Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
www.elsevier.com/locate/cryogenics
a
Cryogenics Section, Raja Ramanna Centre for Advanced Technology (RRCAT), Indore 452013, India
b
Institute of Engineering and Technology, DAVV, Indore (MP), India
Received 7 March 2006; received in revised form 14 March 2007; accepted 24 March 2007
Abstract
Coiled finned-tube heat exchangers have been used in small and medium helium refrigerators/liquefiers, miniature J–T refrigeration
systems for many years. The efficiency of these cryogenic systems strongly depends on the thermal and pressure drop performance
of these heat exchangers. A considerable improvement in the performance of heat exchanger is possible by choosing an appropriate
geometrical configuration for a given process requirement. In the present study, geometry of heat exchanger has been derived taking into
consideration the clearance provided for manufacturing of the heat exchangers and an optimized geometrical configurations have been
find out. The results show the possibility of adjusting the thermal and pressure drop performance by varying the clearance. The predic-
tions of four end temperatures from present design method have been compared with the actual experimental results of one of the
prototypes fabricated in our lab.
2007 Elsevier Ltd. All rights reserved.
0011-2275/$ - see front matter 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.cryogenics.2007.03.010
P.K. Gupta et al. / Cryogenics 47 (2007) 322–332 323
Nomenclature
shell diameter to keep the pressure drop with in stipulated coefficients for shell side and tube side flow. They used their
limits. This will result in lower flow velocity within heat designing method for the Clarendon laboratory helium liq-
exchangers and therefore lower heat transfer coefficient in uefier heat exchangers [6]. Croft and Cosier [7] also
the shell side. Hence, it will require larger unit and as a con- designed a new form of finned-tube heat exchanger by
sequence of this there is a need to optimize the geometry of applying the design method described by Croft and Tebby
heat exchanger to [5]. However, design proposed by Croft et al. does not con-
sider the effect of diametrical clearance on thermal and
• minimize the cool down time of the system, pressure drop performance.
• minimize the refrigeration loss to cool the unit, In the present work, the expressions have been derived
• minimize the radiation loss, and to taking in to account of clearance. These presented expres-
• reduce the cost of the system. sions can be used as design charts for thermal and pressure
drop design of finned-tube heat exchanger. The methodol-
So, fact is that the clearance provided for ease of manufac- ogy prescribed in the present work has been used to com-
turing can be used for adjusting the thermal and pressure pare four end temperatures of one heat exchanger tested
drop performance of finned-tube heat exchangers. To the in our lab. The main system parameters are given in Table
best of the knowledge of present authors, a little informa- 1. These parameters correspond to one of the units fabri-
tion has been published in open literature regarding the cated in our laboratory and may be suitable for medium
designing of coiled finned-tube heat exchangers. Geist size helium liquefiers/refrigerators. Using these charts, the
and Lashmet [4] presented the heat transfer factor and fric- optimization for thermal and pressure drop performance
tion factor for different fin geometries. Croft and Tebby [5] has been carried out for the operating parameters listed
presented the expressions for thermal design and they have in Table 1 as an example. The geometry of heat exchanger
suggested the correlations for calculation of heat transfer has been optimized for the selection of
324 P.K. Gupta et al. / Cryogenics 47 (2007) 322–332
Table 1
Geometry of the coiled finned-tube heat exchanger and its operating
parameters
Geometry of heat Inner tube diameter, di 8.2 mm
exchanger Finned-tube diameter, df 13.5 mm
No. of fins per meter, n 1024
Axial length, L 1000.0 mm
Mean diameter, De 145.0 mm
Fin type Integrated copper low height radial fins
Operating parameters Working fluid Helium
Temperature range 300–90 K
Mass flow rate 3.0 g/s
High pressure 15.0 bars
Low pressure 1.0 bar
The assembly of these heat exchangers consists of the 2. All thermo physical properties have been calculated at
inner shell, outer shell and finned tube. The finned tube is the mean temperature Tm of the individual fluid stream.
