You are on page 1of 227

Experimental and Analytical Analysis of Perimeter Radiant Heating Panels

By Martin Kegel

A thesis presented to the University of Waterloo in fulfillment of the thesis requirement for the degree of Master of Applied Science in Mechanical Engineering

Waterloo, Ontario, Canada, 2006 © Martin Kegel 2006

I hereby declare that I am the sole author of this thesis. This is a true copy of the thesis, including any required final revisions, as accepted by my examiners.

I understand that my thesis may be made electronically available to the public.

ii

Abstract
In recent years the U.S. and Canada have seen a steady increase in energy consumption. The U.S. in particular uses 25% more energy than it did 20 years ago. With declining natural resources and an increase in fuel costs, it has become important to find methods of reducing energy consumption, in which energy conservation in space heating and cooling has become a widely researched area. One method that has been identified to reduce the energy required for space heating is the use of radiant panels. Radiant panels are beneficial because the temperature set points in a room can be lowered without sacrificing occupant comfort. They have therefore become very popular in the market. Further research, however, is required to optimize the performance of these panels so energy savings can be realized. An analytical model has been developed to predict the panel temperature and heat output for perimeter radiant panel systems with a known inlet temperature and flow rate, based on a flat plate solar collector (RSC) model. As radiative and convective heat transfer coefficients were required to run the model, an analytical analysis of the radiative heat transfer was performed, and a numerical model was developed to predict the convective heat transfer coefficient. Using the conventional radiative heat exchange method assuming a three-surface enclosure, the radiative heat transfer could be determined. Numerically, a correlation was developed to predict the natural convective heat transfer. To validate the analytical model, an experimental analysis was performed on radiant panels. A 4m by 4m by 3m test chamber was constructed in which the

surrounding walls and floor were maintained at a constant temperature and the heat

iii

iv . a 0.61m wide panel with 8 passes. simple algorithm which could be used to predict panel performance. and the panel temperature and panel heat output were within experimental uncertainty. concluding that the RSC model is a viable. Two radiant panels were tested.61m wide panel with 4 passes and a 0. The panels were tested at 5 different inlet water temperatures ranging from 50oC to 100oC. The RSC model panel temperature and heat output predictions were in good agreement with the experimental results.output from an installed radiant panel was measured. The RSC model followed the same trends as that in the experimental results.

Veronique Delisle. who funded this construction and Ted. which funded this project for 8 months. His time. In addition. Finally. I would also like to thank all those at Sigma Convectors. Vivek Kansal. Chris Hadlock.Acknowledgements I would like to begin by acknowledging Professor Mike Collins for his excellent guidance and outstanding advice. input and experience is greatly appreciated. Appreciation is extended to Prof. Jim and Clive for their valuable time and support dedicated towards completing the test chamber. D. for their financial support and assistance in constructing the radiant panel test chamber. A special acknowledgement is given out to Dave Herzstein and Anthony Gaspari. I would like thank Professor John Wright for his support to get the project underway. I would like to acknowledge Nathan Kotey. Colin McDonald and Shohel Mahmud for their input and assistance with this project. Inc. Schneider and Prof. I am also grateful for the National Sciences and Engineering Council of Canada who funded this research for the course of one year and the Ontario Graduates Scholarship Program. G. Victor Halder. Cronin for serving as members of committee and for their precious time and inputs to my thesis. v .

vi .Dedication I would like to dedicate this thesis towards the development of installing efficient radiant panel heating applications for the commercial sector.

1 3.2 3.1 3.4 1.3.2 1.Table of Contents Abstract Acknowledgements Table of Contents List of Tables List of Figures Nomenclature iii v vii x xi xiv Chapter 1 1.2.4 Test Results vii .3 Procedures and Data Processing 3.3 2.2 Radiant Panel Test Chamber 19 19 20 22 31 34 35 36 36 Purpose of Test Facility Radiant Panel Test Chamber Construction 3.6 Literature Review 8 8 11 12 16 16 17 Existing Mean Radiant Panel Temperature Models Existing Radiant Interchange Models Existing Convective Heat Transfer Models Currently Accepted Radiant Panel Model Existing Test Facilities for Radiant Heating Radiant Panel Standards Chapter 3 3.4 2.2 Temperature Controllable Walls The Ceiling Test Procedure Test Samples 3.1 1.2 2.1 3.2.5 2.3.3 1.1 2.5 Introduction 1 1 5 5 6 7 Background Industrial Motivation Analytical Motivation Objectives Thesis Outline Chapter 2 2.

