Chapter 1 Introduction

1.1 Introduction of Heat Exchanger

Heat exchangers have several industrial and engineering applications. The design procedure of heat exchangers is quite complicated, as it needs exact analysis of heat transfer rate and pressure drop estimations apart from issues such as long term performance and the economic aspect of the equipment .Whenever inserts are used for the heat transfer enhancement, along with the increase in the heat transfer rate, the pressure drop also increases. This increase in pressure drop increases the pumping cost. Therefore any augmentation device should optimize between the benefits due to the increased heat transfer coefficient and the higher cost involved because of the increased frictional losses. The present paper includes various heat transfer augmentation techniques. Experimental work on heat transfer augmentation using twisted tape and different kind of inserts are carried out. Inserts when placed in the path of the flow of the liquid, create a high degree of turbulence resulting in an increase in the heat transfer rate and the pressure drop. The work 1

includes the determination of friction factor and heat transfer coefficient for various twisted tapes and twisted angles having different twist ratios. Heat exchangers have several industrial and engineering applications. The design procedure of heat exchangers is quite complicated, as it needs exact analysis of heat transfer rate and pressure drop estimations apart from issues such as long term performance and the economic aspect of thee equipment. The major challenge in designing a heat exchanger is to make the equipment compact and achieve a high heat transfer rate using minimum pumping power. Techniques for heat transfer augmentation are relevant to several engineering applications. In recent year the high cost of energy and material has resulted in an increased effort aimed at producing more efficient heat exchange equipment. Furthermore, sometimes there is a need for miniaturization of a heat exchanger in specific applications, such as space application, through an augmentation of heat transfer. For example, a heat exchanger for an ocean thermal energy conversion (OTEC) plant requires a heat transfer surface area of the order of 10000m2/MW Therefore, an increase in the efficiency of the heat exchanger through an augmentation technique may result in a considerable saving in the material cost. Furthermore as a heat exchanger becomes older, the resistance to heat transfer increases owing to fouling or scaling. These problems are more common for heat exchangers used in marine applications and in chemical industries. In some specific applications, such as heat exchangers dealing with fluids of low thermal conductivity (gases and oils) and desalination plants, there is a need to increase the heat transfer rate. The heat transfer rate can be improved by introducing a disturbance in the fluid flow (breaking the Viscous and thermal boundary layers), but in the process pumping power may increase significantly and ultimately the pumping cost becomes high. Therefore, to achieve a desired heat transfer rate in an existing heat exchanger at an economic pumping power, several techniques have been proposed in recent years and are discussed in the following sections [3].

1.2 Heat Transfer Enhancement Technique for Viscous Fluid

Heat transfer effects of variable viscosity and viscous dissipation for heated developing laminar flows in circular tubes have been investigated. Three studies are reported covering a comprehensive range of input data for the case of constant wall heat flux. Initially the program 2

was used to predict the effect on heat transfer of temperature-dependent viscosity via a general temperature power relation. In addition, predictions were made for nine particular fluids covering a range of Prandtl numbers from 0.025 to 12 500, and a range of Brinkman numbers from 1 .8x 10^-10 to 6.8x 10^3. A more detailed study was made for two particular oils covering a range of practical interest. For the, liquids considered their viscosity temperature-dependence resulted in enhancement of heat transfer, whereas for fluids with a Prandtl number <200 the effect of viscous dissipation was negligible, and for fluids of a Brinkman number >2x 10 -2 the outcome was a reduction of heat transfer. A numerical instability problem occurred for situations of very high viscous dissipation which limited the length of duct that could be examined [4].

1.3 Augmentation of Heat Transfer

AUGMENTATION of heat transfer and improvement of power efficiency are the principle goals in the development of heat exchangers. The enhancement of convective heat transfer in single-phase flow of different liquids and low-temperature gases is very important. At present more and more attention is being paid to this problem [l-5]. Here we consider mainly enhancement of forced convection heat transfer in viscous fluid flows. The amount of heat transfer can be increased by various ways; the first one is a proper selection of working fluids, since Nu - Pr. For example, in laminar flow Nu - PY.~ and in turbulent flow over a plate NM - PY. So the heat transfer in liquids flows is several times higher than in air flow. The heat transfer rate is mainly governed by the flow velocity and the flow pattern. Normally a higher flow velocity implies a higher heat transfer coefficient. At the same time, high coolant velocities are obtained at the penalty of increasing pumping power. Depending on velocity and Reynolds number there are laminar or turbulent flows. A turbulent flow means a higher effective viscosity with a considerably higher momentum flux in comparison with laminar flow. This difference is reflected in the heat transfer rate. In a laminar boundary layer over a plate the heat transfer coefficient t( - u- and in a turbulent boundary layer LY- u.~ where u-flow velocity. So from the point of augmentation the turbulent flow is more advantageous [10].

Chapter 2 Classification of Augmentation Techniques

They are broadly classified into three different categories 2.1. Passive Techniques 2.2 Active Techniques 2.3 Compound Techniques.

2.1 Passive Techniques

These techniques do not require any direct input of external power; rather they use it from the system itself which ultimately leads to an increase in fluid pressure drop. They generally use surface or geometrical modifications to the flow channel by incorporating inserts or additional devices. They promote higher heat transfer coefficients by disturbing or altering the existing flow behavior except for extended surfaces. Heat transfer augmentation by these techniques can be achieved by using; (i) Treated Surfaces: Such surfaces have a fine scale alteration to their finish or coating which may be continuous or discontinuous. They are primarily used for boiling and condensing duties. 4

(ii) Rough surfaces: These are the surface modifications that promote turbulence in the flow field in the wall region, primarily in single phase flows, without increase in heat transfer surface area. (iii) Extended surfaces: They provide effective heat transfer enlargement. The newer developments have led to modified finned surfaces that also tend to improve the heat transfer coefficients by disturbing the flow field in addition to increasing the surface area. (iv) Displaced enhancement devices: These are the inserts that are used primarily in confined forced convection, and they improve energy transport indirectly at the heat exchange surface by displacing the fluid from the heated or cooled surface of the duct with bulk fluid from the core flow. (v) Swirl flow devices: They produce and superimpose swirl flow or secondary recirculation on the axial flow in a channel. These include helical strip or cored screw type tube inserts, twisted tapes. They can be used for single phase and two-phase flows. (vi) Coiled tubes: These lead to relatively more compact heat exchangers. It produces secondary flows and vortices which promote higher heat transfer coefficients in single phase flows as well as in most regions of boiling. (vii) Surface tension devices: These consist of wicking or grooved surfaces, which direct and improve the flow of liquid to boiling surfaces and from condensing surfaces. (viii) Additives for liquids: These include the addition of solid particles, soluble trace additives and gas bubbles in single phase flows and trace additives which usually depress the surface tension of the liquid for boiling systems. (ix) Additives for gases: These include liquid droplets or solid particles, which are introduced in single-phase gas flows either as dilute phase (gas-solid suspensions) or as dense phase (fluidized beds).

