Presentation of two thermal models of an innovative patented solar drainpipe

Christian Cristofari, Gilles Notton, Fabrice Motte, Jean-Louis Canaletti and Jean Panighi University of Corsica

UMR CNRS 6134 Scientific Research Center of Vignola

Route des Sanguinaires F20000 AJACCIO – FRANCEcristofari@univ-corse.fr;notton@univ-corse.fr;motte@univ-corse.fr;canaletti@univ-

corse.fr;panighi@univ-corse.fr

Abstract—We developed a new concept of water solar collector integrated into a drainpipe. This collector is made of several serial modules. The drainpipe keeps its water evacuation func-tion. After a brief presentation of the energy situation in France, the new concept of solar collector is described; the ex-periment, the collected data and the first experimental results are presented and discussed. Numerical calculations are per-formed in Matlab® environment, using a finite difference model and an electrical analogy. A second approach using a thermal modeling by Comsol® software is also presented.

Keywords-solar collector; heating; renewable energy.

I. INTRODUCTION The rapid increase in energy consumption in building

sector is seen in many countries. In France, 30 millions of housings use about 50% of final energy and produce 25% of green house gazes. For Europe, 500 millions inhabitants in 160 millions housings consume half the energy. The residen-tial and tertiary sector is the first energy consumer in France (Fig. 1) with 69.4 Mtoe [1] the percentage (43%) stays stable but the absolute value increases (+25% in 1973-2008). In France, energy costs are mainly devoted to domestic heating (72%), followed by lighting and appliances (11%), hot water (11%) and cooking (6%) (Fig. 2) [2].

Figure 1 Energy consumption in France [2]

Figure 2 Part of Energy (housing sector)

A European citizen uses 36 litres of 60 °C hot water daily with tendency for increase in future. The energy to produce hot water is rising slightly because the comfort level sought now is greater than the level accepted in the past. In older buildings, this sector is only 6% of overall energy consump-tion, but with a reduced heating need mainly due to a better thermal insulation, the hot water production represents 30% of energy consumption in a modern housing. Using solar col-lectors is a good and sustainable solution for heating water. They can efficiently provide up to 80% of the hot water needs, without fuel cost or pollution and with minimal O&M expense. The European Union’s solar thermal market has clearly outstripped forecasts with 51.4% growth in 2008, or about 3 238.5 MWth. The collectors that contribute this addi-tional power cover a surface of over 4.6 million m22, which is 1.6 million m2 more than in 2007 [3]. Then, an important renewal in researches for improving and conceiving thermal collectors is occurring.

Introducing innovating and environmentally positive so-lutions is difficult, the obstacles are numerous: financial, technical, psychological obstacles, or too conservative build-ing standards [4]. We must find an innovative concept of heating system easily building-integrated, reducing visual impact (psychological obstacle), easy to install in both new and old houses (technical obstacle), not too costly (financial obstacle) and with an environmental positive solution. Our “basis” idea consists in making actives passive parts of build-ing: in past years, a shutter was transformed into a solar air collector [5], now we develop a water collector integrated in-to a gutter, recovering rainwater and solar radiation.

25.9 31.0 33.040.8

49.0 50.0 50.1 50.9 50.2

47.947.0

38.0

38.2

39.0 39.6 37.2 37.0 36.23.7

4.04.0

4.0

4.0 4.4 4.3 4.2 4.3

69.4 Mtoe41.40%

67.6 Mtoe41.03%

68.0 Mtoe41.01%

66.4 Mtoe41.86%

57.7 Mtoe41.01%

56.2 Mtoe42.03 %

68.0 Mtoe42.61%

67.6 Mtoe42.33%

69.4 Mtoe43.35%

0

20

40

60

80

100

120

140

160

180

1973 1979 1985 1990 2000 2002 2006 2007 2008

MTo

e

AgricultureIndustryTransportsResidential-tertiary

Water heating11%

Cooking6%

Specifical use of electricity

11%Heating

72%

264

2011 International Conference on Environment Science and Engineering IPCBEE vol.8 (2011) © (2011) IACSIT Press, Singapore

II. The new c

and named Hwithout any vso be used osouth into thground level tdrainpipe presnalizations cothe drainpipemodules. One(individual hodrainpipe lengglass, an air stion layer. Firthe inferior inmal contact w

PRESENT

concept of solaH2OSS® presvisual impact. on north orienhe drainpipe).thanks to the serves its roleonnecting the e. An installae module is abouses). The mgth. From top

space, a highlyrst, the cold f

nsulated tube awith the absorb

TATION OF THE

ar water collesents a high The SWC is anted walls (S It is totallydrainpipe inte

e of rainwater house to the

ation includesbout 1 m lengthmodules nump to bottom, iy selective absfluid from theand then in thber.