3.1 5.2 5.7 4.3.3 Heat Transfer Model 39 39 46 52 53 53 55 56 57 58 59 63 70 72 74 Reverse Solar Collector Model Radiant Exchange Model Convective Heat Transfer Coefficient 4.5 4.1 4.2 5.2 4.1 Results 5.3 Summary and Conclusions 88 88 89 90 Summary Conclusions Recommendations References Appendix A ASHRAE Draft Standard 138P Appendix B Radiant Panel Test Booth Calculations Appendix C Experimental Test Data 92 95 121 137 viii .6 4.2.1 4.1.5 4.1.1 6.3.3.4 4.8 Numerical Model Governing Numerical Equations Model Parameters Numerical Simulation Method Data Analysis Grid Refinement Model validation Numerical results 4.3.3.1 5.6 Back Losses Bond Conductance Solution Method Chapter 5 5.1.2.4 4.3 5.2 6.3 4.2 5.2.Chapter 4 4.3.3.2 4.3 Results and Discussion Mean Radiant Panel Temperature Radiant Panel Heat Output Predicted Air Temperature Results Thickness of the Heat Paste Convective heat transfer coefficient Air Temperature 75 75 76 79 81 82 82 84 85 Uncertainties and simplifications in the RSC Model Chapter 6 6.

Appendix D Uncertainty Analysis for Experimental Data and RSC Model 153 Appendix E Solution to the Radiosity Equations 164 Appendix F Grid Refinement. Model Validation and Numerical Results 178 Appendix G Methodology to Determine the Convective Heat Transfer Coefficient 191 Appendix H Effects of Panel Insulation on Panel Performance 201 Appendix I Reverse Solar Collector Model Panel Temperature and Heat Output Results 203 ix .

............List of Tables Table 3...7: Experimental Results..............................................1: RSC Model Results and Experimental Results for the 24-4 Panel....3: Pipe Spacing and Pipe Runs .......................... 84 x ....................................................................... 34 Table 3..1: View Factors of Radiant Panel Test Chamber Surfaces................2: RSC Model Results and Experimental Results for the 24-8 Panel............................. 26 Table 3......................... 24-8............6: Experimental results..................................... 29 Table 3............. 24-4 .......................... 37 Table 3.............................................. 37 Table 4............ 25 Table 3.........................4: Pressure Drops for each circuit............................................................ 27 Table 3............. 58 Table 5.................. 75 Table 5....1: Grid Refinement Results.....2: Incoming Heat Flux for each surface.........................5: List of Data Recorded from Experiment ............................................3: Uncertainty associated with bond conductance................... 76 Table 5....................................

