Professional Documents
Culture Documents
ASME PTC 30-1991 Air Cooled Heat Exchangers
ASME PTC 30-1991 Air Cooled Heat Exchangers
W
O
r
T.
N er
O
t-
g
66
--``-`-`,,`,,`,`,,`---
The total tube-side pressure drop is the sum of APf For fully developed turbulent flow (Rei > approx-
and APent. Nozzleandheaderlossesmayneed to be imately 7,000), there areseveralgoodcorrelations
separately considered. available. For general use, the Petukhov-Popov equa-
.The properties used in these equations and those tion,Eq.(D.61,(Ref. [I 311, is regarded as .the most
in the following paragraphs are usually evaluated at accurate but it is not in convenient form for ratioing
the arithmetic mean bulk temperature on the tube changes in velocity, for example,
side, except for pp,,, which is evaluated atthe inside
wall temperature at the point where thefluid reaches
--``-`-`,,`,,`,`,,`---
0.0668( y)(
y)(
F) ditions:
3.65 +
,+o.o4[ (?)(y)(?)]
Forwater, Figs. D.2aandD.2btaken from Kern
(Ref.[141),are very easy to use.FiguresD.2aand
.(E)"' (D.5)
D.2bmayberepresented
sional equation:
by the following dimen-
67
For transition flow (2100 < Re, 7,000), no ac- vapor with no liquid present has a quality of 1.00,
curate predictions are possible because of the slow- and a totally condensed stream at its boiling point or
ness with which fully-developed velocity and thermal bubble point has a quality of zero.
profiles are achieved and becauseof the strong effect Estimates can be made of the frictional loss using
of the entrance flow geometry. An estimate can be the work of Martinelli andNelson (Ref. [151). The
made by linearly interpolating between the laminar frictional pressure dropthrough the tube for the con-
flow heat transfercoefficient, ,obtainedfrom Eq. densing flow entering as a saturated vapor(x, = 1.O),
(D.5) and the turbulent flow result, h,, , obtained and exiting at a qualityof x, is found from
from Eq. (D.6) or (D.7)) using the equation: -
i APP,TPF = 43P,,Y (D,10)
h, = KL t (h,,, - 7j;.L I( Re' - 2100
4900 ) where TPF indicates two-phase flow and APp,v is the
pressure dropcalculatedfrom Eq. (D.?) assuming
that theflow is all vapor and no,condensation occurs.
for 2100 c Re,< 7000 0.9) The mean two-phasemultipying factor, P,,,is read
from Fig. D.3 as a function of the exit qualityand the
reduced pressure of the vapor,
Any calculation in this range must be regarded as J
highly uncertain.
--``-`-`,,`,,`,`,,`---
P
P, = - (D.11)
PC,
68
1 .o 1.5 2 3 4 5 6 7 8 9 1 0
Velocity Through Tubes, ft/sec
GENERAL NOTE: This chart applies only to a tube 0.62 inside diameter kg., 3/4 in. X 16 BWG).
For other diameters, refer to Fig. D.2b.
69
1.2
'i *I
ò
5 1.0
U ,
.-
0
CI
E0 0.9
ò
-.o-
"
0.5
O.7.
0.4 0.6 0.7 0.8 0.9 1.0
Inside Diameter of Tube, in.
1.5 2.0
GENERAL NOTE: Multiply the value obtained from Fig. D.2a by the above factor.
. .
the tube. Kern's modification (Ref. [I411of Nusselt's and ß is found as a function of I$? in Fig. D.4. The
equation (Ref. 1161) may be used to calculate the Chaddock correlationcorrectsthehorizontaltube
coefficient in this case: Nusselt equation for the relative amount of surface
blanketed by thestratified pool of liquid (through
r i113
which no heat transfer is assumed to occur).
At higher condensing rates, all or a portion of the
L -I
tube may be in annular two-phase flow, in which a
turbulent liquid film covers the entire inner surface
of the tube. A convenient correlation forthis regime
is due to Boyko and Kruzhilin (Ref. [18J).
In this equation, L is the length of the tube and W ,
is the pounds of vapor condensed pertube per hour.
If a U-bend tubeis used, L is the combined lengthof
both straight sections and the U bend.
