1、4 NASA TN 0-2595 GPO PRICE $ OTS PRICE(S) $ Am EXPERIMENTAL LOCAL HEAT-TRANSFER DATA FOR PRECOOLED HYDROGEN i AND HELIUM AT SURFACE I TEMPERATURES UP TO 5300 R by Maynard F. Tuylor Lewis Research Center Cleuehnd Ohio NATIONAL AERONAUTICS AND SPACE ADMINISTRATION 0 WASHINGTON, D. C. 0 JANUARY 1965 Pr
2、ovided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-EXPERIMENTAL LOCAL HEAT-TRANSFER DATA FOR PRECOOLED HYDROGEN AND HELIUM AT SURFACE TEMPERATURES UP TO 5300 R By Maynard F. Taylor Lewis Research Center Cleveland, Ohio NATIONAL AERONAUTICS AND SPACE ADMIN
3、ISTRATION For sole by the Office of Technical Services, Deportment of Commerce, Woshington, D.C. 20230 - Price $1.00 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-EXPERIMENTAL LOCAL HEAT-TRANSFER DATA FOR PRECOOLED HYDROGEN AM) HELIUM AT SURFACE TE
4、MPERATURES UP TO 5300 R by Maynard I?. Taylor Lewis Research Center SUMMARY Local values of heat-transfer coefficients and average friction coeffi- cients were measured experimentally for precooled hydrogen and helium gases flowing through an electrically heated tungsten tube with a length-diameter
5、ratio of 77 for the following range of conditions: local surface temperatures up to 5300 R, inlet gas temperatures from 252 to 325O R, inlet pressures from 37 to 93 pounds per square inch absolute, local bulk Reynolds numbers from 5700 to 48,400, local ratios of surface to bulk gas temperature up to
6、 8, and local heat fluxes up to 2,370,000 Btu per hour per square foot. A comparison of several methods of correlating local heat-transfer coef- ficients was made for several types of wall temperature distributions, and one method was found to work exceedingly well in correlating hydrogen and helium
7、 data with surface to bulk gas temperature ratios up to 8. Average friction coefficients for both helium and hydrogen with the Kq in. ab s - 2 50 !50 to 1000 500 to 1500 40 40 to 100 110 to 850 37 to 93 3F DATA Heat - transfer fluid Air Helium and hydro- gen Helium and hydro - gen Helium Helium Heli
8、um and hydro - gen Helium and hydro - gen Helium and hydro - gen Types of heat - xansfer coef - ficient oeasured iverage Local Local Local and aver- age Local and aver - age Local lverage Local %Unpublished data from Herbert J. Newman of Los Alamos Scientific Lab- oratory . Reference 1 presents cons
9、iderable data showing the effect of surface to fluid temperature ratio on the heat-transfer coefficient for air. Other in- vestigations using helium and hydrogen and extending the range of surface to fluid temperature ratio (refs. 2 and 3) or the range of wall temperature (ref. 4) or both (refs. 5 a
10、nd 6) have been presented. The conditions for which data were obtained in references 1 to 6 and in the present investigation are presented in table I. Reference 3 used an Inconel test section and lowered the inlet gas temperature with a liquid nitrogen ba.th. Inconel limited the wall to fluid bulk t
11、emperature ratio to 4.5 in reference 3, while the room temperature inlet gas and the melting point of the tungsten test section limited the wall to bulk temperature ratio to 5.6 in reference 6. In the present investigation, a tungsten test section was used to obtain high wall temperatures, while the
12、 inlet gas temperature was lowered with liquid nitrogen to obtain surface to bulk fluid temperature ratios as high as 8. mental heat-transfer data from the present investigation are presented along The melting point of The experi- 2 Provided by IHSNot for ResaleNo reproduction or networking permitte
13、d without license from IHS-,-,-with a recommended method for correlation. ExPERlMENTAL APPARATUS The test apparatus, test section, and instrumentation were the same as that described in reference 6 except that a liquid-nitrogen precooler was added to the inlet gas line as shown in figure 1. galion s
14、tainless-steel tanlr in which a nine-turn coil of copper tubing was im- mersed in liquid nitrogen. The liquid level was held constant with a float switch. The tank was insulated with plastic foam. The precooler consisted of a 30- The test section was fabricated and instrumented in the same manner as
15、 the The tungsten test section used in this experiment had one used in reference 6. Section A-A Liquid-nitrogen cooler CD-7889 Figure 1. - Schematic diagram of arrangement of test apparatus. an inside diameter of 0.115 inch, a heat-transfer length of 9 inches, and an entrance length of 14 diameters.
