Heat transfer characteristics and pressure drop of the concentric tube equipped with coiled wires for pulsating turbulent flow

Accepted Manuscript Heat Transfer Characteristics and Pressure Drop of the Concentric Tube Equipped with Coiled Wires for Pulsating Turbulent flow A.E. Zohir, Ali A. Abdel Aziz, M.A. Habib PII: DOI: Reference:

S0894-1777(15)00063-1 http://dx.doi.org/10.1016/j.expthermflusci.2015.03.003 ETF 8424

To appear in:

Experimental Thermal and Fluid Science

Received Date: Revised Date: Accepted Date:

5 September 2014 18 February 2015 2 March 2015

Please cite this article as: A.E. Zohir, A.A. Abdel Aziz, M.A. Habib, Heat Transfer Characteristics and Pressure Drop of the Concentric Tube Equipped with Coiled Wires for Pulsating Turbulent flow, Experimental Thermal and Fluid Science (2015), doi: http://dx.doi.org/10.1016/j.expthermflusci.2015.03.003

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Heat Transfer Characteristics and Pressure Drop of the Concentric Tube Equipped with Coiled Wires for Pulsating Turbulent flow A.E. Zohira, Ali A. Abdel Azizb and M.A. Habibc a b c

Mechanical Engineering Department, Tabbin Institute for Metallurgical Studies, Cairo, Egypt Mechanical Engineering Department, Shoubra Faculty of Engineering, Benha University, Cairo, Egypt

Mechanical Engineering Department, King Fahd University of Petroleum and Minerals, Saudi Arabia

Abstract

The influence of pulsation with different amplitudes on the heat transfer rates in a double-pipe heat exchanger where a coiled wire inserts around the outer surface of the inner tube is investigated. The heat transfer and pressure drop characteristics were investigated for a coiled wire inserts with different pitch values in absence of pulsation. Pulsation frequencies that were produced by using a reciprocating device ranged from 140 to 260 cycles per minute (2.3 to 4.3 Hz). Five different displacement amplitudes were used where the stroke length of the reciprocating piston was varied from 60 to 185 mm. Thermally insulated wires having circular cross section of 2mm diameter, forming a coil of different pitches (p = 6, 12 and 20 mm) are used as turbulators. The investigation is performed for pulsating turbulent water flow in a double-pipe heat exchanger with cold water in the annulus space for counter flows. The experiments are performed for Reynolds numbers ranging from 4000 to 12000. The experimental results indicated that heat transfer rates are enhanced with the increase in the pitch of coiled wire. This was attributed to the combined effect of both pulsation and coiled wires compared with the smooth annulus case (no coiled wire). The highest enhancement ratio in the Nusselt number compared with smooth annulus with no pulsation reached a value of about 12.7 (for the pitch ratio 10, stroke ratio 4.2 and Re = 3856 with pulsating frequency 230 cycles/min) while the friction factor is about 8.7 times at same conditions. An empirical correlation was obtained and expresses the Nusselt number for different study parameters in 13 % deviation. Keywords: Double-pipe heat exchanger, coiled wire, turbulent annulus flow, pulsated flow

1.INTRODUCTION 1

The high performance of heat exchanger is an important and urgent problem for saving material cost and energy use. To decrease the size and material cost of heat exchanger, the high thermal performance of heat exchanger is strongly desired. Passive, active, and compound techniques such as, twisted tape, coiled wires, swirl flow generator are used to enhance the thermal performance of heat exchangers, [1]. Pulsation flows have a special importance in the operation of modern powerproducing facilities and industrial equipment technologies. Pulsation flows are also vital when pipeline faces cavitations where the flow parameters affect the performance of many engineering applications. Most of the previous investigators considered a small number of the operating variables (such as Reynolds number, amplitude, and pulsation frequency) in their studies. Pulsation would alter the thickness of the boundary layer especially near the wall and hence the thermal resistance [2-5]. Pulsating flow studies in a heat exchanger are very important because they help understanding and improving heat transfer characteristics and have been studied previously by many investigators [618], with conflicting conclusions according to the results. Baird et al. [6] conducted a research work to develop a steam-water heat exchanger to improve the performance by using a pulsating water turbulent flow. The cold water was passed upward through a steam jacketed copper tube. Sinusoidal pulsation was produced by an air pulsar, located upstream of the test section. They showed that the maximum enhancement of about 41 % based on the overall heat transfer coefficient for Reynolds number of 8000 is obtained for pulsation frequency ranged from 0.8 to 1.7 Hz while pulsation amplitude varied from 0.0274 to 0.335 m. Karamercan and Gainer [7] carried out an experimental work for heat transfer through a steam-water double-pipe heat exchanger. He aimed to study the effect of pulsator location on the heat transfer enhancement. Pulsation frequencies ranged up to 5 Hz. Five different displacement amplitudes were used and Reynolds number varied from 1000 to 50000. They found that the heat transfer coefficient increases with pulsation when pulsator located upstream/downstream of the heat exchanger with the highest enhancement observed in the transition flow regime. Lemlich [8] studied experimentally the turbulent heat transfer of water in a double pipe, steam-water heat exchanger. The pulsation was generated by an electrical-hydraulic pulsator. Flow conditions corresponding to Reynolds number varied from 500 to 5000 at frequency of 1.5 Hz were maintained. The turbulent pulsating flow increased the overall heat transfer coefficient by up to 80% based on upstream location of pulsator. Better improvement in the overall heat transfer coefficient was achieved for a closer the valve to the test section inlet. Effect of acoustic vibration on forced convective heat transfer was investigated by Lemlich and Hwu [9], which is simpler, is the use of external acoustic vibration effect on air follow in the core of a horizontal, double pipe stream-air heat 2