wound on the inner shell and then it is jacketed by the
outer shell. The dead space between two consecutive coils
has to be filled by some cord as shown in Fig. 1. A diamet- 5. Design and optimization
rical clearance between the shell and tube periphery has to
be provided in order to make the assembly easy. The ther- 5.1. Thermal design
mal and pressure drop performance of heat exchanger is
influenced by this clearance. A part of the cold stream For the usual heat exchanger design problem, where all
passes through this clearance without taking part in heat end temperatures are given, the heat transfer area can be
exchange process, making the heat exchanger ineffective. calculated by using the overall heat transfer coefficient
On the other hand, the pressure drop performance will be based on either hot fluid heat transfer area or cold fluid
improved by increasing the available cross-section area. heat transfer area. In the present study, we will follow
The effect of clearance on the thermal and pressure drop the design procedure described by Croft and Tebby [5].
performance of heat exchangers is considered in the present And therefore, the overall heat transfer coefficient,
study and optimum geometries have been found out for U (W/m K), will be based on axial length of the heat
given operating parameters as given in Table 1. exchanger instead of the heat transfer area of either fluid
If we assume the mass flow rate of the cold stream flow- by applying the concept of wetted perimeter of heat
ing in the shell side is given by m_ c then the actual mass flow exchanger per unit axial length.
rate passing through the fins is Fluid properties are evaluated at mean temperature
m_ c given by
m_ f ¼ ð8Þ
kþ1 T h;in þ T h;out T c;in þ T c;out
where k is bypass area factor and it is given by Tm ¼ or
2 2
Acc where Th, Tc are the respective inlets and outlets tempera-
k¼ ð9Þ
Afc tures of hot and cold streams of heat exchanger.
_ of the hot fluid which has to be
The total heat duty ðQÞ
The Reynolds number will be calculated based on the
actual flow passes through the fins and can be given by removed by exchanging the energy with the cold fluid is
expressed as follows:
ReWOC
Ref ¼ ð10Þ Q_ ¼ C h ðT h;in T h;out Þ ¼ ULDT LMTD ð14Þ
kþ1
where ReWOC is the Reynolds number based on the total where L is the axial length of heat exchanger and DTLMTD
cross-section area available for shell side flow when there is the log mean temperature difference, given by
is no clearance and can be calculated as follows:
DT hot end DT cold end
m_ c Dh DT LMTD ¼ ð15Þ
DT hot end
ReWOC ¼ ð11Þ ln DT
Afc l cold end
The characteristic dimension for the Reynolds number in where DThot end and DTcold end are the hot and cold end
Eq. (11) is the equivalent diameter, or the hydraulic dia- temperature differences
meter Dh DT hot end ¼ T h;in T c;out ð16Þ
4Asc
Dh ¼ ð12Þ and
As =L
m_ c DT cold end ¼ T h;out T c;in ð17Þ
G¼ ð13Þ
Asc The overall heat transfer coefficient U (W/m K) based on
where G is the mass flow rate per unit free-flow area and per unit axial length of heat exchanger can be given by
will be used for calculating the heat transfer coefficients. 1
1 1
In Eqs. (12) and (13) Asc will be equal to Afc, if there is U¼ þ ð18Þ
hi si gho so
not any clearance.