A more rigorous equation used for low condensing
rates is the Chaddock correlation (Ref. [I 71):
(D.14a)
=
r
I
- 5.06 X 10-4
)LrT,,, -7-J"'
D:.75
I
70.142
(D.13b)
L -I
r 1114
70
--``-`-`,,`,,`,`,,`---
where x, and x, are the inlet and exil qualities of the where Pr,,p = cp81,pp1*p
kf,P
stream,. 'respectively, For the special and important
case of totalcondensation of asaturatedvapor
stream, the term in brackets in Eq. (D.14a) reduces Again, research is showing that the transitionfrom
to: laminar to turbulent flow is not as abrupt as,sug-
gested by Fig. D.5 and the actual coefficients in the
transition regime are higher than shown for Rei from
(D.14e) perhaps 800 to 3000.
If the condensing load (or moreexactly, the vapor
flow) is sufficientlyhigh, vapor shear effects cause an
The correlationsof Kern and Chaddock are valid at early transition to turbulence in the condensate film
and a sharp increase in the heat transfer coefficient.
0 low vapor flow rates, where gravity dominates the
flow pattern,andareinvalidat high vapor flow rates,
. wherevaporsheardominates.TheBoyko-Kruzhilin
Undertheseconditions,theBoyko-Kruzhilinequa-
tion given above, Eq. (D.14a) et seq., becomes valid.
correlation operates in exactly the opposite fashion. Again, a conservative and simple procedure for esti-
Comparisonof the fundamentalbases for each equa- mating a condensing coefficient for vertical tubes is
tion indicates that the correlationwhich is valid under to calculate the. coefficient by all threeequations,
a given set of conditions gives a higher heat transfer (D.14), (D.15), and (D.161,andselectthe highest
coefficient than the invalid correlation. Therefore, to coefficient.
determine which type of correlation is applicable in D.3.3.3 Inclined Tubes.Very fewdatahavebeen
a given situation, one may calculate the coefficient published on condensation in downward flow inside
by each method and select the higher.value. In the inclined tubes, though proprietary data and correla-
transition region where the correlations cross, the ac-
tions exist. It is reported that the condensing coeffi-
tual coefficients are found to be greater than those
cient increases significantly (compared to a vertical
predicted by either type of correlation. The values of
tube) in a tube which is inclined from 1 deg. to 20
6, calculated by theseequationsaremeanvaluesfor deg. from the vertical. As the inclination moves to-
the entire tube. Calculation of the local values is be-
ward the horizontal, the coefficient changes toward
yond the scope of this standard;Ref. I191 may be
that for a horizontal tube.
consulted as a typical example.
Nilsson (Ref. [21]) has shown that a very slight (1
to 2 deg.) upward inclination in a horizontal tube can
D.3.3.2 Vertical Tubes. Nusselt (Ref. [I611 also ob- cause substantial reduction in the condensation heat
tained an equation for Condensation under laminar transfer coefficient, presumably because of excessive
condensate film conditions in vertical tubes (corre- pooling of the liquid in the lower end of the tube.
spondinggenerally to low condensingrates).This
equation is
D.3,4 Mean Temperature Difference for Saturated
I- 1113
PureComponentCondensation. Themean tem-
perature difference for condensation of a saturated
pure vapor, assuming a constant overall heat transfer
71
--``-`-`,,`,,`,`,,`---
6
9
O
II
a
Q
I
t
t- 8 9:
O
72
O 1 .o0
0.95
0.90
0.85
0.80
ß
0.75
0.70
O. 65
0.60.
0 0.5 1.0 1.5 2.0 2.5 3.0
*", radians
coefficient, and constant saturation temperature, is sure existing in the vapor space. In this case, it has
given by: been shown that the above-referenced equationsfor
condensing a saturated vapor adequately predict the
EMTD = LMTD = ,t -G \ (D.17) heat transfer rate on the vapor side, if the saturation
temperature of the vapor is used as the process fluid
temperature in Eq. (D.17). It is necessary to include
73
74
Reference i221 shows that the heat transfefflux for These calculationsrequirevapor-liquid equilibrium
cooling a superheated vapor must be higher than the and enthalpy calculations that are usually computer-
condensing flux if the wall is to remain dry. Therefore, based and supplied by the customer. Before any con-
if it is assumed that the vapor is desuperheating in clusions can be drawn about the performanceof the
the wet wall regime from the start (using the simple condenser, mutual agreement must be reached be-
procedure given in the first paragraph), the area cal- tween customer and vendor on the validity of these
culated to be required will either be correct (if the calculations.
