t454

Experimental Investigation and Mathematical Modeling of a Low Energy Consuming Hybrid Desiccant Cooling System for the Hot and Humid Areas of Pakistan

130

Chapter 7 Solar Collector Design for Solar Assisted Desiccant Cooling

7.1 Introduction The last chapter has quantitatively shown the benefit of using solar collector to supply part of thermal energy for the desiccant cooling system. This chapter serves as an essential part of the study to show the feasibility of solar collector to supply part of the thermal energy needed for the desiccant cooling system. Both liquid and air collectors can be used for achieving this end but heat supply system based on the solar air collector offers greater opportunities. This is due to the ease with which the air collector can be integrated in the cooling system. Present chapter shows that thermal analysis of a solar collector for the Pakistani city of Karachi can yield valuable information to reinforce this argument. This chapter illustrates that from modeling point of view it is critical to be able to predict the useful energy gain of the collector under wide range of operating conditions. The low heat transfer properties of air due to its low heat transfer coefficient and low heat capacity necessitate a study of the factors which have an effect on the performance of air collector. The objective of this study is to ascertain two important issues regarding the use of collector. First whether a collector with staggered fins is capable of delivering the amount of temperature and energy needed for regeneration of desiccant wheel. Second area to be explored is to observe what effects the collector slope has on the energy collection ability of the solar air collector. Computation fluid dynamics (CFD) is a suitable technique to perform detailed fluid flow and heat transfer analysis of a phenomenon. This chapter encompasses computational fluid dynamic technique and TRNSYS simulation for the climate of Karachi to predict the performance of solar air collector. Presence of fins is known to have positive effects on the heat transfer in collector absorber. Here an attempt has been made to evaluate the benefits of fins inside the absorber air passage or duct. CFD analysis technique is found to be suitable for micro analysis of the processes taking place inside the solar collector air duct; this analysis will be carried out under steady state working conditions and under constant heat flux to qualitatively predict behavior of the collector. The macro analysis of the collector on TRNSYS provides valuable quantitative information about the behavior of collector in given weather conditions. This chapter suggests that the TRNSYS and CFD analysis techniques of the solar air collector are basically complementary in nature and they reinforce the notion that the solar air collector are a viable means to heat the air for the desiccant wheel regeneration purpose in the weather conditions of Pakistan.

7.2 Pakistan’s Solar Energy Resources In the experimental work performed during this study the heat supplied to actual test plant was obtained from a natural gas heater, however, it was assumed that part of the thermal load for regeneration air heating will be met by a solar air collector array. This methodology necessitated detailed analysis of solar air collector for the climatic conditions of Pakistan.


131

Use of solar radiation for replacing traditional fossil fuels is becoming increasingly common in the world because of feasibility of its application in common life. Solar energy has many advantages such as it is inexhaustible, it is free, and its use does not cause any environmental damage. For space cooling purpose availability of solar energy coincides with the need for cooling. Similarly the summer peak demand of electricity due to extensive use of air conditioners matches with the peak solar irradiance, this situation offers an opportunity to use solar energy in the space cooling system. Solar radiation striking a surface outside the earth atmosphere at a mean sun earth distance is 1395 W/m2 [Duffie, 1991] which is known as solar constant. Any radiation striking a surface on the earth will have a value lower than solar constant due to attenuation in atmosphere. Despite being dilute solar energy nevertheless can be utilized for many useful purposes such heating. In the case of South Asia sunlight is abundantly available during most of the cooling period. These and other encouraging attributes of solar energy made it suitable candidate for use in the desiccant cooling system. Pakistan is a sun-belt country, lying between the latitudes of 24 and 33 degrees, it receives significant amount of sunshine during summer and winter. This favorable geographical position places it among the countries which are endowed with enormous potential for solar energy utilization. The sunshine hours available in Karachi for a cooling season are given in the Figure 7.1; this figure also shows the effects of cloud cover in the form of reduced radiation during the monsoon period, lasting from late June to early September. Figure 7.2 shows the extra terrestrial, total and beam radiation for first week of June for Karachi. It is apparent from these two figures that the amount of radiation available for this city is considerable that can be utilized as a heating source for the desiccant cooling system.


132

0

50

100

150

200

250

300

350

Janu

ary

Febru

ary

March

AprilMay

June

July

August

Septem

ber

October

Novembe

r

Decem

ber

Time (Month)

Suns

hine

Hou

rs (h

our)

Figure 7.1 Monthly mean sunshine hours for Karachi

0

500

1000

1500

2000

2500

3000

3500

4000

4500

5000

3660 3680 3700 3720 3740 3760 3780 3800 3820 3840 3860

Time of year (Hours)

Sola

r Ene

rgy

(kJ/

hour

.m2)

Extraterrestrial radiation

Total radiation on surface

Beam radiation on surface

Figure 7.2 Extraterrestrial, total and beam solar radiation for the first week of June for Karachi


