Integrated collector storage solar systems with asymmetric CPC reflectors

Integrated collector storage solar systems with asymmetric CPC reflectors

Renewable Energy 29 (2004) 223–248 www.elsevier.com/locate/renene Integrated collector storage solar systems with asymmetric CPC reflectors Y. Tripan...

453KB Sizes 3 Downloads 25 Views

Renewable Energy 29 (2004) 223–248 www.elsevier.com/locate/renene

Integrated collector storage solar systems with asymmetric CPC reflectors Y. Tripanagnostopoulos ∗, M. Souliotis Physics Department, University of Patras, Patra 26500, Greece Received 11 September 2002; accepted 2 June 2003

Abstract New types of ICS solar systems were designed and outdoor tests of experimental models were performed. The systems consist of single cylindrical horizontal water storage tanks placed inside stationary truncated asymmetric CPC reflector troughs of different design. We used high emittance absorber surface, low cost curved reflectors, iron oxide glazing and thermal insulation at the non illuminated tank surfaces, aiming towards cost effective ICS systems with satisfactory heat preservation during the night. Four experimental models of different designs were constructed and tested to determine their performance regarding their mean daily efficiency and thermal losses during the night. The new ICS systems were compared to an ICS system with symmetric CPC reflectors of similar construction and dimensions and also to a typical Flat Plate Thermosiphonic Unit (FPTU). Test results showed that the ICS systems with asymmetric CPC reflectors present almost the same mean daily efficiency and better preservation of hot water temperature during the night, compared to the ICS system with the symmetric CPC reflectors. The comparison with the FPTU system confirmed the satisfied daily operation of all ICS systems and their moderate storage heat preservation during the night. Theoretical results showed acceptable thermal performance of all ICS systems regarding annual operation.  2003 Published by Elsevier Ltd. Keywords: Solar water heaters; Integrated Collector Storage systems; Compound Parabolic Concentrators; Thermal performance



Corresponding author Tel./fax: +30-61-997-472. E-mail address: [email protected] (Y. Tripanagnostopoulos).

0960-1481/$ - see front matter  2003 Published by Elsevier Ltd. doi:10.1016/S0960-1481(03)00195-2

224

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

Nomenclature A Aa Ar B C CR Cp DS DT f1 f2 G Gm La LT Lu M RT r ti tf Ta Tm,a Ti Tf Tm,D ⌬Tm,D ⌬Tm,N ⌬t US VT Vw Wa

coefficient of maximum mean daily efficiency aperture area of systems (m2) absorber (receiver) surface area (m2) coefficient of system daily thermal losses (K⫺1 Wm⫺2) coefficient of system daily thermal losses (K⫺2 W2m⫺4) concentration ratio specific heat of water (Jkg⫺1 K⫺1) ICS system depth (m) cylindrical storage tank diameter (m) focal length of the lower parabolic reflector (m) focal length of the upper parabolic reflector (m) incoming solar radiation intensity (Wm⫺2) mean daily solar radiation intensity (Wm⫺2) ICS system aperture length (m) cylindrical storage tank length (m) mean monthly useful thermal energy (MJm⫺2) water mass of storage tank (kg) radius of cylindrical storage tank (m) reflectance initial time (s) final time (s) ambient temperature (°C, K) mean ambient temperature (°C, K) initial storage water temperature (°C, K) final storage water temparature (°C, K) mean water temperature during daily operation (°C, K) mean water temperature difference during the day (°C, K) mean water temperature difference during the night (°C, K) time interval (s) coefficient of thermal losses during the night (WK⫺1) storage tank volume(l, m3) wind speed (ms⫺1) ICS system aperture width (m)

Greek symbols a ar e hd

ICS system acceptance angle (°, rad) absorptance of receiver surface surface emittance mean daily efficiency

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

r j jm y ym y⬘ y⬘m w wm w⬘ w⬘m

225

water density (kgm⫺3) angle of upper parabolic reflector (°, rad) maximum value of angle j (°, rad) angle of main (lower) parabolic reflector (°, rad) maximum value of angle y (°, rad) angle of upper parabolic reflector of system STS (°, rad) maximum value of angle y⬘ (°, rad) angle of involute reflector (°, rad) maximum value of w (°, rad) angle of upper involute reflector of system STS (°, rad) maximum value of w⬘ (°, rad)

1. Introduction The main types of domestic hot water (DHW) solar systems that can effectively cover needs of about 100–200 l of hot water per day are the flat plate thermosiphonic units (FPTU) and the integrated collector storage (ICS) systems. The profit in efficient heat preservation of hot water storage of thermosiphonic systems has resulted in a widespread application of them. ICS systems are less applied solar water heaters because of higher thermal losses of storage tank during the night, although they are cheaper and more aesthetically attractive than the FPTU in building integration. Thermal protection of water storage tanks of ICS systems is not efficient enough, as the total, or a significant part, of the external surface of it is used for the absorption of solar radiation. Single or double glazing, selective absorber surface coatings and transparent insulating materials are used for the thermal protection against the exposure of solar rays on the external storage surface. Opaque thermal insulation can be placed only on the non illuminated parts of the storage tank surface and vacuum thermal protection is considered effective mainly for ICS systems that use cylindrical water storage tanks of small diameter. The direct connection of ICS systems to water mains is considered effective for their practical use, but storage tanks resistant to water pressure are necessary. Among the first studies on ICS systems, Chinnappa and Gnanalingam [1] present a pressurized solar water heater using tubes, which permits the direct connection of the system to the water mains. Tubular water storage tanks are practical enough compared to flat type water storage, but most of the proposed improvements are considered for systems with a flat type storage tank. Garg [2] suggests an insulating cover during the night to keep the water temperature above 40 °C next morning and Garg and Rani [3] and also Prakash et al. [4] improve flat type ICS systems by using a baffle plate inside it. Later, ICS systems with a flat type water storage tank of triangular instead of rectangular geometry are presented by Sokolov and Waxman [5,6] and Ecevit et al.