exchanger. Three different frequencies were imposed on air flowing at Reynolds number ranging from 560 to 5900. The sound vibration frequency was introduced by an electromagnetic driver, located upstream of the test section. They found that the enhancement of Nusselt number reaches about 51% and 27 % in the nominally laminar and turbulent regimes, respectively. The improvement is peaked with both amplitude and frequency. Shuai et al. [10] studied experimentally the enhancement heat transfer for laminar flow with pulsated perturbation, in a co-axial cylindrical tube heat exchanger for Reynolds number 150 to 1000, frequency from about zero to 2 Hz and amplitude of pulsation from 155 mm to 400 mm. A reciprocating pump located upstream the heat exchanger was used to introduce the flow pulsation. They found that the heat transfer coefficient was enhanced with strong-pulsed perturbations by as much as 300%. Many investigations concerning shell-tube heat exchangers were published. West and Taylor [11] carried out experimental study concerning steam-water heat exchanger. The water side was pulsated by upstream reciprocating pump. The results showed that the enhancement of heat transfer coefficient was between 60 to 70% at amplitude ratio of 1.42. Darling [12] conducted a research work to use a pulsation flow to improve the heat transfer coefficient by using a pulsator upstream the heaters. He showed that the heat transfer increases by 90% when applied a pulsation flow with a rate of 160 cycles/min and Reynolds number of 6000. Havemann and Rao [13] found that the increase in the heat flux for pulsations at 5-33 cycles/s is within 42% over the steady flow value under turbulent flow conditions. Keil and Baird [14] observed that the overall heat transfer coefficient increases with the pulsation frequency from 24 to 66 cycles/min to a nearly 100% for steam-water shell and tube heat exchanger. Ludlow [15], studied the convective heat transfer through a double pipe heat exchanger with hot water in the annulus. He found a heat transfer enhancement up to 5.0 times in a tube side for transition flow regime for pulsation frequencies from 10 to 170 cycles/min. Many investigations concerning pulsating flow through heat exchangers were published. A lack of proper understanding of the pertinent variables involved due to the conflicting conclusions of the pulsated flow effect in a heat exchanger. Baird et al., [6], West and Taylor, [11], Darling [12], Havemann and Rao [13], Keil and Baird [14] and Ludlow [15] stated that pulsating flow in a heat exchanger can be substantially improve the heat transfer. Mueller [16] performed a comparative study of theoretical and experimental comparison for heat transfer. He observed that the average Nusselt number was found to be less than the corresponding steady flow Nusselt number for pulsating flows which encompassed a frequency range from 2.3 to 14.9 cycles/min and a Reynolds number range of 53000 to 76000. 3

McMichael and Hellums [17] conducted a theoretical work to investigate the behavior of pulsation for laminar flow. It was found that for laminar flow, the pulsation is generally decreases the heat transfer rate especially at low amplitudes, which has no flow reversal. Martinelli et al. [18], used semi-sinusoidal velocity disturbances to study the heat transfer through a concentric tube heat exchanger. They focused on laminar flow conditions. They found that the overall heat transfer coefficient was increases with velocity disturbances by as much as 10% over the steady flow coefficient at most cases. The pulsation flow and its effect on the heat transfer coefficient were investigated by many researchers. Their analyses showed that there are conflicting results for pulsation effect on heat transfer due to variety of heat transfer control parameters. Some researchers stated that because of the flow pulsation there are increasing in heat transfer [6-15]. The previous experimental results covered variations of pulsating flow results with heat transfer coefficient. However, there is no surprising in view of many parameters combinations of pulsation variables, such as mechanisms of pulse generation, flow regimes, etc. that affect the flow and heat transfer characteristics. In some review papers, the values of heat transfer rates were found to decrease, no increase, little increase, or increase with pulsation as reported in some review papers [16-18]. Gunes et al. [19] used a coiled wire inside a circular tube to investigate the mechanism of heat transfer enhancement and friction factor. The circular tube was fitted with equilateral triangle cross sectioned wire coil which causes swirl flow movement in tube. Promvonge [20] investigated the mechanism of heat transfer enhancement by using a twisted tape together with typical wire coil with uniform pitch ratio. It seems that the combined effect of both wire coil and twisted tape on the heat transfer rate, friction factor and also thermal enhancement factor is much higher than that for wire coil or twisted tape separately, indicating the synergy effect of both devices. The thermal enhancement factor increases with decreasing coil spring pitch ratios. Eiamsa-ard et al. [21] studied experimentally the combined effect of non-uniform wire coil and twisted tape inserts on the thermal performance characteristics. They focused on experimental study in turbulent flow regime of the air flow with Reynolds numbers ranging from 4600 to 20,000. The idea of heat transfer augmentation was to apply compound devices such as twisted tape and coiled wires where a further augmentation is observed for heat transfer rate. Naga Sarada et al. [22] carried out experimental study using twisted tape inserts with different width ratios. Compared to smooth tube, the heat transfer enhancement due to twisted tape was further augmented. Experiments were carried out for plain tube with/without twisted tape inserts at constant wall heat flux and different mass flow rates. They also found that the enhancement in heat transfer for the twisted tape inserts varied from 36 to 48% for full width (26mm) and 33 to 39% for reduced width (22 mm) for all 4

Reynolds number range from 6000 to 13500. Patil and Vijay Babu [23] studied experimentally the heat transfer and friction factor characteristics of laminar flow of ethylene glycol through a square duct with nearly uniform wall temperature conditions and fitted with full-length helical screw element of different twist ratios. The increase in Nusselt number was about 5.44–7.49 and 2.46–4.87 times higher than the corresponding empty square duct due to induced swirl flow and for turbulent flow based on constant flow rate and constant pumping power criteria, respectively. Most of the methods of heat transfer enhancement use the same technique of twisted tapes insert, to force the flow in the channel to rotate. Murugesana et al [24] had found that the tube equipped with vertical wing-cut twisted tapes, horizontal wing-cut twisted tapes, and plain twisted tapes with twist ratios induced swirl flow and turbulence in the flowing fluid that are attractive techniques for forced convective systems. The tube equipped with horizontal wing-cut twisted tapes was significantly higher than those in the tube fitted with vertical wing-cut twisted tapes, plain twisted tapes, and a plain tube. The compression process necessary for the liquefaction of natural gas on offshore platforms generates large amounts of heat, usually dissipated via seawater cooled plate heat exchangers. The present experimental investigation was carried out to study the combined effect of spring-coiled wires and pulsation, on heat transfer rates through heat exchangers. The investigation has been performed for turbulent water stream in a double-pipe heat exchanger for counter flows with pulsating cold water flow in the shell side. The experiments are performed for flows with Reynolds numbers ranging from 4,000 to 14,000. Three different spring coiled wire pitch values were used. Pulsation frequencies, with using a reciprocating device, ranged up to 260 cycles per minute and five different displacement amplitudes were used. Comparisons between the present results and those of others are determined to validate the results. Practical observations and suggestions are recorded. 2.EXPERIMENTAL APPARATUS AND METHOD 2.1.ExperimentalApparatus An experimental apparatus was established for the investigation of the heat transfer characteristics of the turbulent pulsating water flow through a concentric tube heat exchanger where a coiled wire inserts around the outer surface of the inner tube. A schematic diagram of the experimental apparatus is shown in Fig.1. It consists of a test section, a pulsated mechanism, hot water closed loop, and cold water closed loop. A full set of instruments for measuring and control of temperature and flow rate of all fluids are installed at all important points in the circuit. The t e st section is a horizontal concentric tube heat exchanger as shown in Fig. 2, it consists of 5