wall is wet even at the vapor entrance)or conservative (b) Sensibleheattransfereffects(¡.e./ cooling of
(if some portion of the wall is in fact dry). The term the vapor-gas mixture) are always present in multi-
correct means that the calculated area is as close to component condensation in both the vapor-gas and
that actually required as the validity of the correla- liquid phases. The cooling of the vapor-gas mixture
tions permits; conservative means that the calculated tends to be an important and often controlling part
area is larger than would be obtained by a detailed of the heat transfer process.
local
calculation. (c) Mass transfer effects are always presentin mul-
ticomponent condensation. These processes areonly
poorlyunderstoodandmust be treated in a fairly
D.3.6 Subcooling of Condensate, When subcooled arbitrary manner.The specific problem of a single
condensate is required, it is customary to design the condensable vapor with a noncondensable gas can
condenser so that the bottom row or rows of tubes, be handled with some rigor as shown in Ref. [231.
disposed in one or more passes, run full of conden- (d) Physical properties changein both phases, both
sate. The liquid phase heat transfercoefficient canbe as a result of changing compositions and changing
temperatures.
--``-`-`,,`,,`,`,,`---
calculated using the correlations given in para.D.2
and the heat transfer rateby using a corrected LMTD, Usually, careful analysis of multicomponent con-
the correction factors being given in para. 5.5 for the densationproblems (which may includenoncon-
appropriate pass arrangement. densable gases) requires zone-by-zone analysis on a
The averageair temperature leaving the subcooling computer. However, if the condensing temperature
section can be calculated by a heat balance. Without range is relatively small compared to the mean tem-
going to a zone-by-zone analysis (which requires a perature difference, orif only a small amount of con-
computer program for all practical purposes), it is densate is formed,approximatecalculations . of
necessary to assume that the average air temperature sufficient accuracy may be possible (Ref. [24]).
off of the subcooling rowsis equivalent to a uniform In thesecases, the heattransferprocesson the
tube side may be considered to consist of tw6 re-
inlet air temperature to the condensing rows. This is
sistances in series:
of course not the case,and it is usually somewhat
( I ) Sensible heat transfer from the vapor-gas mix-
nonconservative to assume so. In analyzing the per-
ture to the condensate interface,with a typicalvapor-
formance of an existing unit, this factor can be taken
phase heat transfercoefficient h,, calculated from the
.into account qualitatively without a great deal of
correlations in para. D.2, and,
computation.
(2) Convection of the sensible heat from (1) above
75
Q, is the heat duty required to cool the vapor-gas (c) Reduction in number of tubes in successive
mixture: passes in condensing(andsometimessubcooling)
service in order to maintain high vapor velocity and
condensing coefficients. Note however that uniform
distribution of the two phases among the tubes in
later passescan not be expected, and this can lead
to excessive subcooling in sometubesand incom-
QLis the heat duty required for condensation: plete condensation in others.
(d) Multiple servicesmay be handled in a.single
Q, = X W p , c o n d (D.
18d) unit, usually with the sections in parallel on the air
flow (side-by-side in the frame). Different tube sizes
and number of rows may be used in each section.
(e) A single row of tubes may be split between two
Q,, is the heat duty required for cooling the con- passes in order to obtain the same number of tubes
densate: in eachpass,e.g., two passes in five rows of tubes.
(0 A row oftubes may contain asingle tube (or at
most a few tubes) servingas a vent condenseroff of
an air removal point and having a differentinlet and
outlet header connection.
Thermal analysis of types a, b, and c can be carried
- out by a procedure similar to that suggested for sub-
wp," an'cl Wp,,are the average weight flow rates of
vapor and condensate in the condensingprocess, and cooling sections above, and with the same caution
Wp,condis the amount of vapor actually condensed. upon assuming the air inlet temperature to the upper
rows of tubes to be uniform.
Type d can be analyzed straightforwardly foreach
section if the air flow and exit air temperature for
D.5 UNUSUAL PASSARRANGEMENTS each section are measured.
Type e can .be analyzedreasonablycloselyby
For. avariety of reasons,unusualpassarrange- straightforward methods, using the actual numberof
ments of various kinds are often used in air cooled
tubes in each pass for tube-side calculations and ig-
exchangers, The following examples may be cited:
noring the usually slight imbalance in the air tem-
(a) Reduction in number of tubes in successive
perature profile caused by the split pass or passes.
passes,used in cooling viscous liquids in order to
Type f poses no serious problems on the air side
increase thevelocity and maintainturbulent flow
since only a few tubes are involved.The analysis-in-
conditions. side the tube can be carried out by the methods of
(b) Using enhancement devices in one or more of para, D.4 or Ref. [231.
the last passes for the same purpose.