where hi and ho are the heat transfer coefficients defined as
4. Assumptions the rate of heat transfer across unit area of separating wall
for unit temperature difference between gas stream and sep-
Following assumptions are made for carrying out the arating wall (W/m2 K). si and so are the perimeters (surface
analysis. area per unit axial length) as defined in Eqs. (6) and (7)
across which heat is transferred. The thermal resistance
1. The pressure drops due to other effects are negligible in of separating wall is omitted in Eq. (18) as it is small en-
comparison to the core frictional pressure drop. ough as compared to fluid resistances. The fin efficiency g
326 P.K. Gupta et al. / Cryogenics 47 (2007) 322–332
can be assumed 100% for cupper finned tube in Eq. (18) di
f ¼ 0:184Re0:2 1 þ 3:5 ð25Þ
[5,9]. De
The value of heat transfer coefficients for the inner and
outer stream is calculated using the same correlations as For shell side flow:
described by Croft and Tebby [5] and are reproduced here The shell side flow is generally laminar in the coiled
finned tube heat exchanger and the friction factor for the
0:2 1:8
hi ¼ 0:033ch m_ 0:8
h lh d i ð19Þ shell side is given by [11]
and for 400 < Re < 104 f ¼ 1:904Re0:2 ð26Þ
0:2 1:0 0:2
ho ¼ 0:021cc m_ 0:8
f lc Asc s ð20Þ In the present analysis, the pressure drop for the tube side
Eq. (20) can directly be used for calculating the heat trans- and shell side is presented as dimensionless quantity and it
fer coefficient in the shell side as used by Croft and Tebby is defined as follows:
[5] for calculating the overall heat transfer coefficient, U DP
DP ¼ ð27Þ
(W/m K). However, it can also be deduced in terms of Rey- Pc
nolds number based on the actual flow passes through the
fins.
Eq. (20) can be written as 5.3. Optimization procedure
1:0 0:2
ho ¼ 0:021cc m_ f m_ 0:2
f l0:2
c Asc s ð21Þ In any helium refrigerator/liquefier, the multiple num-
The above Eq. (21) can be rearranged as follows: bers of heat exchangers are used. These heat exchangers
0:2 are used at different temperature levels. The typical work-
4m_ f ing temperature range of first heat exchanger is 300–90 K
ho ¼ 0:027cc Gf ð22Þ
lc so for medium size helium refrigerator/liquefier and the typi-
cal thermal size (NTU) of this heat exchanger is 21. The
Eq. (22) can be expressed in terms of Reynolds number as stringent requirements of these heat exchangers are higher
follows: effectiveness and low pressure drops in both of the streams.
ho ¼ 0:027cc Gf Ref0:2 ð23Þ These heat exchangers can be made in different geometrical
configurations such as by changing the mean diameter of
heat exchanger, by opting the different inner diameter of
5.2. Pressure drop design finned tube or by choosing the different fin geometries to
satisfy the fixed thermal size (NTU) requirement and low
The pressure drop design is equally important as the pressure drop design criteria. Furthermore, for the fixed
thermal design of heat exchanger for any cryogenic sys- values of NTU, one can obtain the heat exchanger config-
tems. The tube side pressure drop across the heat exchan- urations in different sizes and weight to fulfill the effective-
ger will reduce the amplitude of high pressure stream ness and pressure drops requirements but these heat
thereby reducing the area of the expansion space in PV dia- exchangers may not be suitable from system performance
gram and the gross refrigeration produced by the refriger- point of view due to higher thermal mass. The geometrical
ator/liquefier. configurations of these heat exchangers much depend on
On the other hand, pressure drop in the shell side is how one has selected the mean diameter of heat exchanger
extremely important for any cryogenic systems such as for the fixed fin geometries or for the fixed mean diameter,
helium liquefier/refrigerator. In a helium liquefier/refriger- how one has selected the finned tube geometries such as
ator (helium normal boiling point 4.2 K and critical pres- inner diameter of finned tube, fin height and number of
sure 2.2 bar), the total pressure drop of the shell side fins. Therefore, it is necessary to choose the right combina-
should not be more than 0.2 bar because of the constraint tions of different geometrical parameters for obtaining the
of critical pressure of the helium. optimum size of heat exchanger for the given NTU.
