high fluid temperatures are re- quired. These high temperatures can b
18、e obtained at relatively low pressures by using alkali metals as working fluids. With the exception of liquid thermal conductivity, water has physical properties similar to the alkali metals. Since experiments on the boiling of alkali metals are difficult and expensive to perform, and since water it
19、self may be of in- terest for cooling applications, a series of experiments on heat transfer and pressure drop for water boiling in tubes has been done at the NASA Lewis Research Center (refs. 1 to 5). Although there have been numerous studies of boiling heat transfer, there is still no generally ap
20、plicable means of prediction available for high-density-ratio fluids such as alkali metals and low-pressure water. This is especially true of the subcooled-boiling regime. Experimental results for high-heat-flux subcooled boiling of low-pressure water in 0.584-centimeter inside diameter tubes were p
21、resented in reference 1. In fully developed subcooled boiling, it was found that the wall temperature was nearly indepen- dent of fluid bulk temperature and mass velocity for a given pressure level and heat flux (the heat flux being in the range of about 1600 to 9100 kW/m ). Existing semiempirical c
22、orrelations (refs. 6 to 8) did not satisfactorily predict the heat transfer at low pres- sures. 2 It is the objective of this study to obtain subcooled- and net-boiling heat-transfer data over a wide range of test variables, including the range covered in the heat- exchanger boiling studies (ref. 2)
23、. The range of interest is from the inception of boiling up to, but not including, the heat-transfer transition often called “burnout. This re- gime or series of regimes is often termed “nucleate boiling“; however, no such termi- nology is used herein since this range may include other regimes, such
24、 as evaporation from the interface with or without nucleation. The correlations of references 6 to 9 for subcooled-boiling heat transfer and references 10 to 14 for net-boiling heat transfer, based primarily on annular-flow models, are compared with the experimental data to de- termine whether or no
25、t any of them are applicable over the range of the experiments. In order to obtain heat-transfer data under conditions comparable to those of the heat-exchanger boiling studies (ref. 2), the present heat-transfer studies employ 1.219- as well as 0.584-centimeter inside diameter tubes. The experiment
26、al test sections of this study are made of Inconel X tubing having a 0.025-centimeter-thick wall. The water flows vertically upward through the straight, circular tube. No inserts are used. Exit qualities as high as 0.65 are obtained. 2 Provided by IHSNot for ResaleNo reproduction or networking perm
27、itted without license from IHS-,-,-APPARATUS Figure 1 shows a schematic diagram of the test apparatus. This is essentially the same test apparatus as that used in reference 1, wherein it is described in more detail. The system is designed to operate at a maximum pressure of 1700 kN/m abs and a maxim
28、um fluid temperature of 450 K. 2 The water is circulated by either of two gear pumps connected in parallel. The flow passes through the coiled stainless-steel electrical preheater to the test-section in- let plenum. From the exit plenum the flow passes through a 5. l-centimeter-diameter pipe to a sp
29、ray condenser. The coolant is supplied to the condenser by a centrifugal pump having a nominal capacity of -6.310- cubic meter per second. From the con- denser the flow passes into a multiple-tube heat exchanger cooled by cooling-tower water. In most cases the condenser is shut off, and the condensi
30、ng is done in the heat exchanger. The power to the test section and preheater is supplied by 70- and 250-kilowatt ac transformers, each regulated by a saturable core reactor. The two power supplies are interchangeable; either one can be connected to the test section or preheater, depending on the he
31、at loads for a particular test. Test Sections The test sections used in this investigation are constructed of 0.584- and 1.219- centimeter inside diameter by 0.25-millimeter wall Inconel X tubing and are from 14.6 to 121.9 centimeters long. This tubing is specially rolled to limit the wall-thickness
32、 de- viation to less than 3 percent of the nominal wall thickness. The test-section inside- surface roughness is measured by a surface analyzer before and after use for a number of the test sections. The maximum surface roughness (peak to valley) is “5 nanometers. Several samples of tubing are used
33、to measure the electrical resistivity of the Inconel X as a function of temperature. The results agree well with reference 15; the electrical resistivity of Inconel X may be considered independent of temperature for the range of this investigation. Figure 2 shows a schematic diagram of a typical tes