5: Comparative plot of the Convective Heat Flux Correlations developed by Min et al.7: Radiant Panel Control Circuit Figure 3.3: 36-6. Radiant Panel and Cold Wall Figure 3.2: Energy Balance on Fin Element 18 20 21 21 22 24 30 32 33 33 36 38 38 40 40 xi .3: Convective Heat Transfer from a Radiant Ceiling Panel (Min et al.1: Commercial Energy use break down for the U.1: Cross section of a Radiant Panel Figure 2..2: Test Booth Construction – Indoor View of Walls Figure 3.1: Radiant Panel Sheet and Tube Dimensions Figure 4.List of Figures Figure 1.2: Cross-section of a Radiant Panel with Heat Conducting Rails 1 3 4 8 11 Figure 2.12: 24-4 and 24-8 Heat Output versus Panel Inlet Water Temperature Figure 4.8: Chamber Interior.4: Mounting PEX Tubing to Exterior Surface of Chamber Wall Figure 3. 1956) 13 Figure 2.11: Mean Panel Temperature versus Panel Inlet Water Temperature Figure 3.5: Chamber Interior. Back-side shown Figure 2. Back. Perimeter Radiant Heating Panel. (1956).3: Test Booth Construction – Outdoor View of Walls Figure 3.S.9: Overall View of Completed Radiant Panel Test Facility Figure 3. 1999) 14 Figure 2..4: Experimental Results for the Convective Heat Transfer Coefficient for a Radiant Ceiling Panel (Awbi et al.1: Cut Away Sketch of Radiant Panel Test Facility Figure 3.10: Surface and Water Temperature versus time Figure 3. 2005) Figure 1. Schutrum and Vouris (1954) and Awbi and Hatton (1999) 15 Figure 2.6: ASHRAE Proposed Radiant Panel Test Chamber Figure 3. Left and Front Wall with Door Figure 3.6: Radiant Panel Test Chamber Enclosure Temperature Control System Figure 3.2: Serpentine and Parallel Radiant Panel Circuits Figure 1. (DOE.

19: Effect of Heat Paste Thickness on Panel Performance 64 65 66 67 68 69 71 73 Figure 5.13: Nusselt Number versus Aspect Ratio Figure 4.2: Predicted Panel Temperature versus Actual Panel Temperature for a 24-4 77 Panel Figure 5.12: Temperature profile for a 0.4: Cut Away Sketch of a Seven Surface Radiant Panel Test Chamber Figure 4.11: 2D and 3D numerical model simulation results versus ASHVE experimental data 62 Figure 4.15: Nusselt versus Rayleigh number Figure 4.25m and a 1.4: Predicted Panel Temperature versus Actual Panel Temperature for a 24-8 78 Panel xii .9: Temperature Profile for the Numerically Simulated ASHVE Test Facility 61 Figure 4.5m panel at 318K Figure 4.14: Rayleigh number versus Aspect Ratio Figure 4.5: Three Surface Enclosure Approximation 43 47 49 Figure 4.3: Experimental and Analytical Model Panel Temperature versus the Mean 78 Water Temperature for a 24-8 Panel Figure 5.Figure 4.18: Effect of Panel Insulation on the Overall Heat Transfer Coefficient Figure 4.3: Radiant Panel Resistance Network Figure 4.10: Side profile and Top View of ASHVE Test Chamber and y Direction Temperature Gradient Profile 62 Figure 4.8: Nusselt versus Rayleigh number for numerical model results and experimental results from the ASHVE laboratory 60 Figure 4.7: Schematic of CFD Model 53 Figure 4.17: Comparison of various Heat Flux Correlations Figure 4.16: Convective Heat flux versus panel temperature and panel size Figure 4.6: Predicted Radiative Heat Flux versus Panel Temperature for Three Analytical Models 51 Figure 4.1: Experimental and Analytical Model Panel Temperature versus the Inlet Water Temperature for a 24-4 Panel 76 Figure 5.

6: Experimental and Analytical Heat Output versus Inlet Water Temperature for a 24-8 Panel 80 Figure 5.Figure 5.8: Convective Heat Transfer versus Panel minus Air Temperature at different back wall temperatures 84 Figure 5.9: Air Temperature versus Panel Temperature for the 24-4 Panel 86 Figure 5.5: Experimental and Analytical Heat Output versus Inlet Water Temperature for a 24-4 Panel 79 Figure 5.10: Predicted Air Velocity Profile with 12oC and 10oC back wall Temperature 87 xiii .61m Panel and the Predicted Air Temperature from the Numerical Model at the Center of the Radiant Panel Test Chamber versus The Panel Temperature 81 Figure 5.7: Experimental Air Temperature Measurements for a 0.