a horizontal water-to-water concentric tubes heat exchanger with counter flow of water, a coiled wire (coated metal) to act as turbulators, pulsating mechanism, and temperatures measuring devices. The pulsated mechanism was constructed of an electric motor, a variable speed transmission and a reciprocating piston (Scotch-Yoke type). The output of the motor could be adjusted to cover a range of 0-1550 rpm corresponding to different piston frequencies of 140, 170, 200, 230, and 260 cycles/min. The pulsation was imposed to the cold water by the reciprocating piston mechanism. The pulsating flow is carried out for different piston stroke lengths (6, 9.5, 12.5, 15.5, and 18.5 cm) and different operating Reynolds number (based on the hydraulic diameter D and inlet cold-water flow h

conditions), ranged from 3856 to 11568. To minimize the heat losses in the system, the hot water is fed through the inner pipe, with cooling water in the outer annulus of the heat exchanger. The outer surface of the heat exchanger was well insulated to minimize convective heat loss to the surroundings, and necessary precautions were taken to prevent leakages from the system. The heat exchanger has a length of 1100 mm. The outer tube has 40 mm outer diameter and 1.5 mm wall thickness. The inner tube has an outer diameter of 12 mm with 1 mm wall thickness. Both tubes are manufactured of brass with thermal conductivity, k, = 109 W/m.K. Signal outputs of thermocouples were indicated by a data logger device, GL800-UM-851 midi logger, which has an accuracy of ± 0.05 % of reading. The coiled wires of different spring pitch values (P = 6, 12 and 20 mm) were coiled adjacent to the outer surface of the inner tube. These pitches correspond to pitch ratio of 3, 6, and 10, respectively. The cold water entering the system, through outer annulus of the heat exchanger, was drawn by a 3/4 hp pump from cold water supply tank which connected to the main water source, and passed to the drain out at downstream. Control valves are incorporated in each of the two streams to regulate the flow. The flow rates are measured using independent Rota-meters that are installed in each line. The measuring of flow rate was checked three times for each experiment by a standard beaker (20 liter)and stopwatch to reduce the uncertainty (± 1.2 %). A hot storage 50-liters tank is equipped with an electric heater and a temperature controller is used to maintain a temperature to within ± 0.5 °C. A 3/4 hp pump and water returns to the storage tank via a baffle arrangement were used to circulate the hot water to the heat exchanger to ensure adequate mixing. The temperature of hot water was kept at about 65 °C ± 0.5°C. 2.2. Experimental Method An experimental methodology was set to investigate the combined effect of roughness (using an insulated coiled coated wire with different pitches) and pulsation with different frequencies and amplitudes on the heat transfer through a concentric tube heat exchanger for the turbulent pulsating 6

water flow. The inlet hot water temperature was kept at about 65 °C and the inlet cold water temperature was about 25 °C. Three different coil pitches were used to study the influence of wire pitches on the convective heat transfer coefficient for turbulent pulsating water flow. During the experiments the flow rate through the inner tube was kept constant at 0.1144 kg/s and the flow rate in the annulus was 0.0663, 0.0994, 0.1326, 0.1657 and 0.1988 kg/s corresponds to Reynolds numbers of 3856, 5784, 7712, 9640, and 11,568, respectively. 2.3 Data Reduction and Experimental Uncertainty The heat transfer coefficient (ho) is determined from the overall heat transfer coefficient and, accordingly, the Nusselt number is calculated as shown by Pethkool et al. [25] and Meter [26]. The major heat losses are assumed to be through insulation only with neglecting other losses, Baughn et al. [27] and Incropera and Dewitt [28]. To account for the temperatures variations, a log mean temperature difference (∆Tlm= (∆Tin - ∆Tout) / (ln (∆Tin/∆Tout)) is used, [28-30]. The heat transfer through the inner tube and annulus of tube-in-tube heat exchanger was determined from inlet and × outlet temperature measurements and mass flow rate using ( Q = m Cp DT ) As usual, this heat may be expressed in terms of a heat transfer coefficient and tube logarithmic mean temperature difference ∆Tlm .The overall heat transfer coefficient, U, can be determined according to heat balance, with negligible heat losses to surrounding air.

Q = AUDTlm

(1)

Convective and overall heat transfer coefficients were deduced and Nusselt numbers were acquired as follows, [28-32]: 1 1 ln (do /di ) 1 = + + p UA ho Ao 2 kL hi Ai

(2)

Where, hi and ho are heat transfer coefficients for hot and cold water, respectively. The areas, Ai and Ao are the inner and the outer surface areas of the inner tube. The diameters, di and do are the inner and the outer tube diameters of the inner tube. U is the overall heat transfer coefficient and k is the thermal conductivity of the tube material. L being the total tube length. An empirical correlation to determine hi for fully developed turbulent flow in pipes was proposed by Gnielinski [30] and is used in the present work. Hence, ho & Nudo can be determined as follows:

Nu di =

hidi (fr/ 8 )( Re di - 1000 ) Pr = k 1 + 12.7(fr/ 8 )1/ 2( Pr 2 / 3-1 )

(3)

The tube-side heat transfer coefficient was obtained from Gnielinski correlation as follows: 7

hi =

(fr/8 )( Redi - 1000) Pr k . 1 + 12.7(fr/8 )1/ 2( Pr 2 / 3 -1 ) di

(4)

where the friction factor for smooth tubes is given by: (5)

fr = [ 0 .79 ln ( Redi ) - 1.64 ] - 2

The Reynolds numbers for the hot and cold fluids are: ReD = VDH/ν

(6)

