International Communications in Heat and Mass Transfer 38 (2011) 4956

Contents lists available at ScienceDirect

International Communications in Heat and Mass Transfer

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i c h m t

Collector efciency of upward-type double-pass solar air heaters with ns attached

Chii-Dong Ho , Ho-Ming Yeh, Tsung-Ching Chen
Energy and Opto-Electronic Materials Research Center, Department of Chemical and Materials Engineering, Tamkang University, 151 Ying-chuan Road, Tamsui, Taipei County, Taiwan 251

a r t i c l e

i n f o

a b s t r a c t
The collector efciency of upward-type double-pass at plate solar air heaters with ns attached and external recycle is investigated theoretically. The double-pass device was constructed by inserting the absorbing plate into the air conduit to divide it into two channels (the upper and lower channels). The double-pass device introduced here was designed for creating a solar collector with heat transfer area double as well as the extended area of ns between the absorbing plate and heated air. Moreover, the advantage of external recycle application to solar air heaters is the enhancement of forced heat convection strength, resulting in considerable device heat transfer performance improvement. This advantage may compensate for the remixing at the inlet which decreases the heat transfer transfer-driving force decrement (temperature difference). 2010 Elsevier Ltd. All rights reserved.

Available online 19 November 2010 Keywords: Solar air collector Heat transfer Collector efciency Internal ns

1. Introduction A at plate solar air heater differs from more conventional heat exchangers in several respects. The latter usually employ a uid to exchange high heat transfer rates using conduction and convection. In solar air heaters, energy is transferred from a distant source of radiant energy directly into air [1,2]. The heat may then be utilized by passing air through a conduit system located between the bottom and absorbing plate. The heated air is subsequently used for space heating and drying [3,4]. In its simplest form, a at plate solar air heater consists of one or more sheets of glass or transparent material situated above an absorbing plate with the ambient air owing either over or under the absorbing [5], so it acts as a black body to absorb heat. The sun's rays pass through the glass and are trapped in the space between the covers and plate or absorbed into the black body. Except for the glass covers, all parts of a solar air heater are well insulated to make the energy loss as small as possible. The glass covers are employed to reduce convection and radiation losses into the atmosphere. In addition to the essential effects of free and forced convection [6], there are many ways to achieve considerable improvement in collection efciency by increasing the transfer area with internal ns attached [7,8], creating turbulence inside the ow channel using bafes [9,10] or designing corrugated surfaces [11,12], and enhancing the convective transfer rate [13]. Recycle-effect applications in the design and operation of equipment with external or internal reux can effectively enhance the heat and mass transfer rate, leading to improved performance such as air lift reactors, were conrmed by many investigators [14,15], loop reactors
Communicated by W.J. Minkowycz. Corresponding author. E-mail address: cdho@mail.tku.edu.tw (C.-D. Ho). 0735-1933/$ see front matter 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.icheatmasstransfer.2010.09.015

[16], draft-tube bubble columns [17] and heat and mass exchangers [1820], which are widely applied to absorption, fermentation, polymerization, and heat and mass transfer operations. Two conicting effects exist in a recycle operation. The rst is increasing uid velocity, resulting in convective heat or mass transfer enhancement. The second is the decrease in driving force (temperature or concentration difference) due to remixing. Considerable improvement in collector efciency of solar air heaters with ns attached is obtained by employing such a double-pass device due to the convective heat transfer rate increment as compensation for the driving force decrement, instead of using a single-pass device and operating at the same total ow rate. The purpose of this study is to investigate the inuence of the external-recycling effect on the performance in an upward-type at plate solar air heater with ns attached. 2. Mathematical model The structure of an upward-type at plate solar air heater with internal ns attached and operated with external recycle is illustrated by the schematic diagram of Figs. 1 and 2. A black absorbing plate was welded at the center of the collector with air ow channels above and below the metal absorber plate. The designed solar air heater consists of two glass covers, an absorbing plate, a recycling channel with well insulation, and a recycle device was situated at the end of upper ow channel. Before entering the upper channel, the uid of the mass ow rate m and inlet temperature Tf,i mix with the uid exiting from the upper channel of the mass ow rate Rm regulated by a blower situated at the end of the upper channel. The overall heat loss coefcient, in which the edge and bottom heat losses were neglected, was estimated by an empirical correlation from an absorbing plate across a static space between two glass covers to ambient surrounding. The steady-state energy balance will be taken under the following assumptions: the

50

C.-D. Ho et al. / International Communications in Heat and Mass Transfer 38 (2011) 4956