dimensionless Pressure.4. l D DH Ebi f F F F’ FR g G Gr h hL J k k-l KL l L & m n n Nu p Pr q q’ q” R R Ra Re S t B B B B B B B B B B B B B B B B B B Surface Area. m Panel length. dimensionless Heat removal factor. W/m2 Grashof number. K Radiant panel aspect ratio (Lpanel/Lceiling). m of H2O Radiosity. m Sutherland equation constant (1. m Mass flow rate. m P P B B B B P P P P P P P P P P B B P P P P xiv . dimensionless Convective heat transfer coefficient. dimensionless Heat transfer.458x10-6 kg/(msK1/2)) Bond conductance. W/m2 friction factor. dimensionless View factor Fin efficiency factor Fin efficiency factor. K) Thickness. dimensionless Gravitational constant (9. W/m2 Thermal conductivity.81 m/s2) Irradiation. W/m Heat flux. m2 Area-weighted surface temperature. Pa Prandtl number.Nomenclature Symbol A AUST AR b b Cb Cp Cµ. dimensionless Pipe length. dimensionless Molecular viscosity constant for the Sutherland Equation (110. W/mK Turbulence model coefficient Loss coefficient. fµ. W Heat Tranfer per unit length. dimensionless Reynolds number. W/m2K Head loss. dimensionless Bond width. kg/s Number of surfaces Number of passes Nusselt number. W/mK Specific heat. W/m2 Radial distance. m Hydraulic Diameter (4xPerimeter/Area) Emissive Power from surface i. J/kgK Coefficients for the k-l turbulence model Diameter. m Gas constant (287 J/kgK for air) Rayleigh number.

radians Reflectivity.in i i-j ins j o out p panel Air Average Bond Back Backloss Convective Cell Conductive Convective Fluid average Fluid inlet Fluid Outlet Section of the radiant panel Incoming flux Inner. density. surface From surface i to surface j Insulation Surface Outer Out Pipe Panel xv . Ns/m2 Turbulence viscosity. m2/s Volumetric thermal expansion coefficient.T u UL v V W W B B Temperature. m P P Greek Letters α β δ ∆ ε µ µt ν θ ρ σ B B Thermal diffusivity. m/s Velocity. K-1 Plate thickness. m/s Tube or Pipe Spacing. W/m2K y-direction velocity. K x-direction velocity. m/s Overall heat transfer coefficient.in f. m Radiant panel width. m2/s Polar angle. kg/m3 Stefan-Boltzman constant (5.out fin flux. Ns/m2 Kinematic viscosity. m Increment Emissivity Dynamic viscosity.67x10-8 W/m2K4) P P P P P P P P P P P P P P P P P P Subscripts a avg b back backloss c cell cond conv f f.

panel-j PEX rad sur w wall Panel to surrounding surface. j PEX Radiative Surrounding Water Wall Abbreviations MRT MWT RSC Mean Radiant Temperature Mean Water Temperature Reverse Solar Collector xvi .

in particular currently uses 25% more energy than it did 20 years ago. (DOE.1 Background In recent years.S residential sector.1: Commercial Energy use break down for the U. The U. and an increase in fuel costs.S.Chapter 1 Introduction 1.S commercial sector. 2005). In the U.S. the U. 2005) 1 . When looking at the statistics for energy consumption. 2005). 32% of the energy goes towards space heating and 11% towards space cooling (DOE. For the U. 2005) (Figure 1.1). and it is forecasted that this trend will continue for the next 20 years (DOE. it has become important to find methods of reducing energy consumption and finding alternative energy sources. a significant portion is spent on heating and cooling of buildings. With a decline in natural resources. 25% of the total energy consumption is used for space heating and cooling alone (DOE. Figure 1.S. and Canada have seen a steady increase in energy consumption.

