Effectiveness of a vertical passive solar panel containing CaCl₂.6H₂O [for room heating by latent heat exchange]

 

 

du P. Calitz^I; D.F. van der Merwe^I; E.A. Bunt^II

^IEnergy Laboratory, Rand Afrikaans University, Johannesburg
^IISchool of Mechanical Engineering, University of the Witwatersrand, P.O. Wits, 2050 Republic of South Africa

 

 

---

ABSTRACT

A small passive vertical wall thermal panel comprising tubes containing CaCh.6H₂O (of melting temperature 29.8°C, i.e. close to ambient temperatures in Johannesburg at the time of the experiment) to transfer heat to a simulated living space ('room') on recrystallisation at night was built and tested. This chemical was chosen on account of its small volume change during phase change and for its suitability in relation to the diurnal temperature range. The chemical was contained in pipes behind glass, with natural or forced convection to distribute heating. Operation was satisfactory, with the temperature of the air discharged to the 'room' raised by up to 12° C higher than ambient, but under the test conditions employed the efficiency (conversion of incident solar energy to heat) was only in the neighbourhood of 12%.

---

 

 

Nomenclature

asolar altitude angle, degree

Apanel area, m²

A''no atmosphere' correction factor, values given in (3).

B''atmosphere' correction factor, values given in (3)

C_pspecific heat, kJ/kg K

ddeclination angle, degree

F_rempirical correction factor

hsolar hour angle, degree

I solar intensity (normal), W/m²

i_iflatent heat, kJ/kg

Llatitude, degree

nnumber of storage tubes

Nnumber of day (Jan. 1 = 1)

qenergy rate, W

ttemperature, °C

Uoverall collector heat transfer coefficient, W/m²K

Vvolume, m³

zazimuth angle, degree

αabsorptivity

ηefficiency

ρdensity, kg/m³

τtransmissivity

 

Subscripts

amb ambient

av average

eff effective

L latent

s sensible

surf collector surface

 

Introduction

Many latent heat solar energy systems have been suggested (see e.g. [1]) involving LiNO3, H₂O, Na₂SO₄.10 H₂O, etc. Calcium chloride hexahydrate was chosen as having a phase change temperature (29.8°C) within the local diurnal temperature range of 15 to 35°C; this chemical was readily available and is non-corrosive, with only a small volumetric change at solid to liquid phase change. The following properties are also relevant:

 

Theory

To provide a comparison with the actual energy absorption, insolation calculations made use of the following equations:[2; 3; 4]

For glass, it is assumed τ = 0.86 and ρ = 0.08. Then, using a value for αof 0.95 (black acrylic paint), (ατ)_eff = 0.522.

The calculation of U [2] involved consideration of the number of panels used, the emissivities of the glass panels and of the collector, an assumed value of the wind velocity past the panels and the thickness and conductivity of the insulation used. A typical value so determined was 3.52 W/m² K - which compares well with experimental values listed in [5] for double glazing (3-4 W/m²K).

 

Experimental set-up

As shown in Figure 1, the solar collector was combined with the heat storage medium, with - to align with a normal wall - 1 or 2 vertical glass panels of dimensions 860 mm (high) x 600 mm (wide), facing North. The depth of the chamber containing tubes and insulation panels (as measured normal to the glass panel(s)) was 270 mm. The collector could be used under conditions of natural or forced air circulation. At night the panel was insulated by a 50 mm thick polystyrene panel inserted between the tubes and the glass. Likewise, during the day, the upper and lower openings to the polystyrene insulated 'room' (of dimensions 1 220 × 60 × 860 mm = 0.63 m³) were blocked to seal in the tubes.

 

 

The experiment was conducted at an altitude of 1 700 m (average barometric pressure: 864 mb) and a latitude of 26°S. The chemical was stored in 7 (out of 12 available) PVC pipes of 63 mm OD, each pipe containing 0.00143 m³ of crystalline CaCl₂.6H₂O; the available energy storage (see Appendix) was 2 585 4- 33 At kJ. The temperature sensors used were of National (semi-conductor) type LM35. Seven temperatures [ambient, 'room', pipes, inside and outside of glass panel(s), and air temperature in the collector chamber (high and low positions)] were read at 30 min intervals. During the period of testing (early summer, i.e. October 24 to November 1), the maximum altitude of the sun (see Appendix) ranged between 76.73° and 79.23° (with consequent high reflection of energy away from the vertical collector); to improve the level of solar energy entering the collector, the latter was mounted on a reflector comprising a flat roof coated with aluminium paint. (Such a reflector is obviously even more necessary as the summer solstice is approached - when the solar altitude is 87.5° at local noon.)

