Reynolds Number Effects on Pressure Loss and … The aerodynamic characteristics of pressure loss and turbulence on four tube-bundle configurations representing heat-exchanger geometries

NASA Technical Memorandum 85807

tNASA-TM-85807 19840017577 } _6"_3[ _""t-_F_"_'_'_'_._...-._..__-.--.....- ,,,

Reynolds Number Effects onPressure Loss and TurbulenceCharacteristics of Four

Tube-Bundle Heat Exchangers

William B. Igoe and Garl L. Gentry, Jr.

LiR RYCOPYJUNE 1984

,JL:i; :. l: "+t

_A ,_'' N.\ -SAL!_,,, .,'_,

H._,p,+q[':1,qN ".'I}::'_'.I_IA

https://ntrs.nasa.gov/search.jsp?R=19840017577 2018-05-25T02:34:35+00:00Z

NASA Technical Memorandum 85807

Reynolds Number Effects onPressure Loss and TurbulenceCharacteristics of Four

Tube-Bundle Heat Exchangers

William B. Igoe and Garl L. Gentry, Jr.

Langley Research Center

Hampton, Virginia

National Aeronautics

and Space Administration

Scientific and TechnicalInformation Branch

1984

SUMMARY

The aerodynamic characteristics of pressure loss and turbulence on four tube-bundle configurations representing heat-exchanger geometries with nominally the sameheat capacity were measured as a function of Reynolds number from about 4000 to400 000 based on tube hydraulic diameter. Two elliptical- and two round-tube

configurations were tested. All four configurations had plate fins.

The results indicate an apparent aerodynamic advantage of the elliptical-tubeshape compared with the round-tube shape both for pressure loss and turbulence char-acteristics. For all four configurations, the pressure loss coefficient decreasedwith increasing Reynolds number at low Reynolds numbers. At high Reynolds numbers,the same trend persisted for the elliptical-tube configurations but was reversed forthe round-tube configurations.

INTRODUCTION

The adoption of cryogenic operation has made it possible to obtain high Reynoldsnumber test conditions at transonic speeds in conventional closed-circuit, fan-driven, continuous-flow wind tunnels (as indicated, for example, in refs. I to 4).Cryogenic operation in nitrogen with liquid nitrogen injected directly into the cir-cuit for cooling in these tunnels has been shown to increase the maximum testReynolds number by about 5 to 7 times that available at ambient temperatures for thesame stagnation pressure and Mach number conditions (ref. I).

The overall consumption of energy in cryogenic tunnels is high, primarilybecause of the energy required to produce the liquid nitrogen which is used for cool-ing (ref. 2). If a conventional chilled-water heat exchanger was used for cooling inthe ambient temperature mode of operation instead of liquid nitrogen injection, ithas been estimated that the consumption of energy in this mode would be reduced by anorder of magnitude. In addition to the advantages in energy consumption, the bene-fits of using a conventional heat exchanger for cooling also include the capabilityto use air as well as nitrogen as the test gas in the ambient temperature mode.

The Reynolds number range over which a cryogenic wind tunnel operates in thecryogenic mode is generally well above that for which a heat exchanger would beneeded for cooling in the ambient temperature mode. The external, or aerodynamic,characteristics of the tube-bundle elements of the heat exchanger such as pressureloss and turbulence generation are functions of Reynolds number as well as other testvariables. These aerodynamic characteristics are important to the efficiency and tothe flow quality of the wind tunnel and must be determined for all operating condi-tions. Consequently, in the development of large cryogenic wind tunnels, it may benecessary to determine these characteristics at Reynolds numbers substantiallygreater than those for which the heat exchangers are designed for heat transferpurposes.

The National Transonic Facility (NTF) at the Langley Research Center is an exam-ple of the kind of wind tunnel under discussion here. It is a cryogenic, fan-driven,continuous-flow wind tunnel with a 2.5-m-square test section. In order to test atlow and moderate Reynolds numbers in air as well as in nitrogen, the NTF has a

conventional chilled-water heat exchanger in the tunnel circuit (ref. 5). The heatexchanger has a design heat capacity corresponding to about 0.3 MW/m2 of frontalarea.