76
--``-`-`,,`,,`,`,,`---
--``-`-`,,`,,`,`,,`---
APPENDIX E - FOULING
E,1 Theexchangerdesignerincorporates a heat should preferablybe performed in the clean condition
transfer fouling resistance to account for the accu- on bóth air-side and tube-sideto minimize the effects
mulations of layers of resistive material on the heat of fouling since fouling cánnot be reliably predicted.
transfer surfacesas the exchanger operates. Thefoul- The fouling resistances used to interpret the test re-
ing resistance is also known as fouling factor, dirt fac- sults shall be agreed upon by the parties to the test
tor, and dirt film. The fouling resistance occurs on prior to the start of the test, see para. 3.2(j).
both the air-side and the process-side heat transfer
surfaces. Unfortunately,. the existing technologydoes
not provide a dependable analytical method for ac- E.3 the influence of foulingon the overall heat trans-
curate prediction of fouling. The purchaser normally fer coefficientwill vary accordingto the relative-mag-
depends on experience in similar services to select nitudes of the fouling resistances and the clean heat
and specify the design fouling resistances. transferresistances. For example, a closed-circuit
treated'water coolermight have a low tube-side foul-
ing resistance of 0.0005 hr.ft*-"F/Btu referencedto the
E,2 Fouling presentduringthe test affects the air-side inside surface. Thismight be approximately5 percent
and process-side heat transfer coefficients and flow of the total heat transfer resistance. In comparison,
pressure drops. Fouling of the air-side surface may this resistancefor aheavy oil cooler might be 0.003
occur from the depositionof air-borne materials such hr.ft2-"F/Btuwhich might be over 20 percent of the
as dust, organic material, seeds, and insects, or from total heat transfer resistance, making a clean condi-
corrosion. It is impossible to accurately predict the tion for testing relatively more important.
effect of such deposits and they must be removed
prior to testing. Fouling of the inside surface of the
tubes is dependent upon the fouling and corrosion E.4 For additionai information on fouling the reader
characteristics of the fluidin the tubes.Testing may refer to Ref. [25].
77
F.2 Since the performance evaluation procedures de- F.3 A detailed survey should be made just prior to
scribed by this Code are based on entering rather than the test, and agreement reachedby theparties to the
ambient air, the-recirculation and/or contamination test on the number andlocation of measurements to
described abovewill not necessarily havea significant be taken to ensure the desired level of accuracy.
effect op the performance capabirity of the ACHE. The
results of a test conducted while the entering air tem- F.4 For more information on this subject the reader
perature is well above ambient, but uniform, should is referred to Refs. [27] through 1361.
(I,
--``-`-`,,`,,`,`,,`---
V
79
APPENDIX G - REFERENCES
--``-`-`,,`,,`,`,,`---
[9]Blasius, H., Forschg. Arb. h g -Wes. No. 131,Ber- r241 Bell, K.J., and Chaly, M.A., AlChE Symp. Series
lin (1913).Cited in Schlicting, H., BoundaryLayer 69, NO. 131, 72-79 (1972).
Theory, 7th ed., McGraw-Hill Book Co., New York [25]Kakac, S., Bergles, A.E., and Mayinger, G., Heat
(I979). Exchangers: Thermal-Hydraulic Fundamentals and De-
[I O] Shah, R. K., and London, A. L., Advances in Heat sign, McGraw-Hill, and Hemisphere Publishing Corp.
Transfer, Supplement1: Laminar Flow Forced Convec- (I 9811.
tion in Ducts, Academic Press, New York (1978). i261 TubularExchangerManufacturersAssociation,
[Il]Hausen,H., VDlZ Beih. Verfahrenstech. 4. 91 Standards of TEMA, latest edition.
(1943).Cited in Ref. [12]. L271 Gunter, A. Y., and Shipes, K. V., “Hot Air Recir-
1121 Jakob, M., Heat Transfer, J. W: Wiley and Sons, culation .by Air Coolers,”AlChE TwelfthNational
New York, Vol. 1 (1949). Heat Transfer Conference, AIChE-ASME, Tulsa,Okla-
homa, August 15-1 8, 1971.
