J. Cadafalch, R. Consul, X. Trias, C. D. Pbrez-Segarra

Centre Tecnoldgic de Transferencia de Calor (CTTC)
Lab. de Termotecnia i Energetica
Universitat Politecnica de Catalunya (UPC)
labtie@labtie. mmt. upc. es, www. cttc. upc. edu

This paper presents a CFD model for the study of a rectangular water store forming part of a integrated collector storage system. The govening equations (mass, momen­tum and energy conservation) are solved assuming bidimensional and laminar flow. Details of the mathematical model (governing equations and boundary conditions) and of the numerical model (discretization of the equations, solver, numerical param­eters…) are described. A set of numerical solutions of a reference case representing 24 hours of exposure to outdoor weather conditions and without draw-off, have been obtained in order to investigate the sensitivity of the numerical solutions on the tem­poral and spatial discretization adopted. A final validation process by comparison of the numerical solutions with experimental data from the litterature and with direct numerical simulation, DNS, (high level simulation that considers turbulence effects), obtained by the authors is also presented.

Introduction

The integrated collector storage ICS is one of the simplest and cheapest solar thermal systems. The first thermal solar systems constructed were ICSs and during all the XXth century these kind of systems were commercial all over the world.

In the ICS the store is directly exposed to the ambient. Therefore, the system heat losses are high, especially during night periods or during periods with low ambient temperature. This technical problem has limited the applicability of the ICSs up to now to low temperature applications in hot climates. Future improvements of the design of ICSs should be focus on solving this problem, for example by means of transparently insulated covers, or using new configurations of the store.

As in all other thermal engineering areas, the use of Computational Fluid Dynamics tech­niques will play a key roll in the development of new components for solar thermal applica­tions. In this work, a two-dimensional transient numerical model for the evaluation of heat transfer and fluid flow in a rectangular storage element of a ICS system is presented. The model solves the governing equations (mass, momentum and energy conservation) assum­ing two dimensional laminar flow. With this model is possible to evaluate a complete cycle of 24 hours of charging (heating up during sunny hours) and discharging process (cooling down) of the ICSs on a standard Personal Computer in a few hours. The mathematical formulation is discussed including both the governing equations and the boundary condi­tions. Details on the numerical model are given, and the required numerical parameters (discretization, numerical schemes, convergence criteria..) are evaluated. In order to vali­date the results of the numerical model, the store elements has been analysed under two different working conditions. In one of them, results are compared to experimental values from other authors. In the other, solutions are compared to numerical solutions obtained by a more detailed direct numerical simulation (DNS) model in which the hypothesis of laminar flow is not assumed.

M. Cucumo — A. De Rosa — V. Ferraro — D. Kaliakatsos — V. Marinelli

Department of Mechanical Engineering — University of Calabria 87030 Arcavacata di Rende (CS) — Italy Phone +390984494603 — Fax +390984494673 E-mail: m. cucumo@unical. it

The analysis of numerous experimental data recorded on a 2.85 kW photovoltaic plant, suitably monitored and installed at the Department of Mechanical Engineering at the University of Calabria, has shown that the cell temperature significantly influences the performances of the photovoltaic modules, with noticeable diminution of efficiency at temperature increase. To check this parameter, a cooling system was researched, to be mounted at the back of the photovoltaic module, which, by means of a suitable air flow conducted through a system of channels, attenuates the effects of the cell temperature. Finally, the economic availability of using air heated by the modules to heat the buildings in the winter period was considered.

1. TECHNICAL CHARACTERISTICS AND PERFORMANCES OF THE PHOTOVOLTAIC FIELD.

The photovoltaic monitored plant, shown in fig. 1 (Cucumo et al., 2003), is a grid — connected type, with a theoretical peak power of 2.85 kW and it was designed and created at the Building Energy Laboratory of the University of Calabria.