226

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

[7,8] and also the use of a thermal diode at the bottom of the system was proposed by Mohamad [9] to reduce thermal losses during the night. In addition to the above, the effective use of transparent insulating material over the absorber of the flat storage tank of ICS systems was proposed by Goetzberger and Rommel [10], Schmidt et al. [11], Garg et al. [12] and Prakash et al. [13]. Regarding ICS systems with a cylindrical storage tank, Schmidt and Goetzberger [14] combine a tubular water storage tank with a reflector of involute geometry and transparent insulating material for thermal losses suppression. Mason and Davidson [15] studied the use of an evacuated hot water storage tank with selective absorbing surface, giving results from the tested evacuated tubular ICS system and Kaptan and Kilic [16] propose the tubular absorbers with baffle plate placed inside them. The compound parabolic concentrator (CPC) type reflectors have been introduced by Winston [17] to improve the performance of low concentration solar collectors. Welford and Winston presented properties on optics of CPC solar collectors [18] and following them, other authors published several studies regarding collectors which were based on the non imagine concentrators. In contrast to CPC type solar collectors, less work has been done regarding ICS systems with CPC reflectors. Kalogirou [19] gives experimental results from an ICS system using two cylindrical storage tanks (of different diameter) inside a symmetric CPC reflector and Smyth et al. [20] suggest the use of a cylindrical vessel with inner sleeves to improve storage tank thermal performance. CPC type ICS systems have been studied in our laboratory aiming to investigate efficient ICS systems. Tripanagnostopoulos and Yianoulis [21] present ICS systems with a horizontal cylindrical storage tank and stationary asymmetric CPC reflector, suggesting the formation of a hot air trap space between absorber and reflector that suppresses storage thermal losses. A second work is by Tripanagnostopoulos et al. [22], where two, connected in series, horizontal cylindrical storage tanks are proposed as alternative designs, which can achieve satisfactory temperature stratification of stored water. In the present paper we give the results of a new group of ICS solar systems based on the use of single cylindrical storage tank, asymmetric CPC reflectors and inverted absorber–reflector geometry. This geometry was suggested by Rabl [23] and Kienzlen et al. [24] for thermal losses suppression of solar thermal collectors with a curved reflector and flat absorber. The new ICS designs aim to achieve satisfactory temperature preservation of hot water during the night. We constructed four ICS prototypes with asymmetric CPC reflectors, which were compared with an ICS system with symmetric CPC reflectors and a commercial FPTU system. This article follows the paper of Tripanagnostopoulos et al. [25] and is an extension of the recent work of Tripanagnostopoulos and Souliotis [26] on ICS systems.

2. Experimental models Most ICS solar systems consist of a vertical tubular water storage tank to achieve an effective water temperature stratification and in these systems the water tempera-

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

227

ture at the storage top is always at a higher level. These devices are designed with curved reflectors of large acceptance angle a, to collect solar radiation from morning to evening and mostly they use involute (a = 180°, concentration ratio CR = 1) reflector geometry. On the other hand, ICS solar systems with a horizontal water storage tank are less effective with water temperature stratification, but they are aesthetically integrated on horizontal building roofs, because of their lower height. In addition, the CPC reflectors of these ICS systems can be of a smaller acceptance angle a (to achieve concentration ratio CR ⬎ 1) and they are suitable for use in low cost modes for the suppression of water storage thermal losses. The studied ICS systems in our laboratory consist of one or two cylindrical water storage tanks, properly mounted in symmetric or asymmetric CPC troughs. The systems with asymmetric CPC reflectors aim to reduce thermal losses by the effective use of inverted absorber–reflector geometry. This method is suggested as a low cost improvement of ICS systems with one horizontal water storage tank, or in systems with two connected in series horizontal tanks. In all cases the CPC reflector has a stationary design, as the significant weight of the ICS system (because of the water in storage) makes their orientation to the sun unpractical. The design principles of the new ICS systems are: (i) The use of single cylindrical storage tank to avoid increase of system cost. (ii) The application of stationary truncated asymmetric CPC reflectors to suppress inverted absorber surface thermal losses of storage tank. (iii) The additional thermal protection of water storage by using thermal insulation on a part of its cylindrical surface. These design principles are aimed towards cost effective ICS systems with improved thermal performance in heat preservation during the night. The presented new ICS systems are called ICS-1, ICS-2, ICS-3 and ICS-4 (Figs. 1–4) and differ in the design of the asymmetric CPC reflector and the thermally insulated part of the cylindrical storage tank. All systems are of horizontal cylindrical absorber, same CPC reflector acceptance angle a = 90°, single glazing and thermal insulation at the non illuminated surfaces of them. The CPC reflectors are truncated enough to achieve a practical system size regarding depth, as we used a reasonable cylindrical storage tank diameter (0.36 m) to obtain a sufficient ratio of stored water volume per aperture area (⬇100 l/m2). In addition, we present the ICS system with symmetric CPC reflectors, called model STS (Single Tank System, Fig. 5), and also the FPTU system, which were used for performance comparison. The systems ICS-1, STS and FPTU were presented in the recently published work [25], but we consider it necessary to include them in the present paper, believing it useful for the reader to get a clearer idea about the comparison of the design and the performance of all systems. In the following paragraphs we give the geometry of all ICS models, which were constructed with cylindrical water storage tank of same diameter DT = 0.36 m and length LT = 1.01 m, which gives a water storage volume VT = 102.8 l. The aperture area Aa of each model depends mainly on the aperture width Wa, as all ICS models

228

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

Fig. 1.