[I31 Petukhov, B. S., and Popov, V. N., Teplofiz. Vy-
sok. Temperature 7, No. 1 (1963).Also discussed by 1281 Collins, G. F., andMathews, R. T., “Climatic
B. S, Petukhov in Advances In Heat Transfer, Vol. 6, Considerations in Design of Air Cooled HeatExchan-
O Hartnett, J. P., and Irvine, T. F., Jr. Eds., Academic
Press, New York (1970).
gers,“Paper 59-A-255,December 4, 1959,
Meeting ASME.
Annual
81
1291CoolingTower InstituteTechnical Subcommittee [34] Haldridge, E. S., and Reed, B. H., ”Pressure Dis-
No, 2: “Recirculation,” C T I Bulletin PFM-110, 1958. tribution on Buildings,” Department of Army, Con-
Also PFM-11OA, Appendix to PFM-11O. tract No. DA-18-064 CML77,August1956,Texas
Engineering Experiment Station, Texas A & M.
[30] Schmidt, W., “Calculations of Distribution of
Smokeand
Waste
Gases in the Atmosphere,” 1351 Haldridge, E. S., and Reed, B. H., “Pressure Dis-
-
Gesundheits Ing. Vol 49, 1926, pp. 425-426. tribution on Buildings-ReportNo. 2,” Department of
Army, Contract No. DA-18-064 CML77, August
[31] Sutton, O. G., “A Theory of Eddy Diffusion in 1956,
Texas
Engineering
Experiment
Station,
the Atmosphere,“ Proc. Roy. Society (London) Ser. A Texas A & M.
Vol. 135, 1932, PP. 143-1 65.
[36] Kosten, C. J., Morgan, J. I., Burns, J. M., and
1321 Bailey,A., and Vincent,N.D. G., “Wind Pressure Curlett, P. L., “OperatingExperienceandPerform-
on Building Including Effects on Adjacent Building,” anceTesting of theWorlds Largest Air Cooled
Journal Institution of Civil Engineer, March 1943, Condenser,” April 27-29,1981,AmericanPower
PP. 243-275. Conference, Chicago, Illinois.
[33] Dryden, H. L., and Hill, C. C., “Wind Pressures
on Structures,” Scientific Papers of Bureau of Stan-
dards, Vol. 20, 1926, p. 697.
--``-`-`,,`,,`,`,,`---
82
--``-`-`,,`,,`,`,,`---
PTC 3,3
(R1 985)
PTC 4.1 - Steam-Generating Units (With 1968 and
1969 Addenda) .. .............. .....,................................I964
(R1 985)
Diagram for Testing of a Steam Generator,
Fig. 1 (Pad of 100)
Heat Balance of a Steam Generator,
Fig. 2 (Pad of 100)
PTC 4.la - ASME Test Form for Abbreviated EfficiencyTest -
Summary Sheet (Pad of 100) ......... .... . ......,........,.......I964
, , ,
PTC 4.1b - ASME Test for Abbreviated EfficiencyTest -
Calculation Sheet (Pad of 100) ........................ ..............1964
PTC 4.2 - . . ........ . ....... ...... .......,......,......,...I969
Coal Pulverizers , , , , ,, ,
(R1 985)
PTC 4.3 - Air Heaters ...... ... ............... ......... ..........................I968
(RI 985)
PTC 4.4 - Gas Turbine HeatRecovery Steam Generators........................I981
(RI 987)
PTC 5 - ReciprocatingSteam Engines ..................,.......................I949
PTC 6 - Steam Turbines .... ....................,......,............,............1976
(R1 982)
PTC 6A - Appendix Ato Test Code for Steam Turbines
(With 1958 Addenda) ...............,,......,.......................1982
PTC 6 - Guidance for Evaluation of Measurement Uncertainty
Report in Performance Tests of Steam Turbines .........................,..I985
PTC 6 s - Procedures for Routine Performance Tests
Report of Steam Turbines ...............,....................,..............I988
PTC 6.1 - Interim Test Code for an Alternative Procedure
for Testing Steam Turbines ...........,......... ...,.............,...I984
PTC 6 on Steam Turbines- Interpretations 1977-1983
PTC 7 - Reciprocating Steam-Driven Displacement Pumps...............,....1949
(R1 969)
PTC 7.1 - Displacement Pumps ............................,...............,......I962
(R1969)
PTC 8.2 - Centrifugal Pumps .. ..........,......,.......,.................,......,1990
--``-`-`,,`,,`,`,,`---
--``-`-`,,`,,`,`,,`---