2.2 Active Techniques

In these cases, external power is used to facilitate the desired flow modification and the concomitant improvement in the rate of heat transfer. Augmentation of heat transfer by this method can be achieved by 5

(i) Mechanical Aids: Such instruments stir the fluid by mechanical means or by rotating the surface. These include rotating tube heat exchangers and scrapped surface heat and mass exchangers. (ii) Surface vibration: They have been applied in single phase flows to obtain higher heat transfer coefficients. (iii) Fluid vibration: These are primarily used in single phase flows and are considered to be perhaps the most practical type of vibration enhancement technique. (iv) Electrostatic fields: It can be in the form of electric or magnetic fields or a combination of the two from dc or ac sources, which can be applied in heat exchange systems involving dielectric fluids. Depending on the application, it can also produce greater bulk mixing and induce forced convection or electromagnetic pumping to enhance heat transfer. (v) Injection: Such a technique is used in single phase flow and pertains to the method of injecting the same or a different fluid into the main bulk fluid either through a porous heat transfer interface or upstream of the heat transfer section. (vi) Suction: It involves either vapor removal through a porous heated surface in nucleate or film boiling, or fluid withdrawal through a porous heated surface in single-phase flow. (vii) Jet impingement: It involves the direction of heating or cooling fluid perpendicularly or obliquely to the heat transfer surface.

2.3 Compound Techniques

When any two or more of these techniques are employed simultaneously to obtain enhancement in heat transfer that is greater than that produced by either of them when used individually, is termed as compound enhancement. This technique involves complex design and hence has limited applications [3].

Chapter 3

Swirl Flow Devices

Swirl flow devices generally consist of a variety of tube inserts, geometrically varied flow arrangements and duct geometry modifications that produce flows. These Techniques include twisted tape inserts, periodic tangential fluid injection and helically Twisted tubes as shown in fig.1.1.

Fig 1.1 Example of (a) full-length twisted tape, (b) regularly spaced twisted tape and (c) smoothly varying (gradually decreasing) pitch full-length twisted tape

3.1 Single-Phase flows

Twisted tape inserts are the most widely used swirl flow device for single-phase flows. These inserts increase the heat transfer coefficient significantly with a relatively small pressure drop penalty as reported by Smithberg and Landis (1964); Lopina and Bergles (1969); Date and Singham (1972); Manglik and Bergles (1992); Manglik and Yera (2002). Twisted tapes can be used in the existing shell and tube heat exchangers to upgrade their heat duties or when employed in a new exchanger for a specified heat duty, significant reduction in size can be achieved. The ease of fitting multiple bundles with tape inserts and their removal makes them useful in fouling situations, where frequent tube-side cleaning may be required. When swirl flow devices are placed inside a circular tube, the flow field gets altered in several ways like an increase in axial velocity and wetted perimeter due to the blockage and partitioning of the flow cross-section, longer effective flow length in the helically twisting partitioned duct and tapes helical curvature induces secondary fluid circulation or swirl. Swirl generation is the most dominant mechanism which effects transverse fluid transport across the tape partitioned duct, thereby promoting greater fluid mixing and higher heat transfer coefficients as shown in fig.1.2.

Fig. 1.2: Twisted tape inserts with different spacer lengths and twist ratios P.Sivashanmugam and S.Suresh (2006) conducted experimental investigation of heat transfer and friction factor characteristics of circular tube fitted with full length helical screw element of different twist ratio (1.95, 2.93, 3.91, 4.89), and helical screw inserts with spacer length 100, 200, 300 and 400 mm as shown above with uniform heat flux under turbulent flow condition. The friction factor for helical twist insert with spacer length 100 mm was found to be very close to the value of that of full length helical twist for all Reynolds number and decreases by 5% for each 100 mm increment space length indicating that there is no much reduction in pumping power. The increase in Nusselt number from twist ratio 4.89 to 1.95 is nearly 30 to 40% for all 9

Reynolds number for full length helical twist where as the decrease in friction factor is about 40 to 45% for various spacer lengths as shown in fig.1.2. They developed empirical equations for Nusselt number and friction factor: Nu = 0.258 (Re) 0.254 (Pr) (Y)-0.242(1+S/Dh)-0.042 f = (Re)-0.384(Y)-0.852(1+S/Dh)-0.047 Various graphical representations obtained by them are as follows:

Fig. 1.3: Variation of Nu w.r.t. Re for twisted tapes with different spacer lengths.

3.2 Boiling
The heat transfer enhancement due to the tape inserts is reflected in the reduced wall temperature along the tube length in a single phase liquid, sub-cooled boiling, bulk boiling and dispersed film boiling. The primary enhancement mechanism is the tape induced swirl, which tend to increase vapor removal and wetting of the heated surface.

3.3 Condensing
Steam condensation in tubes fitted with twisted tape inserts enhanced the heat transfer coefficients by 30% over the empty tube values.

10

3.4 Active Techniques

Rotating surfaces substantially enhance heat transfer coefficient up to 350% for laminar flows in straight tubes rotating around their own axis or a parallel axis (Mori and Nakayama, 1967; McElhiney and Preckshot, 1977; Bidyanidhi et al., 1977).Tang and McDonald, 1971 reported that, with high speed rotation of heated cylinders in saturated\ pools, the convective coefficients are so high that boiling can be suppressed. Depending on oscillation amplitude to tube diameter ratios and vibration Reynolds number, the heat transfer coefficients increase up to 20 times compared with those of stationary tubes (Bergles, 1998). In case of fluid vibrations, improvements of 100 to 200% over natural convection heat transfer coefficients in air were obtained by Sprott et al.,1960;Fand and Kaye,1961;and Lee and Richardson,1965;by generating intense sound fields and directing them transversely to a horizontal heated cylinder. Robinson et al.,(1958);Zhukauskas et al.,1961;Larson and London,1962;Fand ,1965;and Li and

Parker,1967;reported 30 to 45% increase in free convection heat transfer by means of sonic and ultrasonic vibrations. As per Wong and Chon, 1969, ultrasonic vibrations do not promote any improvements in nucleate pool boiling; but they enhance vapor removal and tend to increase critical heat flux (CHF) by 50% as reported by Ornatskii and Shcherbakov, 1959. Mathewson and Smith (1963) investigated the effects of up to 176 dB acoustic field with frequencies in the range 50 to 330 Hz and found laminar film condensation coefficients for isopropanol to be enhanced by about 60% at low vapor flow rates [3].