Thus, the convective heat transfer coefficient hi is determined from equations (4 and 5) and used in equation (2) to determine the convective heat transfer coefficient, ho. The average Nusselt number for roughened inner tube with pulsating flow ( Nu ) of the present study were normalized by the Nusselt number for plain tube-in-tube heat exchanger (bare tube without wire or pulsation) ( Nuow ) to indicate the combined effect of pulsation with different amplitude and a coiled wire inserts by obtaining the enhancement ratio ( Nu / Nuow ). All fluid properties for cold and hot water flows were taken at the average fluid temperature. A more precise method of estimating uncertainty in experimental results has been presented by Kline and McClintock, which is described in Holman [31]. The method is based on careful specification of the uncertainties in the various primary experimental measurements. Assume that the result dependent variable (R) is a given function of the independent variables x1, x2, x3, ……………, xn. Thus: R = R (x1, x2, x3, ……………, xn). Let uR is the uncertainty in the result and u1, u2, u3, …….., un are the uncertainties in the independent variables. The uncertainty in the result is given as: 2 éæ ¶R ö2 æ ¶R ö2 æ ¶R ö ù uR = êçç u1 ÷÷ + çç u2 ÷÷ + ...................... + çç un ÷÷ ú êëè ¶x1 ø è ¶x2 ø è ¶xn ø úû

1 2

(7)

Table 1 presents the calculated uncertainty values for the measured quantities. 4. RESULTS AND DISCUSSION 4.1 Verification of Experimental Results Before carrying out heat transfer and friction factor measurements of wire coiled pipe with pulsating flow, the plain tube heat exchanger measurements (without using wires) were carried out to normalize the present experimental results and to compare with those from the proposed correlations [32] for Nusselt number and proposed correlation by Blausius [33] for friction factor. Nu = 0.0194 Re 0.8351 Pr 1 / 3 (

fr = 0.448 Re -0.275

m 0.14 ) mw

Jaco correlation [32]

(8)

Blasius correlation [33]

(9)

8

Figures 3 and 4 illustrate a comparison of Nusselt number and friction factor of plain tube obtained from present work with those from proposed correlation in the given range of Reynolds number. The present experimental results for the plain tube are in good agreement with the predicted results from proposed correlation with maximum deviation of 19% and 15% for Nusselt number and friction factor, respectively. Comparison between the present experimental data with the data reported in the literature for the steady flow has been done in order to validate the experimental data. 4-2 Average Friction Factor 4-2-1 Effect of Reynolds number on the friction factor The friction factor results for rough annulus with wire pitch to diameter ratio (P/dw) of 3, 6, and 10 for different Reynolds numbers are shown in Fig. 5. It is noticed that the friction factor decreases for different coiled wire pitches as Reynolds number increases. The reduction in wire pitch leads to increase in the friction factor due to swirl flow generated by wire coiled on inner tube and reaches the maximum for the pitch of 6 mm. This is due to the high contact surface area of the coiled wire for small pitch and number of turns. Hence, more pressure drop and friction factor for pitch 6 mm were noted in comparison to the case of coiled wire of pitch 12 mm and 20 mm. 4-2-2 Effect of wire pitch on the friction factor Fig. 6 presents the effect of the pitch to wire diameter ratio on the friction factor for different values of Reynolds number. The friction factor decreases as the pitch ratio increases for a given Reynolds number. This may be explained by the fact that the flow resistance and the turbulence level decreases as the wire pitch increases. A more significant effect of the pitch ratio on the friction factor is observed for low values of Reynolds number. 4-3 Average Nusselt Number 4-3-1 Effect of Reynolds number on the average Nusselt number ratio

The average Nusselt number ratio or the enhancement ratio was calculated as the ratio of the measured Nusselt number for roughened tube-in-tube heat exchanger to the Nusselt number of a plain tube heat exchanger (bare tube without wire or pulsation). The average Nusselt number ratio ( Nu / Nuow ) versus pulsating frequency (f) at pitch ratio (P/dw) 3.0 for different Reynolds numbers and stroke ratios (S/Di) is shown in Fig.7. The variations of ( Nu / Nuow ) are presented for different values of Re and S/Di. The maximum values of ( Nu / Nuow ) differ with frequency, Reynolds number, and stroke ratio. The highest increase in the enhancement ratio for roughened tube in shell9

tube heat exchanger compared with smooth annulus with no pulsation reaches to about 7.8 times (case of Re=7712, f=140, and S/Di=1.6) and 10.8 times (case of Re=9640, f=230, and S/Di=5.0). Figures 8 and 9 illustrate the effect of Reynolds number on the enhancement ratio ( Nu / Nuow ) , at P/dw=6.0 and P/dw=10, respectively for different values of frequency and stroke ratio. The highest increase in the Nusselt number for the pitch ratio 6 in the order of 9.3 times for stroke ratio 1.6 and Re = 7712 with frequency 200 cpm. The highest enhancement ratio in the Nusselt number reached to a value of 12.5 (for the pitch ratio 10, stroke ratio 5.0 and Re = 3856 with pulsating frequency 230 cycles/min) compared with smooth annulus with no pulsation. To indicate the combined effect of roughened tube and pulsation, the maximum augmentation of the heat transfer process takes place at the lowest values of pulsating frequency and the maximum enhancements in Nusselt number reaches to 12.7, 10.8, and 9.3 times for a roughened tube-in-tube heat exchanger with pulsation and at pitch ratios 10, 3, and 6, respectively in comparison with a plain tube-in-tube heat exchanger (bare tube without wire or pulsation). It is noticed from Figs.7 to 9 that as the pitch ratio and stroke ratio increase the average Nusselt number and thus the enhancement ratio increase. 4-3-2 Effect of wire pitch on the average Nusselt number ratio

The effect of wire pitch on the average Nusselt number ratio for roughened outer surface of the inner pipe at S/Di=1.6 for different values of Reynolds number and pulsation frequency is illustrated in Fig.10. The average Nusselt number increases as the pitch to wire diameter ratio increases for different values of frequency and Reynolds numbers. This may be explained as the wire pitch increases, the turbulence generation by pulsation is extended freely along the roughened surface and the heat transfer enhancement increases also. It is concluded that the enhancement ratio ( Nu / Nuow ) increases for different values of frequency as the Reynolds number increases to 7712. Then, further increase of Reynolds number causes the decrease of ( Nu / Nuow ) to a minimum value and at Reynolds number of 9640 the enhancement ratio increased to a maximum value at Re=11568. This may be attributed to the variations of turbulence level produced from the imposed frequency against the natural frequency of flow in annulus. The highest increase in the enhancement ratio of 10.2 times was produced for high Reynolds number of 11568 with a low pulsation frequency of 170 cpm and high pitch ratio of 10. This is because of the decreasing of the interruptions to the development of the boundary layer of the fluid flow near the wall of the inner tube at high pitch. 4-3-3 Effect of frequency on the average Nusselt number ratio