Cross section of experimental model ICS-1.

are of same aperture length La = 0.99 m. The design of the reflector was based on a smaller radius RT (0.17 m instead of 0.18 m), as indicated in Figs. 1–5, to avoid optical errors from the tangent solar rays. In Table 1 we give the corresponding values of the basic parameters of all ICS systems regarding aperture width Wa, aperture area Aa, system depth DS, ratio of water volume per system aperture VT / Aa, absorber surface exposed to solar radiation Ar and system concentration ratio CR = Aa / Ar. In the same table we include the parameters Aa and VT / Aa for system FPTU. Experimental ICS models were about one meter in length but commercial ICS systems can be constructed longer (1.5–2.0 m) to have a larger volume of water per ICS unit (150–200 l). The constructed models were tested with two reflector materials, innox for moderate reflectance (r = 0.68) and aluminized mylar for higher reflectance (r = 0.85) and aiming to low cost absorbers we used iron oxide glazing and high emittance mat black paint (ar = 0.95, e = 0.9). The absorber of the flat plate solar collector (FPTU) was of the same values of ar and e, in order to compare solar systems with the same absorber properties. All horizontal cylindrical storage tanks are of the same volume, have an entrance (bottom) and an exit (top) for the water and we have put vertically Cu-CuNi thermocouples (TCs) at three positions inside each cylindrical storage tank (up, middle, down) to measure water temperature variation. The CPC reflectors were truncated in a similar way in all experimental

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

Fig. 2.

229

Cross section of experimental model ICS-2.

models, considering the same glazing slope (45°) and air gap (~7 cm) between absorber and glazing. In all models we used polyurethane with a mean thickness of 0.05 m to insulate thermally the storage tank and the external surfaces of the CPC reflectors as well as the two side flat reflectors of each system. 2.1. Experimental model ICS-1 In Fig. 1 we show the cross section of the experimental model ICS-1. The main reflector of model ICS-1 consists of the parabolic (AB) and the involute (BC) parts. The small reflector part (C⬘A⬘) at the system top is parabolic. The storage tank is covered by thermal insulation on 1/4 of its cylindrical surface, using an additional fraction of 1/4 of it to form a hot air trap space between the corresponding storage surface and the involute reflector. The distribution of solar radiation on the absorber surface is almost uniform and this results in moderate water temperature stratification. System ICS-1 is aimed mainly at achieving a satisfactory thermal performance during the night, based on the efficient thermal protection of the water storage tank. The intersection between axis BB⬘ and DD⬘ of the corresponding parabolas consists of the acceptance angle a of the system (a = 90°). We consider RT and DT = 2RT, the radius and the diameter of the cylindrical storage tank, w the angle of the involute reflector part (BC) and y, j the angles of the parabolic reflector parts

230

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

Fig. 3.

Cross section of experimental model ICS-3.

Fig. 4.

Cross section of experimental model ICS-4.

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

Fig. 5.

231

Cross section of experimental model STS.

(AB) and (C⬘A⬘) correspondingly. The sections OC and OE determine the maximum angle w, which is taken as wm = 180° in this model. The sections EB, EA and ED, EA⬘ determine the maximum angles ym and jm (rim angles). The parabolic part (AB) has focal length f 1 = [BE] = pRT and rim angle ym = 56°. The corresponding parameters of the parabolic part (C⬘A⬘) are the focal length f2 = [DE] = RT√2(1 + cos45°) / 2 and rim angle jm = 70ⴰ. The geometry of the curved reflectors in Fig. 1, relative to the rectangular axis system O,x,y (with O being the center of the cylindrical tank) and using the parameters RT, w, y and j is the following: Table 1 Basic parameters of all studied solar water heaters System

wa (m)

Aa (m2)

DS (m)

VT/Aa (lm⫺2) Ar (m2)

CR = Aa / Ar

ICS-1 ICS-2 ICS-3 ICS-4 STS FPTU

0.95 1.11 0.94 0.81 0.95

0.94 1.11 0.93 0.80 0.94 2.47

0.79 0.79 0.57 0.57 0.52

109.4 92.6 110.5 128.5 109.4 55.5

1.12 1.13 1.11 1.14 1.11

0.84 0.98 0.84 0.70 0.84

232

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

Parabolic part (AB)

x ⫽ ⫺RT[1 ⫹ 2π siny / (1 ⫹ cosy)] y ⫽ ⫺RT 2π cosy / (1 ⫹ cosy)



with 0ⱕyⱕym

(1)

Involute part (BC)

x ⫽ RT (cos w ⫹ w sinw)



y ⫽ ⫺RT (sinw⫺w cos w)

with 0ⱕwⱕwm

(2)