 

Results

Figure 2 shows a typical diurnal insolation pattern, using calculated values of I for a horizontal surface. (The 'triangular' shape shown is typical of a fixed flat collector - whereas a tracking collector would yield an insolation curve that is much more flat-topped for much of the day.) Table 1 summarises the results obtained; in this table η_avwas determined by averaging the value of ηat half-hourly intervals over the number of such intervals in the heating period.

 

 

Some notes on these results follow.

Day 293

Two glass sheets were used and Figure 3shows a plot of ηversus as the day advances. (η can be >1early in the morning, when t_amb >- t_Surf)- An estimate of the energy reflected from the roof was 3 963 kJ for the day - which is thus of the total energy received. This shows the importance of the reflector when the orientation of the sun is high in the sky for much of the day. Taking into account a value for the diffuse reflectivity of 0.7, the energy actually absorbed by the collector was thus only about 13% of the total incident energy.

 

 

Day 294

Only one glass sheet was now used, with a consequent increase in energy absorbed (but this arrangement is likely to be associated with a greater loss to the surroundings in winter). In consequence, the value of (τα)_eff increased to 0.7.

Day 295

A higher ambient temperature was measured during this test, with consequent reduced natural convection and 'room' temperature.

Day 296

During this test the collector was not opened to the 'room'. The higher 'room' temperature (over ambient) is ascribed to heat loss by conduction.

Day 303

A fan was now used to provide forced convection and a change of shape of the temperature curves is now evident in this and the following test; these conditions representa considerable operating improvement. The morning temperature reversal did not occur, while 'room' temperatures were greater than ambient throughout the day - and all temperatures were noticeably closer together throughout the 24 h period.

Day 304

The fan was again used. More energy was used on this occasion in heating the 'room' than was received by insolation - making use of heat stored the previous day.

Figure 4 (a) to (f) shows plots of the various temperatures recorded. These curves roughly follow that of Figure 2 (peaking at noon), but then flatten out, with the temperatures decreasing only slowly through the night. In general, using natural convection, the 'room' temperature was found to be less than ambient before llh30, and greater than ambient after llh30, thus meeting the obvious requirements for such an apparatus. However, under forced convection conditions, all internal temperatures were higher than ambient in nearly every reading throughout the period of measurement.

The use of one sheet of glass (at least in the summer) was found to be approximately 20% more effective than the use of two sheets in transmitting solar energy.

 

Conclusions

Operation of the solar collector fulfilled its purpose, namely, that of 'room' heating after dark by making use of latent heat exchange. Using natural convection, the 'room' temperature was greater than ambient after about noon (after which the differential was several degrees K). With forced convection, a differential of about 7 K between 'room' and ambient temperatures was maintained throughout the 24 h period of diurnal operation (the best difference being obtained after dark). In fact, forced convection operation was generally superior in all respects. The chemical selected was also entirely satisfactory in respect of melting temperature, but the main disadvantage of the design was the vertical orientation of the panel; this made virtually essential the use of an external ground reflector (and particularly so in summer). Even so, however, the absorption of available solar energy was estimated to be only about 13%. The use of one sheet of glass (in place of two) was also found to reduce input energy transmission losses by about 20%.

 

References

[1] Selvidge M &¿ Miaoulis IN. Evaluation of reversible hydration reactions for use in thermal energy storage. Solar Energy, 1990, 44, 173.         [ Links ]

[2] Stoecker WF & Jones JW. Refrigeration and air conditioning, 2nd edn. McGraw-Hill, 1982.

[3] ASHRAE Handbook and Product Directory, Applications vol. Chap. 58, 1978.

[4] Moon P. Proposed standard solar radiation curves for engineering use. Journal of Environmental Science, 1940, 230, 583.         [ Links ]

[5] Howell JR et ai Solar/Thermal Energy Systems; analysis and design. McGraw-Hill, 1982.

 

 

Received December 1993
Final version July 1994

 

 

Appendix

Specimen calculations

Day 297 (Oct 24)

i.e. the sun is south of the equator.

Day 805 (Nov 1)

i.e. the sun is south of the equator.

Available energy storage

Latent heat storage capacity

Sensible heat

 

→ Available storage for a solar panel of area 0.516 m² = 2 585 + 33∆i kJ (or 5010 + 64∆t kJ/m²).