Nomenclature Ac Af Af,b B Cp De F FR H h1 hw I I0 k ks L n Nu Qu R Re Ta Tf Tf,i Tf,o 0 Tf, i Tf,m Tp Tp,m t Ut u V w1 w2 z surface area of the absorbing plate, m2 total surface area of ns, m2 total cross-sectional area of ns, m2 width of the air tunnel in the solar collector, m specic heat of air at constant pressure, J/kg K equivalent diameter of the air tunnel, m efciency factor of the solar air heater, dened in Eq. (7) heat-removal factor for the solar air heater height of the air tunnel in the solar collector, m convective heat transfer coefcient for uid owing over a at plate, W/m2 convective heat transfer coefcient between glass cover and ambient, W/m2 improvement in collector efciency solar radiation incident, W/m2 thermal conductivity of air, W/m2 K thermal conductivity of absorbing plate, W/m2 K collector length, m mass ow rate of air, kg/s number of ns attached Nusselt number useful gain of energy carried away by air per unit time, W reux ratio Reynolds number ambient temperature, K temperature distribution for the air ow, K inlet air ow temperature, K outlet air ow temperature, K mixed air ow temperature at the beginning of the upper channel, dened in Eq. (18), K mean air ow temperature, K absorbing plate temperature, K mean absorbing plate temperature, K thickness of the n, m loss coefcient from the top of absorbing plate to the ambient, W/m2 K mean air ow velocity, m/s wind velocity, m/s distance between ns, m height of the ns, m axis along the ow direction, m

temperature of the absorbing plate, bottom plate and bulk uids are functions of the ow direction only, and both the glass covers and uids do not absorb radiant energy. Except for the glass covers, all parts of the solar air collector outside surface and the recycling ow channel are well insulated. 2.1. Temperature distribution for the uid in the ow direction The steady-state energy balance for differential sections of the absorbing plate, bottom plate and uid are, respectively.     2 I0 = h Tp Tf + Ut Tp Ta h i dT   f = h1 B Tp Tf m1 + RCp dz 1

where the dimensionless quantity and n efciency f are dened with collector surface area Ac and total surface area of ns Af as   = 1 + Af f Af ;b = Ac f = tanhMw2 = Mw2 in which M = p h1 2L + 2t = ks Lt , from Eq. (1) 5 3

h  i 2 Tp Tf = I0 Ut Tf Ta = h1 + Ut Substituting Eq. (5) into Eq. (2), one has  j  k F B Ut Tf Ta I0 2 dTf h i =0 + dz m1 + RC

p

Where F = h1 = h1 + Ut Eq. (6) can be easily integrated for the boundary condition Tf = Tf ;i
o

Greek letters absorptivity of the absorbing plate g Emissivity of glass cover p Emissivity of absorbing plate Collector efciency f n efciency 0 obtained without recycle air viscosity, kg/s m air density, kg/m3 the Stefan Boltzmann constant, W/m2 K4 transmittance of glass cover dimensionless quantity dened by Eq. (3) Subscript a g f i o p ambient glass cover uid inlet outlet absorbing plate

at

z=0

The result is F Ut Bz = exp m1 + RCp Tfo;i Ta I0 2 = Ut Tf Ta I0 2 = Ut " # 9

Eq. (9) is the temperature distribution of the bulk uid along the ow direction. Thus, the uid temperature at the outlet is readily obtained from Eq. (9) by substituting the condition: Tf = Tf, o at z = L. The result is Tf ;o Ta I0 2 = Ut " # F Ut Ac = exp m1 + RCp Tfo;i Ta I0 2 = Ut

10

Superscript o mixed

where Ac = BL, surface area of the absorbing plate.

C.-D. Ho et al. / International Communications in Heat and Mass Transfer 38 (2011) 4956

51

I0 (insolation) Ut Glass cover

Glass cover

Air in

T f0,i

T f ,i

(R+1)m

H Tp Absorbing plate

m T f ,o

Air out

T f ,o
Recycling channel z=0 Insulation
Fig. 1. Schematic diagram of a solar air heater.