Parabolic part (C⬘A⬘)



x ⫽ RT [ ⫺ 1 ⫹ (1 ⫹ 冑2) cos j / (1 ⫹ cos j)]

y ⫽ RT(1 ⫹ 冑2) sin j / (1 ⫹ cos j)

p with ⱕjⱕjm 4

(3)

2.2. Experimental model ICS-2 The second ICS system with asymmetric CPC reflectors is shown in cross section in Fig. 2. Model ICS-2 has the same main CPC reflector (ABC) as system ICS-1, but a larger curved reflector (C⬘A⬘) at the system top, which makes it have a wider aperture width compared to system ICS-1. In addition, the upper reflector part of system ICS-2 provides an effective density of solar radiation on the higher part of the storage tank, giving satisfactory water temperature stratification. The main system reflector consists of the parabolic (AB) and the involute (BC) parts. The larger parabolic part (C⬘A⬘) at the system top, compared with that of model ICS-1, gives an increased aperture width Wa = [AA⬘] and the thermal insulation covers the 1/8 of the storage cylindrical surface. The intersection between axis BB⬘ and C⬘F of the corresponding parabolas consists of the acceptance angle a of the system (a = 90°). Focal length f 1 = [BE] = pRT and angles w, wm and y, ym are same to those of model ICS-1 and the rim angle of the upper parabolic part (C⬘A⬘) is jm = 58ⴰ. The corresponding focal length of

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

233

the upper parabolic part is f2 = [C⬘E⬘] = RT√2. The geometry of the curved reflector (ABC) is same to that of model ICS-1 (Eqs. (1) and (2)). Regarding the upper part (C⬘A⬘) we consider the rectangular axis system O,x,y and using the parameters RT and ϕ we have the following equations: Parabolic part (C⬘A⬘)



x ⫽ ⫺RT(冑2 / 2⫺2冑2 cos j / 1 ⫹ cos j)

with 0ⱕjⱕjm

x ⫽ ⫺RT(冑2 / 2⫺2冑2 cos j / 1 ⫹ cos j)

(4)

2.3. Experimental model ICS-3 Model ICS-3 is shown in cross section in Fig. 3. This model has thermally insulated the 1/4 of the total cylindrical surface at the middle of the storage, its main reflector is smaller in length and the system has smaller depth than systems ICS-1 and ICS-2. The upper reflector part (C⬘A⬘) is the same as model ICS-2, aiming towards the achievement of satisfactory water temperature stratification. The acceptance angle a is determined in the same way as in model ICS-2. The focal length [BE] of the parabolic reflector part (BA) f 1 = 3pRT / 4 and the focal length [C⬘E⬘] of the upper parabolic part (C⬘A⬘) is the same as that of model ICS-2. Angles j, jm are the same as model IC-2, but ym and wm are 58.24° and 135° respectively. The geometry of the main curved reflector (ABC) regarding the rectangular axis system O,x,y and the parameters RT, w, y, j is the following: Parabolic part (AB)



x ⫽ ⫺RT[1 ⫹ (3p / 4) siny / (1 ⫹ cos y)] y ⫽ ⫺RT (3p / 4)cos y / (1 ⫹ cos y)

with 0ⱕyⱕym

(5)

Involute part (BC) p p x ⫽ ⫺RT[sin(w⫺ )⫺w cos(w⫺ )] 4 4 p p y ⫽ ⫺RT[cos(w⫺ ) ⫹ wsin(w⫺ )] 4 4



with 0ⱕwⱕwm

(6)

234

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

2.4. Experimental model ICS-4 The fourth ICS system with asymmetric CPC reflector is the model ICS-4, which is shown in Fig. 4 and has a small parabolic reflector (C⬘A⬘) at the system top the same as model ICS-1 and thermal insulation that covers 3/8 of its storage cylindrical surface. Both models ICS-3 and ICS-4 are of smaller system depth than models ICS1 and ICS-2 and the main reflector (ABC) of model ICS-4 is of the same geometry to that of model ICS-3, consisting of a parabolic part (AB) and an involute part (BC). Model ICS-4 has the smallest aperture width Wa = [AA⬘] among the four models, but it has the best thermal protection regarding the storage cylindrical surface. The focal length f 1 = [BE] of the main parabolic reflector is the same as model ICS-3 and the focal length f 2 = [DE] of the upper parabolic reflector part is the same as model ICS-1. Angles ym, wm are the same as those of model ICS-3 and jm the same as model ICS-1. 2.5. Comparative models STS and FPTU We consider that the four ICS systems with the asymmetric CPC reflector should be compared to an ICS system with symmetric CPC reflector and also to a FPTU solar water heater. The cross section of the ICS system with symmetric CPC reflector (model STS) is shown in Fig. 5, consists of two parabolic (AB), (DA⬘) and two involute (BC), (C⬘D) parts, has acceptance angle a = 90o and 1/4 thermally insulated storage tank. The focal lengths of the two parabolas are f 1 = [BE] = f 2 = [DE⬘] = pRT / 2. Angles ym, wm are, respectively 63.91°, 90° with the same values for y⬘m, w⬘m of the upper reflector part. The geometry of the reflector parts (ABC) and (C⬘DA⬘) considering the rectangular axis system O,x,y and the parameters RT, w, y, y⬘, w, w⬘ are the following: Parabolic part (AB)

x ⫽ ⫺RT[1 ⫹ π siny / (1 ⫹ cosy)] y ⫽ ⫺RT π cosy / (1 ⫹ cosy)



with 0ⱕyⱕym

(7)