34、t section. The ratio of unheated length to inside diameter is 0.8 for test section 480-100M (table I) and 15 for all other test sections. (The test-section numbers give the inside diameter in thousandths of an inch followed by the ratio of heated length to inside diameter; the letter indicates the o
35、rder of testing.) Power leads are connected to copper flanges that are silver-soldered to the tube. High temperature solder is used, and care is taken to ensure good contact and to avoid fillets in the heated portion of the test section. The end flanges are bolted 1 3 I Provided by IHSNot for Resale
36、No reproduction or networking permitted without license from IHS-,-,-to the inlet and exit plenums and electrically insulated by Teflon and asbestos gaskets and seals. The exit plenum is fixed to the structural frame, while the inlet plenum is allowed to slide longitudinally to provide for thermal e
37、xpansion. The test-section as- sembly is enclosed in a transparent shield to minimize external convective currents and to provide protection in case of failure. Instrumentation Fluid temperatures. - - Plenum chambers (fig. 3) are provided at the test-section in- let and exit in order to obtain (at l
38、east in the all-liquid case) true bulk-temperature read- ings. The inlet- and exit-plenum temperatures, as well as other fluid temperatures around the loop are measured by copper-constantan thermocouples and recorded by a multipoint self -balancing potentiometer. This instrument is periodically cali
39、brated with a standard potentiometer. The error in the bulk temperatures is estimated to be *O. 3 K. Pressures. - For most of the data reported herein, the fluid pressures at the inlet and exit plenums are measured with strain-gage-type transducers having a range of 0 to 690 kN/m and a stated accura
40、cy of 1/4 percent of full scale. The signal is recorded on a multichannel oscillograph. A 30-centimeter, 0- to 690-kN/m Bourdon tube gage is connected to the exit plenum. In some cases, the test-section pressure drop is also measured with a 0- to 207-kN/m differential gage connected across the press
41、ure taps drilled through the copper flanges of the test section. Valves are provided to damp large amplitude gage pressure fluctuations or to shut off the gage completely. Under steady conditions, the correspondence between the transducers and gage readings is considered good. 2 2 2 Flow rate. - The
42、 flow rate is measured by one of three turbine-type flowmeters with overlapping ranges, All flowmeters are calibrated prior to their installation. The flow- meter signal is read from a frequency-converter indicator and on a digital frequency counter, both of which are calibrated by a signal generato
43、r. Power. - Because a distorted ac output is obtained at low heating power levels (see ref. 1) , a dynamometer-type wattmeter is used to measure the power input to the test section. The accuracy of this true rms instrument is 0.1 percent of full scale. A true rms, electronic-tube voltmeter is used t
44、o measure the test-section voltage drop and thus provide a check on the wattmeter. Test-section wall temperatures. - The outside wall temperatures of the heated tube are measured by a number of 36-gage Chromel-Alumel thermocouples spot-welded in one longitudinal plane to the outside surface of the t
45、ube at various axial locations. The milli- volt signal is recorded on a self-balancing potentiometer with a special filtering and iso- lation circuit to eliminate ac voltage pickup. Two 36-gage Chromel-Alumel thermo- 4 Provided by IHSNot for ResaleNo reproduction or networking permitted without lice
46、nse from IHS-,-,-couples are spot-welded on the outside surface of the tube, both 0.635 centimeter from the exit end of the heated length, but circumferentially spaced 180 apart. These ther- mocouples are connected in parallel to yield a single average surface-temperature read- ing that is continuou
47、sly recorded on a separate self-balancing potentiometer and serves as an overtemperature control. PROCEDURE Prior to filling, the system is flushed with deionized water, purged with nitrogen, and then evacuated. It is then filled with deionized, deaerated water. The water is cir- culated through the
48、 system and further degassed by boiling in the test section and venting to the atmosphere from the top of the condenser. Generally, water samples are taken and analyzed daily on a gas analyzer for gas content. The dissolved gas content is main- tained at or below 3 parts per million (by weight). The
49、 desired conditions for each run are established by adjusting the power to the preheater and test section and setting the pump speed and system pressure at selected values. When the mean inlet and exit bulk temperatures become constant with time, the data for that run are taken. Temperatures are automatically recorded on strip charts. Flow rate, power, and pressures are read from gages. In order to check the wall the