Rm z=L

2.2. Collector efciency and uid outlet temperature Once the inlet and outlet temperatures are known, the useful gain may be estimated from the relation   o Q u = m1 + RCp Tf ;o Tf ;i or   Q u = mCp Tf ;o Tf ;i Substitution of Eq. (10) into Eq. (11) results in h  i 2 o Q u = FR Ac I0 Ut Tf ;i Ta where the heat-removal factor is j kn j ko FR = m1 + RCp = Ut Ac 1 exp F Ut Ac = m1 + RCp 14 13 12 11

Finally, the collector efciency may be dened as Q u Useful gain of energy carried away by air I0 Ac Total solar radiation incident

15

h   i 2 o = FR Ut Tf ;i Ta = I0

16

o Inspection of Eq. (16) shows that the mixed inlet temperature Tf, i because the recycling is not specied a priori. Mathematically, one more relation for the mixing effect at the inlet is needed to determine this value. For this reason, let take an energy balance at the inlet with the inlet temperature Tf, i as the reference temperature:

      o mCp Tf ;i Tf ;i + mRCp Tf ;o Tf ;i = m1 + RCp Tf ;i Tf ;i or  h i o Tf ;i = Tf ;i + R = 1 + R T f ;o Tf ;i

0 Substituting Eq. (18) into Eq. (16) to eliminate Tf, i, we have

17

18

n    o 2 = FR Ut = I0 R = 1 + R Tf ;o Tf ;i + Tf ;i Ta

19

There is an alternative denition of the collector efciency along Eq. (15), i.e.   mCp Tf ;o Tf ;i I0 Ac

20

Flow channel Recycling channel

Substituting Eq. (20) into Eq. (19) to eliminate Tf, o, one has the expression of in terms of the known inlet temperature Tf,i, instead of 0 the unknown mixing temperature Tf, i, as h   i 2 FR Ut Tf ;i Ta = I0  h i = 1 + FR Ut Ac = mCp R = 1 + R

Glass covers

Fins Absorbing plate

Insulation

21

Fig. 2. Solar air heater with internal ns attached.

52 Table 1 Physical properties of air at 1 atm [3]. T (K) 273 293 313 333 353 (kg/m3) 1.292 1.204 1.127 1.059 0.999 Cp (J/kg K) 1006 1006 1007 1008 1010

C.-D. Ho et al. / International Communications in Heat and Mass Transfer 38 (2011) 4956

k (W/m K) 0.0242 0.0257 0.0272 0.0287 0.0302

105 (kg/m s) 1.72 1.81 1.90 1.99 2.09

Once the collector efciency is determined, the uid outlet temperature is readily obtainable from Eq. (20), i.e.   Tf ;o = Tf ;i + I0 Ac = mCp 2.3. Mean uid and absorbing plate temperatures in terms of The mean uid and absorbing plate temperatures are needed for calculating the heat transfer coefcients. The mean uid temperature may be dened as 1 L T zdz L 0 f 23
Fig. 4. Collector efciency for I0 = 1100 W/m2.

22

Tf ;m =

Substituting Eq. (9) into Eq. (23) and integrating, one has
o Tf ;i

Tf ;m =

+ I0 = Ut FR 1 FR = F

24

Another expression of in terms of mean absorbing plate temperature Tp, m is dened as   2 = Ut Tp;m Ta = I0 26

o Substituting Eqs. (18) and (20) into Eq. (24) to eliminate Tf, i and Tf, o yields

Tf ;m = Tf ;i

  + R = 1 + R I0 Ac = mCp + I0 = Ut FR 1 FR = F 25

Equating Eqs. (20) and (26), we have    mCp Tf ;o Tf ;i = Ta + I0 = Ut Ac Ut 

2

Tp;m

27

Fig. 3. Collector efciency for I0 = 830 W/m2.

Fig. 5. Outlet air temperature for I0 = 830 W/m2.

C.-D. Ho et al. / International Communications in Heat and Mass Transfer 38 (2011) 4956

53

following the basic procedure of Hottel and Woertz [23]. For the horizontal collector shown in Fig. 1   2 Tp;m =520 1 Ut = 2 30:431100 = T + h > > p;m > T T w> > > > > a > >4 p;m > > 5 > > ; : 2+f    2 2 Tp;m + Ta Tp;m + Ta +  1 h i + 2 2 + f 1 + 0:133p = g 2 p + 2 0:00591hw 30 where f = (1 + 0.089hw 0.1166hwp)(1 + 0.07866 2). In the study of solar air heaters and collector-storage walls, it is necessary to know the forced convection heat transfer coefcient between two at plates. For air, the following correlation [24] may be derived from Kays data for fully developed turbulent ow with one side heated and the other side insulated: Nu = h1 De = k = 0:0158Re
Fig. 6. Outlet air temperature for I0 = 1100 W/m2.
0:8

8 > > > > > > > <

91 > > > > > > > =

31

The characteristic length is the equivalent diameter of the duct De = 4HB = 2B + H ; 28 32

Substituting Eq. (22) into Eq. (27) to eliminate Tf, o results in   2 Tp;m = Ta + I0 = Ut