10

Figure 11 illustrates the effect of pulsating frequency on the average Nusselt number ratio at S/Di=1.6 and P/dw=3.0 for different values Reynolds number. It is noticed from the figure that there is a more significant effect of frequency on the ( Nu / Nuow ) only at Re = 7712. The highest values of average Nusselt number ratio takes place at the lowest values of pulsating frequency at 140 , 170, and 200 cpm and at Re = 7712, and consequently, more heat is transfer from the inner tube to the cold water stream. The maximum augmentation of the heat transfer process takes place at Re=7712 and frequency 140 cpm and the Nusselt number enhancement reaches to 7.8 times of that for plain tube without pulsating flow. For all values of pulsating frequency the ( Nu / Nuow ) decreases sharply to a minimum value as the Reynolds number increases from 7712 to 9640 and rises again by about 68% at Re =11568. Figure 12 shows the effect of pulsating frequency on the average Nusselt number ratio, at P/dw = 6.0 and S/Di=1.6. For different Reynolds numbers, the Nusselt number enhancement increased with an increase in the pitch ratio. The highest increase in the enhancement of 9.3 times was produced for the inner pipe with a coiled wire at a pitch ratio 6.0, frequency 200 cycles/min and Re =7712. The enhancement decreased to about 2.15 times, and then increased to about 7.95 times. It is noticed from the figure that there is a more significant effect of Reynolds number on the

( Nu / Nuow ) at P/dw = 6.0 in comparison with the pitch ratio 3.0. Figure 13 illustrates the effect of pulsating frequency on the enhancement ratio ( Nu / Nuow ) , at P/dw=10 and S/Di=1.6. It is noticed from the figure that the maximum enhancement ratio is 10.2 times at high frequency values of 170 and Re=11568. 4-3-4 Effect of stroke length on the average Nusselt number ratio

The variation of the average Nusselt number with Reynolds number at P/d w=3 and frequency of 140 cycles/min for different stroke ratios (S/Di) is shown in Fig.14. The figure depicts that at low and high Reynolds number (Re = 3856 and 11568) there is a small significant effect of stroke ratio on the enhancement ratio ( Nu / Nuow ) , otherwise for Reynolds number at ( 5784, 7712, and 9640), high variations in ( Nu / Nuow ) is noticed for different values of S/Di. This may be due to higher turbulence level generated for layer wire to inner diameter ratio. At stroke ratio 2.6, the highest increase in the enhancement was produced for the inner pipe with a coiled wire at a pitch ratio 3.0, frequency140cycles/min, and Re = 7712. Figure 15 shows the variation of average Nusselt number ratio with Reynolds number for the coiled pipe with P/d w = 6 and pulsating frequency 140 for five different stroke ratios. It can be concluded that the Nusselt numbers for the tube coiled wire at P/dw = 6 are higher than that of plain tube. From Fig. 16, it is found that the lower limit and upper limit of 11

the Nusselt number ratio increases with coiled wire pitch and the ( Nu / Nuow ) for the roughened tube is highest for a given Reynolds number and stroke ratio when compared with plain tube. Also it creates the turbulence and whirling motion to the water which is flowing in the annulus. Sufficient frequency and amplitude increase the level of turbulence, which leads to improved convection heat transfer. 4-4 Correlation Development Based on the present experimental results, the average Nusselt number enhancement ratio ( Nu / Nu ow ) was correlated as a function of Reynolds number, pitch to wire diameter ratio, stroke length to shell tube inner diameter ratio, and dimensionless angular pulsating frequency as follows:

Nu / Nuow = 50(Re )

-0.642

( P / d w )1.411 (S / Di )0.082 (w * )0.008

(10)

Where: 3856 ≤ Re ≤ 11568, 3 ≤ p/dw ≤ 10, 1.6 ≤ S/Di ≤ 5.0, 883 ≤ ω* ≤ 4290, with maximum deviation of 13%. The dimensionless angular pulsating frequency = ( w D h ) , where ω is the angular U *

um frequency of imposed pulsation in rad/s and U * = 0.1989 in m/s. Figure 17 shows a comparison 0.125 Re

between the present experimental results and the results using the present correlation (10). 5- Conclusions Heat transfer and pressure drop characteristics through an annulus of tube in tube-heat exchanger with coiled wire on the outer surface of the inner tube have been studied. Three tube-in-tube heat exchangers were used with coiled wires with pitch of 6 mm, 12 mm, and 20 mm. Experimental setup was a plain tube-in-tube heat exchanger to be used for comparison of augmentation techniques. The following are the main conclusions: 1- As Reynolds number increases, the average Nusselt number and thus the enhancement ratio increases except at value of Re = 9640. 2- A long inserted coiled wire pitch increases the heat transfer enhancement ratio for pulsation flow while a small pitch length results in a small value with larger pressure loss. 3- A maximum enhancement in the average Nusselt number ratio reaches to about 12.7 times for a roughened tube-in-tube heat exchanger with pulsation at (P/dw =10, S/Di=4.2, Re = 3856, and f=230 cpm) in comparison with a plain tube-in-tube heat exchanger (bare tube without wire or pulsation). 4- As the pitch ratio increases the average Nusselt number and thus the enhancement ratio increases. 5- As stroke ratio increases the average Nusselt number and thus the enhancement ratio increases. 12

6- The effect of pitch length on swirl flow generation and pressure loss was more significant than that of frequency amplitude. 7- A convective heat transfer correlation was deduced from experimental results that predict average Nusselt numbers ratio in relationship with the pitch ratios, stroke ratio, angular pulsating frequency, and Reynolds number.

Nomenclature A C dw d D f fr h k L

m×

Nu P Pr Q Re S

heat transfer area (m2) specific heat capacity (kJ/kg K) wire diameter (m) inner tube diameter (m) outer tube diameter (m) pulsating frequency (cycle/minute) friction factor convective heat transfer (W/m2 K) thermal conductivity (W/m K) tube length (mm) mass flow rate (kg/s) Nusselt number pitch (mm) Prandtl number (µCp/k) heat transfer rate (W) Reynolds number stroke length (mm)

T U V

temperature (°C) overall heat transfer coefficient (W/m2K) average axial velocity (m/s)

Greek Symbols

ΔTlm µ ν ε Subscripts c e h i o s w

logarithmic mean temperature difference (K) dynamic viscosity (N s/m2) kinematic viscosity (m2/s) heat exchanger effectiveness cold fluid hydraulic hot fluid inlet/inner outlet/outer steady the coiled wire