Involute part (BC)

x ⫽ ⫺RT (sinw⫺w cosw)



y ⫽ ⫺RT (cosw ⫹ w sinw)

with 0ⱕwⱕwm

(8)

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

235

Involute part (C⬘D)

x ⫽ RT (cosw⬘ ⫹ w⬘sinw⬘) y ⫽ RT (sinw⬘⫺w⬘cosw⬘)



with 0ⱕω⬘ⱕω⬘m

(9)

Parabolic part (DA⬘)

x ⫽ RT π cos y⬘ / (1 ⫹ cos y⬘) y ⫽ RT[1 ⫹ π sin y⬘ / (1 ⫹ cos y⬘)]



with 0ⱕy⬘ⱕy⬘m

(10)

The experimental models are distinguished in two groups regarding the use of innox reflector as, ICS-1A, ICS-2A, ICS-3A, ICS-4A and STS-A (group A) or aluminized mylar reflector as, ICS-1B, ICS-2B, ICS-3B, ICS-4B and STS-B (group B), considering the same as all other system parameters. In addition to STS model we used for the performance comparison a commercial FPTU system with Aa = 2.47 m2, VT = 137 l and VT / Aa = 55.5 l / m2. The absorber of the collector was of the same absorptance (ar = 0.95) and emittance (e = 0.9) to those of the ICS systems, in order to compare solar systems with the same absorber properties. In Fig. 6 we present the front side view of the four studied ICS systems ICS-1, ICS-2, ICS-3, ICS-4 as well as of systems STS and FPTU.

Fig. 6.

Front side view of all studied systems.

236

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

3. Experimental study 3.1. Measured and calculated parameters The five ICS systems and the FPTU system were tested outdoors regarding their thermal performance during the day (mean daily efficiency, hd) as well as during the night (coefficient of thermal losses, Us) with system operation without water drain. We measured the incoming solar radiation G (Wm⫺2) by a Kipp & Zonen pyranometer, the wind speed Vw (ms⫺1) by an A100R anemometer and the temperatures T (°C) of stored water and ambient air by using type T (Cu-CuNi) thermocouples. The water temperature was calculated from the three thermocouples, which were vertically placed inside the water storage. All measurements were collected by a CR10X data logger and tests during the day were performed from 6:30 up to 18:30 (noon at Patras is 12:30) and during the night from 18:30 till 6:30 next morning. We consider Ti and Tf the mean initial and final storage water temperature respectively for the operation of systems during the corresponding time intervals. The mean daily efficiency hd is calculated by the formula: hd ⫽ MCp(Tf⫺Ti) / Aa Gm

(11)

where M is the water mass in the cylindrical storage (in kg), Cp is the specific heat of water (Cp = 4180 Jkg⫺1K⫺1) and Gm is the mean daily incoming solar radiation from 6:30 to 18:30 on aperture plane of systems. The mean daily efficiency hd is determined as a function of the ratio ⌬Tm,D / Gm, with ⌬Tm,D = (Ti + Tf) / 2⫺Tm,a and Tm,a the mean ambient temperature, by second degree polynomial fitting: hd ⫽ A ⫹ B(⌬Tm,D / Gm) ⫹ C(⌬Tm,D / Gm)2

(12)

The coefficient A represents the mean daily efficiency of the system considering operation of it for value (Ti + Tf) / 2 = Tm,a and B, C the thermal losses parameters of the system during the daily operation. Regarding thermal losses during the night we can calculate the coefficient of water storage thermal losses Us from the formula: Us ⫽ (r CpVT / ⌬t) ln [(Ti⫺Tm,a) / (Tf⫺Tm,a)]

(13)

where VT is the water volume of system storage, rCp = 4180 kJm K for water, Tm,a is the mean ambient temperature during the night and ⌬t = tf⫺ti = 43,200 s, the time from 18:30 (ti) to 6:30 (tf) next morning. The temperatures Ti and Tf are the initial (18:30) and the final (6:30) mean storage water temperature of the system. The coefficient Us is calculated for various initial water temperatures in the system storage and can be linear fitted as a function of the temperature difference during the night ⌬Tm,N = Ti⫺Tm,a. ⫺3

⫺1

3.2. Experimental results In Figs. 7 and 8 we give the experimental points and the diagrams of all tested systems, regarding their mean daily efficiency hd as a function of the ratio ⌬Tm,D /

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

237

Fig. 7. Mean daily efficiency results of the ICS systems of group A (innox reflector) including test results of systems STS-A and FPTU for comparison.

Fig. 8. Mean daily efficiency results of the ICS systems of group B (aluminized mylar reflector) including test results of systems STS-B and FPTU for comparison.

Gm. The diagrams of Fig. 7 correspond to ICS systems of type A (innox reflector) and Fig. 8 to ICS systems of type B (aluminized mylar reflector). In Figs. 9 and 10 we present the results of the system operation during the night, regarding thermal losses coefficient US as a function of the temperature difference ⌬Tm,N. The results from ICS systems of type A are shown in Fig. 9 and of type B in Fig. 10. From these results we see that for the daily operation, the ICS systems of type A with the asymmetric CPC reflectors present almost the same thermal performance to that of the ICS system with the symmetric CPC reflector (Fig. 9) for values of

238

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

Fig. 9. Experimental results of the thermal losses coefficient during the night of the ICS systems of group A (innox reflector) including test results of systems STS-A and FPTU for comparison.