2003).K. Barker discovered that small hot 2 . U. A constant temperature can be maintained through electric resistance elements or by circulating water or air through tubes mounted on the backside of the surface. when they constructed their buildings with hollow floors to allow hot gases from the central fire to travel under the floor (Adlam. surfaces in the conditioned space can be warmed or cooled. Sir John Stone set up a system in London. 1949). In 1908 radiant heating was introduced on a commercial basis (Adlam. The temperature set points impact occupant comfort and are established through convective heating systems. Professor Arthur H. controlling the solar heat gain to a room by using shading devices. A radiant panel is a temperature controlled surface in a conditioned space. while maintaining a lower than normal ambient air temperature. Radiant heating dates back to the Romans in 50 BC. behind or under grilles.Over the last few years energy conservation in space heating and cooling has become a widely researched area. in which 50% or more of the heat transfer is through radiation (ASHRAE. 1949). radiant panels can be used. such as baseboard radiators and ducted HVAC systems. or by reducing the heating and cooling load in a building by adjusting temperature set points. giving way to a radiant heating system. by placing several large pipes. To adjust the temperature set points without sacrificing occupant comfort. Radiant panel systems have been in existence for centuries. carrying hot water from a boiler. The heating source did not change until 1790 when hot water was introduced as the working fluid. The tiles above the floor would warm up. Energy can be saved by improving the insulating values in walls and windows. The benefits of such systems are that through the radiant heat transfer.

. believing cooling and heating loads could not be met. Figure 1.2a. In Germany. radiant ceiling panels are constructed from copper tubing bonded. The principles behind radiant heating have since then remained the same. In North America however. pressed or clipped onto an aluminum extruded sheet or a heat conducting rail glued onto the back-side of the panel (ASHRAE. Depending on the design constraints and desired energy output.3. Tin Tin Tout a) Serpentine Circuit b) Parallel Circuit Tout Figure 1. 1994). Imperial units are always used in North America to specify radiant panel sizing. or as a parallel circuit. 1989). Today. 2004). Thus far. 6 pass radiant panel. more than 50% of newly constructed buildings and residences incorporate radiant panel systems (Kilkis et al. believing the North American climate was not suitable for such systems (Buckley. Europe has been the leader in utilizing these systems for a wide variety of commercial and residential applications (Simmonds.2b. A 36-6 serpentine radiant panel manufactured by Sigma Convectors is shown in Figure 1. so for a typical 36 inch. the copper tubing can be run in a serpentine circuit. designers have rarely used radiant panel systems.2: Serpentine and Parallel Radiant Panel Circuits The panels are specified based on their width and number of passes. 3 . 1996). a 36-6 panel would be specified. Figure 1.water pipes embedded in plaster or concrete made an efficient heating system.

it is important that the heat output of a radiant panel be estimated accurately within the context of the room and building. When designing radiant heating or cooling systems. 4 . Properly designed systems can produce long term energy savings of up to 30% (Conroy.Figure 1. In office buildings and schools. In order to achieve the desired comfort level. thermally stable environment. temperature and location of the panel are important factors affecting the performance of the system and the thermal comfort of the occupants. Other applications include swimming pools. residential buildings and general space heating or cooling in industrial buildings (Shoemaker. leading to a recent emergence of radiant heating/cooling applications in North America.3: 36-6. 2001). however. can occur if the panels are not properly sized. 1989). copper tubing ceiling panels are used as perimeter heaters to balance out the heating load and minimize mechanical equipment noise. If the correct panel size is not chosen. Thermal discomfort. perimeter radiant panel heating and cooling systems are ideal. providing a draft free. energy savings may not be realized and comfort levels not satisfied. In hospitals. 1954). the size. Typical residential radiant panel systems are grid coils in the floor or copper tubing systems in the ceiling. Back-side shown Recent studies have shown that human comfort can be obtained by radiant transfer in any type of climate (Buckley. Perimeter Radiant Heating Panel.

Sigma Convectors currently manufacturers various air and hydronic heating and cooling systems. manufacturers of these systems simply approximate the required panel sizing for panel heating/cooling 5 .2 Industrial Motivation The number of manufacturers producing radiant panels is growing every year. 2005). Their ultimate objective with this test booth was to be able to showcase their products to potential customers. they contacted the University of Waterloo to help develop a radiant panel test booth. 1996). leading to various products promising performance capabilities unattainable in actual scenarios (Kochendörfer. In the early 1990’s. Sigma Convectors. leading to the need for North America to develop a standardized test and rating method as well. 2004). many companies strive towards selling low-cost and high performance panels. Currently.1. One manufacturing company.3 Analytical Motivation Various theoretical/experimental results show that radiant systems have the potential to achieve energy savings up to 30% (Laouadi. Similar to Europe. With an increase in market demand for radiant panels. Currently. manufacturers have rated their products with insufficient descriptions of testing conditions. the use of radiant panels in North America is increasing. 1. With the increasing need to verify heating and cooling outputs from heating and cooling panels. With no standardized test in place. Europe saw a 20% increase of radiant panel manufacturers in one year. has predicted an increase of 300% in sales over the next year in radiant heating panels alone (Kramer. and to show them in progress testing of their radiant panels. 1994). a European rating standard was developed (DIN 4715.