11

Chapter 4

Terminology of Enhancement Technique

4.1 Important Definitions

In this section a few important terms commonly used in heat transfer augmentation work are defined.

4.1.1 Thermo hydraulic performance

For a particular Reynolds number, the thermo hydraulic performance of an insert is said to be good if the heat transfer coefficient increases significantly with a minimum increase in friction factor. Thermo hydraulic performance estimation is generally used to compare the performance of different inserts such as twisted tape, wire coil, etc., under a particular fluid flow condition.

12

Table.2.1 Notation of symbols

4.1.2 Overall enhancement ratio The overall enhancement ratio is defined as the ratio of the heat transfer enhancement ratio to the friction factor ratio. This parameter is also used to compare different passive techniques and enables a comparison of two different methods for the same pressure drop. The overall enhancement ratio is defined as (Nu=Nu0) (f =f0)1=3 where Nu, f, Nu0 and f0 are the Nusselt numbers and friction factors for a duct configuration with and without inserts respectively. The friction factor is a measure of head loss or pumping power.

13

4.1.3 Nusselt number The Nusselt number is a measure of the convective heat transfer occurring at the surface and is defined as hd/k, where h is the convective heat transfer coefficient, d is the diameter of the tube and k is the thermal conductivity.

4.1.4 Prandtl number The Prandtl number is defined as the ratio of the molecular diffusivity of momentum to the molecular diffusivity of heat, n=a.

4.1.5 Pitch Pitch is defined as the distance between two points that are on the same plane, measured parallel to the axis of a twisted tape.

4.1. Twist ratio, y The twist ratio is defined as the ratio of pitch to inside diameter of the tube y H/di, where H is the twist pitch length and d is the inside diameter of the tube [11].

14

15

16

Chapter 5

Laminar Flow And Heat Transfer In Plain Circular Tubes

The most commonly used methods for predicting fully developed laminar flow heat transfer coefficients inside smooth, round tubes can be derived from first principles, which can be found in most textbooks and hence this theory is not reproduced here. For a uniform heat flux wall boundary condition (H) for fully developed laminar flow (both thermally and

hydrodynamically), the Nusselt number is given by:

The Nusselt number NuH is based on the tube diameter di and apt is the convective heat transfer coefficient for a plain tube. As can be seen, the heat transfer coefficient is not a function of the Reynolds number or the Prandtl number. Similarly, for a uniform wall temperature wall boundary condition (T) for fully developed laminar flow, the Nusselt number is given by:

17

Again, the heat transfer coefficient is not a function of the Reynolds or Prandtl numbers. The tubular Reynolds number Re is defined as:

& 2s and is obtained by dividing the mass flow rate in kg/s by the cross- sectional area of the tube in m2.The mass velocity m is in kg/m The Prandtl number Pr is obtained from the physical properties of the fluid and is defined as:

The hydrodynamic entrance length zeh to arrive at fully developed hydrodynamic flow can be estimated by the following expression: The value of 0.03 is used to arrive within 5% of the fully developed laminar friction factor while a value of 0.05 is used to arrive within 1.4% of that value, as is cited in Lienhard and Lienhard (2006). For the thermal developing length when the velocity profile is already fully developed before the heat transfer zone begins, the thermal entrance length zet starting from that location is given by the expression:

Thus, highly viscous flows, such as oils, practically never become fully developed. The value of 0.034 is used to arrive within 5% of the fully developed laminar heat transfer for a constant wall temperature condition while a value of 0.043 is used for a constant heat flux condition. When the velocity and temperature profiles develop simultaneously, the value tends to range between 0.028 and 0.053 depending on the wall condition and the Prandtl number. All of these values are those recommended in Lienhard and Lienhard (2006). Shah and London (1978) give the analytical solutions to many laminar flow situations for hydrodynamically developing flows, thermally developing flows, simultaneously developing flows, etc. The most comprehensive work in the literature is for circular ducts for thermally developing flow. For practical design, Shah and London (1978) recommend the use of the following expressions for the local value of NuT,z for a constant wall temperature boundary (T):

18

The term z* is a form of the Graetz number based on the length from the entrance of the

channel z, given as: This expression includes the Graetz number, which is defined as:

For a uniform wall heat flux boundary condition (H), Shah and London (1978) recommend using the following expressions for the local Nusselt number NuH,z:

For simultaneously thermally and hydrodynamically developing flow, Shah and London (1978) recommend the use of the following equation for a constant wall temperature boundary condition (T) for the local Nusselt number NuT,z as shown in fig.1.4.

Similarly for a uniform heat flux boundary condition (H) for simultaneously developing flow, they recommended:

19

Fig.1.4 shows the heat transfer simulations for single phase flow inside a circular tube for the different boundary conditions and flow developments. The Nusselt numbers in this graph are plotted versus the dimensionless length z* used above, where z* written in terms of the Reynolds and Prandtl numbers is.

This diagram illustrates several important aspects of thermal entry regions. First of all, as the distance from the channel entrance is shortened (while keeping the Reynolds number, Prandtl number and tube diameter constant), the influence of the thermal boundary layer becomes more significant. Secondly, this is also true if the Reynolds number, Prandtl number or channel diameter is increased. The Hausen (1943) equation is also often cited for the uniform wall temperature boundary condition (T) in a developing laminar flow: In this expression, z is the length from the entrance of the tube and it gives the mean Nusselt number up to length z, NuT. The older and simpler expression of Seider and Tate (1936) is also often quoted, for a combined entry length for a uniform wall temperature condition (T), that which gives the mean Nusselt number as

20

This expression can be used for both heating and cooling conditions and includes the fluid property effects through the bulk to wall viscosity ratio. It is applicable when 0.48 < Pr < 16700 and 0.0044 < ( bulk/wall)0.14 < 9.75. Bulk physical properties in the above expressions should be taken at the mean fluid temperature between inlet and outlet. Gnielinski (1995) proposed a complete set of equations for the mean Nusselt number for laminar, transition and turbulent flows based on a large experimental database. In the laminar regime (Re < 2300), the mean value of the Nusselt number up to length z was correlated with the Graetz number for the uniform temperature and uniform heat flux boundary conditions. For the uniform wall temperature condition (T), using the definition of Gz above, Gnielinski (1995) proposed the following asymptotic correlation for the mean Nusselt number NuT:

The individual values of NuT,1, NuT,2 and NuT,3 are obtained as follows:

For the uniform heat flux boundary condition (H), Gnielinski (1995) proposed two expressions for the mean Nusselt number NuH, where the larger of the two values of NuH from these expressions is used:

The values of NuH,1 and NuH,2 are obtained as follows:

In these expressions at very low Reynolds numbers, each tends to the Nusselt number corresponding to the respective theoretical value for fully developed flow. For fluid property 21

effects for the other methods, using the recommendations of Lienhard and Lienhard (2006), it is most common to correct these expressions by multiplying the heat transfer coefficient by the
ratio of (bulk/ wall)0.11 for heating liquids (Twall > Tbulk) but ( bulk/wall)0.25 for cooling liquids (Twall < Tbulk).