In order to provide comparative data needed to guide the selection of the heat-exchanger geometry for the NTF, several candidate tube-bundle configurations, eachdesigned for nominally the same heat capacity, were tested for the aerodynamic char-acteristics of pressure loss and turbulence. None of the heat-transfer characteris-tics of the heat exchangers were simulated, and all tests were conducted under

adiabatic flow conditions. Some results of tests at low Reynolds number are reportedin reference 6.

The present tests were conducted in the 20- by 60-cm test section of the Langley0.3-Meter Transonic Cryogenic Tunnel (TCT) at Reynolds numbers from about 3.8 x 105to 2.8 x 107 per meter. The free-stream Mach number was varied from about 0.01

to 0.10, the stagnation pressure was varied from about 1.2 to 5.0 atm, and the stag-nation temperature was varied from about 100 to 300 K. The Reynolds number based ontube hydraulic diameter varied from approximately 4 × 103 to 4 × 105.

SYMBOLS

a speed of sound

length

M Mach number, U/a

q dynamic pressure, pU2/2

R Reynolds number, pU_/_

U free-stream velocity

u' root-mean-square (rms) turbulent-velocity component in streamwisedirection

x,y,z streamwise coordinate, lateral coordinate parallel to tube axis, andlateral coordinate normal to tube axis, respectively

APt stagnation pressure loss across heat-exchanger tube bundle

dynamic viscosity of test gas

p mass density of test gas

TEST APPARATUS

The current configuration of the 0.3-m TCT with a 20- by 60-cm test section

(fig. 1) is described in reference 7. It is a closed-circuit, fan-driven, pressur-ized tunnel capable of continuous operation. The test gas is nitrogen, and themethod of cooling is by injection of liquid nitrogen into the circuit. This methodof cooling, and the consequent exhausting of gaseous nitrogen to the atmosphere tomaintain a constant pressure, limits the minimum operating pressure of the tunnel to

2

about 1.2 atm. At the time of the tests, the maximum pressure limit on the pressure

shell was 5.0 atm, although since that time this limit has been raised to 6.0 atm

(ref. 7).

The temperature limits of the tunnel range from a low of 77 K (the liquefaction

temperature of nitrogen) to about 327 K. This wide range of temperatures permits a

wide range of Reynolds numbers to be obtained at low dynamic pressures.

The tube-bundle heat-exchanger models were installed in the 20- by 60-cm test

section of the tunnel with the tubes spanning the 20-cm dimension as shown in the

photograph of figure 2 and the sketch of figure 3. Normally, the test section is

ventilated with longitudinal slots in the top and bottom walls. The side walls are

usually solid except when provisions are made for boundary-layer removal. For the

present tests, the slotted walls were replaced with solid walls, and no test-sectionventilation was used.

MODELS

The cryogenic feature of the 0.3-m TCT allowed high Reynolds number test condi-

tions to be reached without imposing severe loading conditions on the test models.

The low dynamic pressures allowed a simplified method of construction for the models

of the various tube-bundle heat exchangers.

Since heat transfer was not included in the tests, it was not necessary to con-

struct the tube bundles of actual tube hardware. Consequently, for simplicity, all

the models were constructed with thin, flat brass sheets for the plate fins with

aluminum washers aligned between the fins to simulate the tubes. The fin spacing,

tube diameter, and round or elliptical tube shapes were provided by using washers of

appropriate size and shape. The tube bundles were held together by aluminum stud

bolts through the centers of the washers. All the materials used were compatible

with the cryogenic temperature environment to which they were exposed.

Geometrically, the models were made the same size as the tube bundles they were

intended to represent. Streamwise, the full depth of the tube bundle was simulated.

Laterally, the number of tubes represented was sufficient to fill the 60-cm dimension

of the test section. The length of the tubes spanned the 20-cm dimension of the test

section.

Four tube-bundle heat-exchanger models were tested. The configurations of the

models are shown in figure 4. All the configurations had in-line tube arrays with

plate fins. Two of the configurations (elliptical 1 and 2) had elliptically shaped

tubes, and the other two configurations (baseline and six-row) had round tubes. Both

of the elliptical-tube configurations and the baseline configuration had four rows of

tubes streamwise. The other round-tube configuration had six rows of tubes stream-

wise. The two configurations designated elliptical 1 and 2 differed only in the size

of the tubes.