133

7.3 Solar Collector A solar collector is a type of heat exchanger which converts solar radiation falling on its absorber surface into heat that is used to raise the temperature of a fluid [Duffie, 1991]. Fluids used to remove the heat from the collectors include water, air and thermal fluids. Collectors may have different shapes and design to achieve different fluid temperatures. Solar collectors can be mainly classified into two types: concentrating and non-concentrating. They are further categorized on the basis of their concentrator optical properties and operating temperature that can be obtained at the receiver. Concentrating systems are used to produce higher temperatures. Fixed orientation flat plat collectors are simple in design and are most generally used for medium temperature use. Concentrating type collectors use reflection or refraction of solar radiation to focus it on a smaller area thus increasing its flux density and hence its ability to achieve higher temperatures, Table 7.1 explain the temperatures that can be achieved from different collector designs and Figure 7.3 illustrates efficiency curves of different collector designs. Table 7.1 Characteristics of typical solar collectors Solar collector technology

Max operating temperature (ºC)

Concentration ratio

Tracking Reference:

Flat plate Collectors

Tambient + 100

1 No tracking [Duffie, 1991]

Solar fresnel reflector technology

260-400 8-80 One-axis

Parabolic trough collectors

260-400 8-80 One-axis

Parabolodial dish concentrators

500-1200 800-8000 Two-axis

[Luzzi, 2004]

Due to its low cost and simple design and operation the most frequently used type of solar collectors is flat plat collector; it is usually none concentrating and has fixed south facing orientation. It consists of a black colored absorber plate, transparent cover for decreasing radiation and convection losses and side and back insulation for lessening convection and bottom losses. Figure 7.4 illustrate the input energy and different form losses for a collector. The functional part of a collector is its absorber which has passages for the fluid to carry away the heat produced by the absorption of solar radiation. An absorber is always heated from the top side by solar radiation which is not a good direction from the point of view of heat transfer. Flat plate glazed collectors are used when moderate temperatures are required these collectors utilize both beam and diffused radiation for heating the fluid and are non tracking or fixed orientation type. Since diffuse light is available during cloudy days a flat plate collector can supply heat during overcast days albeit at reduced rates. Based on the cover, flat plate solar collectors are of two types.


134

• Unglazed solar air collector • Glazed solar air collector

Unglazed collectors have the simplest design and are used in situations where lower temperatures are acceptable such as swimming pool heating and space heating. Since these collectors operate at lower temperatures they have lower thermal losses, absence of transparent cover also eliminates much of the optical losses consequently these collectors tend to have higher efficiencies compared to glazed collector. A glazed solar collector on the other hand has a transparent cover over the sun facing side, this cover allows the visible sun light to pass through it and absorbed by the black colored absorber plate. The glass cover is opaque to long wave length thermal radiation emitted by the hot absorber plate. Consequently because of these two features of the cover heat is trapped inside the collector in a process commonly referred to as green house effect. The transparent cover also reduces convection losses from the collector front. The overall impact of these factors is relatively high operating temperature of glazed collector in comparison to unglazed collectors.

7.4 Practical Considerations for Solar Collectors Mechanical design of a solar collector has direct bearing on its thermal performance. A properly designed and fabricated solar collector should have following feature to attain maximum possible efficiency and service life [Solarexpert, 2005].

7.4.1 Glazing Materials The function of the glazing is to permit maximum possible short wave solar radiation and minimize the leakage of long wave thermal radiation from the absorber collector. The glazing should be transparent having a high transmittance for incident short wave solar radiation and low transmittance for thermal long wave radiation emitted by collector. For longest life and maintained transmittance, the most appropriate glazing material is tempered plate glass. It has above 90 percent transmittance for short wave solar and very low transmittance for long wave thermal radiation emitted by the absorber. Of the various grades of tempered plate glass, low-iron glass has the highest transmission and lowest reflection of sunlight. These properties result in significant increases in collector efficiency. Although low-iron glass is slightly more expensive than ordinary tempered glass but the cost premium for low-iron glass is smaller than the increase in efficiency. Plastic glazing of various types is some times used on some solar collectors to reduce weight and cost, but it may reduce performance and lifetime. Plastics inside a well sealed collector may deteriorate rapidly and will outgas, depositing a haze of condensed oily liquid on the inside surface of the glazing. Such haze will seriously reduce the collector efficiency by blocking solar radiation. Plastic used in a collector may also result in limitations or restrictions of collector use in high fire-risk zones by local building and safety departments.


135

7.4.2 Insulation Solar collector must be insulated form all sides to minimize the heat losses and increase thermal efficiency. Different materials are used for insulation, best one are those which have low thermal conductivity low density, long life and low cost. An otherwise well-designed solar collector will experience excessively high stagnation temperatures that will cause polymer based insulations like polyurethane and polystyrene to outgas and rapidly destroy the effectiveness of the collector. Urethane and closely related products may be prohibited in collectors in fire hazard areas, due to their ability to produce toxic fume at high temperature.. When these materials are used in solar collectors, they should be used underneath a substantial blanket of other insulation material, such as binder-free fiberglass to reduce the hazard of exposure to high temperatures, and should have an intervening tight vapor barrier. When fiberglass is used, a larger thickness is required than would be needed for urethane or related products. Case insulation is not the only important thermal insulation in a collector, the absorber plate and connecting tubing penetrating the enclosure must be thermally insulated from the case at all points of support. Heat losses can be high if either the absorber or tubing touches the case or is supported through heat-conducting materials to the case.