A single exponent of 0.14 is also found in many textbooks. Instead, for heating of gases
(Tbulk/Twall)0.47 is used while no correction is used when cooling of a gas (Twall < Tbulk). These terms

the ratio
correct

the heat transfer coefficient for the effect of the variation in viscosity between the bulk

fluid temperature and that at the tube wall. The pressure drop in a tube of length L is given by the expression:

For laminar fully developed flow, the following expression can be derived from first principles for the friction factor, adding then the viscosity ratio effect added:

In the study by Marner and Bergles (1985), they recommended a value of m = -0.12 for cooling of a fluid and m = -0.25 when heating a fluid for: circular tubes, internally finned tubes and tubes with twisted tape inserts. For heat transfer in a plain tube, an internally finned tube and a tube with a twisted tape insert,
heating and cooling of liquids, they found the term ( bulk/wall)0.14 worked well for their data for both

finding however significantly different correlating expressions for the

Nusselt number for heating and cooling that are described later in this chapter. These above expressions for heat transfer and pressure drop, however, do not take into account any effects of natural convection on the flow, which can increase these values by two to three times. The above expression can also be written for isothermal flows in terms of the Poiseuille number Po as:

This is valid for a circular channel but varies for different channel cross-sectional geometries. For example, the value for a rectangular channel varies from 14.2 to 24, depending on the aspect ratio of the channel, while for an equilateral triangle the value is 13.33 and for an ellipse the value varies from 16 to 19.7, depending on the aspect ratio.

22

For a developing hydrodynamic boundary layer, the friction factors are higher. From a physical viewpoint, because of the reduction in the fluid velocity near the wall in the developing boundary layer, continuity requires that the fluid in the centre core must accelerate, which in turn thins the boundary layer. Once the opposing boundary layers meet at some distance down the length of the channel, the flow is fully developed. In the entrance region, however, the thinning of the boundary layer and the acceleration of the fluid in the core region give rise to an increase in shear. For a more detailed description, see for example White (1991).

5.1 Transition of Flow and Entrance Shape Effects on Heat Transfer in Plain Tubes
Before discussing heat transfer enhancement of laminar, single-phase flows inside tubes, the effects of the tube inlet configuration on heat transfer and flow regime transition thresholds need to be addressed. For a summary of this topic, the reader is referred to Section 5.2 elsewhere in this book. In generally, laminar flow is normally assumed to occur at Reynolds numbers below 2300 (some also cite 2100); however, this threshold value is a function of the inlet shape among other things and methods for its calculation are given in Section 5.2. This topic has also been addressed by Gnielinski (1995) with propositions for the flow transition regimes and heat transfer methods.

5.2 Laminar Heat Transfer with Wire Mesh Inserts

Wire mesh inserts developed by Cal Gavin Ltd. have been successfully applied to laminar flows now for several decades in a wide range of applications and are an excellent option for enhancing heat transfer in laminar flows. These inserts tend to yield much higher heat transfer and pressure drop augmentation ratios than twisted tape inserts. However, these are not described here since the design methods are not in the public domain; the reader is referred to Cal Gavin Ltd. for a description of the inserts and thermal design information.

23

5.3 Laminar Heat Transfer in Internally Finned Tubes

Internally finned tubes for application to laminar flows have not been widely investigated in the literature. Marner and Bergles (1985) surveyed the literature at that time, finding only a handful of experimental studies. For example, Watkinson, Milleti and Kubanek (1975) and Soliman and Feingold (1977) carried out tests on specific fin geometries. Choudhury and Patankar (1985) and MacArthur and Patankar (1985) numerically investigated laminar flow and heat transfer inside internally finned tubes and spiral annular passages, respectively. In the first study, they noted significant buoyancy (natural convection) effects in a developing laminar flow, primarily on heat transfer but less so on pressure drop. In the second study, with fins in the annulus of a double-pipe heat exchanger, they analyzed the effect of the twist (or spiral) of the fins on pressure drop and heat transfer relative to straight fins, finding that for some conditions the Poiseuille number (Po = f Re) in spiral fins was doubled while the heat transfer was up to five times higher, both created by the secondary flow effects. This work points to the high heat transfer augmentation effects that may potentially be created by internal fins, either inside tubes or in annuli. In perhaps the detailed experimental study, Marner and Bergles (1985) tested one internally finned tube with a highly viscous fluid (Polybutene 20). The tests were done under constant wall temperature conditions. Their 25.1 mm internal diameter, finned tube had 16 spiral fins of 0.66 mm thickness and 2.08 mm height with a spiral fin pitch of 88.9 mm. The isothermal friction factors for this tube were 1.7 times those of a comparable diameter plain tube. For heat transfer, the finned tube provided values 3 to 4 times higher than a plain tube when heating the fluid while only a marginal improvement when cooling. During heating, they cited the increased heat transfer surface area of the fins, a beneficial viscosity ratio effect and a swirl effect as being the primary reasons for the higher performance. For cooling this viscous fluid, they instead noted that this fluid apparently tended to freeze in the region between the fins, negating the influence of the fins, so that the principal effect on enhancement was only the effective reduction in the tube diameter. Correlations were proposed for the friction factor and heat transfer for this tube, plus viscosity ratio effects of the wall temperature, but these correlations are specific to this one tube and cannot be generalized for use with other internally finned tubes.