— 30 Modules;

— 72 cells connected in series, 100 cm2 Eurosil cells;

— Upper covering: tempered glass with high transmittance;

— Sheathing material: EVA (Ethylene Vinyl Acetate);

— Rear covering: Tedlar/polyester/Tedlar;

— Frame: Anodised aluminium;

— Junction box: weatherproof (8IP500), glass — fibre reinforced;

— N° 1 inverter

— N° 1 instrument board and protections;

— N° 1 interface to network board;

— N° 1 datalogger,

Tab. 1 — Components of the photovoltaic plant

The photovoltaic field, mounted on a variably inclinable trellis, has a gross surface of 27 m2 and a useful surface of 25.7 m2; it is subdivided into 3 strings of 10 modules, connected in series, with a maximum string voltage of 340 V (420 V at open circuit).

The photovoltaic modules used, each contain 72, 103 x103 mm polycrystalline silicon cells in series, with a reference efficiency of t|r = 11,11 %.

Table 1 shows the components of the field and the technical specifications of the modules, each 0.996 x 0.904 m, while tab. 2 shows the electrical specifications in the reference conditions.

The manufacturing company of the photovoltaic modules supplied for each module the experimental values of the maximum power, the short circuit current (Icc), open circuit voltage (Vca), maximum power current (Imp) and the maximum power voltage (Vmp). These values, were certified by JRC (Joint Research Centre) of Ispra (Italy).

Once the characteristic curves of the single modules are known, taking into account their series-parallel connections, the characteristic curve (I, V) of the photovoltaic field was obtained. In the reference conditions (TR= 298 K, GR= 1000 W/m2) and of maximum power, the reference efficiency of the field was 10.45 % while the maximum power supplied was 2.687 kW.

Figure 2 shows the characteristic curve (I, V) and the power curve (P, V) of the photovoltaic field, evaluated in the reference conditions.

V (volt)

Fig. 2 — Characteristic curve (I, V) and curve (P, V) of the field.

Figures 3 and 4 show the experimental data of the photovoltaic field efficiency in February and April 2003. A marked oscillation in efficiency can be noted, owing both to the effect of radiation and the temperature of the modules.

The Evans correlation (Evans, 1981) reported in literature

Л = hR [1_P(tc _tr) + Y log G] (1)

shows how efficiency decreases with cell temperature increase and with a decrease in solar radiation. According to this equation, if the cell temperature doubles in value from 25 to 50 °C, efficiency decreases by about 12%, whereas if solar radiation is halved from 1000 to 500 W/m2, efficiency decreases by 4%. Therefore the influence of temperature is greater than that of solar radiation.

The study of an air-cooling system for photovoltaic modules to keep their temperature low to improve their performances is considered interesting.

CAVICAL1 was instrumented with nine thermocouples type T as shown in Fig. 3; seven located at the copper cone, in the wall in contact with water (T1 — T7), to measure the temperature distribution there, and two more to measure the inlet and outlet temperatures of the cooling fluid. A DAS system with other instruments was used to control DEFRAC and to run different set of experiments. In each experiment, mass flow rate, direct incident solar radiation and temperatures were measured and recorded. The experiments were done at the middle of the day and typical runs have duration of 10 minutes.

2. Numerical model

A numerical simulation was performed to evaluate the thermal behavior of the cavity calorimeter. One feature that the simulation allows is the evaluation of the convective energy losses due to natural convection heat transfer from the inner walls to the inner air in the cavity and from the cavity aperture to the atmosphere air. The phenomena that occur inside the air cavity is a free convection driven by the buoyancy force caused by density differences developed due to temperature gradient in the air fluid from the calorimeter wall and the fluid [3].

A three-dimensional numerical model was developed using the computer code FLUENT. A geometric mesh was set on the program according to the geometry of the calorimeter as shown in Fig. 3, together with the thermal properties of the materials (solids and fluids) involved into the experiments.

Mass, momentum and energy equations were solved simultaneously in the fluids domain (water and air) and in the solids domain of the system. The most important assumptions in the mathematical formulation of the cavity calorimeter were the following: the flow is in steady state, the fluid is radiatively non-participating, the Boussinesq approximation is valid, air physical properties are temperature dependent and solid properties are not temperature dependent. The conservation equations in primitive variables are solved using the finite element method.