This is because the sponsoring company manufactures and distributes radiant heating panels. Often times. To validate the analytical model with experimental results obtained from the radiant panel test chamber. To construct a radiant panel testing facility based on existing and potential standards. the focus of this thesis was on radiant heating. To numerically model the test chamber to predict the natural convective heat transfer from a ceiling mounted radiant panel over various panel temperatures and sizes. 3. The specific objectives of this thesis are: 1. 4. the panels specified are not optimized for the required application. 2. water flow rate and inlet water temperature. 6 .applications from experimental data.4 Objectives Although space heating and space cooling energy savings are equally important to the North American climate. It is therefore ideal to develop an accurate prediction model capable of helping building and system designers to specify and size radiant systems. and limited research has been done in heating applications. tube spacing. 1. To develop an analytical model for predicting the mean radiant ceiling panel temperature and heat output based on size.

1.5 Thesis Outline Chapter 2 presents a literature review on the radiant ceiling panel topic. Chapter 5 presents the validation of the analytical model. Chapter 3 discusses the construction of the radiant panel test facility and the results obtained. Theoretical models. 7 . Discrepancies between the analytical model and experimental results will be discussed. Chapter 4 will present a CFD model used to predict the natural convective heat transfer inside the test chamber and show derivations to calculate the radiative and conductive heat transfer coefficients. radiant panel test chambers and potential or existing standards will be discussed. In particular any models used to predict the mean radiant panel temperature will be examined as well as any correlations developed to predict the convective heat transfer from a heated ceiling panel. conclusions and recommendations resulting from this research. Chapter 6 presents the summary. In addition to the analytical model. Chapter 4 presents the development of the analytical model used to predict the panel temperature and heat output. The obtained values will be compared to current accepted models.

Following this. a fin efficiency was calculated. Fin width qlosses Ceiling Room qconv qrad Figure 2. 2.1: Cross section of a Radiant Panel 8 .1 Existing Mean Radiant Panel Temperature Models The earliest model used to predict the radiant panel temperature was developed in 1943 (Heid and Kollmar. convective and conductive heat transfer coefficients. Finally. particular emphasis will be placed on research done to determine the radiative and convective heat transfer coefficients for the system. This enabled the authors to predict the mean radiant panel temperature based on the mean water temperature (MWT). This Chapter will begin with a literature review of existing radiant panel prediction models. With the knowledge of the radiative. using the assumption that the radiant panel operated like a plate fin with heat losses from the upper surface (Figure 2. a review of any existing radiant panel test chambers will be discussed as well as any standards currently used to rate radiant panels.1). 1943).Chapter 2 Literature Review As a significant portion of this thesis is dedicated to the development of an analytical model used to predict the panel heat output and mean radiant panel temperature.