E SOLAR GUTT

ctor (SWC) pabuilding inte

arranged so it SWC being oy invisible froegration (Fig.3evacuation. TSWC are hid

s several conh and 0.1 m in

mber depends it is composesorber and an e tank flows the upper tube i

TER atented gration can al-

oriented om the 3). The The ca-dden in nnected n width on the

ed by a insula-

through in ther-

tesindrtwstaan

An experimsting the thermcreasing perf

rainpipe compwo rows (Fig. ant by a contrnd cools it in th

Figure 3 A

III. TH

mental wall is bmal behavior, formances by

prises 18 seria4). The input

rol loop whichhe other case.

H2OSS® module

HE EXPERIMENT

built in Ajacciovalidating a t

y parametersal modules (at fluid temperah heats the flu

e

TATION o with 3 objecthermal modes adjustmentsabout 2m²) spature is taken uid if it is too

ctives: el and s. A

plit in con-

cold

265

Figur

Every mitemperature, hrate and inputule). The flow

Fig. 5 shobient temperaneous efficien

Fig

The maximis 9°C. Tin isneous efficienpidly after no

0

10

20

30

40

50

60

70

06:00 0

Tem

pera

ture

(°C

), Ef

ficie

ncy

(%)

Win

d sp

eed

(km

/h)

Input temperatAmbient tempe

re 4 Experience

inute are collhumidity, wint and output f

w rate was fixeows inlet and ature, wind spncy defined by

.(Cp.Q.ρ=η

gure 5 Experim

mum gap betws not constantncy, up to 60%oon because th

08:24 10:48

ture Outpuerature Wind

e and temperature

lected: solar nd speed and fluid temperated at 0.120 m3

outlet collectoeed, solar irra

y:

TT( inleoutlet −

mental results for t

ween inlet andt in this exper% at the steadyhe wall is sou

13:12Time (h)

ut temperaturespeed

e control loop

irradiance, adirection, fluitures (for each.h-1. or temperatureadiance and in

Ac.I)et

the collector

d outlet temperiment. The iny-state, decrea

uth-east oriente

15:36 18:00

Instantaneous effSolar irradiance

ambient id flow h mod-

es, am-nstanta-

(3.1)

eratures nstanta-ases ra-ed. We

meafutin18izepathen

(EWre

offictio(FThulmwh

0

100

200

300

400

500

600

700

800

900

1000

Sola

r irr

andi

ance

(W/m

²)

ficiency

Effic

ienc

y

measured tempach module. Tul length has nnues to increa8 serial module, it will be narallel modulee fluid temper

ntering in the a

Figur

The efficienEuropean stan

We calculated tlation coeffici

=η

with Φ the f the SWC, Taciency and K tonal SWC witFig 7). The avhis difference es. The therm

more importanthen the reduce

η = -15.08 TR2

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.015 0.02

Effic

ienc

y

peratures locathe profile is li

not been reachase, thus we cles. If the outpecessary to m

es. In the inputrature increaseabsorber.

re 6 Temperatur

ncy (η) at statndard) vs. reduthe linear regreient at 0.96 (3

=Tr

Tr.085.15−=

solar irradiancamb the ambiethe thermal loth simple glas

verage value ois due to the

mal looses ont and so the ed temperatur

Tr + 0.83= 0.9642

0.025 0.0Reduced

ted on the upinear (Fig. 6). ed because thcan install effput temperatur

modify the cont tube located es by 1°C for

re evolution vs. t

tionary state iuced temperaession and we.3):

Φ− ambm TT

K83.0r −=+

ce, Tm the avent temperaturooses [7] highess and highly of K is usuallygeometry of t

n the sides ofperformances

e increases.