24

5.4 Laminar Heat Transfer in Spirally Fluted Tubes

These spirally fluted tubes (and corrugated tubes) have not seen a significant research effort in the literature. Bergles (1981) presented a detailed survey of heat transfer characteristics of deep spirally fluted tubes. Another review is available in Garimella and Christensen (1992). Spirally fluted tubes for laminar flows were also addressed in the review by Bergles, Webb and Junkhan (1979). With respect to experimental studies, according to the aforementioned reviews by Bergles, for laminar flows inside a Turbotec spirally fluted tube, heat transfer was increased by up to two times compared to a plain tube when heating the test fluid (Alta-Vis-530) but no improvement was found when cooling the same fluid. Rozalowski and Gater (1975) made tests on a deeply grooved flexible hose in laminar flow with two highly viscous fluids (Zerolene SAE-50 and Alta-Vis-530), finding up to three times heat transfer augmentation during heating, no augmentation when cooling Alta-Vis-530 but up to 100% higher performance when cooling the Zerolene. Even earlier than that, Kalinin and Yarkho (1971) studied deeply deformed tubes with transverse ridges for laminar flow with water, which from limited data showed that three of the seven configurations actually reduced heat transfer rather than augmented it. With respect to pressure drop in laminar flows, several experimental studies with deep spirally fluted tubes are available. Rozalowski and Gater (1975) found pressure drops to be about four times those of the equivalent diameter plain tube. Kliebe (1978) found isothermal pressure drops from 2 to 4 times higher for the Turbotec tube with Alta-Vis-530. For a large selection of spirally fluted tubes used as the inside tube of an annulus with water as the test fluid, Garimella and Christensen (1992) found pressure drop multipliers to range from 1.1 to 2.0 and also noted that the transition to turbulent flow occurred at a much lower Reynolds number in the range from 310 to 1000. Based on the wide variety of spirally fluted geometries and shapes, it is not possible apparently to provide a general method for predicting their heat transfer and pressure drop characteristics. Methods have been proposed in some of the above-mentioned experimental papers, but each method is limited to the particular type of flute shape and usually only one tube sample. Thus, no prediction methods are presented here for spirally fluted tubes [1].

25

5.5 Disturbance of the Boundary Layer by Roughness Elements

Nunner [6] maintained that surface roughness generated vortices mainly on the boundary of the turbulent core, which had no effect on the viscous sub layer and this method for augmentation of high Pr numbers fluids is not effective. Our analysis shows that in air the main contribution to thermal resistance is in the turbulent boundary layer, but in liquids with Pr >> I, it is mainly in the viscous sub layer. Therefore, heat transfer should be enhanced by the action of surface roughness in the wall layers. We encounter a new problem here, that of choosing the suitable height, shape and location of the surface roughness elements for each fluid in relation to its type and physical properties. Thus, in order to reach the necessary heat transfer enhancement in gases, the whole wall layer must be disturbed. But in liquids with high Pr only the viscous sub layer should be disturbed. The higher the Pr number, the shorter are the surface elements needed for a significant heat transfer enhancement. One has to bear in mind that in liquids it is a transition region of roughness that gives most intensive heat transfer rates. In air, with a growth of height k of roughness elements, a continuous increase of the heat transfer is observed. In transformer oil, according to the measurements by Driiius et ui. [9], a maximum increase of the heat transfer coefficient, as described by the Stanton number, is observed at a roughness k = 0.22 mm. A further increase of roughness up to k = 1.4 mm results in a decrease of St (Fig. 2). In liquids it is a transition region of roughness that gives most intensive heat transfer rates. The dimensionless height k+ is introduced to describe the roughness regime : (11)

Where k is height of roughness. Measurements suggest that k+ < 5 implies surface elements submerged in the viscous sub layer and having no influence on the heat transfer rate. At k+ > 70 a full roughness effect is observed. The range of k+ from 5 to 70 covers transitional region of its effect. Heat transfer curves obtained with rough-surface plates suggest different influence of k+ on heat transfer with a variation of Pr. In air a continuous increase of the heat transfer is observed with a growth of k+ (Fig. 3). In transformer oil a maximum increase of the heat transfer coefficient, as described by the Stanton number, is observed at a transition region roughness 26

effect at k < 70. A further increase of k results in a decrease of St. In water at Pr = 5.4 a maximum increase of the heat transfer coefficient, as described by the Stanton number, is also observed at a transition region of roughness at k+ < 70. A further increase of k+ results in a decrease of St [9]. Therefore, measurements confirm the suggestion that in viscous liquid flows the heat transfer

Fig.1.5 Heat transfer from a rough surface at different Pr. fi**-thickness of the momentum loss. Augmentation by roughness elements is also possible, but in this case low roughness elements should be used. The process of heat transfer enhancement inside rough-surface pipes, especially in the entrance region, is similar to the above-mentioned case of rough surface plates. General tendencies of the heat transfer augmentation techniques by rough surfaces are surveyed in Fig. 4. Here for J/f0 up to 3 or 4 the heat transfer augmentation due to the surface roughness is nearly proportional to the increase of hydraulic drag [lo]. Transverse ribs are more effective then sandtype roughness. This supports the speculation over regular Separation. The experiments suggest that on specific surfaces the heat transfer augmentation can even exceed the increase of hydraulic drag as shown in fig.1.5.

27

5.6 Free Stream Turbulization

Fig.1.6..Heat transfer in a flat channel

Fig.1.7 The effect of free stream turbulence

Artificial flow turbulization is one of the effective means of heat transfer enhancement. Main flow turbulization has a significant effect on heat transfer in a laminar boundary layer. It has been established that an increased flow turbulence exerts a strong effect on the hydrodynamics and heat transfer even in the case of a developed turbulent flow in the boundary layer on the plate. An analysis shows that this influence is exerted through the disturbance of the outer zone of the boundary layer while the wall region remains unaffected. But, as mentioned, with an increase of Pr the viscous sublayer shows main resistance to heat Transfer. So increased flow turbulence should exert a strong effect on heat transfer in the case of air flow, but there should be less effect for viscous liquids, since the wall region of the turbulent boundary layer is very conservative with respect to external effects. This was supported by experiments on the plate by Sian- Eiauskas and PediSius [I I]. They show (Fig.1.6) that increased flow turbulence exerts a strong effect on the heat transfer (.St) in the case of a developed air flow in the turbulent boundary layer. But as we see from Fig.1.6 this effect in water flows is less. In Fig. 5 St, is the Stanton number for non-turbulent flow. Analysis of these data shows that free stream turbulization influence is exerted through the disturbance of the outer zone of turbulent boundary layer while the wall region remains unaffected as shown in fig1.6 and fig.1.7.

28

5.7 Enhancement of Heat Transfer From Tubes In Cross-Flow

Fig.1.8 Local heat transfer from a rough-surface cylinder at Re = Ix IO in water flow.

Tubes constitute important elements of heat transfer surfaces. Therefore great attention is paid to the enhancement of heat transfer in the case of flow over tubes. At low Reynolds numbers the heat transfer at the front part of the tube is maximum and it decreases downstream with the growth of the laminar boundary layer thickness on the tube. In the region downstream of the separation of the laminar boundary layer at 84 the heat transfer gradually increases (curve at Tu = 1% in Fig. 6). A high free stream turbulence aids the formation of the turbulent boundary layer on the tube surface, thus enhancing the heat transfer. As shown in Fig. 6. At Rc = 6.8 x 104, in the flow of water, an increase of turbulence intensity from I to 6.8% is not only accompanied by an increase of the heat transfer at the front part of the cylinder, but the laminar boundary layer, instead of separating at C/I = 90, becomes turbulent and separates at 4 = 140; this also adds to the heat transfer augmentation [12]. Our measurements [12] show that with the increase of free stream turbulence from 1.2% to 15%, the average heat transfer from tube in cross-flow increases as shown in fig.1.8.