Fig. 10. Experimental results of the thermal losses coefficient during the night of the ICS systems of group B (aluminized mylar reflector) including test results of systems STS-B and FPTU for comparison.

ratio ⌬Tm,D / Gm⬎ 0.04 KW⫺1m2. Considering the ICS systems of B type we observe that the better reflector contributes to the performance increase of ICS systems and model with symmetric CPC aluminized mylar reflectors (STS-B) achieves in performance the level of mean daily efficiency that corresponds to system FPTU. During the night the use of asymmetric CPC reflectors contributes to lower values of Us, giving the systems better performance in preservation of hot water temperature compared to the system with symmetric CPC reflectors. The aluminized mylar reflector increases daily efficiency, but contributes also to slightly higher thermal losses during

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

239

the night because this surface has higher values of emissivity and reflectivity than these of innox reflector surface. Among the studied ICS systems with asymmetric CPC reflectors, models ICS-1 and ICS-4 present higher mean daily efficiency in the larger range of values of the ratio ⌬Tm,D / Gm and lower thermal losses coefficient Us for the experimentally studied range of ⌬Tm,N. This performance is explained by the better thermal protection of the cylindrical storage tank, for operation during the day and night and by the contribution of the main curved reflector (ABC) to the effective storage water heating during the day. Additionally to the diagrams of Figs. 7–10 we give in Table 2 the calculated values for the mean daily efficiency hd and the coefficient of thermal losses Us during the night, regarding systems with innox reflector (ICS-1A, ICS-2A, ICS-3A, ICS-4A) and aluminized mylar reflector (ICS-1B, ICS-2B, ICS-3B, ICS-4B). Also, we include for comparison the corresponding results of system STS (A and B type) and FPTU. Considering water temperature rise during the day, model ICS-2 presents the best performance among all tested ICS systems and this is due to the larger aperture width and the lower ratio of water storage volume per aperture area. On the contrary, this model presents the highest thermal losses among asymmetric ICS systems because of the less (1/8) thermal insulated cylindrical tank surface. On the other hand, model ICS-4 gives lower water temperature rise, because of the higher ratio of water storage volume per aperture area, but the better thermal protection of the storage tank results in a satisfied preservation of the water temperature during the night. We performed tests for a sequence of 3–5 days without water drain to determine the water temperature variation. The most important results regarding four days operation are shown in the diagrams of Figs. 11 and 12 for ICS-2 (A and B type) and in Figs. 13 and 14 for ICS-4 (A and B type). We give the results regarding temperature rise for the above Table 2 Experimental results of all tested systems System

Mean daily efficiency hd

Thermal losses coefficient US (WK⫺1)

ICS-1A ICS-2A ICS-3A ICS-4A STSA ICS-1B ICS-2B ICS-3B ICS-4B STSB FPTU

hd=0.60–4.96(⌬Tm,D/Gm)⫺5.54(⌬Tm,D/Gm)2 hd=0.50–3.46(⌬Tm,D/Gm)⫺10.11(⌬Tm,D/Gm)2 hd=0.54–4.61(⌬Tm,D/Gm)⫺4.95(⌬Tm,D/Gm)2 hd=0.60–6.01(⌬Tm,D/Gm)⫺0.32(⌬Tm,D/Gm)2 hd=0.59–5.33(⌬Tm,D/Gm)⫺0.59(⌬Tm,D/Gm)2 hd=0.63–4.61(⌬Tm,D/Gm)⫺10.10(⌬Tm,D/Gm)2 hd=0.54–4.13(⌬Tm,D/Gm)⫺4.35(⌬Tm,D/Gm)2 hd=0.59–5.16(⌬Tm,D/Gm)⫺0.56(⌬Tm,D/Gm)2 hd=0.64–5.45(⌬Tm,D/Gm)⫺2.25(⌬Tm,D/Gm)2 hd=0.69–4.43(⌬Tm,D/Gm)⫺11.00⌬Tm,D/Gm)2 hd=0.69–4.71(⌬Tm,D/Gm)⫺3.20(⌬Tm,D/Gm)2

Us=4.05+0.009⌬Tm,N Us=4.54+0.009⌬Tm,N Us=4.22+0.009⌬Tm,N Us=3.96+0.009⌬Tm,N Us=4.69+0.009⌬Tm,N Us=4.25+0.009⌬Tm,N Us=4.57+0.009⌬Tm,N Us=4.43+0.009⌬Tm,N Us=4.03+0.009⌬Tm,N Us=4.88+0.009⌬Tm,N Us=1.74+0.009⌬Tm,N

240

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

Fig. 11. Variation of storage tank mean water temperature of models ICS-2A and STS-A for a sequence of four 24-h operation of them without water drain.

Fig. 12. Variation of storage tank mean water temperature of models ICS-2B and STS-B for a sequence of four 24-h operation of them without water drain.

devices in comparison with model STS. From these diagrams we see that at noon the water temperature rise of model ICS-2 is about 3 °C higher (Figs. 11 and 12) and of model ICS-4 about 3 °C lower (Fig. 13) than that of the comparative model STS. In addition, the water temperature drop of model ICS-4 (Figs. 13 and 14) is less than that of models ICS-2 and STS. In the case of the B type reflector, the achieved hot water temperature level of model ICS-4 is the same as that of the

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

241

Fig. 13. Variation of storage tank mean water temperature of models ICS-4A and STS-A for a sequence of four 24-h operation of them without water drain.