Acknowledgement: The authors wish to acknowledge the support received from King Fahd University of Petroleum and Minerals. References [1] Liu S. and Sakr M., A comprehensive review on passive heat transfer enhancements in pipe exchangers. Renewable and Sustainable Energy Reviews 2013, (19): 64–81. [2] Togun H., S.N. Kazi, Badarudin A. ''A Review of Experimental Study of Turbulent Heat Transfer in Separated Flow''. Australian Journal of Basic and Applied Sciences, 2011, 5(10): 489-505. 13

[3] Zheng, J., Zeng, D., Wang, P. and Gao, H. ''An experimental study of heat transfer enhancement with a pulsating flow''. Heat transfer-Asian Research, 2004, 33(5): 279-286. [4] Hatami, R., ''Pulsating flow in a heat exchanger''. Journal of Warme and Stoffubertragung, 1980, (14): 109-118. [5] Krishnan, K. N. and Sastri, V. M. K., “Pulsating flows in heat exchangers - an experimental Study”. Warme - und Sroffubertragung, 1982, (16): 169-173. [6] Baird, M. H. I, Duncan, G. J., Smith, J. I. and Taylor, J., “Heat transfer in pulsed turbulent flow”. Chemical Engineering Science, 1966, (21): 197-199. [7] Karamercan, O. E. and Gainer J. L., “The effect of pulsation on heat transfer”. Ind. Eng. Chem. Fundam., 1979, 18(1): 11-15. [8] Lemlich, R. (1961): “Vibration and pulsation boost heat transfer”. Chemical Engineering, 1961, 68(10): 171–176. [9] Lemlich, R. and Hwu, C., “The effect of acoustic vibration on forced convective heat transfer”. A. I. Ch. E. J., 1961, (21): 197-199. [10] Shuai, X., Cheng, S. and Antonini, G., “Pulsation effects on convective heat transfer in the laminar flow of a viscous fluid”. The Canadian Journal of Chemical Engineering, 1994, (72): 468 - 475. [11] West, F. B. and Taylor, A. T.: “The effect of pulsation on heat transfer in turbulent flow of water inside tubes”. Chemical Engineering Progress, 1952, (48): 39-43. [12] Darling, G. B., "Heat Transfer to Liquid in Intermitted Flow Petroleum"., London, 1959, (22):177-182. [13] Havemann, H. A., Rao, N. N., “Heat transfer in pulsating flow”. Nature 7(4418): 41, 1954. [14] Keil, A. H,. Baird, M. H. I., "Enhancement of heat transfer by flow pulsations". Ind., Eng. Chem. Process Des. Dev., 1971, 10:473. [15] Ludlow, J. C., MS. Thesis, University of Virginia, Charlottesville, Va., 1975. [16] Muller, W. K., ''Proceedings fifth Midwestern conference of fluid mechanics''. p 146, Ann Arbor, Mich., 1957. [17] McMichael, W. J., Hellums, J. G., AIChE J., "The Effect of Pulsations on Heat Transfer for Laminar Flow". 1975, 21:743. [18] Martinelli, R. C., Boelter, L. M. K., Weinberg, E. B. and Yakal, S., “Heat transfer to a fluid flowing periodically at low frequencies in a vertical tube”. Transaction of ASME, 1943, 789-798. [19] Gunes S., Ozceyhan V. and Buyukalaca O., Heat transfer enhancement in a tube with equilateral triangle cross sectioned coiled wire inserts, Experimental Thermal and Fluid Science 2010, 36 (6) 684–691. [20] Promvonge P., Thermal augmentation in circular tube with twisted tape and wire coil turbulators, Energy Conversion and Management 2008, (49): 2949-2955. [21] Eiamsa-ard S., Nivesrangsan P., Chokphoemphun S., and Promvonge P., Influence of combined nonuniform wire coil and twisted tape inserts on thermal performance characteristics. International Communications in Heat and Mass Transfer 2010, (37): 850–856. 14

[22] Naga Sarada S., Sita Rama Raju A.V., Kalyani Radha K., and Shyam Sunder L., Enhancement of heat transfer using varying width twisted tape inserts. International Journal of Engineering, Science and Technology 2010, 2(6): 107-118. [23] Patil S. V. and Vijay Babu P. V., Laminar Heat Transfer Augmentation through A Square Duct and Circular Tube Fitted with Twisted Tapes. Experimental Heat Transfer, 2014, (27):124-143. [24] Murugesan P, Mayilsamy K. & Suresh S. Heat Transfer in a Tube Fitted with Vertical and Horizontal Wing-Cut Twisted Tapes., Experimental Heat Transfer, 2012, (25): 30-47. [25] Pethkool S., Eiamsa-ard, S., Kwankaomeng, S. and Promvonge, P., Turbulent heat transfer enhancement in a heat exchanger using helically corrugated tube. International Communications in Heat and Mass Transfer, 2011, (38): 340-347. [26] Meter, J. V., Experimental investigation of a printed circuit heat exchanger using super critical carbon dioxide and water as heat transfer media. MS. Thesis, Kansas State, University, Manhattan, Kansas, 2008). [27] Baughn, J. W., Hoffman, M. A., Launder, B. E., and Takahashi, R., Turbulent heat transport in circular ducts with circumferentially-varying heat flux". ASME J of heat Transfer, 1984, (106): 64-70. [28] Incropera, F., and Dewitt, P.D., Introduction to heat transfer. Third edition, John Wiley & Sons Inc., 1996. [29] Kays, W.M., and Crawford, M.E., Convective heat and mass transfer". Third ed., McGraw-Hill, Singapore, 1993. [30] Gnielinski, V., New equation for heat and mass transfer in turbulent pipe and channel flow. International Chemical Engineering, 1976, (16): 359-368. [31] Holman, J. P., Experimental methods for engineers. Fourth Editions, McGraw- Hill Book Company, New York, 1984. [32] J. Dirkey and J. P. Meyey, Convection in Concentric Annular Regions for Turbulent Flow of Liquid Water. R & D Journal, 2003, 19 (2) incorporated into The SA Mechanical Engineer. [33] Schlichting, H., "Boundary Layer Theory", 7th ed., McGrew- Hill, New York, 1979.