29

Fig.1.9 Mean heat transfer of a rough-surface cylinder in a 1.25x0.935 bundle in transformer oil flow [13]. K= Nu-PT-~.

about 40% in the subcritical flow regime (Re < 2 x IO), and about 55% in the supercritical flow regime (Re > 2 x 10). The behavior of heat transfer is also similar in external flows over roughsurface tubes. Suitable surface elements may provide significant heat transfer enhancement from tubes in cross-flow. A two- and three-fold increase of local heat transfer is observed on rough surfaces, as compared to smooth ones ; fig.1.9. The presence of roughness introduced a sharp change of the hydrodynamic pattern over the surface. The average heat transfer [l] suggests that the onset of the critical flow regime is strongly dependent on surface roughness, and occurs at lower Re values and at larger roughness heights. At low Re the laminar boundary layer covers surface elements completely and the average heat transfer is not influenced by surface roughness. In the higher range of Pr and Re the boundary layer thickness is considerably smaller than the height of roughness elements, and average heat transfer is again independent of surface roughness. An optimal surface roughness must be found for each specific case and a given flow regime [10].

30

Chapter 6

Modified Heat Exchangers

Extended surfaces: addition of fins to the outer surface of the copper tube A total of 42 fins were placed on the inner tube. These fins were placed in seven banks of six fins each. The distance between consecutive banks was 116 inch, with the fins at the closest allowable distance to the entrance and exit of the flow. The length of each bank of fins was chosen to be 2 inches. The six fins in each bank were radially spaced, in increments of 60 degrees, on the outside surface of the inner tube. The fins were made of copper due to its high thermal conductivity. The maximum allowable fin height based on the outer diameter of the tube and the inner diameter of the shell was calculated to be 0.477 inch. Given this constraint, a fin height of 7 16 inch was chosen for ease of manufacturability, assembly, and material availability. The fin thickness was chosen to be 0.025 inch, since material of this gage was readily available and inexpensive. The base of each fin was soldered to the inner copper tube. Plumbing solder was used because of its relatively high thermal conductivity, low cost, high strength, and lack of lead content (for toxicity reasons). The inner copper tube with the attached fins and a portion of the shell removed.

31

These fins not only increase the surface area, but they also increase turbulence, and induce a mixing effect inside the water jacket. This was accomplished by bending the top-rear corner of the fin towards the center of the tube. The increase in turbulence and the induced mixing tend to increase the heat transfer coefficient. Obstruction devices: annular rings or disks Orifices or other temporary restrictions to flow often induce turbulence. Our approach, therefore, was periodically to restrict the flow with annular rings or disks. A schematic of the disk/ring is shown in Fig. 4. The opening in the disk is in the center. This forces all the fluid to pass through a narrow opening near the surface of the copper tube. Because this opening is much smaller than the rest of the annulus, liquid velocity is high through the opening. Also, the rapid expansion on the opposite side of the opening triggers the transition from laminar to turbulent flow. The 162 H. I. AbuMulaweh International Journal of Mechanical Engineering Education 31/2 Fig. 2 Fin dimensions. Outside edges of the disks are sealed to the PVC pipe to force the fluid through the narrow opening near the copper tube. The resulting high liquid velocity and flow turbulence tend to increase the heat transfer coefficient. The problem is then how to install and support the annular disks inside the PVC pipe. The method that was chosen was to make the disk an integral part of a thin- Heat transfer enhancement methods walled tube that slipped inside the PVC pipe as shown in fig.1.10.

Fig.1.10 Heat exchanger in cross-section with disks. All dimensions are in inches.

32

6.1 Displacement Device

A spiraled rod inside the copper tube the spiraled rod, shown in cross-section Fig.1.11, was basically a 0.125-inch diameter stainless steel S316 rod with pins inserted into it.

Fig.1.11 Spiraled rod. The diameter of the pins was 0.0938 inch and the angle of rotation of the pins was 45 degrees. They were arranged at 0.5-inch intervals along the axial length. The spiraled rod should enhance the heat exchanger for three reasons. First, the pins act as triggers and promoters of turbulence. Secondly, secondary flow develops as the flow field is spiraled inside the annulus. And thirdly, the spiraled rod reduces the hydraulic diameter of the heat exchanger. All three effects turbulence, secondary flow, and reduction in the hydraulic diameter tend to increase the heat ransfer coefficient (i.e., enhance the heat exchanger).

Fig.1.12 Twisted tape inside the copper tube.

33

6.2 Swirl flow device

A twisted-tape insert the thin twisted helical tape, shown in Fig.1.12, was made of an aluminum sheet (1.6 mm thick and 12.7 mm wide). The tape was twisted eight times. There was a small clearance between the twisted tape and the walls of the copper tube to allow easy insertion of the tape. Recently, the work by Manglik and Bergles [8,9] on twisted tapes presented an updated understanding of the heat transfer for both laminar and turbulent flows. The blockage caused by the finite tape thickness increases the average velocity. Heat transfer enhancement may occur for two reasons. First, the tape reduces the hydraulic diameter of the inner copper tube; this tends to increase the heat transfer coefficient, even for zero tape twist. Secondly, the twist of the tape causes a tangential velocity component. This causes the speed of the flow to increase, particularly near the wall. The enhancement in the heat transfer is a result of the mixing by the secondary flow and the increased shear stress at the wall [9].

34

Chapter 7 Heat Transfer And Pressure Drop Augmentation For Laminar Flow In Spirally Enhanced Tubes
This paper work is done by Sara Rainieri, Angelo Farina and Giorgio Pagliarini.

The effect of the internal helical ridging on the heat transfer coefficient and friction factor is Experimentally investigated for laminar internal flow of a Newtonian fluid in the case of uniform wall heat flux. Five different geometries are considered. The results are compared with the results of previous investigations and with the predictions for the smooth tube in order to point out the heat transfer enhancement and the friction factor increase in the laminar flow field.