Fig. 14. Variation of storage tank mean water temperature of models ICS-4B and STS-B for a sequence of four 24-h operation of them without water drain.

comparative model STS in the fourth day (Fig. 14). The performance of the other two ICS systems with the asymmetric CPC reflectors (ICS-1 and ICS-3) and regarding the same operation, is observed between the presented performance of models ICS-2 and ICS-4 (not shown in figures).

242

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

4. Annual performance of all studied systems

The presented ICS systems are new types of solar water heaters and their advantages are lower cost construction, easier installation and simpler operation, compared to the usual thermosiphonic systems with flat plate collectors (FPTU). According to the Greek experience in fabrication of DHW solar systems, the cost of commercial ICS system units is less (almost half) than that of same ratio of water storage mass per aperture area of a FPTU system, and this is very important for the penetration in the market of ICS type solar thermal systems. In practical applications the minimum water temperature level of DHW systems is about 35 °C, which is higher than that of the water mains temperature (~20 °C) and water drain from the storage tank is not considered useful when the water temperature drops below 35 °C. During the night, there is an additional temperature drop because of the storage tank thermal losses and therefore the initial water temperature in the morning could be about 25 °C, considering a mean ambient temperature of about 15 °C during night. Based on the above assumptions, the daily operation of DHW systems most days in the year is expected to be in the range of 25–55 °C, taking into account the seasonal change of the weather conditions. We calculated the mean daily efficiency for each month under the weather conditions of Patras (latitude 38.25°N). We consider all systems are of the same aperture area and the calculation of their daily performance is based on the corresponding ratio of storage water volume per aperture area of each system. In addition, we calculated the useful thermal energy of all systems for all months of the year, considering each system of one square meter in aperture area. The calculations were derived for several system mean daily operating temperatures Tm,D, but we present the results for two characteristic values: Tm,D = 35 °C and Tm,D = 45 °C. We give these results considering that they correspond to two usual operation modes regarding the initial (in the morning) storage water temperature. The Tm,D = 35 °C corresponds to a low initial operating water temperature (20–25 °C in the morning, rising to 45–50 °C in the evening) and the Tm,D = 45 °C to a medium initial operating water temperature (30–35 °C in the morning, rising to 55–60 °C in the evening). In Figs. 15 and 16 we present the calculated annual performance regarding mean daily efficiency hd (Fig. 15) and useful thermal energy Lu (MJm⫺2, Fig. 16) for each month. These results correspond to A type ICS models and to system FPTU for a mean daily operating water temperature level of 35 °C. In Figs. 17 and 18 we present the annual performance of B type ICS systems for operation at the same temperature. In Figs. 19–22 we give the corresponding annual performance of all studied models for an operating water temperature level of 45 °C. The mean daily operating water temperature levels Tm,D ⬍ 35 °C and Tm,D ⬎ 45 °C are not usual enough in the application of DHW solar systems and therefore they are of less interest to be shown. The results show that ICS systems of B type (aluminized mylar reflector) present thermal performances closer to system FPTU than all ICS systems of A type (innox

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

243

Fig. 15. Annual performance regarding mean daily efficiency of ICS models of type A for mean daily operating water temperature level of 35 °C.

Fig. 16. Annual performance regarding useful thermal energy of ICS models of type A for mean daily operating water temperature level of 35 °C.

reflector). This means that the use of a reflector of higher reflectance is necessary to achieve efficient ICS systems in water heating during day. Most interesting from these calculations is that model STS-B presents almost the same performance as that of model FPTU. In addition, the diagrams show that for Tm,D = 35 °C all studied systems (ICS and FPTU) present positive thermal performance all year. In winter and for Tm,D = 45 °C, the calculations (based on mean values

244

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

Fig. 17. Annual performance regarding mean daily efficiency of ICS models of type B for mean daily operating water temperature level of 35 °C.

Fig. 18. Annual performance regarding useful thermal energy of ICS models of type B for mean daily operating water temperature level of 35 °C.

of mounthly solar radiation and ambient temperature) show that the ICS systems present negative thermal performance and system FPTU almost zero. The above results can be considered acceptable regarding ICS system construction and are mainly explained by the higher storage thermal losses of ICS systems compared to system FPTU. This makes it clearer that a more effective heat protection mode is necessary for a further performance improvement of ICS systems. The main modifications that could easily be applied are the use of a selective absorber, for the

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

245

Fig. 19. Annual performance regarding mean daily efficiency of ICS models of type A for mean daily operating water temperature level of 45 °C.

Fig. 20. Annual performance regarding useful thermal energy of ICS models of type A for mean daily operating water temperature level of 45 °C.

reduction of radiation thermal losses and the mounting of additional glazing, for the suppression of convection and radiation thermal losses. Recently, we applied these modifications to an ICS system similar to model STS and test results showed that the improvement of performance regarding thermal losses coefficient is about 30% for the selective absorber and about 40% for the additional glazing. Considering that asymmetric CPC reflectors present better performance than symmetric CPC reflectors in system thermal losses for water tempera-

246

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

Fig. 21. Annual performance regarding mean daily efficiency of ICS models of type B for mean daily operating water temperature level of 45 °C.

Fig. 22. Annual performance regarding useful thermal energy of ICS models of type B for mean daily operating water temperature level of 45 °C.

ture preservation during the night as well as during the day for higher water temperature rise, it makes us optimistic for the achievement of improved ICS systems. Therefore, optimization of these systems could make it possible for the development of practical and cost effective solar water heaters, which would be suitable for a wider application in future.