Boundary conditions for the cooling fluid are: Constant temperature, pressure and mass flow rate at the calorimeter inlet. Physical properties (density, viscosity, thermal conductivity) are not temperature dependent. Boundary conditions for the solid are: External walls are assumed adiabatic. Internal walls (in direct contact with the fluid) are considered as a heater walls and modeled as a boundary conditions with a different heat flux profiles. The calorimeter was modeled with tetrahedrons finite elements (1 millions for air in the cavity, 0.9 millions solid calorimeter walls and 0.5 millions water cooling fluid). Steady state solution of the Navier-Stokes equations with conjugated heat transfer calculation was performed.

To simulate the interaction between the air inside of the cavity with the air outside, an atmospheric air domain, bigger than the air domain inside the calorimeter was considered, as it is shown in Figure 4. It was assumed that the inlet velocity boundary condition for the atmospheric air volume was 1 m/s going form left to right, as if it was win velocity.

In order to get confidence with the simulation, some sensitivity analysis where performed, such as the influence of the cooling flow rate, the calorimeter material (steel, cooper, etc.), the velocity and direction of wind, heat flux profile of incident irradiance, and after that, a comparison with experimental data was carried out.

As an example of typical numerical results, figure 5 shows the steady state temperature field in a central longitudinal plane of the calorimeter. It is possible to see how the heat is absorbed and transferred. The hot zone corresponds to a region on the copper wall and on the air close to it. The coldest zone corresponds to the water field and the solid field which are far from the hot spots. The stratification of the air inside the cavity and in the aperture is very clear; the air tends to go up due to buoyancy forces.

It is possible to observe the energy losses due to the effect of the natural convection mode in the calorimeter aperture and the movement of the air due to the density dependence with the temperature. The air with higher temperature rises to the top of the cavity and there is cooled down by the wall in contact with the water, so the maximum temperature in the calorimeter wall occurs at different position in the region where the highest radiative flux is absorbed.

Figure 6 shows the temperature profile in the calorimeter heated wall and in the fluid inside the cavity. As the figure shows, the maximum wall temperature occur at 8 cm from the aperture and the maximum air temperature near 12 cm. In the upper cone cavity thermal equilibrium occurs, the wall and fluid temperature are the same.

An easy to use user interface in the Windows environment using Borland Delphi has also been underdeveloped. It is aimed at providing the users with a friendly environment for the selection of different simulation options. The simulation may be achieved through a selection of different options, including different modes of the still, weather data, design or simulation options and other input variables. The output window summarises the simulated performance of the solar still and the heat-exchangers, and this data can be viewed in separate files. Some of the interface windows are shown in Figures 1 and 2. The current user interface is operational and has very basic features. To upgrade the interface, several in-built functions and graphical display features may be easily incorporated in the future.

As is testified by the technical research of the issue since two years ago, selecting the exact goal and making good research plan is the first ink to fulfill the research successfully, which is more important than implementing the technical scheme. The research adheres to many correct principles in the process of formulation and implementation including fitting the urgent need of social production, ensuring the scientificality, practicability, prediction and innovativeness. The project has achieved prominent economic benefit and big technical breakthrough, the main source of which is to declare, formulate and set up the task on the research and development of the protection-styled aquatic cultivation industry of solar pond.

6. The existing problem

Though the overall technical level has reached the designing target and it has made prominent economic benefit, many problems should be studied and solved in the subsequent industrialized development. The problems are as follows:

6.1 The leakage-prevention problem of solar pond

The key to this problem is to develop the lining materials which are of low cost, high-intensity and ageing-resistance.

6.2 The odor-removing problem of solar pond

Tasks to solve the problem are to select good underwater antiseptic materials, to study and develop the processing devices for optimizing the water quality of solar pond and to develop the technique of water-purification of the solar pond cellular project.