7.1 Introduction
The introduction of roughened surfaces obtained with internal helical ridging or transverse ribbing, has revealed a successful method in enhancing heat transfer in tubes and, therefore, in reducing the heat exchanger size. Although most prior studies have focused on the turbulent 35

flow, the laminar range is of particular interest in a wide variety of engineering situations, also including heat exchangers for viscous liquids in chemical process and food industry. When the flow is laminar because of small dimensions, low flow rates, or highly viscous fluids, like the ones processed in food industry, the use of smooth tubes gives poor performances and new configurations are needed to enhance the rate of heat transfer and, therefore, to reduce heat exchanger size and cost. The helical corrugation has revealed a type of roughness very suitable for dairy products, since fouling may be caused by sharper ridges. Tubes having a transverse repeated-rib roughness with rectangular cross section have been analyzed first by Webb et al. By using the law of the wall similarity and heat-momentum analogy he proposed some correlations where the friction factor is expressed as a function of the dimensionless geometric parameters and the Reynolds number, while the Stanton number is described as a function of the same variables and the Prandtl number as well. Gee and Webb extended this study to the effect of the rib helix angle. Withers developed correlating equations, based on the heat-momentum transfer analogy, for the heat transfer and pressure drop in tubes having simple and multiple helical internal ridging. These correlations were derived by experimental data obtained with water in the Reynolds number range 10000-120000 and Prandtl number range 4-10. Heat transfer enhancement of up to 2.5 to 3 was reported. Heat transfer augmentation in a spirally fluted tube was investigated by Yampolski. He showed that the swirl in the flow due to the helical flutes enhances turbulent exchange on both the inside and the outside of the tube without causing a considerable increase in the friction factor. Other experimental data on pressure drop and heat transfer coefficient for turbulent flow inside doubly-fluted tubes were obtained by Richards et al. Twelve different geometries have been analyzed and the outcome shows that only some of them yield an improved performance, expressed by the Authors in terms of heat exchanger volume reduction. Garimella et al.expressed the performance of spirally enhanced tubes in terms of a single non dimensional geometric parameter, the severity. Again turbulent flow is considered and it is shown that for severity values between 0.001 and 0.01 the heat transfer augmentation is accompanied by relatively low friction factor increase, thus confirming the efficiency of this kind of geometries. Garimella and Christensen have investigated the fluid flow in annuli formed by placing spirally fluted, indented and ribbed tubes inside a smooth outer tube. Detailed temperature profile measurements and flow visualization tests were performed for the laminar, 36

transitional and turbulent flow in order to better understand the development of the swirl in the bulk flow. It was found that the fluted inner tubes are the most efficient in promoting the secondary flow and hence in enhancing the convective heat transfer. The same Authors suggested also correlating equations for the friction factor and the Nusselt number for the laminar, transitional and turbulent flow in spirally fluted annuli.

7.2 Experimental Apparatus

Fig. 1.13 - Experimental apparatus The test section together with a schematic drawing of the flow loop are shown in fig.1.13 The prescribed condition of uniform heat flux has been obtained by Joulean dissipation in the tube wall. The heated section is 1.84 m long and it is preceded by an unheated development approach section of about 1 m. In the case of the smooth tube this allows the fluid to reach the complete hydrodynamic development before entering the heated section for flow rates corresponding approximately to Reynolds numbers of up to 1500. The whole length of the heat transfer section is thermally insulated to minimize the heat exchange to the environment. The wall temperature has been measured trough 45 type T copper constantan thermocouples, previously calibrated. They were attached circumferentially to the external surface of the tube, fixed with a layer of self bonding tape and electrically insulated from the tube itself. The inlet 37

and outlet temperature has been measured by two thermocouple probes directly immersed in the fluid. The bulk temperature at any location in the heat transfer section has been calculated from the power supplied to the tube and heat losses trough the insulation, whose thermal resistance had been measured in a previous calibration of the apparatus. The data acquisition software provided to update the data coming from all the channels and to plot them as a function of time on a screen. The program provided besides to hold the heat flux constant by varying the output of the power supply according to the effective resistive load of the tube wall. The power supplied to the test section was selected for the various flow rates to limit the difference between the inlet and outlet bulk temperature to about 2.5 C. In these conditions the effect of the variation of the fluid properties with temperature was assumed to be negligible. As working fluid a mixture containing approximately 50 per cent by weight of ethylene glycol in water has been used. The dynamic viscosity of this Newtonian fluid has been measured with a capillary-tube viscometer, type Ubbelohde. The Reynolds number range investigated is 200 < Re < 2000. Flow rates were obtained by measuring the time needed to fill a volumetric flask placed at the outlet of the test section, while the magnetic flow-meter incorporated in the loop was simply used as a flow indicator. Pressure drop along the whole length of the tube was obtained by measuring the level reached by the fluid in two piezometric tubes in isothermal condition. Further details about the experimental apparatus and the data acquisition system can be found in as shown in fig.1.13.

7.3 Results
Approximately 35 thermocouples probes have been distributed along the whole length of the heated section externally on the upper and lower crest of the corrugation. In order to measure also the local wall temperature variations, at some axial positions a thermocouple has been attached to the crest of the corrugation and other two probes have been placed in the preceding and in the following trough within a short distance. The internal wall temperature has been obtained by assuming a uniform heat generation in the tube wall and then calculating the temperature drop through the wall thickness. The temperature measured by the thermocouples placed in the external trough results lower than the one measured by the thermocouple placed in the adjacent external crest. In this region the local heat transfer coefficient results lower, probably due to some fluid stagnation [2].

38

Chapter 8

Applications of Heat Exchangers And Enhancement Technique

8.1 Refrigeration and Air-Conditioning System Applications

The benefits of heat transfer augmentations in these systems are well-known to the industry and their use is the norm rather than the exception. Still, their selection, optimization and proper implementation are critical factors in order to realize the maximum benefit. For in tube boiling (or condensation), the aluminum star inserts that were once widely used have nearly been completely abandoned in favor of internally micro finned tubes. These micro finned tubes provide heat transfer augmentation similar to that of the star inserts but at a fraction of their increase in two-phase pressure drop. Wolverine Tube Inc. is a leading manufacturer of micro fin tubes. With air passing over air-conditioning coils, the newest louvered plate designs for "compact" heat exchangers provide a substantial increase in heat transfer compared to older designs. Thus, the inside coefficient is now often augmented with micro fins. Use of an internal 39

augmentation in turn reduces the length and weight of the unit, which reduces the number of louvered external plates required, increases air velocity and thus the air-side coefficient too. For evaporation in flooded evaporators, many external augmentations are available commercially, most of which are normally produced in doubly-enhanced versions (i.e. with tube-side enhancement for the chilled water/cooling water/brine).

8.2 Refinery and Petrochemical Plant Applications

Several examples of heat transfer augmentation interventions in chemical plants and refineries to look for are:

8.2.1 Single-Phase Exchangers: 1. Use of integral low finned tubes in heat exchangers when the limiting thermal resistance is on the shell-side; 2. Use of tube inserts (wire mesh or twisted tape types) are highly effective in laminar flows inside tubes.