03 0.035 temperature (K.m²/W)

pper tube betwThe maximal

e temperatureficiently more re begins to st

nfiguration in uinto the insula18 m length b

the length

is plotted (Figature, Tr (3.2)e obtain with a

BTr.K +

verage temperre, B the opticer than for conselective absoy 5 W.m-1.K-

the H2OSS® mf the modules decrease ra

0.04 0.045

ween l use- con-than

tabil-using ation, efore

g. 7), ) [6]. a cor-

(3.2)

(3.3)

rature al ef-nven-orber 1 [6]. mod-s are

apidly

0.05

266

Figure 7 Efficiency for our collector and for conventional solar ones:

without glass (a); with one (b) and two glasses (c), with a selective absorber and one glass (d) and vacuum collector (e)

IV. THERMAL MODELS The aim of the presented thermal modelling is to improve

the performances of the collector. The first step is to get ac-curacy a good accordance between numerical and experi-mental results, to be able, in a later step to optimize the thermal properties. In this paper, we present only the accura-cy part.

Two complementary thermal models have been devel-oped. The first one using a Matlab® environment, and the second one a Comsol Multiphysics software. The Matlab model, complex to develop, offers a huge flexibility of all the model parameters. One the other hand, the Comsol model is relatively simple to build and allows a good visualization of the thermal phenomenon occurring inside the collector. The concordance between the thermal results of the two models has been checked.

A. Thermal model under Matlab® environment

We present a bi-dimensional model with the thermal transfers composed of a serial assembling of one-dimensional elementary models. Each model is based on a nodal discretisation. The domain is broken up into 52 ele-mentary isotherm volumes, and for each node, we write a thermal balance equation using an electrical analogy (Fig. 8) where temperatures, flows, flow sources and imposed tem-peratures are assimilated to potentials, currents, current gene-rators and voltage generators. The three different types of thermal resistances represent the convection, the conduction and the radiation exchanges. Thermal properties are constant. This model uses as input physical parameters: Total solar ir-radiance Φ, ambient temperature Tamb, air speed in front of the collector v, ground and sky temperatures and cold fluid temperature. It is impossible to give all the equations, due to a lack of space, but the thermal balance for the first elemen-tary model (Fig.8), upper left, is detailed in Eq. (4.1).

( )( )

cd1contactcd

112

cd

124sky

41skyside1transglass

cvcd

1amb

cvcd

1amb4sky

41sky11glassglass1

11glassglass

RRRTT

RTTTT.f.A..

RRTT

RRTTTT.f.A.....A

dtdT.V.Cp.

1 ++−+−+−+

+−

++−+−−=

−

−

σε

σεφαρ

(4.1)

with A area, α absorption coefficient, Φ total solar irra-diance, ε emissivity coefficient, f geometrical factor, T tem-perature and R thermal resistance.

Figure 8 Electrical analogy of the collector

For the circulating fluid’s thermal equation, there were 3 equations with 4 unknown factors. In order to solve this sys-tem, the outside fluid temperature is estimated using the NUT equation (Eq. 4.2). It corresponds to the temperature profile of a fluid circulating inside a homogeneous tube (Ta) with an internal surface Sfc, at steady state [7]

( ) NUT

ae/fcas/fc eTTTT −−+= (4.2)

fcfcfcafc CpmShNUT ./= (4.3)

Using Eqs. (4.2) and (4.3), we can calculate the outside fluid temperature from the inside fluid and tube temperatures. The 52 thermal equations for the 52 elements were devel-oped and are solved using a direct implicit method.

The values of some thermal parameters are difficult to determine particularly the unknown thermal resistances dues to physical contact between the various element parts, and the heat transfer coefficients (due to the complexity of the solar collector geometry); thus, we tested empirically various numerical values for these resistances in such a way that we obtain a good accuracy. The adjusted values of these parame-ters are checked to be sure that they are physically acceptable. In order to find the best coefficients, the root mean square er-ror (RMSE) between experimental and numerical outside fluid temperature values was calculated and minimized and the optimized values are recorded. Fig. 9 shows an experi-mental verification for a given day for the outside water tem-perature.