6.3 The support-assembling problem

The combined translucidus support should be developed in the shallow-styled solar pond and the conservatory-modeled solar pond.

6.4 The problem of monitoring and controlling water quality

Study and research the automatic monitoring equipments of multiple channel and high performance to monitor the physico-chemical target of the underwater water quality.

6.5 The technology-integrating problem

In order to achieve the sustaining development and to make constant innovations, technology of conservatory project, technology of solar pond project, technology of aquatic biological science and technology of processing water quality should be integrated to work.

Bibliography

1 F. Zangrando(U. S.A.),On the hydrodynamics of salt-gradient solar ponds,323,VOLUME 46,NUMBER6,1991

2 S. Folchitto(ltaly),Seawater as salt and water source for solar ponds,343, VOLUME46, NUMBER6,1991

3 K. Kanayama, H.lnaba, H.Baba(Japan),experiment and analysis of practical-scale solar pond stabilized with salt gradient, VOLUME46,NUMBER6,1991

4 F. Collado and P. Lowrey(U. S.A.),Temperature, thermal efficiency, and gradient performance from two seawater-SZ solar ponds,361, VOLUME46,NUMBER6,1991

5 S. J.Kleis and L. A.Sznchez(U. S.A.),Dependence of sound velocity on salinity and temperature in saline solutions,371, VOLUME46,NUMBER6,1991

6 G. Lesino and L. Saravia (Argentina),Solar ponds in hydrometallurgy and salt production,377, VOLUME46,NUMBER6,1991

7 E. Wilkins(U. S.A.),Operation of a commercial solar gel pond,383, VOLUME46, NUMBER6,1991

8 M. K.Smith and T. A.Newell(U. S.A.),Simulation and economic evaluation of a solar evaporation system for concentrating sodium chloride brines,389, VOLUME46, NUMBER6,1991 Vlume 46, 1991-List of Contents and Author lndex,401, VOLUME46,NUMBER6,1991

9. Jinlong Cao, Scheme of Sunlight Project of Fifteen Coasts, Collected Works of Discussion on Clean Energy Technology of China, Oct.2000

It is already well known that thermal solar systems provide benefits for the environment. Nevertheless in the present investigation this was confirmed once more by the short energy payback times. The minimal values for the solar domestic hot water systems were

1.3 years (system H13) and 2.0 years for the solar combisystems (system C2).

In addition the development of the system costs is quite positive. Figure 3 shows average values of the system costs (including VAT and installation) for solar domestic hot water systems (SDHW) and solar combisystems (COMBI).

It is obvious that the price degradation observed in the past for SDHW systems nowadays also appears for solar combisystems.

Although thermal solar systems, especially with regard to single and double family houses, are usually not sold in order to save money it is important to know the price of one kilowatt hour of solar energy. Therefore figure 4 shows the energy savings and the heat prices of the systems investigated.

The cost accounting was performed on the basis of the annuity method (interest rate 4 %, lifetime 20 years) without taking into account subsides. Due to the fact that the systems C9, C10 and C11 were combisystems with an integrated gas burner the heat prices of these systems are not included since they are not direct comparable with the other systems.

On the basis of theoretical considerations it can be expected that the heat prices will increase with increasing (fractional) energy savings, and therefore decreasing system efficiency. This consideration is valid for solar domestic hot water systems as well as for solar combisystems. However figure 4 shows that this effect can (up to now) not be observed. This indicates that heat prices and system costs are primarily determined by other parameters such as the system technology or the individual cost structure of the manufacturers or traders. On the basis of this fact it can in general be concluded that most of the systems are still far away from a cost minimum.

Figure 4 shows also that the minimum heat price is in the range of 0,12 to 0,13 €/kWh. Taking into account the present German subsides of 110 €lm collector area leads to heat prices of approximately 0,10 €/kWh. This value is very close to the current price of heat generated with individual oil or gas boilers.