Recognizing that radiant cooling panels were constructed very 9 . that it was difficult to evaluate their numerical model because no exact analytical solution existed. could be calculated as follows: B B q panel = π D 1 1 + ln( o ) hw Di 2k p Di (To − T f ) 1000 W (2.Approximately 40 years later. 1986). This enabled them to investigate the effects of tube spacing and the convective heat transfer coefficient on the panel performance. (2001) presented a novel approach to predicting the mean radiant panel temperature. They found. Using a 2D numerical model. qpanel. the authors showed that the heat output from the panel. (1994) developed an analytical model to estimate the heating capacity based on pipe spacing and mean water temperature. as computational power improved significantly.. Conroy et al. Using the practical fin ceiling model from Heid and Kollmar (1943). however. the authors modelled the transient and steady state temperature distribution across the panel. Their numerical results were in good agreement with some of their experimental results.1) where: hw is the convective heat transfer coefficient of the water kp is the conductivity of the pipe To is the outer surface temperature of the pipe Tf is the average water temperature W is the pipe spacing Do and Di are the outer and inner pipe diameters respectively B B B B B B B B B B B B The authors then used the fin efficiency model from Heid and Kollmar (1943) to predict the panel temperature based on heating load. the model was never validated with experimental results. Although they presented a case study to show the effectiveness of radiant floor heating. a finite difference algorithm was developed to study the heat transfer characteristics of a radiant ceiling panel (Zhang and Pate. Kilkis et al.

in ) A panel FRU L (1 − FR ) (2. however. In comparison to the classical fin-tube model developed by Heid and Kollmar (1943). Hadlock (2004) studied the possibility that the flat plate solar collector model could be used for radiant heating applications. the authors showed that using the solar collector model of Hottel. however.2) & where: m is the mass flow rate of the panel Cp is the specific heat of water Tf. Hadlock showed that the flat plate collector model was within 13% of the predicted heat output.mean = T f . with the previously presented solutions or any experimental data.2) (Xia et al. leading to the desire to develop a test chamber to determine actual radiant panel heat outputs under known conditions.. 2005). Whillier and Bliss (1958) the mean panel temperature could be calculated as follows: T p .2.out − T f . As the analytical models were compared to heating performance curves provided by Sigma Convectors (Sigma. In 2005 the solar absorber model was further improved by modeling radiant cooling panels with heat conducting rails (HCR) (Figure 2. the results were only within 40% of actual values.similarly to flat plate solar collectors. They never verified this model.out are the inlet and outlet fluid temperatures respectively Apanel is the surface area of the panel Fr is the heat removal factor UL is the overall heat transfer coefficient B B B B B B B B B B B B The authors went on to use this model to show how effective radiant cooling ceiling panels could be in the North American climate.in + & mC p (T f . 2004). Similar to the previously developed models. an equation was developed that was used to predict the radiant panel 10 . It accounts. The analysis is very similar to the existing solar collector model.in and Tf. for the heat transfer through the heat conducting rail in Equation 2.

2: Cross-section of a Radiant Panel with Heat Conducting Rails Each of the preceding models require an overall heat transfer coefficient that includes radiative and convective heat transfer.2 Existing Radiant Interchange Models To determine the net radiative heat transfer in an enclosure. " 4 q rad = ∑ σF panel − j (T panel − T j4 ) j =1 n (2. Sections 2. Again.temperature based on a mean water temperature (MWT).3 will review existing methods and correlations used to calculate an overall heat transfer coefficient. this method was not validated. minimizing the need for complex view factor calculations. 2.2 and 2. 11 . the fundamental radiative heat transfer equation can be used.3) where: σ is the Stefan – Boltzman constant n is the number of surfaces Fpanel-j is view factor between panel surface and surrounding surface j Tpanel is the radiant panel temperature Tj is the surrounding surface temperature B B B B B B Several models have been developed to simplify the net radiative heat transfer calculation. Panel Tubing HCR Clip Figure 2.