D iff D iff

1 2 3 4 5 6 7 8 9 10 11 12 13

18 19

33 20

34

21

35

22

37

23

38

25

39

40

41

26

42

27

43

57

58

59

28

44

30

46

47

31 32

48 49 50 51 53 54 55 56 60 62 63 64

65 66 78 79

1

2

3 4

5 6

78

10

13

14

1 5

16

17

2021

22

2526

33

34

3536

38

39

4041

43

44

7414 15 16

24 29

36 45

6152

67 68 69 70 71 72 73 74 75 76 77

80

81 82 83 84 85 86 87 88 89

90

91 92 93 94 95 96 97

9 11

1 2

1 8 1 9 2 3 24

2 8 29

3031

3 2 37

42

4546

47

48

49

5051

52

53

54

5556

57

58

59

6061

62

63

64

6566

67 6

8

69

70

71

7273

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93 9

4

95

96

97

98

99

1 00

1 01

1 02

1 03

1 04

1 05

1 06

1 07

1 08

1 09

1 10

1 11

1 12

1 13

1 14

1 15

1 1

6

1 17

1 18

1 19

1 20

1 21

1 22

1 23

1 24

1 25

1 26

1 27

1 28

1 29

1 30

1 31

1 32

1 33

1 34

1 35

1 36

1 37

1 38

1 39

1 40

1 41

1 42

1 43

1 44

1 45

1 46

1 47

1 48

1 49

1 50

1 51

1 52

1 53

1 55

1 56

1 57

1 58

15 4

1 60

1 6

1

16 2

1 63

1 66

1 69

1 72

1 75

1 78

16 4

16 5

16 716 8

17

017 1

17

317 4

17 6

17 7

1 80

1 88

17 9

18

1

1 82

1 83

1 84

1 86

18

5

18 7

1 89

19 019 1

1 92

19 319 4

1 95

19 619 7

1 98

19 920 0

2 01

20 220 3

2 04

20 5

2 06 2 0

7

20

8

20 9

2 10

2 11

2 12

2 17

2 13

2 14

21

5

21 6

2 20

21 821 9

2 22

22 1

2 24

22 3

2 26

22 5

2 28

22

7

22 9

2 30

23

1

2 32

2 34

242

23 3

23 5

2 36

2 37

2 38

2 40

23 924 1

24 4

2 43

2 45

24 6

2 47

24

8

2 50

25 1

2 53

24 9 25 2

25 4

2 55

2 56

2 57

2 58

25 9

2 60

2 65

2 70

26

1 26 6

26

2 26

7

27 1

27 2 27 3 27 4 27 5 27

6 27 7 27

9 28 0

28

4 28 9

29 3

2 82

2 87

2 92

28 3 28

8

2 85

2 86

2 90

2 91

2 63

2 64

2 68

2 69

28 1

27

8

2 942 95

2 96

3 00

3 01

3 02

3 03

3 04

3 05

3 06

3 07

3 11

3 123 13

3 10

3 09

2 97

2 98

29

9 30 8

31 4

3 15

3 20

31 631

7

3 18

3 19

32 1

32 2 32 3 32

4 32 5 32 6 32 7 32 8

32 9

33 2

33

6

3 35

3 30

33 1

3 33

3 34

3 373 38

3 39

3 43

3 40

3 41

34 2 34 4 34 6 34 8

3 45

3 47

35 0

3 49

35 2

3 51

3 53

3 57

35

4

3 55

3 56

3 583 59

36 0

36 3 36

8 37

3 37 8 38 3 38 8 39 3

39

7

39 238 738 2

3 96

3 91

3 86

3 81

37

7

3 76

3 71

3 66

3 61

36 2 36 7 37 2

3 64

3 65

3 69

3 70

3 74

3 75

3 79

3 80

3 84

3 85

3 89

3 90

3 94

3 95

15 9

2 7

Thermal capacity

Radiation : global, diffuseD iff

Conductive resistance

Contact resistanceMaterial change

Convective resistance 1

1

2

3 4

7 9

1

1

2

3 4

7 9

Detail of first elementarymodel (upper left)