It was intended that all the tube-bundle heat-exchanger models would represent

configurations with substantially the same heat capacity. After the dimensions of

the smaller elliptical tube were selected, it was discovered that, from a heat-

transfer consideration, this configuration would require a water circulation velocity

inside the tube that was higher than was considered desirable. The larger elliptical

tube allowed a lower water velocity. For completely comparable internal flow condi-tions (same water velocity), the smaller elliptical-tube configuration would there-fore have fallen somewhat short (about 75 percent) of the required heat capacity.

INSTRUMENTATION

The instrumentation for the present tests consisted primarily of the instrumen-tation normally in use at the 0.3-m TCT (refs. 4 and 7). I_ should be noted that thecapacities of the pressure transducers regularly used at the tunnel are sized for thefull range of pressures associated with high Mach number (near-sonic) operation athigh total pressures. Because of the low speeds of the present tests, in someinstances lower capacity, more sensitive pressure transducers were used.

The tunnel free-stream total pressure is normally measured in the settling cham-ber (downstream of the anti-turbulence screens) with a differential pressure gagereferenced to vacuum with a capacity of about 700 kPa. For the present tests, thetotal pressure was measured with a pitot probe located just upstream of the testsection with the same kind of pressure gage installation and capacity.

The free-stream static pressure is normally measured with sidewall static-pressure orifices located near the front of the test section with a pressure gageinstallation and capacity the same as those used for the total pressure. For thepresent tests, a more sensitive pressure gage with a capacity of about 3 kPa wasconnected to the same static-pressure orifices and referenced to free-stream totalpressure.

The pressure drop across the tube-bundle heat-exchanger models was obtained fromthe difference in total pressures measured upstream and downstream of the models.The upstream total pressure was obtained from the free-stream total-pressure pitottube, and the downstream total pressure was measured with a total-pressure rake.This rake, which is shown in figure 5, is normally used for surveying the wake behindairfoil models in the test section with differential pressure gages referenced tofree-stream total pressures; the capacity of these gages is about 140 kPa. For thepresent tests, the farthest outboard total-pressure tube on the rake was connected toa sensitive differential-pressure gage referenced to free-stream total pressure. Thecapacity of this gage is about 50 kPa.

For the heat-exchanger models installed in the 0.3-m TCT test section, the dis-tance from the rear face of the tube bundle to the total-pressure-rake tubes was

52.7 cm for the elliptical I and 2 configurations, 48.9 cm for the baseline config-uration, and 38.7 cm for the six-row configuration. These distances expressed interms of tube spacing are 13.8, 12.8, and 7.6, respectively. Although the surveyrake is normally capable of vertical (z-direction) translation, for this test therake was fixed at the tunnel centerline. Consequently, the pressure-tube readings onthe rake represented single-point measurements. A vertical survey yielding a spatialaverage would have been preferred, but this was not possible because of hardwarelimitations. However, flow uniformity measurements of reference 6 for similar tube-bundle configurations at low Reynolds numbers indicate that single-point measurements

of total pressure loss, expressed in terms of Apt/q, should not differ from a spa-tially averaged measurement by more than ±0.2.

The hot-wire probes consisted of crossed wires of 5-_m, platinum-coated tungstenwire with an effective length-diameter ratio of about 250. The hot wires were oper-ated in the constant temperature mode for turbulence measurements and in the constant

4

current mode for measurements of temperature spottiness. The output signals from the

hot-wire probes were recorded on magnetic tape for off-line analysis, and on-line rms

readings were also recorded. Only the rms turbulence data are presented in this

report. The tunnel stagnation temperature was measured in the settling chamber withthe standard tunnel instrumentation consisting of a platinum resistance thermometer.

RESULTS AND DISCUSSION

The pressure loss characteristics of the various tube-bundle configurations are

presented in the form of the difference in stagnation pressures ahead and behind the

models Ap t divided by the free-stream dynamic pressure q. The turbulence charac-teristics are presented as the rms turbulent-velocity components u', v', and w'

divided by the free-stream velocity U. For convenience, the Reynolds numbers are

given as unit Reynolds numbers of the flow rather than the Reynolds number based on

the hydraulic diameter of the tubes. This is done because the main intention of the

tests is to obtain comparative data on the performance of tube bundles of different

geometries under the same flow conditions.

Effect of Mach Number

During the tests, it was necessary to vary velocity as well as temperature and

pressure in order to obtain the wide range of Reynolds numbers desired. The small

change in Mach number caused by the change in velocity was not expected to have sig-

nificant effects on the aerodynamics of the various tube bundles.