16、 copper- c ons t ant an thermocouples with a liquid- nitrogen cold junction. The inlet gas temperature was measured with METHOD OF CALCULATION The chemically frozen (chemical reaction term not included) transport and thermodynamic properties of hydrogen and helium used in the calculations of the hea
17、t-transfer and friction coefficients in this investigation were precisely the same as those used in reference 6, as were the physical properties of tung- sten and molybdenum. The average friction and local heat-transfer coefficients were calculated by the method used in reference 6. Local heat-trans
18、fer coefficients were ap- 3 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-. proximated by dividing the test-section length into 10 equal increments and evaluating average coefficients for those small increments. Coefficients for the first and last
19、increment were not used because of the large end losses. RESULTS AND DISCUSSION Axial Wall Temperature Distributions Four axial outside wall temperature distributions, two for uncooled inlet gas and two for precooled inlet gas, are shown in figure 2 as a function of distance from the test-section en
20、trance. Temperature measurements, including thermocouple and optical pyrometer readings for each run, are also shown. Ex- perimental data including local h, %, and Tw for runs 1 to 23 (uncooled runs) are listed in table I1 of reference 6, while Run Gas flow, Entrance Total heat ?.-I W, temperature,
21、transferred, lblhr T11 QIS, c OR BtuI(hr)(sq ft) Figure 2. -Comparison of wall temperature distributions for cooled and un- cooled inlet hydrogen based on flw rate and maximum wall temperature. runs 32 to 52 (precooled runs) are summarized in table I1 of this report. (All symbols are defined in the
22、appendix.) Fig- ure 2 contains a compari- son of run 17 with run 51 and run 18 with run 52. The runs compared have the same flow rate and maximum wall temperature. It can be seen from figure 2 that there is an increase in the surface temperature near the entrance of the tube for the runs with cooled
23、 inlet gas over that of the runs where the in- let gas is not cooled. The increase is a result of two factors. First, the ratio of surface to bulk fluid temperature is increased by lowering the fluid temperature. This is accompanied by a de- crease in the heat-transfer coefficient, which tends to in
24、crease the surface temperature further. Second, the effect of in- creasing the ratio of sur- face to bulk fluid tem- perature is magnified by the increased electrical resistivity of tungsten at higher temperatures. The large axial temperature 4 Provided by IHSNot for ResaleNo reproduction or network
25、ing permitted without license from IHS-,-,-gradients at the entrance and the exit of the test section are the result of conduction losses to the connecting flanges, the mixing tanks, and the elec- tric a1 connectors. The heat-transfer parameters for the four hydrogen runs in figure 2 will be shown a
26、nd discussed in the section Heat-Transfer Coefficients. Fri c t i on Coefficients As in reference 67 only average friction coefficients were measured. The friction coefficients for hydrogen and helium are shown in figure 3. The line representing the K neither agrees with the predicted line, however.
27、 The hydrogen friction data of this investigation fall below the hydrogen data of reference 6. The conclusion that must be drawn from figure 3 is that there is a need for further study of friction coefficients for conditions where the physical properties and density vary greatly in both the radial a
28、nd axial directions. Heat-Transfer Coefficients In the present investigation as in reference 6, only local heat-transfer coefficients were calculated. The results of reference 6 indicate that, while some local heat-transfer data can be correlated to within 210 percent by using the following equation
29、 The data with large axial gradients in heat flux and surface temperature near the test-section entrance introduced deviations of as much as 30 percent from the correlation line. axial gradients in heat flux and surface temperature nearer the test-section entrance deviate as much as 60 percent from
30、the correlation line (see fig. 4). Data of the present investigation that have greater Reference 7 investigates the various methods of correlating hydrogen heat- transfer data proposed in references 2, 8, 9, and 10. ing equations were proposed: The following correlat- where C2 is 0,048 for hydrogen
31、and 0.020 for helium along with a determination of reference temperature for evaluating the prop- erties used in heat-transfer equations. The reference temperature Tx is de- fined by Both the hydrogen and helium heat-transfer data of this investigation and 7 Provided by IHSNot for ResaleNo reproduct
32、ion or networking permitted without license from IHS-,-,-(b) Helium. Figure 5. - Correlation of local heat-transfer coefficients using equation (4). 8 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-(bl Helium. Figure 6. - Correlation of local heat-t