7.4.3 Enclosure The enclosure is used to contain insulation, provide support for the absorber and glazing, and to protect the collector from heat loss due to convection. Enclosure also serves as barrier against moisture which can get into the insulation from rain and dew. Enclosures are usually made from various types of materials; some commonly used materials are aluminum sections with sheet aluminum back, galvanized steel and painted mild steel. The desirable features of a good collector enclosure are its weather resistance, dimensional integrity, strength and leak tightness. For a leak tight design the number of joints and seams are usually minimized and completely sealed. Aluminum is a good metal from the point of view of its light weight and corrosion resistance but its use in polluted environment can shorten its life.

7.4.4 Selective Surface To achieve maximum possible air temperatures a solar collector must have high absorptance for solar radiation and low emittance for thermal radiation or long wave radiation. A surface which has higher absorptance for short wave solar radiation and low emittance for long wave thermal radiation is called a selective surface [Tiwari, 2002]. A flat plate collector operating at medium temperature emits all energy at wave length in excess of 3µm. A collector with selective surface will thus have minimal radiation losses and will thus be able to achieve higher operating temperatures and efficiency. For collectors intended to operate in the upper end of the medium temperature range and in the high temperature range, selective absorber coating can improve efficiency by decreasing radiative losses from the aperture. An effective selective coating is black chrome, applied by electroplating process over a nickel base. If applied on a material other than copper, the plating must be applied to both sides to avoid corrosion. Good black chrome plating on nickel base has proven stable and not susceptible to high stagnation temperatures or aging. Figure 7.5 shows that the efficiency of


136

collectors with selective coated absorbers and black painted absorbers; it is obvious that selective coating results in a better efficiency.

7.5 Solar Air Collector Solar air collectors are effective devices to harness solar radiation for space heating and cooling and other purposes. Detailed thermal investigation of a solar air collector was first performed by Whillier [Whillier, 1964], who showed that the efficiency of solar air collector can be improved by fins. Solar air collectors may have a number of designs however they can be divided into two categories according to design of their absorber; porous absorber and non-porous absorber. A porous absorber design allows the solar radiation to penetrate some depth of the absorber and heats the air flowing though the absorber. The absorber in this type may be fashioned from slit or expended metal sheet, overlapped glass plates, and transpired honeycomb. Air collector with overlapping glass plate absorber was first analyzed by Selcuk [Selcuk, 1971]. Because of the need for greater mass of the absorber, these collectors tend to be massive and hence have thermal inertia or lag. These collectors also tend to have larger air pressure drop due to the shape of the absorber. The implication of the larger pressure drop is that such collectors require more fan power to maintain air flow. Non-porous absorbers on the other hand are fabricated from metal sheets formed to make a passage for air flow.. The air may flow between the transparent collector cover and absorber plate or in a channel below the absorber plate. The channel usually has fins or corrugation or roughness to increase heat transfer between the absorber and air. Since these collectors can be made from light gage metal sheets they tend to be lighter with smaller pressure drop Solar air collectors because of their simple construction and low cost are extensively used in the world for heating purposes. Presently there are 1.16 million square meter air heaters installed in the world supplying around 1.1 GW of thermal energy [REW, 2006]. A solar air collector is simple in construction and can be fabricated from relatively cheap materials which may not necessarily by expensive corrosion resistant alloys.

7.5.1 Advantages of Using Solar Air Collector Both liquid and air type solar collectors can be used to heat regeneration air for the desiccant cooling system. In comparison to air collectors the liquid collectors can achieve higher temperatures and efficiencies because of superior thermal properties of water. Furthermore liquid type collectors have a greater variety of designs that allows higher temperatures to be attained. However for regeneration air heating a liquid type solar heater would require an extra heat exchanger and pump adding to the system complexity and cost. Moreover the stagnation temperatures of a concentrating type collector can be very high that requires a drain back system or some other safety measures to ensure that if this situation arises it does not cause any damage or loss of performance. An air collector on the other hand does not need to be pressurized and also there is no requirement for an intermediate heat exchanger as same air can be sent through the solar collector and returned to the


137

system. Since desiccant cooling system is usually located on roof tops it is easier to locate air collector near the plant. It is also possible to integrate the solar air collectors in the roof of the room housing the desiccant cooling plant this will result into a shorter ducting length and small pressure drop. Since the air is at ambient pressure the collector and accompanying duct work can be made of lighter material. Furthermore air is lighter than water consequently the operating collector load on the structure is lower. Leakage in the case of liquid type solar collector can cause problem and is usually repaired by shutting the system, while in the case of air heater leakage, except for a loss of some heat, does not pose any major problem.