The Reynolds number for the rectangular duct is then dened by 4HB = 2B + H m1 + R = BH De u 2 m1 + R = = B + H 33

Eqs. (24) and (28) are the expressions of Tf, m and Tp, m in terms of . 2.4. Heat transfer coefcients The convective heat transfer coefcient hw for air owing over the outside surface of the glass cover depends primarily on the wind velocity V. McAdams [21] obtained experimental result as hw = 5:7 + 3:8V 29

Re =

2.5. Calculation method for and Tf, o The procedure for the calculation of theoretical values of collector efciency and outlet uid temperature will now be described. First, with known collector geometries (L, B, H, n, t, w1, w2) and system properties , CP, , , k, ks, P, g), as well as the given operating (, conditions I0 ; Ta ; V; m; R; Tf ;i , a temporary value of is estimated from Eq. (21) once TP, m is assumed. The value of TP, m is then checked using Eq. (28) and the new value of TP, m may be obtained. If the

where the units of hw and V are W/m2K and m/s, respectively. An empirical equation for the loss coefcient from the top of the solar collector to the ambient Ut was developed by Klein [22]
Table 2 The improvement of collector performance for I0 = 830 W/m2 and n = 12. Tf,i (K) 288 R m = 0.01 kg/s Tpm (K) 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5 370.89 356.31 347.77 342.06 337.93 334.78 372.49 358.56 350.42 344.96 341.02 338.01 374.08 360.81 353.05 347.86 344.09 341.23 Tfm (K) 292.29 296.44 298.62 299.99 300.94 301.64 297.13 301.13 303.24 304.56 305.49 306.17 301.97 305.82 307.85 309.13 310.02 310.68 Tf,o (K) 296.48 299.23 300.73 301.69 302.37 302.88 301.17 303.82 305.27 306.21 306.87 307.36 305.85 308.40 309.81 310.72 311.36 311.83 (%) 28.59 37.84 42.90 46.15 48.44 50.14 27.53 36.46 41.36 44.52 46.74 48.39 26.46 35.07 39.80 42.86 45.01 46.62 I (%) 0.00 32.36 50.07 61.43 69.41 75.36 0.00 32.43 50.24 61.69 69.75 75.77 0.00 32.51 50.41 61.95 70.09 76.17 m = 0.015 kg/s Tpm (K) 361.01 345.34 336.81 331.33 327.49 324.62 363.04 348.08 339.93 334.70 331.02 328.28 365.07 350.82 343.05 338.06 334.55 331.93 Tfm (K) 291.48 294.58 296.09 297.00 297.61 298.05 296.36 299.35 300.81 301.69 302.28 302.71 301.23 304.11 305.52 306.37 306.95 307.37

m = 0.02 kg/s Tf,o (K) 294.91 296.76 297.70 298.28 298.67 298.96 299.66 301.45 302.36 302.93 303.31 303.59 304.40 306.14 307.02 307.57 307.94 308.22 (%) 34.94 44.30 49.04 51.96 53.94 55.39 33.66 42.73 47.34 50.17 52.12 53.54 32.37 41.13 45.60 48.36 50.26 51.65 I (%) 0.00 26.79 40.36 48.69 54.37 58.52 0.00 26.91 40.61 49.04 54.81 59.03 0.00 27.04 40.86 49.40 55.25 59.53 Tpm (K) 353.50 337.68 329.51 324.42 320.90 318.32 355.87 340.76 332.95 328.08 324.71 322.24 358.23 343.83 336.38 331.73 328.51 326.14 Tfm (K) 290.95 293.41 294.54 295.21 295.64 295.95 295.85 298.22 299.32 299.97 300.39 300.69 300.74 303.03 304.10 304.72 305.13 305.43 Tf,o (K) 293.87 295.21 295.85 296.23 296.49 296.68 298.65 299.95 300.58 300.96 301.21 301.39 303.44 304.70 305.31 305.68 305.93 306.10 (%) 39.54 48.57 52.90 55.49 57.22 58.47 38.11 46.88 51.11 53.64 55.34 56.57 36.66 45.16 49.28 51.75 53.42 54.63 I (%) 0.00 22.85 33.81 40.36 44.74 47.90 0.000 23.01 34.11 40.76 45.22 48.45 0.00 23.17 34.40 41.15 45.70 48.99

293

298

54

C.-D. Ho et al. / International Communications in Heat and Mass Transfer 38 (2011) 4956