In order to illustrate this anticipated insensitivity to small changes in Mach

number, the pressure loss data Apt/q as a function of y (the probe position onthe rake) are shown in figure 6 for each configuration at a nearly constant Reynolds

number over a range of Mach numbers from about 0.05 to 0.10. In fibre 6, theReynolds number for the elliptical 1 configuration is about 25 x 10 per meter; for

the other configurations it is about 13 x 106 per meter. Within the normal scatter

of the data, the pressure loss is uniform for the several probes over the span of the

rake.

The data from figure 6 for the farthest outboard probe of the rake have been

replotted as a function of Mach number in figure 7 to show that the pressure losscoefficients do not vary significantly with Mach number. From these results, it has

been concluded that varying Reynolds number by varying velocity in this low Mach

number range does not introduce any extraneous Mach number effects.

Effect of Reynolds Number

As mentioned earlier, the Reynolds numbers of the tests varied from about

3.8 x 105 to 2.8 x 107 per meter. This wide range was only possible by varying

velocity or Mach number as well as pressure and temperature. The Reynolds number

range at constant Mach number was somewhat more limited. Pressure loss data for a

nearly constant Mach number of about 0.05 are presented in figure 8 for the four

tube-bundle configurations. With minor exceptions for the baseline and six-row con-

figurations, once again the pressure loss is uniform over the span of the rake.

The data from figure 8 for the farthest outboard probe on the rake (y = 11.4 cm)

have been replotted as a function of unit Reynolds number for the four configurations

5

in figure 9. Data for higher and lower Mach numbers have been included to increasethe Reynolds number range.

Within the range of the tests, both the elliptical I and 2 configurations show acontinuing reduction of pressure loss coefficient with increasing Reynolds number.The two round-tube configurations show a reversal to a rising trend at the higherReynolds numbers, with the effect being more pronounced for the baseline configura-tion, which also had the lowest porosity (ratio of projected open area to total areaof tube bundle).

The magnitude of the pressure loss coefficient in figure 9 varies from the low-

est for the elliptical 1 configuration (which had the highest porosity) to the high-est for the baseline configuration (which had the lowest porosity). A comparison ofthe pressure loss characteristics of the elliptical 2 and six-row configurations,which had nearly the same porosities, shows an apparent aerodynamic advantage for theelliptically shaped tubes.

Turbulence Characteristics

The root-mean-square (rms) longitudinal components of turbulent velocity u'/Umeasured upstream and downstream of the tube-bundle heat-exchanger models are pre-sented in figure 10. The downstream turbulence varies from lowest to highest roughlyin the order elliptical I, elliptical 2, baseline, and six-row configurations. Thehigh value of turbulence measured for the six-row configuration may be at least par-tially due to the greater streamwise depth of this tube bundle. The turbulence-measuring station was a fixed distance (54.3 cm) downstream from the front face ofall the tube-bundle heat-exchanger models. Consequently, for a tube bundle with

greater streamwise depth, the distance from the last row of tubes to the measuringstation was less. The streamwise decay of turbulence increases with this distance.Since this distance was less for the six-row configuration than for the others, theturbulence for this configuration had less flow length over which decay could takeplace. Consequently, the higher level of turbulence indicated for this configurationmay not be wholly attributable to the characteristics of the tube bundle itself, butmay be at least partially due to less decay of the turbulence.

The distance from the rear face of the tube bundle to the hot wire was 39.1 cmfor the elliptical 1 and 2 configurations, 35.3 cm for the baseline configuration,and 25.1 cm for the six-row configuration. In terms of tube spacing, these distanceswere 10.3, 9.3, and 4.9, respectively. Based on turbulence decay data obtained atlow Reynolds numbers for similar tube-bundle configurations (ref. 6), it is estimatedthat the difference in decay distance (35.3 cm for the baseline configuration com-pared with 25.1 cm for the six-row configuration) could cause a difference of 20 per-cent more reduction for the configuration with the greater decay length. Since thestreamwise depths of the elliptical I and 2 configurations were more nearly equal tothe baseline configuration, their decay distances were also more nearly equal(39.1 cm compared with 35.3 cm), and so their turbulence levels may be compareddirectly.

CONCLUDING REMARKS

The aerodynamic characteristics of pressure loss and turbulence on four tube-bundle configurations representing heat-exchanger geometries with nominally the sameheat capacity were measured as a function of Reynolds number from about 4000 to

6

400 000 based on tube hydraulic diameter. Two elliptical- and two round-tube

configurations were tested. All four configurations had plate fins.