Useful Energy

Solar Radiation

Re-radiation

Reflection

Convection Bottom Losses

Figure 7.4 Heat losses in a flat plate air heater

Figure 7. 3 Efficiency curves for different classes of solar collectors, [REW, 2006]


138

Insulation Fins

Transparent Cover

Absorber plate

Figure 7.6 Solar air collector absorber plate and fins

0

0.10.2

0.30.4

0.5

0.60.7

0.80.9

1

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2Tin - Tamb / G (K . m2 /W )

Effic

ienc

y G = 800 W/m2 α = 0.96 εBlack = 0.92 εSelective= 0.07 m = 100 g/s

Selective Surface

Black paint

Figure7.5 Efficiencies of black painted and selective coating absorber air collectors [Morhenne, 1990]


139

7.6 Thermal Analysis of Solar Air Collector It has been observed that the heat transfer coefficient between the absorber plate and working fluid of solar air collector is generally low. It is attributed to two contributory factors; first, the formation of a very thin boundary layer at the absorber plate surface commonly known as viscous sub-layer [Alok, 2006] and second, the low heat capacity of air [Lin, 2006]. The predominant mechanism of heat transfer through the boundary layer is conduction, since the conductivity of air is very low hence it can be said that this boundary layer works as an impediment to heat transfer from the absorber plate to bulk air flow. To increase the heat transfer between absorber and air two variables can be improved i. e., heat transfer coefficient and heat transfer area [Goldstein, 1976]. The convective heat transfer coefficient can be increased by providing artificial roughness on the heat transferring surface [Webb,1987] this technique tends to increase the heat transfer rate between absorber plate and air [Liou, 1993]. The other technique is to increase the heat transfer area which could result in an overall increase in heat transfer rate. One technique of increasing heat transfer area is to use fins below the absorber plate. In view of the fact that absorber channel receives heat from the top side; since solar collector is heated from this side by the sun. Air receiving heat from the under side of the absorber top plate tends to cling and move along top side with little heat transfer between top and underlying layers of air. This stratification results into a heat transfer value that is lower than what would bottom heating will yield. Fins can be helpful in overcoming this situation by conducting heat from the top side to the lower layers of air where it is conveyed to air by convection; thus improving solar air collector’s efficiency. Although the application of fins in a conventional solar air heater has been known to be an efficient method of enhancement of thermal efficiency of solar air heater and several experimental studies in this area have been carried out; literature search in this area revealed that few CFD studies have been done to investigate the effect of slope on the overall heat transfer in an inclined duct with staggered fins. In this work, an attempt is made to numerically predict effect of staggered fins, low flow velocity and solar collector slope angle on the heat transfer effects. Following section gives a general brief overview of modeling procedure in Fluent computational fluid dynamics software.

7.7 Modeling Procedure in CFD A Computational fluid dynamics model is based on the concept of dividing the solution domain into very large number of sub-zones. For each node and a given boundary condition mass, momentum, and energy conservation equations are solved, consequently the fluid temperature and velocity at that point are determined. Any small change occurring in the properties of the fluid at a point is used in the input to next node point. This procedure is repeated for all grid points; the end result of this computation is that all the properties, like temperature, velocities etc at all the points in the domain are eventually known. Since this procedure involves solution of very large number of equations iteratively, it is best done using the computation power of modern digital computers. A computer


140

performs extremely quick calculations of large number of equation giving detailed results in relatively short time. The estimation of temperature, pressure and velocity at each point in the domain can be easily accomplished using computational fluid dynamic software and results can be obtained in visual form. CFD modeling was originally developed for industrial application. Today it is used in research work, product development, and in almost all sphere of activity where a detailed picture of phenomena involving heat transfer and fluid flow, phase change etc is desired. A number of software based on CFD codes have been developed, few of them are: Fluent, Flovent, Phoenics, and CFX. Each software is usually supported by supplementary software for different applications, such as domain model preparation, mesh generation etc. Presently CFD techniques are increasingly used to model flow through solar collectors [Lee, 2001].

7.7.1 Using Fluent Software The analysis of a process in Fluent involves many steps in sequence to obtain the desired results. Figure 7.7 shows the steps needed to complete a Fluent analysis. Gambit is pre-processor software which facilitates the use of Fluent software. Its CAD interface facilities include Drawing two and three dimensional model, Defining their boundary conditions Generating the calculation mesh The calculation mesh is exported to Fluent software which is mainly used as a calculation tool. These three steps are explained briefly in the following sections. Other details can be found by referring to Gambit help menu. [Fluent Inc, 2004]

7.7.2 Defining Boundary and Continuum Conditions The Graphical User interface (GUI) of Fluent software is mainly a CAD interface with different commands. These commands have been set to draw the main components of any domain such as a solar air collector consisting of such components as collector absorber plate, fins and edges. In the case of complicated geometry it is also possible to use any CAD software such as AutoCAD to prepare the model. The model can then be imported in Fluent environment as DXF file. It is possible to draw the two dimensional models using edges and faces and three dimensional model using faces and volumes. Generally speaking two dimensional models are used for simple cases where it is anticipated that the fluid properties will not change in three directions. Two dimensional models require relatively small computational power and time because of smaller number of cells present in the domain. However two dimensional models give unrealistic results for complex flows where properties change in all directions (x, y, z, axes). Three dimensional analysis however significantly increases the computation time as more cells are present and more equations have to be solved - consequently convergence takes a longer time. Three dimensional CFD analysis gives a better picture of the processes taking place in the solution domain and is used where detailed investigation is required [Versteeg, 1995].