8.2.2 Condensers: 1. Substitution of a plain U-tube bundle of an overhead condenser with a low finned tube bundle (or installation of inserts in an existing tube bundle) to increase the overall heat transfer coefficient, which reduces the distillation tower operating pressure and temperature, and saves energy in heating the tower's feed to a lower operating temperature; 2. Using low finned tubes as opposed to plain tubes can often reduce the number of shells in multiple shell condensers by a factor of two, especially for condensing multi-component vapors on the shell-side.

8.2.3 Air Coolers: For cooling viscous fluids in air-cooled heat exchangers, tube inserts inside high finned tubes can significantly reduce the number of parallel units and the plot size required (some air cooler manufacturers have already been building units with wire mesh type inserts, for instance). 40

8.3 Air-Separation and Liquefied Natural Gas Plant Applications

Plate-fin heat exchangers are widely used in these facilities. While plate-fin exchangers allow very close temperature approaches, their long and tortuous flow paths can have larger pressure drops than shell-and-tube units. Thus, the real gain in overall system efficiency must be looked at. Enhanced boiling tubes and enhanced condensing tubes are used to advantage in these facilities in their refrigeration system reboiler-condensers, with the heating fluid-side also augmented with fins.

8.4 Applications to Lubricating Oil Coolers

In tube heat transfer augmentations are particularly suited to increasing heat transfer coefficients in lubricating oil coolers. These coolers normally come as part of package or skid mounted units for compressors, turbines, motors, engines, etc. or are mounted on large construction equipment. Hence, a lighter weight, more compact cooler has many economical advantages. Normally an increase in tube-side heat transfer performance is obtained with an insert while meeting the same pressure drop limitations as for the plain tube unit. This is achieved by using fewer tube passes.

8.5 Power Plant Operations

Integral low finned tubes are becoming widely used in power plant main condensers. The external fins for condensing steam must be optimized for the right fin density and height. Use of external fins shifts the controlling thermal resistance to the cooling waterside. Thus, low finned tubes with internal ribs are particularly suitable for this application, especially when utilizing alloys. Corrugated tubes are also beneficial in these applications.

41

8.6 Geothermal and Ocean-Thermal Power Plant Applications

The viability of geothermal and ocean-thermal power plants depends in part on the use of small temperature approaches in the heat exchangers in order to increase the cycle efficiency. Low temperature approaches result in low log-mean-temperature-differences and thus large heat exchanger surface areas [8].

42

Chapter 9

Summary and Conclusions

An investigation has been made of the effects on heat transfer of allowing for temperature dependence of viscosity and viscous dissipation. The first effect consistently increases Nu, the increase being larger for higher heat fluxes and higher viscosity temperature coefficients. It is negligible at low Pr. Viscous dissipation is negligible for Pro < 200. The constant heat flux Brinkman number is a reliable criterion, and viscous dissipation becomes apparent at Br > 2 x 10 -~. For very high Br, a numerical instability problem limited the length of duct which could be studied.
Laminar flow forced convection to a Newtonian fluid in spirally enhanced tubes has been experimentally investigated. The experimental data show that in the spirally enhanced geometries the Transition to the turbulent flow can occur at Reynolds number values much lower than 2000. This is in good agreement also with the results obtained for annuli formed with spirally fluted inner tubes. This early transition is accompanied by a significant heat transfer enhancement which assumes values between 1.1 and 6 in the Reynolds number range 300-1800. Despite the limited number of geometry Tested, it can be concluded that the cross-corrugated tubes exhibit the highest heat transfer increase over the smooth tube values.

The following are conclusions of the present work: 1. Periodically fully developed flow and heat transfer in a square duct with twisted tape insert are computed using numerically generated non-orthogonal, non staggered grids. The novel feature of 43

the numerical method is the use of smoothing pressure correction to eliminate the problem of check-board prediction of pressure.

2. The laminar flow friction factor data for 406Re6 1100 and 1:56Y 610 are correlated by f Re 361 0:15Sws 1=1:3 where the definition of Sws. Agreement with the experimental data is within 10%.

3. The laminar local Nusselt number shows axial variation occur over H=2 length in a repetitive manner, with the ratio of maximum to minimum Nusselt number varying between 1.02 and 1.17. Higher values of the ratio are encountered at higher Re and Pr. The maximum local Nusselt number occurs at a cross-section where the tape is aligned with the diagonal of the duct. In turbulent flow, the axial variation of Nu is not significant [5].

44

References

1. Enhanced Single-Phase Laminar Tube-side Flows and Heat Transfer, Engineering Data Book III by Wolverine tube, Inc.

2. Sara rainieri, angelo farina and giorgio pagliarini, Experimental Investigation Of Heat Transfer And Pressure Drop Augmentation For Laminar Flow In Spirally Enhanced Tubes, dipartimento di ingegneria industrial v.le delle scienze, 43100 parma.

3. Ranjit Gouda& Amit Bikram Das, Some Experimental Studies On Heat Transfer Augmentation For Flow Of Liquid Through Circular Tubes Using Twisted Angles And Tapes, Department of Chemical Engineering National Institute of Technology Rourkela 2008.

4. M. W. Collins and M. Keynejad, Effects of temperature dependence of viscosity and viscous dissipation on laminar Flow heat transfer in circular tubes.

5. S. Ray a, A.W. Date b, Friction and heat transfer characteristics of flow through square duct with twisted tape insert.

6. F.S.K. Warnakulasuriyas, W.M. Worek, Heat transfer and pressure drop properties of high viscous solutions in plate heat exchangers.

7. Wolverine Tube Heat Transfer Data Book, Heat Transfer and Pressure Drop Inside Tubes.

45

8. Wolverine Tube, Inc. Engineering Data Book, Design Consideration For Enhanced Heat Exchangers.

9. Hosni I. Abu-Mulaweh, Experimental comparison of heat transfer enhancement methods in heat exchangers, Department of Engineering, Indiana University Purdue University Fort Wayne, Fort Wayne, IN 46805, USA.

10. a. zukauskas, Enhancement of forced convection heat transfer in viscous fluid flows, lithuanian academy of sciences, vilnius 2600, Lithuania.

11. A Dewan1_, P Mahanta1, K Sumithra Raju1 and P Suresh Kumar2, Review of passive heat transfer augmentation Techniques, 1Department of Mechanical Engineering, Indian Institute of Technology, Guwahati, India 2Department of Ocean Engineering and Naval Architecture, Indian Institute of Technology, Kharagpur, India.

46