The amplitude of the pressure drop (DP), either shell In the present study, the first heat exchanger of
side or tube side, per unit working length through the heat NTU = 21 has been chosen as an example for the optimi-
exchanger is given by zation purpose while keeping the operating parameters,
fG2 as described in Table 1, fixed. The expressions derived in
DP ¼ ð24Þ the preceding sections have been used for the optimization
2qDh
purpose. The calculations are performed for different val-
In Eq. (24) the value of friction factor, f, has to be calcu- ues of bypass factors, mean diameters, inner tube dia-
lated as follows: meters, fin height and number of fins to obtain the
For the tube side: corresponding pressure drops in tube side and shell side
For turbulent flow inside a smooth tube of any cross- and surface area requirements for the fixed values of
section, the friction factor was calculated by the empirical NTU and mass flow rate. These calculations were carried
equation as suggested by Timmerhaus and Flynn [10] out to study the effect of clearance, for selecting the suitable
P.K. Gupta et al. / Cryogenics 47 (2007) 322–332 327
6.1. Comparison of experimental results with the present Fig. 2 shows the maximum allowable clearance that can
analysis be provided for the manufacturing of a heat exchanger for
different fin height and number of fins. The figure shows
Recently, Gupta et al. [8] presented detailed description that the maximum allowable clearance is either increased
of experimental set-up and the results obtained from one of with the fin height for given number of fins or increased
the prototypes designed and manufactured in our lab based with the lesser number of fins for given fin height. This
on the geometrical parameters given in Table 1. The pres- can be attributed as either the larger fin height or the lesser
ent analysis considers the effect of leakage of flow through number of the fins can be used to reduce the effect of leak-
clearance provided for the ease of manufacturing and it is age of flow through the clearance. For an example, the
one of the important causes to deteriorate the performance maximum allowable clearance required for 2 mm fin height
of coiled finned-tube heat exchangers. The predictions of for 1024 fins per meter (26 fins per inch) is 2.77 mm.
four end temperatures based on the present analysis con- Generally, 1.5–2 mm diametrical clearance is sufficient for
sidering the clearance effects, are compared with the actual the assembly of a medium size heat exchanger during fab-
results published earlier [8]. rication. Therefore, in this case the bypass area will be less
Table 2 gives the details of experimental parameters for as compared to the cross-sectional area offered by the fins
which results have been compared. The quantitative com- and the maximum flow will be diverted through the fins.
parison of four end temperatures of a heat exchanger with Hence, it will reduce the effect of leakage of flow on the
the experimental results has shown in Table 3. It is thermal performance of the heat exchanger. In another
observed from the experimental results that the present case, the maximum allowable clearance required for
analysis show good agreement with theoretical results. It 1.5 mm fin height for 1024 fins per meter (26 fins per inch)
can be noted from Table 3 that the heat exchanger effec-
tiveness can be increased from 91.7% to 95.8% by reducing
the clearance from 1.2 mm to 0.3 mm and establish the
importance of the effect of clearance for designing of such
Table 2
Details of experimental parameters
Working fluid Nitrogen
Hot end temperature 296.0 K
Cold end temperature 151.0 K
Hot stream pressure 15.0 bar
Cold stream pressure 1.0 bar
Mass flow rate 9.5 g/s
Number of transfer units, NTU 12.0
Overall heat transfer coefficient 142.2 W/m K
Ratio of the finned side to tube side area 4.4
Fig. 2. Maximum allowable clearance as a function of fin height and
Manufacturing clearance, c 1.2 mm
number of fins.
328 P.K. Gupta et al. / Cryogenics 47 (2007) 322–332
Fig. 3. Heat transfer coefficient correction factor as a function of bypass Fig. 5. Non-dimensional pressure drops and axial length as a function of
area factor. mass flow rate.
P.K. Gupta et al. / Cryogenics 47 (2007) 322–332 329
Fig. 8. Non-dimensional pressure drops and surface area as a function of Fig. 9. Non-dimensional pressure drops and surface area as a function of
inner tube diameter. fin height.