141

Defining boundary and continuum conditions of a model is a shared task between Fluent and Gambit preprocessor programs. Generally initial definition is performed in Gambit. Considering three dimensional modeling type of each entity such as pressure, at the external boundary of the domain should be defined. Furthermore every entity inside the domain should be defined as fluid, porous or solid. The fluid and solid zones to be modeled are sketched and the boundaries of the domain are assigned with respective boundary conditions which can be flow inlet, flow exit, symmetry, rotational axis etc. Therefore boundaries of the system should be placed at regions where conditions are known.

7.7.3 Generation of Mesh Gambit software is used for generating and analyzing grid. Boundary types can also be defined in Gambit; this task can also be defined in the Fluent preprocessor. Fluent can use unstructured as well as body fitted structured meshes with all types of mesh elements, such as triangular and quadrilateral elements in two dimensional, and tetrahedral, hexahedral, pyramid, and wedge elements in three dimensional analysis. This software is also capable of adapting all types of meshes during the solution. This allows one to refine the resolution in areas of significant gradients in order to prevent high numerical errors. After the pre-processing part has been finished the simulation setup can be stored in a case file. This file includes information on the grid file, the boundary conditions and the physical as well as computational models of the run. It is possible using Fluent software to simulate airflow under the influence of various parameters such as air velocity, location and size of opening etc. however it is crucial to define the program settings correctly which requires a good back ground knowledge of the software. Some relevant software settings are as follows: 3D, Single Precision Segregated Solver Implicit formulation Steady state scheme Boussinesq Model for modeling density k-ε turbulence model Thermo-physical properties In the present case the specific heat, viscosity and thermal conductivity values of air were kept constant at T = 328 K.

7.7.4 Turbulence Model Most air flow models in solar air collectors consider airflow to be turbulent; the reason for this being the very low value of air viscosity. Interaction with fins inside the solar collector also induces turbulence. No universal turbulence model exists that caters for all ranges of flow, internal or


142

external. The choice of turbulence model is not only dictated by the type of flow but also by the availability of computational power and the accuracy desired.

7.7.5 Residual Monitoring After completing all the relevant settings Fluent starts to perform the calculation in an iterative manner until a sufficient tolerance, defined by the user, is achieved. This means that solution will converge after attaining that minimum error. The calculation time increases when a smaller error is defined. There residual settings for the analysis were single precision, i.e. 10-3 for the continuity and 10-6 for energy equation.

Define the modeling goals

Create model geometry (Gambit or other CAD software)

Generate grid or mesh

Set up Solver and Physical model

Compute and monitor the solution

Examine and save the results

Figure 7.7 Steps involved in the CFD analysis procedure


143

7.8 Modeling Solar Air Collector – Fluent Analysis Fluid flow characteristics greatly influence the heat transfer mechanism in a collector absorber region. To fully appreciate these aspects a complete picture of the flow and heat temperature variations inside the domain must be available. The flow pattern and turbulence intensity are principle aspect influencing the local heat transfer coefficients. Researchers [Slanciauskas, 2001] and [Davidson, 1997] have shown that peak in local heat transfer coefficient occurs at reattachment point. Hence detailed information of flow pattern including flow direction change, plumes due to buoyancy effects and dead zones within the domain was desired for this analysis.

7.8.1 Solution Domain The solution domain for the CFD analysis was adapted from the experimental details of solar air collector as described by Kurtbas et al [Kurtbas, 2006]. The collector absorber was a rectangular duct with length of 1960 mm, width 930 mm and height of 60 mm. The duct had 32 fins of 0.012 m2 area each with a width of 200mm and height of 60 mm. In their experimental work Kurtbas et al placed fins in staggered fashion in the absorber duct. Because of the symmetrical nature of the solution domain a quarter portion of the collector was used for the domain to save computational time and power. In the experimental details, the thickness of the heated plate was only 1 mm, which was very small in comparison to the surface area normal to the heat flow. Hence the Biot number was also very small, that means that the internal resistance could be neglected in comparison to convective resistance. This permitted a uniform heat flux of 400W/m2 to be applied on absorber top surface, neglecting the conduction resistance within the plate. A 3-D analysis of heat transfer and air flow through the quarter portion of the solar air collector duct, inclined at 24º and 45º, with parallel fins, insulated bottom side was done on Fluent 6.2 CFD software.

7.8.2 Mesh Density Meshing of the domain was done using Fluent complementary software Gambit 2.2, with non-uniform quad grid of 1 mm grid size and 1.01 node size ratio. This size was suitable to resolve the laminar sub-layer. A general practice is to use fine mesh size in the area where greater details are desired such as fins and coarse mesh in areas of small changes in domain geometry. Non uniform high density mesh in inter fin region and on edges and corners were employed while low density mesh was adapted in the smooth wall regions. The mesh density for th CFD analyses was kept at 200 x 200 x 16. This helped in economizing the number of cells needed for acceptable computational time and computer memory, Figures 7.8 and 7.9 show the grid pattern for the flow domain. Quad mesh is very suitable for quadratic upstream interpolation for convective kinetics (QUICK) scheme [Fluent, 2005]. QUICK is a second-order discretization scheme which yields fairly accurate results on a lower cell count when flow is aligned with grid. This saved both computational effort and


144

time. An inlet air velocity of 0.03 m/s and outlet pressure equal to atmospheric pressure was used. Because of the scope of the study no attempt was made to validate the model with experimental results, good judgment based on knowledge of the basic underlying processes helped in ascertaining the quality of the results.