D iff D iff

1 2 3 4 5 6 7 8 9 10 11 12 13

18 19

33 20

34

21

35

22

37

23

38

25

39

40

41

26

42

27

43

57

58

59

28

44

30

46

47

31 32

48 49 50 51 53 54 55 56 60 62 63 64

65 66 78 79

1

2

3 4

5 6

78

10

13

14

1 5

16

17

2021

22

2526

33

34

3536

38

39

4041

43

44

7414 15 16

24 29

36 45

6152

67 68 69 70 71 72 73 74 75 76 77

80

81 82 83 84 85 86 87 88 89

90

91 92 93 94 95 96 97

9 11

1 2

1 8 1 9 2 3 24

2 8 29

3031

3 2 37

42

4546

47

48

49

5051

52

53

54

5556

57

58

59

6061

62

63

64

6566

67 6

8

69

70

71

7273

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93 9

4

95

96

97

98

99

1 00

1 01

1 02

1 03

1 04

1 05

1 06

1 07

1 08

1 09

1 10

1 11

1 12

1 13

1 14

1 15

1 1

6

1 17

1 18

1 19

1 20

1 21

1 22

1 23

1 24

1 25

1 26

1 27

1 28

1 29

1 30

1 31

1 32

1 33

1 34

1 35

1 36

1 37

1 38

1 39

1 40

1 41

1 42

1 43

1 44

1 45

1 46

1 47

1 48

1 49

1 50

1 51

1 52

1 53

1 55

1 56

1 57

1 58

15 4

1 60

1 6

1

16 2

1 63

1 66

1 69

1 72

1 75

1 78

16 4

16 5

16 716 8

17

017 1

17

317 4

17 6

17 7

1 80

1 88

17 9

18

1

1 82

1 83

1 84

1 86

18

5

18 7

1 89

19 019 1

1 92

19 319 4

1 95

19 619 7

1 98

19 920 0

2 01

20 220 3

2 04

20 5

2 06 2 0

7

20

8

20 9

2 10

2 11

2 12

2 17

2 13

2 14

21

5

21 6

2 20

21 821 9

2 22

22 1

2 24

22 3

2 26

22 5

2 28

22

7

22 9

2 30

23

1

2 32

2 34

242

23 3

23 5

2 36

2 37

2 38

2 40

23 924 1

24 4

2 43

2 45

24 6

2 47

24

8

2 50

25 1

2 53

24 9 25 2

25 4

2 55

2 56

2 57

2 58

25 9

2 60

2 65

2 70

26

1 26 6

26

2 26

7

27 1

27 2 27 3 27 4 27 5 27

6 27 7 27

9 28 0

28

4 28 9

29 3

2 82

2 87

2 92

28 3 28

8

2 85

2 86

2 90

2 91

2 63

2 64

2 68

2 69

28 1

27

8

2 942 95

2 96

3 00

3 01

3 02

3 03

3 04

3 05

3 06

3 07

3 11

3 123 13

3 10

3 09

2 97

2 98

29

9 30 8

31 4

3 15

3 20

31 631

7

3 18

3 19

32 1

32 2 32 3 32

4 32 5 32 6 32 7 32 8

32 9

33 2

33

6

3 35

3 30

33 1

3 33

3 34

3 373 38

3 39

3 43

3 40

3 41

34 2 34 4 34 6 34 8

3 45

3 47

35 0

3 49

35 2

3 51

3 53

3 57

35

4

3 55

3 56

3 583 59

36 0

36 3 36

8 37

3 37 8 38 3 38 8 39 3

39

7

39 238 738 2

3 96

3 91

3 86

3 81

37

7

3 76

3 71

3 66

3 61

36 2 36 7 37 2

3 64

3 65

3 69

3 70

3 74

3 75

3 79

3 80

3 84

3 85

3 89

3 90

3 94

3 95

15 9

2 7

Thermal capacityThermal capacityThermal capacity

Radiation : global, diffuseD iffRadiation : global, diffuseD iff

Conductive resistanceConductive resistance

Contact resistanceMaterial change Contact resistanceMaterial change

Convective resistanceConvective resistance 1

1

2

3 4

7 9

1

1

2

3 4

7 9

1

1

2

3 4

7 9

1

1

2

3 4

7 9

Detail of first elementarymodel (upper left)