3 Conclusions

Compared to the previous comparison test carried out in 1998 solar technology made one more step towards professionalism. Most of the investigated products convinced due to good quality and performance. This was indicated e. g. by the fact that during reliability and durability testing only one collector failed with a major failure. Furthermore no significant lacks of safety such as underdimensioned expansion vessels, have been noticed. This comparison test showed that by now thermal solar systems are well introduced to the market and are a serious technology for the generation of heat for domestic hot water and space heating.

/1/ Stiftung Warentest, Berlin, Test solar domestic hot water systems: "Eine Technik zum Erwarmen", Consumers’ magazine "test", 4/2002, pages 56 — 61, April 2002

/2/ Stiftung Warentest, Berlin, Test solar combisystems: " Sonne tanken",

Consumers’ magazine "test", 4/2003, pages 69 — 73, April 2003

/3/ DFS hot water comfort test. Described in: Stryi-Hipp, G., Kerskes, H., DrQck, H., Bachmann, S.: Abschlussbericht des Projekts "Testverfahren fQr Solaranlagen zur kombinierten Brauchwassererwarmung und Raumheizung (Kombianlagen)", Institut fQr Thermodynamik und Warmetechnik (ITW), Universitat Stuttgart, 2001

Acknowledgement:

The results presented in this paper have been obtained by the engineers of the Test and Research Centre for Thermal Solar Systems (TZS) located at ITW: S. Bachmann, S. Fischer, M. Hampel, H. Kerskes, M. Peter und E. Streicher. This work was partly financed by the Deutsche Bundesstiftung Umwelt. The authors gratefully acknowledge this support.

A very significant parameter for solar thermal plants is the temperature level of a process; it has a major influence on the energy supply system. The second important characteristic number of a process is the energy consumption. The combination of these two values for all important processes in a single sector of industry has not been investigated yet and was thus the innovative part of this project.

Results

Those industry sectors which were actually to be examined for solar thermal use more closely, were

• food and beverage industry,

• textile industry,

• chemical industry and

• production of plastic goods.

One important result from all investigations in industry sectors and in enterprises is that the analysis for solar thermal supply also depends strongly on location and specific parameters (e. g. power supply, product range, mode of operation).

General statements about the appropriateness of the solar thermal process heat within an industry sector do not permit conclusions on individual enterprises.

The total annual requirement for energy of the Austrian industry is approx. 264 PJ (incl. approx. 67 PJ for steam generation). The low-temperature heat ratio (under 100 °C) is slightly over 10 PJ. After consideraton of possible heat recovery a little more than 4 PJ for solar thermal process heat remain of interest.

Conclusions

There are some industry sectors, in which solar thermal process heat seems to be of interest as an auxiliary system to supply energy. But only after all aspects of energy
efficiency (insulation, power supply system, heat recovery) have been regarded in detail, follows as last step the replacement of fossil sources of energy by renewable forms of energy. The supply by renewable forms of energy in an energetically inefficient system is economically and ecologically useless.

How an individual enterprise is able to assess and optimize its existing energy system and how during this optimization the integration of solar process heat takes place, is topic of the follow-up project SolProBat.

The result of this study the potential for the solar thermal supply of industrial processes in the Austrian industry — has only to be seen as an intermediate result. The results and pilot plants focused in the follow-up projects can initiate a new market for solar plants in the industry within the next years.

The interest of the producing trade and also the solar technique industry is however already very large and demands in consequence further development in the areas of the collector development (costs and/or higher temperatures), the integration of solarthermal plants into existing heat supply systems as well as the development of financing concepts, which make a conversion possible by the industry. At present a broad conversion is braked by solar plants for the process heat supply particularly by the short amortization periods of maximally three years, demanded by the industry. The first estimations of the potentials for solar thermal plants in industry and trade result in a portion of 3,0 % of the industrial for Austria. This corresponds to a demand of over 4 PJ on basis of the energy consumption of 1998, for their covering a collector surface of approx.. 2,5 million is square meters necessary. As comparison to the fact it is marked that the entirely installed collector surface in Austria in the year 2001 approx.. 2 million square meters amounted to.