De = 145.0 mm. Fig. 8 shows that the tube side pressure side. Similarly, the pressure drop for tube side decreases
drop decreases drastically as the tube diameter increases, due to the fact that the total finned-tube length required
but shell side pressure drop and surface area required for for the given heat duty and tube diameter decreases and
the given heat duty increases. This is due to the fact that hence total pressure drop of the tube decreases. From
the inner heat transfer coefficient decreases as the tube Fig. 9, we can conclude that there is no advantage of
diameter increases and hence reduces the over all heat increasing the fin height beyond 1.8 mm as there is no mea-
transfer coefficient. This will be resulted in the increase of surable reduction in the pressure drops performance but if
heat exchanger size (axial length) and increase the shell side one selects the fin height more than 1.8 mm, the surface
pressure drop. Therefore, the best practice for any designer area requirement will be more for the same operating
is to first select the inner tube diameter of the finned tube parameters with out gaining in pressure drop performance.
according to the acceptable pressure drop in the tube to
keep the shell side pressure drop with in stipulated limits 6.9. Effect of number of fins on the pressure drop
with minimum surface area. For an example, if we select performance and sizing
the inner tube diameter di = 8 mm for which the tube side
pressure drop is 0.5 bar and shell side pressure drop is Figs. 10–12 show the non-dimensional shell side, tube
0.05 bar approximately for the operating parameter side pressure drop and surface area requirement as a func-
described in Table 1. It could be noted here that 0.5 bar tion of number of fins for the mean diameter of 120.0 mm,
tube side pressure drop in the medium capacity helium liq- 145.0 mm and 175.0 mm. These figures show that the tube
uefier/refrigerator can be tolerated but on the other hand if side pressure drop decreases with the number of fins while
emphasis is given to reduce the tube side pressure drop fur- shell side pressure drop increases with the number of fins.
ther by increasing the tube diameter, the surface area This can be explained as the more tube length is required
requirement will increase accordingly and the system for the given heat duty for lesser number of fins which will
becomes unnecessarily bulky. Therefore, the selection of increase the total tube side pressure drop for given finned-
the larger diameter of inner finned tube has to be avoided. tube diameter. On the other hand the shell side pressure
7. Conclusions
Acknowledgements
The design and optimization of coiled finned-tube heat
exchangers with the consideration of clearance provided The authors are grateful to Dr. Parthasarathi Ghosh,
for the ease of manufacturing have been presented. The Mr. Rupul Ghosh and Mr. R.C. Sharma for their useful
thermal and pressure drop performance of a coiled discussions, suggestions and fabrication of heat exchanger
finned-tube heat exchanger depend on the clearance during carrying out this work.
332 P.K. Gupta et al. / Cryogenics 47 (2007) 322–332
References [7] Croft AJ, Cosier J. A new form of finned-tube heat exchanger.
Cryogenics 1970(June):239–40.
[1] Collins SC. A helium cryostat. Rev Sci Instr 1947;18:157–67. [8] Gupta Prabhat, Sharma RC, Kush PK. Development of cross-
[2] Barron RF. Cryogenics systems. 2nd ed. Oxford: Oxford University counter flow heat exchanger for cryogenic temperature range. In:
Press; 1985. Proceedings of national seminar and conference on cryogenics and its
[3] Atrey MD. Thermodynamic analysis of Collins helium liquefaction frontier application. Bengal Engineering College; Howrah, India,
cycle. Cryogenics 1998;38:1199–206. 2004. p. 125–7.
[4] Geist JM, Lashmet PK. Miniature Joule–Thomson refrigeration [9] Hewitt GF. Heat exchanger design hand book. New York: Begell
systems. Adv Cryo Eng 1960;5:324–31. House Inc; 1998.
[5] Croft AJ, Tebby PB. The design of finned-tube cryogenic heat [10] Timmerhaus KD, Flynn TM. Cryogenic process engineering. New
exchangers. Cryogenics 1970(June):236–8. York, USA: Plenum Press; 1989.
[6] Croft AJ, Bligh BR. The Clarendon laboratory helium liquefier. [11] Trumpler PR, Dodge BF. Heat transfer and friction factors for edge-
Cryogenics 1961(June):231–4. wound coil fins. AIChE 1947;43:75–84.