33、ransfer coefficients using equation (51. 9 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-100 80 60 40 a 101 I/L I I I Ill1 I I I IIIII Id 104 105 Bulk Reynolds number, Reb I GD/ub (bl Helium. Figure 7. - Correlation of local heat-transfer coefficie
34、nts using equation (81. 10 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-% 3 z Id 104 Modified film Reynolds number, Ref - (GD/Tflf) (a) Using equation (31. ld 101 1/1 I I IIlll I I I I I Ill1 Id 104 Id Bulk Reynolds number, R4 - GDlb (b) Using equ
35、ation (8). Figure 8. - Correlation of hydrogen runs 17, l8, 55 and 52 11 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-reference 6 correlated by equation (4) are shown in figure 5. be no improvement over the film correlation, merely a shift of the
36、greatest amount of scatter from low to high Reynolds number. There appears to I Equation (7) could not be used because the range of Reynolds numbers pre- sented in reference 10 was not low enough to include the uncorrelated data of either this investigation or reference 6. Reference 7 shows that thi
37、s method does not correlate high Reynolds number data as well as equation (3) does. Equation (5) attempts to correlate heat-transfer data by removing the properties from the conventional heat-transfer equation to simplify calcula- tions. The heat-transfer data for hydrogen and helium are shown in fi
38、gure 6. Both the hydrogen and helium data of reference 6 and this investigation fall as high as 40 percent above the correlation line. This cannot be corrected by in- creasing the constant in equation (5) because the data of reference 3 fall considerably below the correlation line. C2 I A very good
39、correlation can be obtained for the hydrogen and helium heat- transfer data of reference 6 and the present investigation with equation (6), as shown in figure 7. The exponent of the surface to fluid bulk temperature ratio was decreased from 0.29 + 0.0056 to 0.29 + 0.0019 L/D giving which is useful f
40、or test sections where the length-diameter ratio is as high as 250, such as in reference 3. Both exponents worked equally well for data pre- sented in this investigation and reference 6. The hydrogen data correlate better than the helium data with 90 percent of the hydrogen data falling within +lo p
41、ercent while 80 percent of the helium data correlate to within 210 percent of the correlating line. The physical properties and density are evaluated at the bulk temperature. In this investigation and in reference 6, the maximum bulk temperature is about 2800 R, which is less than the temperature at
42、 which dissociation occurs at the pressures involved. To show more clearly the trend of heat-transfer parameters when evalua.ted by means of equations (3) and (8), the parameters for the two noncooled and the two precooled wall temperature distributions of figure 2 are shown in figure 8. Figure 8(a)
43、 shows the parameter evaluated by using equation (3), and fig- ure 8(b) shows the parameter evaluated by using equation (8). The improved correlation obtained by using equation (8) is quite striking. SUMMARY OF F3SULTS The following results were obtained in an investigation of heat transfer to hydro
44、gen and helium at pressures of 37 to 93 pounds per square inch flowing through a tungsten tube at surface temperatures up to 5300 and ratios of sur- face to bulk fluid temperature up to 8: 1. Some local heat-tra.nsfer data agree to within +lo percent when cor- 12 Provided by IHSNot for ResaleNo repr
45、oduction or networking permitted without license from IHS-,-,-related by using the Dittus-Boelter equation and chemically frozen (chemical reaction term not included) viscosity, thermal conductivity, and specific heat. These physical properties and density were evaluated at either the film or the su
46、rface temperature. Data obtained with large axial gradients in heat flux and surface temperature and large ratios of wall to fluid bulk temperature near the test-section entrance introduce deviations as great as 60 percent from the cor- relation equation. 2. A much improved correlation can be achiev
47、ed for all the data by using - Lo. 29+0. Ool (L/D) 1 where my, is the bulk Nub = 0.021 Re;-Pr: (a/Tb) Nusselt number, R% is the bulk Reynolds number, Pq, number, Tw is the wall temperature, Tt, is the bulk temperature, L is the distance from the test section inlet, and D is the inside diameter of th
48、e test section. 90 percent of the hydrogen data correlate to within 210 percenf., vhile 89 per- cent of the helium data correlate to within +lo percent. The plrrysical prop- erties and the density were evaluated at the bulk temperature. is the bulk Prandtl The hydrogen data correlate better then the helium data doj 3. Friction coefficients without heat addition are in good agreement with the K Tx E x (Tw - Tt,) + Tt, absolute viscosity of gas, lb/(hr) (ft) 0 R Subs c rip t s : av average for complete test section I 14 Provided by IHSNot for ResaleNo reprod