The results indicate an apparent aerodynamic advantage of the elliptical-tubeshape compared with the round-tube shape both for pressure loss and turbulence char-acteristics. For all four configurations, the pressure loss coefficient decreasedwith increasing Reynolds number at low Reynolds numbers. At high Reynolds numbers,the same trend persisted for the elliptical-tube configurations but was reversed forthe round-tube configurations.

Langley Research CenterNational Aeronautics and Space Administration

Hampton, VA 23665May 15, 1984

REFERENCES

1. Kilgore, Robert A; Goodyer, Michael J.; Adcock, Jerry B.; and Davenport,Edwin E.: The Cryogenic Wind-Tunnel Concept for High Reynolds Number Testing.NASA TN D-7762, 1974.

2. Polhamus, E. C.; Kilgore, R. A.; Adcock, J. B.; and Ray, E. J.: %_neLangley Cryo-genic High Reynolds Number Wind-Tunnel Program. Astronaut. & Aeronaut.,vol. 12, no. 10, Oct. 1974, pp. 30-40.

3. Kilgore, Robert A.; Adcock, Jerry B.; and Ray, Edward J.: Simulation of FlightTest Conditions in the Langley Pilot Transonic Cryogenic Tunnel. NASATN D-7811, 1974.

4. Kilgore, Robert A.: Design Features and Operational Characteristics of theLangley 0.3-Meter Transonic Cryogenic Tunnel. NASA TN D-8304, 1976.

5. Howell, Robert R.; and McKinney, Linwood W.: The U.S. 2.5-Meter Cryogenic HighReynolds Number Tunnel. High Reynolds Number Research, Donald D. Baals, ed.,NASA CP-2009, 1977, pp. 27-51.

6. Johnson, William G., Jr.; and Igoe, William B.: Aerodynamic Characteristics atLow Reynolds Numbers of Several Heat-Exchanger Configurations for Wind-TunnelUse. NASA TM-80188, 1979.

7. Ray, Edward J.; Ladson, Charles L.; Adcock, Jerry B.; Lawing, Pierce L.; and Hall,Robert M.: Review of Design and Operational Characteristics of the 0.3-MeterTransonic Cryogenic Tunnel. First International Symposium on Cryogenic WindTunnels, Proceedings of the Symposium held in the Department of Aeronautics andAstronautics at the University of Southampton, England, Apr. 3-5, 1979,pp. 28.1 - 28.15.

co

9.14Zlm

_ _--1.219m /-Plenum& f----LN2

J /- Screensection / testr / / section / injection

i! n-t..

• _ _

1.7 _

Fan section _ r

Nace,,e [T_ \ [T_ "_ Drive _-_

__ Sliding joints __]]'_ anchorSeCti°n _-_7_/_,._t-- !;/,lrliTunnel _/ /\ \ ///

Jl ' ' '

Figure I.- The Langley 0.3-Meter Transonic Cryogenic Tunnel with 20- by 60-cmtest section.

L-77-8237.1

Figure 2.- Tube-bundle heat-exchanger model installed in 20- by 60-cm test sectionof the Langley 0.3-Meter Transonic Cryogenic Tunnel.

-72.39 -31.75 Sta. 0 78.74

Pitot f Hot-wire I

prob20._3__ probe

"//////2" ///,h\\'_/ /.,_\,_,,_ _.-_._////

• _ Flow_

/L.227 =-

Tube bundlej H - _--Total-pressure-80.01 rake

Plan view

Figure 3.- The 20- by 60-cm test section of Langley 0.3-Meter Transonic CryogenicTunnel showing typical tube-bundle heat-exchanger model installation. Alldimensions in centimeters.

I0

___ 15.240

1.905 11.43_ "_--L905

,1.826

(D O O O---_/

0 0 0 0

0 0 0 0

0 0 0 0

Q 0 0 0

0 0 0 0

0 0 0 0

FI0w--" 0 0 0 0

C_ 0 C_ 0 _ "_-'_

O CZ> O O j _ Platefins,=_ .0256thick<D <Z> tED tED _ '

; I rO (D CD O IL _ ' ' '

i J'

o tED <D o I' _ ,_ I I j') i [

1.826 per2.54

Z4 rows in-line, 3.81x 3.81Porosity= 0.713

(a) Elliptical 1 configuration.