Table 3 The improvement of collector performance for I0 = 1100 W/m2 and n = 12. Tf,i (K) 288 R m = 0.01 kg/s Tpm (K) 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5 393.46 376.06 365.60 358.48 353.28 349.28 394.90 378.15 368.08 361.24 356.23 352.39 396.34 380.24 370.57 363.99 359.18 355.49 Tfm (K) 293.45 298.88 301.79 303.63 304.92 305.87 298.29 303.56 306.39 308.19 309.44 310.37 303.13 308.24 310.99 312.74 313.96 314.87 Tf,o (K) 298.79 302.48 304.53 305.85 306.79 307.49 303.47 307.05 309.05 310.35 311.26 311.95 308.15 311.62 313.57 314.83 315.73 316.40 (%) 27.45 36.81 42.03 45.40 47.78 49.56 26.63 35.73 40.81 44.11 46.44 48.18 25.81 34.64 39.59 42.80 45.08 46.78 I (%) 0.00 34.12 53.13 65.43 74.11 80.59 0.00 34.18 53.26 65.64 74.38 80.93 0.00 34.23 53.39 65.85 74.66 81.27 m = 0.015 kg/s Tpm (K) 381.74 362.58 351.86 344.90 339.96 336.26 383.60 365.16 354.85 348.15 343.39 339.82 385.47 367.75 357.84 351.39 346.81 343.38 Tfm (K) 292.47 296.56 298.59 299.81 300.64 301.24 297.34 301.32 303.29 304.49 305.30 305.89 302.21 306.07 307.99 309.16 309.95 310.53 Tf,o (K) 296.87 299.40 300.69 301.49 302.04 302.44 301.61 304.07 305.34 306.12 306.66 307.06 306.35 308.74 309.98 310.75 311.28 311.66 (%) 33.85 43.48 48.42 51.47 53.56 55.08 32.86 42.24 47.07 50.06 52.11 53.62 31.85 40.98 45.70 48.63 50.65 52.12 I (%) 0.00 28.43 43.04 52.05 58.21 62.72 0.00 28.54 43.25 52.36 58.60 63.17 0.00 28.64 43.47 52.67 58.99 63.62 m = 0.02 kg/s Tpm (K) 372.64 352.97 342.56 336.00 331.45 328.08 374.84 355.91 345.88 339.56 335.17 331.92 377.04 358.85 349.20 343.11 338.88 335.75 Tfm (K) 291.82 295.08 296.60 297.50 298.09 298.51 296.71 299.88 301.37 302.25 302.83 303.24 301.59 304.68 306.13 306.99 307.56 307.97 Tf,o (K) 295.58 297.42 298.32 298.85 299.21 299.47 300.36 302.16 303.04 303.56 303.92 304.17 305.14 306.89 307.75 308.27 308.62 308.87 (%) 38.55 47.92 52.47 55.19 57.01 58.33 37.43 46.59 51.05 53.72 55.52 56.82 36.30 45.23 49.60 52.23 54.01 55.29 I (%) 0.00 24.32 36.11 43.16 47.89 51.31 0.00 24.45 36.37 43.52 48.33 51.80 0.00 24.59 36.63 43.87 48.76 52.30

293

298

calculated value of TP, m is different from the assumed value, continued calculations by iteration are needed until the last assumed values meet the nally calculated values, and thus the corresponding value of are also nally obtained. Once the correct value of is obtained, the outlet uid temperature is readily calculated from Eq. (22).

for collector efciency and outlet air temperature were obtained. The results are plotted in Figs. 36 and also given in Tables 25. 3.2. Improvement in device performance by recycle The improvement in collector efciency I by operating with recycling is best illustrated by calculating the percentage increase in collector efciency based on operating without recycling I= 0 0 34