7.8.2 Turbulence Settings Standard k-ε turbulence model is the most widely used turbulence model due to its versatility. This model was used with the following settings: Inlet: Turbulence Kinetic Energy: 1 m2/s2

Turbulence dissipation rate: 1 m2/s3 Model constants are as follows: Cmu =0.09 C1-ε =1.44 C2- ε =1.92 Turbulent Kinetic Energy Pr = 1 TDR Pr = 1.3 Energy Pr = 0.85 Wall Pr = 0.85 Default settings were used for k-ε model.

7.8.4 Governing Equations The general transport equation in integral form can be written as [Fluent, 2005]

dVSdAVdAdVt VAAV ∫∫∫∫ +∇Γ=+∂∂ φφρφρφ ,

Unsteady + Convective = Diffusion + Generation. The above equation is the general transport equation and can be easily converted into flow and energy equations by the general variable Φ with: Φ = 1 will give continuity equation Φ = u, will give x- momentum Φ = v, will give y-momentum Φ = ho, will give energy equation


145

Boussinesq approximation model was used to obtain faster convergence of the solution. This approximation model assumes that the density of the fluid is constant in all solved equations except for the buoyancy term in momentum equation. For achieving rapid convergence a transient strategy was adapted.

Figure 7.8 Meshing in quarter portion of the air collector absorber region to save on computation time and memory


146

Figure 7.9 Unequal meshing in the domain - walls have fine and base has coarse grid


147

7.9 Results and Discussion - Fluent The CFD analysis was carried out in two phases. In phase 1, collector slope angle of 45º was chosen. In the second phase the slope of 24º was selected in light of the optimal solar gain for Karachi latitude 24º. The purpose was to find out the effect of slope angle on the performance of the collector. A closer inspection of the physics of heating by an inclined overlaying surface revealed that, under free convection conditions, increase in angle promotes heat transfer. 0 degree inclination or horizontal plate is the worst case scenario, while vertical plate is the best case Since heating was carried out from a surface overlaying the absorber, lower convective heat transfer was anticipated as compared to when the absorber was heated by an underlying plate. Top heating of absorber results into inefficient heating of air. Hot air due to its low density clings to under side of the top surface while relatively cool air because of high density stays in the lower regions. This stratification effect is clearly shown in 45º and 24º slope angles of the absorber in Figures 7.11 and Figure 7.17 Figure 7.12 shows the surface created in the mid region to illustrate the temperature of the air as it was heated during its passage through the collector air duct. It is obvious from this picture that highest temperatures are present the in middle region of the absorber. For a natural convection situation at lower angle 24º, a relatively lower heat transfer should be expected as compared to the higher angles of 45º as buoyancy forces are more dominant for steeper angles. This is true if the flow of air is determined by free convection only. The flow in the present case is low velocity forced flow hence effects of bouncy are not totally absent but play a part in the heat transfer process along with the forced convection effects. It can be noted from the Figure 7.10 for 45º absorber slope and Figure 7.15 for 24º slope that buoyancy in the flow reaches a higher level of strength at steeper angle as apparent from rising plumes of hot air. The reason for this type of flow behavior is the higher buoyancy induced flow at steeper angle. It is apparent that the sweep action of the bulk flow velocity is unable to dislodge the dead regions or diminish their size. This can be attributed to low flow velocity. Low flow velocity however resulted in a longer resident time for air in the heating zone which caused a larger temperature increase. This also suggest that the ideal inlet and outlet duct positions should be diagonally opposite so as to have maximum temperature gain, better heat collection at the outlet and reduced pressure losses It can also be noticed in both cases, from Figure 7.13 for 45º and Figure 7.16 for 24º , that fins are effective in transferring heat from top heat absorber plate to lower strata of air and hence raising air temperature. The meandering path lines of the air elements are clearly shown in Figure 7.14 as air being heated from top side passes along and through the gaps between fins. Left side shows the orderly entrance to side of collector. Longer streak lines imply longer residence time which results in air attaining a higher temperature in the collector at lower velocity. Figure 7.18 illustrates the air temperature variation along the height of the air passage in the absorber. It is obvious from the graph that the temperature of the air increases near the underside of the absorber the reason being the top heating of the absorber plate.


148

Figure 7.10: Contours of velocity magnitude for 45ºcollector inclination, the buoyancy effects are apparent as the lateral velocity is low, Inclination 45 degrees

Figure 7.11 Contours of temperature for 45º slope of collector, the air in contact with the underside of the absorber plate


149

(a)

(b)

Figure 7.12 Surface created in the mid region (a) to show the temperature of the air as it was heated during its passage through the collector air duct (b), highest temperature in middle region, collector slope was 45º


150

Figure 7.14 The velocity streak lines show the meandering path of streak lines as air being heated from above goes through the gaps between fins. Left side shows the orderly entrance to side of collector, collector slope 45 º

Figure 7.13 Air temperature in the middle of the absorber plate is highest, collector slope was 45º


151

Figure 7.15 Temperature contours on the mid plane (created virtually to view results) of the heater 30mm from the top. Rising plumes of air indicate that buoyancy effects are present, collector slope 24º