Figure 4.- Geometry of tube-bundle heat-exchanger models. All dimensions

in centimeters.

11

15.240

1.905_ _iI.43_ _ i.i051.826

O O O 0-_-_-0000

0000

OOOO

0000

OOOO

0000Flow--_ 0000

O O O O 60.803 ....._1< _1!71_O O O O ) 1.135

O O O O _-Flatefins, --_-

<D CD <D (_ .

3.81

(_ CD O C2_ f /

_ <D (_ CD _ rz1.826 ,

_4rows in-line, 3.81x3.81Porosity= 0.631

(b) Elliptical 2 configuration.

Figure 4.- Continued.

12

0 0 0 0

0 0 0 0I

O O O 0

0 0 0 0

0 0 0 0Flow

--"- 0 0 0 0 60.803 J_

0 0 0 O _ Platefins, _.

6,thick 1.59 diam

0 O 0 0I

0 0 0 0(

0 0 0 0

(c) Baseline configuration.

Figure 4.- Continued.

13

i 29.210 _/_25.400 _i 1.905

O Q Q ® Q 0- _46j

O O O O O Q

O O O O O O

O 0 O O O O

0 0 O O O OFlow

---- 0 0 O (D O O 6o.803

55.88 __0 0 O O O O _?,i_s,thick 1.59diam

0 O O Q O O ,"x

0 O 0 O O Q-Ti.08

0 0 0 0 0 0 -I

0 O O O (D Q 6Z_p__54

O (D © © ©-- ,6 rowsin-line, 5.08x 5.08Porosity= 0.623

(d) Six-row configuration.

Figure 4.- Concluded.

14

Temperatureprobe

1.27

r

10.16 c_

Flow 12 95 ,

I

(Do

Y

l 2.54_'----Wall

_1.91_

Figure 5.- Rake used to measure downstream total pressure. All dimensionsin centimeters.

15

3.0

M

2.6 051060 25.3072 25.9082 25.8

2.2 088 25.i

APtlq 099 25,0

1.4 _._: ir_i_ _

....!i....0 2 _ 6 8 i0 12

(wall)y, cm


40

3.6 M• 048• 063 13.2• 065 12.7

3. 2 .077 12.6•083 12.3

APtlq

2.8

2.4

2.0

0 2 4 6 8 i0 12(wall)

y, cm


Figure 6.- Pressure loss coefficient at nearly constant Reynolds number forvarious Mach numbers.

16


50

:_x:.>:::x;l:::_..t:t:: :::l>:Ix:;{..:-:l:;:{_!_ii_il_I!_l_N_l_'_l_4_..-M R, per meter

4.6 :::::::::::::::::::::::::::::::::::::::::::::::::::::: -'-_'_::::::::::::::::::::::::::::::::::::::::::::::::::::::0 .049 14.1 x 106 =_..

,-, . 063 14.3 _

::::::::::::::::::::::::::::::::::::::::................ z_ .077 13.9 i_;_

_i_r. _';t U [ t:; i:;_ :*'r: :-_..: 7..:; ::;: ::-_= :::: ;_;: :::3 _ .7,_:;::_:IV +_ _ _'_:

0 2 4 8 10 12

(wall) y, cm



17

R, permeter

Elliptical1 o 25Xl06Elliptical2 _ 13Baseline _ 13Six Row I_ 13

10 --

8--

6- _ AAPt/q _

4- _L-_. _ _

2 -

, I I I , I l I , I0 .02 .04 .06 .08 .I

M

Figure 7,- Pressure loss coefficient as a function of Math number at constantReynolds number for y = 11 .4 cm.

18

3.4 R, per meter M

o L68x106 .052o 2.26 .049

3.0 <> 3.37 .049A 6.80 .050t_ Ii.O0 .050r,, 25.30 .051

Z6

APtlq 2.2

L8

1.4

LO0 2 4 6 8 10 12

(wall) y, cm



Figure 8.- Pressure loss coefficient at nearly constant Mach number forvarious Reynolds numbers.

19


5.8 R, per meter M

0 1.77x106 .051 ;_:;:ii_[] 3.58 .049 !tiiil!i::i_!i

5.4 <> 6. 15 .047 l_;i__Iii!A 7.13 .049 r;'r_ 14.23 .050 i!ii{3 2Z44 .050 _::::

5. 0 n 28.47 .050

4.2 ....