3. Results and discussions 3.1. Theoretical predictions of collector efciencies and outlet temperatures The method for theoretical predictions of collector efciencies from Eq. (21) as well as the use of the empirical equations for calculating the heat transfer coefcients has been described in the previous work [7]. The improvement in performance of a solar air heater by operating with external recycling may be illustrated numerically using Table 1. The following design and operating conditions: A c = LB = 0.36 m; B=nw1 =0.6 m; H=0.05 m; I0 =830 and 1100 W/m2; L=B=0.6 m; m =0.01, 0.015 and 0.02 kg/s; n=12 and 24; Ta =283 K; Tf,i =288, 293 and 298 K; t=0.001 m; V=1 m/s; w1 =0.05 and 0.025 m; =0.95; g =0.94; p =0.95; =5.6710 8 W/m2 K4; =0.875. By substituting the specied values into the appropriate equations, theoretical predictions
Table 4 The improvement of collector performance for I0 = 830 W/m2 and n = 24. Tf,i (K) 288 R m = 0.01 kg/s Tpm (K) 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5 360.49 345.99 338.29 333.41 330.00 327.47 362.55 348.72 341.36 336.70 333.44 331.02 364.61 351.43 344.42 339.98 336.87 334.56 Tfm (K) 293.30 297.80 299.94 301.21 302.07 302.68 298.10 302.45 304.52 305.76 306.58 307.18 302.91 307.10 309.10 310.29 311.09 311.67 Tf,o (K) 298.46 301.03 302.31 303.09 303.62 304.01 303.08 305.57 306.81 307.57 308.09 308.47 307.69 310.10 311.30 312.04 312.54 312.91 (%) 35.27 43.93 48.24 50.86 52.65 53.95 33.98 42.36 46.54 49.10 50.85 52.12 32.67 40.77 44.82 47.31 49.01 50.25 I (%) 0.00 24.56 36.78 44.23 49.29 52.99 0.00 24.67 36.99 44.52 49.66 53.40 0.00 24.78 37.20 44.82 50.02 53.82 m = 0.015 kg/s Tpm (K) 349.76 335.13 327.86 323.42 320.39 318.17 352.29 338.33 331.38 327.13 324.23 322.10 354.83 341.52 334.89 330.83 328.05 326.02 Tfm (K) 292.17 295.42 296.87 297.69 298.24 298.62 297.02 300.17 301.57 302.37 302.90 303.27 301.87 304.91 306.26 307.04 307.55 307.92

The theoretical values of I for the system of the present interest were also calculated using Eq. (34), and the results are also listed in Tables 25. The device improvements in the collector efciency are substantially improved by operating with recycling, as indicated from these tables and Figs. 36. The enhancement increases with increasing reux ratio R, especially for operating at lower air ow rate m with higher inlet temperature Tf, i. For instance, the improvements in collector efciency for m = 0:01 kg = s, n = 12, Tf, i = 298 K and R = 5 are 76.17 % and 81.27 % for I0 = 830 and 1100 W/m2, respectively. Considerable improvement of performance is achievable if a doublepass device is employed, instead of using a single-pass one.

m = 0.02 kg/s Tf,o (K) 296.26 297.88 298.63 299.08 299.37 299.58 300.96 302.54 303.27 303.71 304.00 304.20 305.66 307.19 307.91 308.33 308.61 308.82 (%) 41.75 49.95 53.75 55.99 57.48 58.54 40.25 48.22 51.93 54.12 55.59 56.64 38.74 46.46 50.08 52.22 53.65 54.68 I (%) 0.00 19.63 28.74 34.10 37.66 40.22 0.00 19.78 29.01 34.46 38.09 40.70 0.00 19.93 29.28 34.81 38.51 41.17 Tpm (K) 342.08 327.97 321.27 317.26 314.57 312.62 344.96 331.47 325.06 321.22 318.64 316.77 347.83 334.97 328.84 325.17 322.70 320.91 Tfm (K) 291.45 293.98 295.06 295.66 296.05 296.32 296.33 298.78 299.82 300.41 300.79 301.05 301.20 303.58 304.59 305.16 305.52 305.78 Tf,o (K) 294.85 295.97 296.46 296.75 296.94 297.07 299.61 300.70 301.18 301.47 301.65 301.78 304.36 305.42 305.90 306.18 306.36 306.49 (%) 46.15 53.70 57.05 58.98 60.24 61.14 44.52 51.88 55.17 57.07 58.32 59.21 42.87 50.03 53.25 55.11 56.34 57.22 I (%) 0.00 16.36 23.62 27.80 30.55 32.50 0.00 16.53 23.92 28.18 30.99 32.99 0.00 16.70 24.20 28.55 31.42 33.47