Figure 7.16 Temperature contours on the mid plane of the heater 30mm from the top, collector slope 24º


152

Air temperature in the middle region of absorber

0

20

40

60

80

100

120

140

0 10 20 30 40 50 60 70

Average Distance from top plate, mm

Tem

pera

ture

, C

Figure 7.18 Temperature variation of air along the thickness of the air passage in the absorber

Figure 7.17 Temperature profile for the top heated absorber slope 24º


153

7.12 TRNSYS Simulation of Solar Collector To confirm the findings of CFD analysis of the solar collector that the staggered fin solar air collector based on Kurtabas et al design can deliver required air temperature a simulation of a similar collector was carried out in TRNSYS software. The solar collector model used in the TRNSYS simulation project was based on Hottel-Whillier steady state model. The collector model Type73 based on this theory was selected from the library of the TRNSYS components. Important feature of Type 73 were the flexibility to adjust several parameter of the collector, these included the optical properties of the selective surface of the absorber, number of covers and series or parallel arrangement possibility. Some important parameters of the collector are given below in table 7.2 and Figure 7.19 shows a view of the solar collector simulation project. The type 73 solar collector was simulated in TRNSYS for the weather condition of Karachi. For this purpose a project was created in Simulation Studio with Type73 collector, weather data reader and radiation processor and output file printer. TMY2 weather file for Karachi was generated in METEONORM software and was used in the Type 109 data reader and radiation processor. The collector was simulated for time period of three days, with a simulation time step of 1 hour. The results of simulation were plotted using plotter type 65. It is obvious from the observation of Figure 7.20 and 7.21 that for same collector geometry the collector inclined at 24º collects larger amount of energy than the one inclined at 45º. Similarly the outlet temperature of air follows the same trend. The cause is the larger energy captured by the collector when it is receives the sun’s radiation at angle approaching normal to absorber top plate. This is only possible when the collector is inclined at an angle equal to local latitude which in case of Karachi is 24º.


154

Table 7.2 Parameters for the solar air collector model used in TRNSYS simulation No. Parameters Value

Chosen Unit Comments

1 Number of collectors in series 1 -- The numbers of collectors in series arrangement

2 Collector area 12 m2 Total area of collector array–for achieving suitable outlet air temperature

3 Fluid specific heat 1.007 kJ/kg.K Specific heat of air flowing through collector array

4 Collector fin efficiency factor 0.7 _ Measure of heat transfer from fin to maximum possible heat transfer from the fin

5 Bottom and edge loss coefficient

3 kJ/hr.m2

.K Loss coefficient of bottom and edges of the collector per unit aperture area

6 Absorber plate emittance

0.09 _

7 Absorber plate absorptance

0.95 _

From page 208, table 4.7.1 for selective absorber surface [Duffie, 1991]

8 Number of covers 1 _ one glass cover found to be suitable

9 Index of refraction of cover 1.526 _ Index of refraction of glass was used

10 Extinction coefficient thickness product

0.0026 _ The product of extinction coefficient and thickness of the glass.


155

Figure 7.20 Collector outlet air temperature and useful energy gain for the

climatic conditions of Karachi in three days of May, Ac = 12 m2, Collector slope 24º

Figure 7.19 View of TRNSYS Simulation Studio showing the solar air collector simulation project for weather conditions of Karachi


156

Figure 7.21 Collector outlet air temperature and useful energy gain for the climatic conditions of Karachi in three days of May, Ac = 12 m2, collector slope 45º


157

7.13 Conclusions Solar air collector heat transfer and flow characteristics in the presence of staggered fins were analyzed using computational fluid dynamic method. Shear stress transport k-ε turbulence model was employed for the flow situation in the finned collector absorber heated from top side. The analysis showed that regions around and between the fins promoted heat transfer from the absorber plate to air. The results showed a dependence of buoyant flow on low flow velocity and slope of collector. It was also demonstrated that at low flow velocity inclination of solar absorber affected the temperature and velocity profile with buoyancy playing a part in determining the flow pattern and the heat transfer rate. Stratified flow was found in the duct, top side being at highest temperature and bottom of duct at lowest temperature. This is attributed to the fact that collector was heated from the top consequently the air in touch with the underside of the collector absorber top plate was hottest and lightest. The density gradient with highest density in the bottom part and lowest at the top promoted and sustained stratification. The buoyancy effects were observed to be more pronounced at the 45º inclination than at 24º inclination. The bouncy induced flow through the space between the fins resulted because flow velocity was not very high to promote rapid mixing. Close examination however revealed that the difference in temperature and hence heat transfer was not very large between the collector slope angles of 24 º and 45º. The presence of staggered fins in the absorber resulted in the heat transfer to lower cooler air. Fins also helped in increasing the residence time of air consequently increasing air temperature gain. Presence of fins also supported to the mixing of air masses at different temperature thus increasing the thermal energy yield. It has been shown that the presence of fins in the absorber channel enhances heat transfer:

• By increasing the heat transfer area, • By promoting mixing of air • By increasing residence time of air in the absorber channel

The actual position of a solar collector used in summer is that the sun is at its high altitude in the summer months. The maximum heat flux can therefore be captured by a collector which is inclined in such a way as to receive sun rays at normal angle. To receive sun rays at right angle a solar collector should have a slope equal to local latitude, which in case of Karachi is 24º. This brings into play two conflicting factors influencing the useful energy output of a solar air collector: the larger inclination angle favoring a more efficient collector from the point of view of heat transfer and smaller angle supporting the capture of more solar energy during summer months in the climatic conditions of Pakistan. Examination of the results solar energy gain was larger at 24º slope than the larger heat transfer gain due to 45º collector slope revealed that was because of these reasons that an angle of 24º was chosen for simulation purpose in TRNSYS.