267

B

Figure

We calculteorological cRMSE is 1.9 the used thermconsidered su

B. Thermal mment

Comsol Mment using fifor the most pthermal finiteare about 550sented in Figdimensions viature dependedimensional mstudy the edgepoint is that wand space meself. The tempreferring to th

9 Experimenta

late the RMSEconditions an% (0.37°C).

mal sensors onufficient.

model under C

Multiphysics iinite elementspart, preloadee elements, bu00 elements f

g. 10, and 500iew, Fig 11. Tent. We develmodel. The 3Des effects in th

we did not modmory consumperatures of th

he experimenta

Figure 10 2 D

al verification for

E for 17 daysnd physical e

Considering nly is close to

Comsol Multip

is a simulatio method. Thed. In this modut with a realfor the bi-dim0000 elementThe thermal prloped a bi-dim

D model was dhe length direcdel the fluid b

ming .We focushe inside of thal data.

Dimensions Mesh

r an average day

s, with variablentries. The athat the accur

o 0.2°C, this r

physics® envir

on software ene used equatiodel, we considly small size

mensional viewts along the eroperties are tmensional anddeveloped in oction. One imecause it was sed on the colhe tubes are im

h view

les me-average racy of result is

ron-

nviron-ons are, der iso-: there

w, pre-entire 3 temper-d a tri-order to

mportant to time

llector i mposed

ob

firsoresidtemm

bo

We study obtained are pre

The white a

rst observationorber seem to ctional. The sde of the absomperature gra

most important

On the 3D oth length side

Figure 11 Dim

nly stationaryesented in Figs

arrows represens that could bbe too wide b

second point isorber and the adient in this pof the collecto

model, despite of the collect

mensions Mesh vie

y configurations. 12-13.

nt the total hebe done is thatbecause the hes that the insuexterior is rea

part is very imor.

te there is an tor, we can se

ew

ns. The result

eat flux. One ot the fins of theat flux isn’t uulation betweeally too small

mportant, one o

insulation pae that the edge

ts we

of the he ab-unidi-en the . The of the

art on es ef-

268

Figure 12 2D model - Thermmal repartition insside the collector

Fiigure 13 3D moddel - Thermal repand X

artition inside theXZ section

e collector: XY seection

fects exist but are not too significant which justify the utiliza-tion of a 2D-thermal model as the Matlab one.

V. CONCLUSION A new concept of flat plate solar water collector highly

building integrated was presented. The collector is made of several modules in serial position. The particularities of the collector are that it is integrated into a drainpipe and totally invisible from the ground level. It can be installed on both new and old buildings, and on individual or collective habita-tions. An experiment was implemented and promising first experimental results were presented. At low reduced temper-ature values, the thermal performances are close to conven-tional ones. However it is necessary to optimize the shape of this collector in order to improve the thermal insulation. Two thermal models have been developed and the obtained results are very close to the experimental ones which validates them. The next step will be to use these two models to optimize the performances of the solar collector.

ACKNOWLEDGMENT The authors would like to thank the Corsica Territory

Collectivity for their financial supports.

REFERENCE [1] Ministère de l’écologie, de l’énergie, du développement durable et de

la mer (2010) Les consommations finales énergétiques par secteur économique, corrigée du climat en 2008. www.statistiques.developpement-durable.gouv.fr.

[2] Besson D (2008) Consommation d’énergie: autant de dépenses en carburant qu’en énergie domestique. INSEE première, 1176.

[3] Observ’ER (2009) Solar thermal barometer, Systèmes Solaires, n°191, June 2009

[4] Intrachooto S, Horayangkurab V (2007) Energy efficient innovation: Overcoming financial barriers, Building and Environment, 42(2): 599-604.

[5] Canaletti JL, Notton G, Cristofari C (2008) New concept of solar air heater integrated in the building. ISJAEE, 5:39-44.

[6] Duffie JA, Beckman WA (1980) Solar engineering on thermal processes; John Wiley & Sons. New York.

[7] Incropera FP, Dewitt DP (1996) Fundamentals of heat and mass transfer, 4th edition, Wiley.

269