15

Table 1 values of the uncertainty of the measured quantities Relative Uncertainty,

Variable, Parameter Mass flow rate

% ± 1.2

Reynolds number

± 1.2

Logarithmic mean temperature difference

± 2.4

Mean Nusselt number

± 4.7

16

Figure Captions Fig. 1: Experimental apparatus Fig. 2: Test section Fig. 3: Comparison of average Nusselt number with previously published results for the plain tube. Fig. 4: Comparison of friction factor with previously published results, (Pethkool et al. [25]), for the plain tube. Fig. 5: The average friction factor versus pitch ratio for different Reynolds number (without pulsation). Fig. 6: The average friction factor versus Reynolds number for plain and rough annulus with different pitch ratios (without pulsation) Fig 7: The average Nusselt number ratio versus frequency for rough annulus at P/dw= 3 for different stroke ratios and Reynolds numbers. Fig 8: The average Nusselt number ratio versus frequency for rough annulus at P/dw= 6 for different stroke ratios and Reynolds numbers. Fig. 9: The average Nusselt number ratio versus frequency for rough annulus at P/dw= 10 for different stroke ratios and Reynolds numbers. Fig. 10: The average Nusselt number ratio versus Reynolds number for wire coiled with different pitches and different frequencies at stroke ratio 1.6. Fig. 11: The average Nusselt number ratio versus Reynolds number for wire coiled at P/dw = 3.0 with different frequencies and S/Di = 1.6. Fig. 12: The average Nusselt number ratio versus Reynolds number for wire coiled at P/dw = 6.0 with different frequencies and S/Di = 1.6. Fig. 13: The average Nusselt number ratio versus Reynolds number for wire coiled at P/dw = 10 with different frequencies and S/Di = 1.6. 17

Fig. 14: The average Nusselt number ratio versus Reynolds number for wire coiled at P/d w=3.0and frequency =140 cpm. Fig. 15: The average Nusselt number ratio versus Reynolds number for wire coiled at P/dw=6.0 and frequency =140 cpm. Fig. 16: The average Nusselt number ratio versus Reynolds number for wire coiled at P/dw=10and frequency =140. Fig. 17: Comparison between the experimental values of ( Nu / Nuow ) with the values computed from correlation.

6

1- Hot water tank 2- Electric heater 3- Centrifugal pump 4- Ball valve 5- Rotameter 6- Thermocouple 7- Voltage regulator 8- Test section 9- Cold water tank 10- Centrifugal pump 11- Condensing unit 12- Receiver tank

8 5 4 2

3

7 1 11

10 9

12

Fig.1: Experimental apparatus.

18

Fig.2: Test section.

19

120 Present work

Jaco Dirkey [32]

Plain tube in tube

Nusselt Number

100

80

60

40

20 0

2000

4000

6000

8000 10000 12000 14000

Re Fig. 3: Comparison of average Nusselt number with previously published results for the plain tube. 20

0.12 Present work Blasius Eq. Plain tube in tube

Friction Factor

0.10

Pethkool

0.08 0.06 0.04 0.02 0.00 0

2000

4000

6000

8000 10000 12000 14000

Re Fig. 4: Comparison of friction factor with previously published results, (Pethkool et al. [25]), for the plain tube. 21

2.5

Friction Factor

2.0

Re = 3856

Re = 5784

Re = 9640

Re = 11568

Re = 7712

No pulsation 1.5

1.0

0.5

0.0 0

2

4

6

8

10

Pitch to wire diameter ratio (P/dw) 22

12

Fig. 5: The average friction factor versus pitch ratio for different Reynolds numbers (without pulsation).

2.5

P/dw= 3 P/dw = 6 No pulsation

P/dw = 10

Plain tube

Friction Factor

2.0

1.5

1.0

0.5

0.0 0

5000

10000 Re

23

15000

Fig. 6: The average friction factor versus Reynolds number for plain and rough annulus with different pitch ratios (without pulsation). 14.0

14.0

Re = 3856 Re = 9640

Re = 7712

Re = 3856 Re = 9640 12.0

Pitch ratio = 3.0 Stroke ratio = 1.6

10.0

Nu/Nuow

Nu/Nuow

12.0

Re = 5784 Re = 11568

8.0

Pitch ratio = 3.0 Stroke ratio = 2.6

8.0

6.0

4.0

4.0

2.0

2.0

0.0

100

150

200

250

300

100

150

200

Frequency (cpm)

12.0

300

(b) S/Di = 2.6

Re = 3856 Re = 5784 Re = 9640 Re = 11568 Pitch ratio = 3.0 Stroke ratio = 4.2

14.0

Re = 7712

Re = 3856 Re = 5784 Re = 9640 Re = 11568 Pitch ratio = 3.0 Stroke ratio = 5.0

12.0

10.0

Re = 7712

Nu/Nuow

10.0

8.0

8.0

6.0

6.0

4.0

4.0

2.0

2.0

0.0 100

250

Frequency, (cpm)

(a) S/Di = 1.6 14.0

Nu/Nuow

Re = 7712

10.0

6.0

0.0

Re = 5784 Re = 11568

0.0 150

200

250

300

Frequency, (cpm)

100

150

200

250

Frequency, (cpm)

(c) S/Di = 4.2

(d) S/Di = 5.0 24

300

Fig. 7: The average Nusselt number ratio versus frequency for rough annulus at P/dw= 3 for different stroke ratios and Reynolds numbers.

14.0

14.0 Re = 3856 Re = 9640

Re = 7712

Re = 3856 Re = 9640 12.0

Pitch ratio = 6.0 Stroke ratio = 1.6

10.0

Nu/Nuow

Nu/Nuow

12.0

Re = 5784 Re = 11568

8.0

Pitch ratio = 6.0 Stroke ratio = 2.6

8.0

6.0

4.0

4.0

2.0

2.0

0.0 100

150

200

250

300

100

150

Frequency (cpm)

200

250

300

Frequency, (cpm)

(a) S/Di = 1.6

(b) S/Di = 2.6 14.0

14.0 Re = 3856 Re = 9640 12.0

Re = 5784 Re = 11568

Re = 3856 Re = 9640

Re = 7712

12.0

Pitch ratio = 6.0 Stroke ratio = 4.2

Re = 5784 Re = 11568

Re = 7712

Pitch ratio = 6.0 Stroke ratio = 5.0

10.0

Nu/Nuow

10.0

Nu/Nuow

Re = 7712

10.0

6.0

0.0

Re = 5784 Re = 11568

8.0

8.0

6.0

6.0

4.0

4.0

2.0

2.0

0.0

0.0 100

150

200

250

100

300

150

200

Frequency, (cpm)

Frequency, (cpm)

(c) S/Di = 4.2

(d) S/Di = 5.0 25

250

300

Fig. 8: The average Nusselt number ratio versus frequency for rough annulus at P/dw= 6 for different stroke ratios and Reynolds numbers.