293

298

C.-D. Ho et al. / International Communications in Heat and Mass Transfer 38 (2011) 4956 Table 5 The improvement of collector performance for I0 = 1100 W/m2 and n = 24. Tf,i (K) 288 R m = 0.01 kg/s Tpm (K) 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5 381.12 363.39 353.74 347.54 343.19 339.93 383.01 365.96 356.67 350.71 346.52 343.39 384.90 368.52 359.60 353.87 349.84 346.82 Tfm (K) 294.81 300.74 303.61 305.33 306.48 307.32 299.61 305.38 308.17 309.85 310.98 311.79 304.41 310.01 312.72 314.36 315.46 316.26 Tf,o (K) 301.44 304.94 306.71 307.79 308.53 309.07 306.05 309.46 311.18 312.24 312.97 313.49 310.65 313.97 315.65 316.69 317.39 317.91 (%) 34.18 43.09 47.58 50.33 52.20 53.57 33.18 41.85 46.24 48.93 50.77 52.12 32.16 40.60 44.88 47.52 49.32 50.64 I (%) 0.00 26.06 39.19 47.24 52.72 56.73 0.00 26.15 39.37 47.49 53.04 57.09 0.00 26.24 39.55 47.74 53.35 57.45 m = 0.015 kg/s Tpm (K) 368.05 349.74 340.44 334.71 330.77 327.88 370.42 352.80 343.85 338.32 334.53 331.75 372.80 355.86 347.25 341.93 338.28 335.59 Tfm (K) 293.40 297.72 299.67 300.79 301.52 302.04 298.25 302.46 304.35 305.45 306.17 306.68 303.09 307.18 309.03 310.10 310.81 311.30 Tf,o (K) 298.70 300.94 301.99 302.61 303.02 303.31 303.40 305.58 306.61 307.22 307.62 307.92 308.09 310.22 311.23 311.82 312.22 312.51 (%) 40.83 49.36 53.36 55.71 57.28 58.40 39.66 48.00 51.92 54.24 55.78 56.89 38.47 46.61 50.45 52.73 54.26 55.35 I (%) 0.00 20.89 30.67 36.44 40.27 43.03 0.00 21.03 30.91 36.76 40.66 43.46 0.00 21.16 31.15 37.08 41.04 43.89 m = 0.02 kg/s Tpm (K) 358.50 340.58 331.92 326.70 323.18 320.62 361.23 343.97 335.62 330.59 327.19 324.72 363.95 347.36 339.31 334.46 331.18 328.80 Tfm (K) 292.50 295.87 297.32 298.13 298.65 299.02 297.37 300.66 302.07 302.87 303.38 303.75 302.24 305.44 306.83 307.60 308.11 308.46 Tf,o (K) 296.93 298.48 299.17 299.57 299.83 300.02 301.67 303.20 303.88 304.28 304.54 304.72 306.42 307.91 308.58 308.97 309.23 309.41 (%) 45.39 53.30 56.83 58.86 60.19 61.13 44.11 51.87 55.34 57.35 58.67 59.61 42.82 50.41 53.83 55.81 57.12 58.05 I (%)

55

293

298

0.00 17.42 25.19 29.66 32.59 34.68 0.00 17.57 25.45 30.00 32.99 35.13 0.00 17.73 25.71 30.34 33.39 35.57

3.3. The effect of operating temperatures on collector efciency The effects of the air inlet temperature Tf, i and incident solar radiation I0 on the collector efciency as well as the air outlet temperature Tf, o of double-pass solar air heaters with ns attached have been investigated theoretically. As seen from Figs. 5, 6 and Tables 25, either Tf, o or Tf, m increases as R increases, but decreases when air ow rate m increases. The same tendency of the mean plate temperature Tp, m with the operating conditions is also observable in Tables 25, while the mean temperature difference between the absorbing plate and owing air, (TP, m Tf, m), decreases with increasing R, increasing m, increase n, or increasing Tf, i, while increases when I0 increases. 4. Conclusions The equations for theoretical prediction of the useful gain Qu and collector efciency as well as outlet temperature Tf, o in a solar air heater with ns attached under external recycle have been well derived from the energy balances with the recycle ratio as a parameter. The double-pass type solar air heaters proposed in the present study have the extended heat transfer area and the strengthened convective heat transfer coefcient, leading to improved thermal performance on the new device. The effect of recycle ratio on the collector efciency of double-pass type solar air heaters with ns attached has been investigated theoretically. The calculation of , Tf, o and Qu may be generally based on Eqs. (21), (22) and (12), respectively. The improvements in collector efciencies of a double-pass type solar air heater with ns attached operating at the same air ow rate, based on a single-pass device of the same working dimension, were calculated by Eq. (34), and the results are presented in Tables 25. The collector efciencies of both device, and 0, increase with air ow rate m, number of ns attached n and incident solar radiation I0. Accordingly, collector efciencyI increases as m, n and I0 increase. Considerable improvement in collector efciency is obtainable if the operation is carried out with external recycling. The enhancement increases with increasing reux ratio, especially for operating at lower air ow rate with higher inlet air temperature. It is shown that the desirable effect of increasing the uid velocity using the recycling operation to overcome the undesirable driving force decreases (temperature difference) for heat transfer due to remixing at the inlet. We can see in Tables 25 that more than 80% improvement in collector efciency is obtained using the recycling operation.