158

A TRNSYS collector model Type73 simulated for Karachi summer climatic conditions produced outlet air temperature similar to the CFD predicted temperatures thus showing agreement between two methods.


159

References Alok Chaube, P.K. Sahoo, S.C. Solanki, “Analysis of heat transfer augmentation and flow characteristics due to rib roughness over absorber plate of a solar air heater” Renewable Energy Vol. 31, pp 317–331, 2006 Duffie, J.A., Beckman, A. B., “Solar Engineering of Thermal Process”.2nd Ed, John Wiley & Sons, New York, 1991 Fluent 6.2 Documentation, User’s Guide 2005, Fluent Inc. Goldstein, L., Sparrow, E. M., “Experiments on the transfer characteristics of a corrugated fin and tube heat exchanger configuration, Transaction of the ASME, Journal of Heat Transfer Vol. 98, pp 26–34, 1976 Jaurker, A.R., Saini, J.S., Gandhi, B.K., "Heat Transfer and Friction Characteristics of Rectangular Solar air Heater Duct Using Rib-grooved Artificial Roughness” Solar Energy, Vol. 80, pp 895–907, 2006 Karwa R, Solanki SC, Saini JS. Thermo-hydraulic performance of solar air-heaters having integral chamfered rib roughness on absorber plates, Energy 2001;26:161–76 Kurtabas, I., Emre, T., “Experimental Investigation of Solar Air Heater with Free and Fixed Fins: Efficiency and Exergy Loss.” Int. Jr. of Science and Technology, vol. 1, No., 1 pp 75-82, 2006 Lee C.K., Abdel-Moneim S.A., “Computational Analysis of Heat Transfer in Turbulent Flow Past a Horizontal Surface with 2-D Ribs”, Int Commun. Heat Mass Transfer, Vol. 28 No. 2, pp 161–70, 2001 Lin, W., Gao, W., Liu, T, “A Parametric Study on the Thermal Performance of Cross-Corrugated Solar Air Collectors”, Applied Thermal Engineering, Vol. 26, pp 1043–1053, 2006 Liou T-M, Hwang J-J, Chen S-H. Simulation and measurement of enhanced turbulent heat transfer in a channel with periodic ribs on one principal wall. Int. Jrn. Heat Mass Transfer, 36, pp 507–17, 1993 Luzzi, A., Lovegrove, K., Solar Thermal Power Generation, Encyclopedia of Energy, Vol. 5, 2004 Morhenne, J., Fiebig, M., “Development and Testing of a series of Optimized Modular Solar Air Heaters for Heating and Drying” Projekt 0335003E6 Institute for Thermo and Fluid Dynamics, University of Bochm, 1990 Renewable Energy World, vol. 9, No 4, pp 84 , 2006 Rau G, Cakan M, Moeller D, Arts T., “The effect of periodic ribs on the local aerodynamic and heat transfer performance of a straight cooling channel”. ASME 1998;120:368–75.


160

Tanda G. Heat transfer in rectangular channels with transverse and V-shaped broken ribs. Int J Heat Mass Transfer, Vol. 47, No. 330, pp 229–43, 2004 Chaube, A., et al. / Renewable Energy, Vol. 31, 317–331, 2006 Fluent 6.0 User’s Guide, vol. 2; 2001. Selcuk, M. K., Thermal and Economic Analysis of the Overlapped Glass Plate Solar-Air Heaters”, Solar Energy, vol. 13, No. 165, 1971 Selcuk, M.K., “Solar Air Heater and their Application” Solar Energy Engineering, Academic Press, New York, 1977 Slanciauskas A. “Two Friendly Rules for the Turbulent Heat Transfer Enhancement”, Int J Heat Mass Transfer, Vol. 44, pp 2155–61, 2001 Solarexpert web site: http://www.solarexpert.com/heatpanel.html, downloaded on 20 March, 2005 Tiwari, G.N., Solar Energy: Fundamentals, Design, Modeling and Applications, Narosa Publishing House, New Delhi, 2002 Versteeg, H. K, Malalasekera, W. “An introduction to computational fluid dynamics”. Longman; 1995. Willier, A., “A Black Painted Solar Air Heater of Conventional Design”, Solar Energy, No. 8, Vol. 31, 1964 Webb, R.L., Eckert, E.R.G., “Application of Rough Surface to Heat Exchanger Design”, Int. J. Heat Mass Transfer, Vol. 15, pp 1647–1658, 1997

Date post:	15-Jan-2016
Category:	Documents
Upload:	engr-muhammad-sohail
View:	216 times
Download:	0 times

t454

Documents