16.0

16.0 Re = 3856 Re = 9640 14.0

Re = 5784 Re = 11568

Re = 3856 Re = 9640

Re = 7712

14.0

Pitch ratio = 10 Stroke ratio = 1.6

12.0

Re = 7712

Pitch ratio = 10 Stroke ratio = 2.6

Nu/Nuow

Nu/Nuow

12.0

10.0

10.0

8.0

8.0

6.0

6.0

4.0

4.0

2.0

2.0

0.0

0.0 100

150

200

250

300

100

Frequency (cpm)

14.0

Re = 3856 Re = 9640

Re = 5784 Re = 11568

200

250

300

(b) S/Di = 2.6 16.0

Re = 7712

Re = 3856 Re = 9640 14.0

Pitch ratio = 10 Stroke ratio = 4.2

Re = 5784 Re = 11568

Re = 7712

Pitch ratio = 10 Stroke ratio = 5.0

12.0

Nu/Nuow

12.0

10.0

10.0

8.0

8.0

6.0

6.0

4.0

4.0

2.0

2.0

0.0

0.0

100

150

Frequency, (cpm)

(a) S/Di = 1.6 16.0

Nu/Nuow

Re = 5784 Re = 11568

150

200

250

100

300

150

200

250

Frequency, (cpm)

Frequency, (cpm)

(c) S/Di = 4.2

(d) S/Di = 5.0 26

300

Fig. 9: The average Nusselt number ratio versus frequency for rough annulus at P/dw= 10 for different stroke ratios and Reynolds numbers.

16.0

P/dw = 3.0

P/dw = 6.0

16.0

P/dw = 10.0

P/dw = 3.0

Stroke ratio = 1.6, Frequency = 140 cpm

12.0

12.0

10.0

10.0

Nu/Nuow

14.0

Nu/Nuow

14.0

8.0

6.0

4.0

4.0

2.0

2.0

0.0 0

5000

10000

15000

0

Re

(a) f = 140 cycles/min 16.0

5000

Re

10000

15000

(b) f = 170 cycles/min 16.0

P/dw = 3.0 P/dw = 6.0 P/dw = 10.0 Stroke ratio = 1.6, Frequency = 230 cpm

14.0

14.0

12.0

12.0

10.0

10.0

Nu/Nuow

Nu/Nuow

P/dw = 10.0

8.0

6.0

0.0

P/dw = 6.0

Stroke ratio = 1.6, Frequency = 170 cpm

8.0

8.0

6.0

6.0

4.0

4.0

2.0

2.0

0.0

P/dw = 3.0 P/dw = 6.0 P/dw = 10.0 Stroke ratio = 1.6, Frequency = 260 cpm

0.0

0

5000

10000

15000

0

Re

(c) f = 230 cycles/min

5000

Re

10000

(d) f = 260 cycles/min 27

15000

Fig. 10: The average Nusselt number versus Reynolds number for wire coiled with different pitches and different frequencies at stroke ratio 1.6.

16.0

f=140

14.0

f=170

f=200

f=230

f=260

Pitch ratio = 3.0, Stroke ratio = 1.6

12.0

Nu/Nuow

10.0 8.0 6.0 4.0 2.0 0.0 0

5000

10000

15000

Re Fig. 11: The average Nusselt number ratio versus Reynolds number for wire coiled at P/dw = 3.0 with different frequencies and S/Di = 1.6.

28

16.0

f=140

14.0

f=170

f=200

f=230

f=260

Pitch ratio = 6.0,Stroke ratio = 1.6

12.0

Nu/Nuow

10.0 8.0 6.0 4.0 2.0 0.0 0

2000

4000

6000

8000

10000

12000

14000

Re

Fig. 12: The average Nusselt number ratio versus Reynolds number for wire coiled at P/dw = 6.0 with different frequencies and S/Di = 1.6.

29

16.0

f=140

14.0

f=170

f=200

f=230

f=260

Pitch ratio = 10,Stroke ratio = 1.6

12.0

Nu/Nuow

10.0 8.0 6.0 4.0 2.0 0.0 0

5000

Re

10000

15000

Fig. 13: The average Nusselt number ratio versus Reynolds number for wire coiled at P/dw = 10 with different frequencies and S/Di = 1.6.

30

14.0 S/di = 1.6 12.0

S/di = 2.6

S/di = 3.4

S/di = 4.2

S/di = 5.0

Pitch ratio = 3.0, Frequency = 140 cpm

Nu/Nuow

10.0 8.0 6.0 4.0 2.0 0.0 0

2000

4000

6000

8000

10000

12000

14000

Re Fig. 14: The average Nusselt number ratio versus Reynolds number for wire coiled at P/dw=3.0and frequency =140 cpm.

31

14.0

S/di = 1.6

12.0

S/di = 2.6

S/di = 3.4

S/di = 4.2

S/di = 5.0

Pitch ratio = 6.0, Frequency = 140 cpm

Nu/Nuow

10.0 8.0 6.0 4.0 2.0 0.0 0

2000

4000

6000

8000

10000

12000

Re Fig. 15: The average Nusselt number ratio versus Reynolds number for wire coiled at P/dw=6.0 and frequency =140 cpm.

32

14000

14.0

S/di = 1.6

S/di = 2.6

S/di = 3.4

S/di = 4.2

Pitch ratio = 10, Frequency = 140 cpm

12.0

S/di = 5.0

Nu/Nuow

10.0 8.0 6.0 4.0 2.0 0.0 0

2000

4000

6000

8000

10000

12000

14000

Re Fig. 16: The average Nusselt number ratio versus Reynolds number for wire coiled at P/dw=10 and frequency =140.

33

16.0

(Nu/Nuow)Experimental

14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 0.0

2.0

4.0

6.0

8.0

10.0

(Nu/Nuow)Calculated

34

12.0

14.0

16.0

Fig. 17: Comparison between the experimental values of ( Nu / Nuow ) with the values computed from correlation.

35

Highlights

Ø Study is aimed to investigate experimentally the influence of pulsation with different amplitudes on the heat transfer rates in a concentric tube equipped with coiled wires.

Ø The investigation is performed for pulsating turbulent water flow in a double-pipe heat exchanger with cold water in the annulus space for counter flows.

Ø The combined effect of both pulsation and coiled wires on heat transfer rate was more significant compared to smooth annulus.

Ø Pulsation with high frequency and high amplitudes in a concentric tube equipped with coiled wires enhances heat transfer rates significantly.

36