Acknowledgement The authors wish to thank the National Science Council of the Republic of China for its nancial support. References
[1] H.P. Garg, R.S. Adhikari, Performance evaluation of a single solar air heater with n-subcollectors connected in different combinations, International Journal of Energy Research 23 (1999) 403414. [2] M. Belusko, W. Saman, F. Bruno, Performance of jet impingement in unglazed air collectors, Solar Energy 82 (2008) 389398. [3] J.A. Dufe, W.A. Beckman, Solar Engineering of Thermal Processes, Wiley, New York, 1991. [4] T. Koyuncu, Performance of various design of solar air heaters for crop drying applications, Renewable Energy 31 (2006) 10731088. [5] C.H. Liu, E.M. Sparrow, Convectiveradiative interaction a parallel plate channelapplication to air-operated solar collectors, International Journal of Heat and Mass Transfer 23 (1980) 11371146. [6] H.M. Yeh, T.T. Lin, The effect of collector aspect ratio on the collector efciency of at-plate solar air heaters, Energy 20 (1995) 10411047. [7] H.M. Yeh, C.D. Ho, J.Z. Hou, Collector efciency of double-ow solar air heaters with ns attached, Energy 27 (2002) 715727. [8] Mohd.Y.Hj. Othman, YatimB. , SopianK. , Mohd.N. Abu Bakar, Performance analysis of a double-pass photovoltaic/thermal (PV/T) solar collector with CPC and ns, Renewable Energy 30 (2005) 20052017. [9] A. Hachemi, Thermal performance enhancement of solar air heaters, by a fan-blown absorber plate with rectangular ns, International Journal of Energy Research 19 (1995) 567578. [10] H.M. Yeh, C.D. Ho, C.Y. Lin, Effect of collector aspect ratio on the collector efciency of upward type bafed solar air heaters, Energy Conversion and Management 41 (2000) 971981. [11] L. Goldstein Jr., E.M. Sparrow, Experiments on the transfer characteristics of a corrugated n and tube heat exchanger conguration, Journal of Heat Transfer 98 (1976) 2634. [12] W.F. Gao, W.X. Lin, T. Liu, C. Xia, Analytical and experimental studies on the thermal performance of cross-corrugated and at-plate solar air heaters, Applied Energy 84 (2007) 425441. [13] C.D. Ho, C.W. Yeh, S.M. Hsieh, Improvement in device performance of multi-pass at-plate solar air heaters with external recycle, Renewable Energy 30 (2005) 16011621. [14] M. Atenas, M. Clark, V. Lazarova, Holdup and liquid circulation velocity in a rectangular air-lift bioreactor, Industrial and Engineering Chemistry Research 38 (1999) 944949. [15] J. Korpijarvi, P. Oinas, J. Reunaneu, Hydrodynamics and mass transfer in airlift reactor, Chemical Engineering Science 54 (1999) 22552262. [16] E. Santacesaria, M. Di Serio, P. Iengo, Mass transfer and kinetics in ethoxylation spray tower loop reactors, Chemical Engineering Science 54 (1999) 14991504. [17] K.I. Kikuchi, H. Takahashi, Y. Takeda, F. Sugawara, Hydrodynamic behavior of single particles in a draft-tube bubble column, Canadian Journal of Chemical Engineering 77 (1999) 573578. [18] H.M. Yeh, S.W. Tsai, C.L. Chiang, Recycle effects on heat and mass transfer through a parallel-plate channel, AIChE Journal 33 (1987) 17431746.

56

C.-D. Ho et al. / International Communications in Heat and Mass Transfer 38 (2011) 4956 [22] S.A. Klein, Calculation of at-plate loss coefcients, Solar Energy 17 (1975) 7980. [23] H.C. Hottel, B.B. Woertz, Performance of at-plate solar-heat collector, Transactions of the American Society of Mechanical Engineers 64 (1942) 91104. [24] W.M. Kays, M.E. Crawford, Convective Heat-and-Mass Transfer, McGraw-Hill, New York, 1980.

[19] C.D. Ho, H.M. Yeh, W.S. Sheu, An analytical study of heat and mass transfer through a parallel-plate channel with recycle, International Journal of Heat and Mass Transfer 41 (1998) 25892599. [20] C.D. Ho, H.M. Yeh, R.C. Wang, Heat-transfer enhancement in double-pass at-plate solar air heaters with recycle, Energy 30 (2005) 27962817. [21] W.H. McAdams, Heat Transmission, McGraw-Hill, New York, 1954.