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

247

5. Conclusions Integrated Collector Storage systems with single horizontal cylindrical storage tanks and stationary truncated asymmetric CPC reflectors were designed, constructed and tested. The systems were compared to an ICS system with symmetric CPC reflector and to a Flat Plate Thermosiphonic Unit. The proposed ICS systems differ mainly in the thermal protection of their cylindrical storage tank and the design of each ICS system is analytically given in the paper. All ICS systems were experimentally studied with black painted absorber and two types of reflector material. The performed tests aimed at the determination of the mean daily efficiency and the coefficient of thermal losses during the night for system operation without water drain. The results showed that the use of asymmetric CPC reflectors contribute to better water heat preservation during the night compared to the system with symmetric CPC reflector. All tested ICS systems achieved mean daily efficiency close to system FPTU. The ICS system with symmetric CPC reflectors of aluminized mylar type presents mean daily efficiency almost the same as that of system FPTU. All systems were studied regarding their annual performance under the weather conditions of Patras. The results showed that the thermal performance of the ICS systems with CPC reflectors is acceptable, but aiming to be of the same or better performance than system FPTU, regarding operation during the day and the night, further improvement in the thermal protection of their water storage tank is necessary. References [1] Chinnappa JVC, Gnanalingam K. Performance at Colombo, Ceylon, of a pressurized solar water heater of the combined collector and storage type. Solar Energy 1973;15:195–204. [2] Garg HP. Year round performance studies on a built-in-storage type solar water heater at Jodhpur, India. Solar Energy 1975;17:167–72. [3] Garg HP, Rani U. Theoretical and experimental studies on collector / storage type solar water heater. Solar Energy 1982;29:467–78. [4] Prakash J, Garg HP, Datta G. Effect of baffle plate on the performance of built-in storage type solar water heater. Energy 1983;8:381–7. [5] Sokolov M, Vaxman M. Analysis of an integral compact solar water heater. Solar Energy 1983;30:237–46. [6] Vaxman B, Sokolov M. Experiments with an integral compact solar water heater. Solar Energy 1985;34:447–54. [7] Ecevit A, Al-Shariah M, Apaydin ED. Triangular built-in-storage solar water heater. Solar Energy 1989;42:253–65. [8] Ecevit A, Chaikh Wais MAM, Al-Shariah AM. A comparative evaluation of the performances of three built-in-storage-type solar water heaters. Solar Energy 1990;44:23–36. [9] Mohamad AA. Integrated solar collector-storage tank system with thermal diode. Solar Energy 1997;61:211–8. [10] Goetzberger A, Rommel M. Prospects for integrated storage collector systems in Central Europe. Solar Energy 1987;39:211–9. [11] Schmidt C, Goetzberger A, Schmid J. Test results and evaluation of integrated collector storage systems with transparent insulation. Solar Energy 1988;41:487–94.

248

Y. Tripanagnostopoulos, M. Souliotis / Renewable Energy 29 (2004) 223–248

[12] Garg HP, Hrishikesan DS, Jha R. System performance of built-in-storage type solar water heater with transparent insulation. Solar & Wind Technology 1988;5:533–8. [13] Prakash J, Garg HP, Hrishikesan DS, Jha R. Performance studies of an integrated solar collectorcum-storage water heating system with tranparent insulation cover. Solar & Wind Technology 1989;6:171–6. [14] Schmidt C, Goetzberger A. Single-tube integrated collector storage systems with transparent insulation and involute reflector. Solar Energy 1990;45:93–100. [15] Mason AA, Davidson JH. Measured performance and modeling of an evacuated-tube, integral-collector-storage solar water heater. Journal of Solar Energy Engineering 1995;117:221–8. [16] Kaptan IN, Kilic A. A theoretical and experimental investigation of a novel built-in-storage solar water heater. Solar Energy 1996;57:393–400. [17] Winston R. Principles of solar concentrators of a novel design. Solar Energy 1974;16:89–95. [18] Welford WT, Winston R. High collection nonimaging optics. San Diego USA: Academic Press Inc, 1989. [19] Kalogirou S. Performance enhancement of an integrated collector storage hot water system. Renewable Energy 1998;16:652–5. [20] Smyth M, Eames PC, Norton B. A comparative performance rating for an integrated solar collector / storage vessel with inner sleeves to increase heat retention. Solar Energy 1998;66:291–303. [21] Tripanagnostopoulos Y, Yianoulis P. Integrated collector-storage systems with suppressed thermal losses. Solar Energy 1992;48:31–43. [22] Tripanagnostopoulos Y, Souliotis M. Nousia Th. Solar ICS systems with two cylindrical storage tanks. Renewable Energy 1998;16:665–8. [23] Rabl A. Comparison of solar concentrators. Solar Energy 1976;18:93–111. [24] Kienzlen V, Gordon JM, Kreider JF. The reverse flat plate collector: A stationary, nonevacuated, low technology, medium-temperature solar collector. Journal of Solar Energy Engineering 1988;110:23–30. [25] Tripanagnostopoulos Y, Souliotis M, Nousia Th. CPC type integrated collector storage systems. Solar Energy 2002;72:31–43. [26] Tripanagnostopoulos Y, Souliotis M. ICS solar systems with asymmetric CPC reflectors. In Proceedings (CD, Session of Solar Thermal Applications) of WREC VII, Cologne, Germany 2002.