To predict the performance of solar collectors, it is necessary to evaluate the radiation exchange between a surface and the sky. The sky can be considered as a blackbody at some equivalent sky temperature Ts so that the actual net radiation between a horizontal flat plate and the sky is

Q = zAa(T4 — Ts4) (2.1)

The equivalent blackbody sky temperature, equation 2.2, accounts for the fact that the atmosphere is not at a uniform temperature and that the atmosphere radiates only in certain wavelength bands. Several relations have been proposed to relate Ts for clear skies to measured meteorological variables. Berdahl and Martin (1984) determined such a relationship by relating the effective sky temperature to the dew point temperature, dry bulb temperature, and the number of hours from midnight t by the following equation

1

Ts =7p.711+0.0056^ +0.0000737p2 +0.013cos(15f)]4

The prediction of solar systems performance requires knowledge of the amount of solar energy absorbed by the collector absorber plate. The solar energy incident on a tilted collector consists of three different distributions: beam radiation; diffuse

radiation, and ground-reflected radiation. In this study the absorbed radiation is calculated by isotropic sky model [1]:

(2.3)

Assuming no directional dependence of є and a, the following relationships are valid. The total emittance is found by integrating over wavelengths from zero to infinity [2]:

(2.4)

The total absorptance for a surface for a given incident spectrum is found by integrating over wavelengths from zero to infinity [2]:

(2.5)

The experimental investigations as well as the simulation study show that the solar gains under low-flow and high-flow operation are quite similar. On clear and cloudless days the solar system gain under high-flow operation exceeded the solar gain under low-flow opera­tion by up to 5.7 %, while for unsettled irradiation the thermal performance of the low-flow system was up to 4.6 % better. Under intermittent sunshine the low-flow operation benefits from a reduced number of switching cycles and from longer operation intervals. Periods of low irradiance are better utilized under low-flow operation. If the electric power consump-

tion of the collector pump is taken into account and the savings of primary energy are re­garded, then the low-flow operation comes generally out on top.

The analysis of the temperature distribution in the domestic hot water stores showed that the better stratification caused by low flow rates is responsible for the lower return tem­peratures in the collector loop. Thus the mean collector loop temperatures under low-flow and high-flow operation are nearly equal and the similar solar gain of both systems is ex­plainable.

This investigation shows that for small solar domestic hot water systems with an internal heat exchanger a low-flow rate in the collector loop could be advisable, at least if a uniform flow distribution inside the collectors is guaranteed by the construction. Furthermore the development of pumps that are optimized for low flow rates would make low-flow operation even more advantageous than it is already now.

The tank is divided into three parts, where the upper part, middle part and lower part correspond to flow through from flap 3, flap 2 and flap 1 respectively. The power supply to each part of the tank and the temperatures measured in the stratification inlet pipe is shown in Fig. 7. The inlet temperature is shown in Fig. 5. The power supply to each part of the tank is a result of the water flowing through the inlet, the downward flow in the tank, the tank heat loss and the downward thermal conduction. From the figure it can be seen that the power supply to the upper part, flow through flap 3, is high in the first 10 minutes after which flap 2 opens and the power supply in the middle part of the tank is then highest. Through the whole experiment water is sucked in through flap 1 and the power supply to the lower part of the tank is first of all caused by the downward flow in the tank towards the outlet at the bottom of the tank. The temperature in the inlet pipe is higher at the bottom of the pipe than in the middle and the top of the pipe and only slightly higher in the middle than in the top. This also indicates that cold water enters the lower inlet.

An additional experiment was conducted in order to verify that water was sucked in through flap 1. The test conditions were the same as in the previous experiments but flow

through the middle and lower inlets, flap 1 and 2 were physically prevented. The flow was only allowed through the upper opening, flap 3. The inlet temperature and the temperatures in the pipe with and without flap 1 and flap 2 closed is shown in Fig. 8. From the figure it is obvious that the temperature was practically the same in the whole pipe when flow through flap 1 and flap 2 was physically prevented.