• _, , _ .Lii!i

0 2 4 6 8 I0 12(wall)

y, cm



2O

100

50

20

fl Pt' q10

5

2

R, per meter

o Elliptical 1(> Elliptical 2~ Six rowl::. Baseline

Porosity

0.713.631.623.525

Figure 9.- Pressure loss coefficient as function of unit Reynolds numberfor y = 11.4 cm.

14--

12 --I

O Upstream[] Downstream

10 --

08 -- []

D[][]

U'-G 06-- D

04--[]

02 -- OO O

0 0 0 0 0o oo

0 ' _ r v I llrl w I I , i _II ; I ; ; , rit{

105 106 107 108R, per meter

(a) Elliptical I configuration.

Figure 10.- Longitudinal turbulence characteristics of tube-bundle heat-exchanger models.

• 14 --

• 12 --

0 Upstream[] Downstream

• 10 --E3

[]

[]• 08 --

U' []

D•06 -- [] []

[13

• 04 -- 0

• 02 -- o© 0 0

o oo o o °80 l l _ I I _ _II l i l I w f l_l I , , f I ,,,I

105 106 107 108R, per meter


Figure I0.- Continued.

• 14 --

• 12 --

© Upstream[] Downstream

• i0 --

[] [] []

U' .08 -- [] [] m__ [7 '

U o E_nn DD

• 06 --

•04 --

•02 --0 0

0 0

0 I I I l I ,Ill I I I I I fal! I I I I I IIII

105 106 107 108R, per meter


Figure I0.- Continued.

• 14 --[]

0 Upstream[3 13 Downstream

• 12 m [] []

[][]

• 10

• 08

U'

.06 m

•04

•02 _ o

o o

0 I i i i L I LiJ l i l i , ,,,1 I i l i i i,,l

105 106 107 108R, per meter



1. Report No. I 2. GovernmentAccessionNo. 3. Recipient'sCatalogNo.NASA TM-85807 I

4. Title and Subtitle 5. Report Date

REYNOLDS NUMBER EFFECTS ON PRESSURE LOSS AND TURBULENCE June 1984

CHARACTERISTICS OF FOUR TUBE-BUNDLE HEAT EXCHANGERS 6. PerformingOrganizationCode

505-31 -53-08

7. Author(s) 8. PerformingOrganizationReport No.

William B. Igoe and Garl L. Gentry, Jr. L-15721

10. Work Unit No.9. PerformingOrganizationNameand Address

NASA Langley Research Center 11 Contractor GrantNoHampton, VA 23665

13. Type of Report and Period Covered12. SponsoringAgency Name and Address

Technical MemorandumNational Aeronautics and Space Administration

14. SponsoringAgency CodeWashington, DC 20546

5. SupplementaryNotes

16. Abstract

The aerodynamic characteristics of pressure loss and turbulence on four tube-bundle

configurations representing heat-exchanger geometries with nominally the same heat

capacity were measured as a function of Reynolds numbers from about 4000 to 400 000

based on tube hydraulic diameter. Two configurations had elliptical tubes, the other

two had round tubes, and all four had plate fins. The elliptical-tube configurations

had lower pressure loss and turbulence characteristics than the round-tube configura-tions over the entire Reynolds number range.

,17. Key Words (Sugg_ted by Author(s)) 18. Distribution Statement

Finned tubes Wind tunnel Unclassified - UnlimitedHeat exchangersPressure loss

Turbulence

Reynolds number effects Subject Category 02

19. S_urity _a_if.(ofthisreport) 120. S_urityClassif.(ofthis_) 21. No. of Pages I 22. Dice

Unclassified I Unclassified 26 I A03

ForsalebytheNationalTechnicalInformationService,Sprinsfield,Virsinia2216! NASA-Langley,]98e

National Aeronautics and THIRD-CLASS BULK RATE Postage and Fees PaidSpace Administration National Aeronautics and E _ _

Space Administration

Washington, D.C. NASA-4Sl20546

Official Business

Penalty for Private Use, $300

POSTMASTER: If Undeliverable (Section l S8Postal Manual) Do Not Return

Documents

Reynolds Number Effects on Pressure Loss and … The aerodynamic characteristics of pressure loss and turbulence on four tube-bundle configurations representing heat-exchanger geometries