In contrary, two temperature levels were measured throughout the whole experiment when all three flaps were in operation.

The power supply to the tank and the temperatures in the inlet pipe in the two experiments are shown in Fig. 9. The figure shows that the power supply to the upper part of the tank is almost the same in both experiments while the power supply to the remaining part of the tank differs in the experiments. The most significant difference between the experiments is the power supply to the lower part of the tank after about 26 minutes. At this time only a third of the tank volume has been replaced. After 34 min, nearly the total heat supply is moved to the bottom part of the store for the case that all flaps are in operation, whereas the heat is almost moved completely to the middle storage part, if the flow through the two lower flaps is prevented. Further the figure shows that the inlet temperature from the upper and middle flaps is several degrees lower when all three flaps can move freely.

Conclusion

A marketed stratification inlet pipe was investigated by means of Particle Image Velocimetry (PIV), a non-intrusive optical method and by temperature measurements inside and outside the inlet pipe. The pipe consisting of three compound pipes was built into the centre of a glass tank with a volume of about 140 litres. The functioning of the pipe was investigated for a typical operation condition where the tank was heated from about 20 °C to about 40 °C with a volume flow rate of about 2 l/min. The volume flow rate used in the experiment is close to a typical volume flow rate that develops from mixed (natural — forced) convection in the pipe when a compact heat exchanger is integrated below the stratification inlet device.

It was fount that thermal stratification was built up in a good way with the used test conditions, but also that effects of inertia influence the flow trough the stratification pipe and thereby the thermal stratification. If the inlet temperature was low compared to the tank temperature and slowly increased to the set temperature level, more cold water was sucked into the pipe from the lower flap at the beginning of the measurement.

Finally it was found that small amounts of cold water were sucked in through the flap at the bottom of the tank during the whole experiment that lasted for 50 minutes. This lead to a several degrees reduced inlet temperature at the top of the tank.

Further detailed investigations are needed, before the function of the stratification inlet pipe is full elucidated.

Reference

Loehrke R. I., Holzer J. C., Gari H. N., SharpM. K. (1979). Stratification enhancement in liquid thermal storage tanks. Journal of Energy, Vol. 3, No. 3, pp. 129-130.

Gari H. N., Loehrke R. I. (1982). A controlled buoyant jet for enhancing stratification in a liquid storage tank. Journal of Fluids Engineering, Vol. 104, pp. 475-481.

Davidson J. H., Carlson W. T., Duff W. S. (1992). Impact of component selection and operation on thermal ratings of drain-back solar water heaters. Journal of Fluids Engineering, Vol. 116, pp. 130-136.

Krause Th., Kuhl L. (2001). Solares Heizen: Konzepte, Auslegung und Praxiserfahrungen Shah L. J. (2002). Stratifikationsindlobsrar. Department of Civil Engineering, Technical University of Denmark, DTU.

Shah L. J., Morrisin G. L., Behnia M. (1999). Characteristics of Vertical Mantle heat Exchangers for Solar Water Heaters. Solar Energy, Vol. 67, No 1-3, pp 79-91.

Shah L. J. (2001). Heat Transfer Correlations for Vertical Mantle Heat Exchangers. Solar Energy, Vol. 69, No. 1-6, pp 157-171.

Knudsen S. (2003). Analysis of the flow structure and heat transfer in a vertical mantle heat exchanger. iSES International Solar World Congress, Gothenburg, Sweden, CD — ROM P6 55

Jordan U., Furbo S. (2004). Impact of inlet devices on the thermal stratification of a storage tank. EuroSun European Solar Energy Conference, Freiburg, Germany. Paper. Raffel M., Willert C., Kompenhans J. (1998). Particle Image Velocimetry. A Practical Guide. Springer Berlin. ISBN 3-540-63683-8.

SHAPE * MERGEFORMAT