Using the MRT equation developed by Walton (1980). To improve the accuracy of the MRT method. but is more accurate than the previous uncorrected MRT method. (1989). which predicted the radiant interchange between surfaces in a room. 12 .Walton (1980) presented an algorithm. that this method is only applicable for cases where the temperature difference between the radiant panel and surrounding enclosure surfaces is not great. known as the mean radiant temperature (MRT) method.4) This method required more computational time because of view factor calculations. the panel heat flux was shown to be: ⎡⎛ n ⎢⎜ ∑ A j T j ⎜ j ≠i = σ ⎢⎜ n ⎢ ⎢⎜ ∑ A j ⎜ ⎢⎝ j ≠i ⎣ 4 ⎤ ⎞ ⎟ ⎥ n 4⎥ ⎟ ⎟ − ∑ F panel − j T j ⎥ j ≠i ⎟ ⎥ ⎟ ⎥ ⎠ ⎦ " q rad (2.3 Existing Convective Heat Transfer Models Limited research has been done to determine the convective heat transfer coefficient from radiant ceiling panels because it is typically deemed to be negligible. temperature gradients below the panel were measured for three different sized rooms. and adding a correction component accounting for exact view factors. This method assumes a fictitious surface represented the surrounding enclosure with a similar temperature and emissivity as the enclosure it replaced. The earliest study on convective heat transfer was done in 1956 in the American Society of Heating and Ventilating Engineers (ASHVE) environment laboratory by Min et al. 2. however. an MRT correction method was presented by Steinman et al. To determine the convective heat transfer from a radiant ceiling panel. This method simplifies a system greatly because only two surfaces need to be analysed. (1956). It has been shown.

. 1956) A convective heat transfer correlation was developed. 1.45 11.85 1.9 1.2 11.65 1.3: Convective Heat Transfer from a Radiant Ceiling Panel (Min et al. The results obtained from the experiment are plotted in Figure 2.6 Figure 2.6 11. P P P P The walls and floor were maintained at a constant temperature and the entire ceiling was used as a radiant panel.3 was found to equal 0.5x12x12 Chamber Size Correlation 1.25 11.8 Log(Nu) 1.04.3.24 (2.5) where: Ta is the air temperature (1.25 / DH 0.35 11.5 could be simplified to: 13 .5x12x8 Chamber Size 24.7 1.3 11.55 11.75 24. Schutrum and Vouris (1954) concluded that the effects of room size on panel performance of heated ceiling panels was relatively small and Equation 2.5m above the floor) DH is hydraulic diameter of the radiant panel (4xPerimeter/Area) B B B B The R2-value (correlation coefficient) from Figure 2.5 11.2(T panel − Ta )1.4 Log(Ra) 11. reflecting P P the wide range of scatter in the plot and significant error associated with the correlation. " q c = 0.without any air infiltration and ceiling panel temperatures ranging from 32oC to 65oC.

however..4: Experimental Results for the Convective Heat Transfer Coefficient for a Radiant Ceiling Panel (Awbi et al. however. 1999) 14 .8 0.138(T panel − Ta )1. 2 1. They gave no mention.25 (2. of the exact radiant panel setup.2 0 0 5 10 15 20 25 30 35 40 Surface to Air Temperature (K) Figure 2. Awbi and Hatton (1999) questioned the accuracy of the measurements of the previous experiments and constructed their own test chamber. It was also known that one wall was maintained at a much colder temperature than the other surfaces.8 1." q c = 0.4 0.6 1.4) and a heat transfer coefficient was correlated shown in Equation 2.6 0.6) The authors conclusions. The test results for the various heated panel sizes showed great variation (Figure 2.4 Chamber (Large Plates) Chamber (Small Plates 2) Chamber (Small Plates 1) Chamber (Small Box) Chamber (Small Plates 3) Awbi and Hatton (1999) 1.7.2 hc (W/m2K) 1 0. were based only on completely heated ceilings. They used the most accurate temperature measurements techniques available to determine the natural heat transfer coefficient and their tests covered a wide range of heated ceiling conditions and chamber sizes. The air temperature in the chamber was maintained at 20oC by adjusting the wall and floor P P temperatures.

(1956).5: Comparative plot of the Convective Heat Flux Correlations developed by Min et al. the total heat transfer can be off by as much as 60% if the incorrect correlation is used.133 DH 0.601 (2. contrary to the conclusions from Schutrum and Vouris (1954).025m. (1956). 90 80 70 Convective Heat Output (W/m ) 2 60 Min et al. Comparing the correlated convective heat transfer correlations from Min et al. Although Figure 2. it can be concluded that the room size and panel size affect the natural convection coefficient for perimeter radiant ceiling panels.7) The correlation coefficient for Equation 2.87.4 shows a wide range of scatter. It is therefore necessary to accurately determine the convective heat transfer from a radiant ceiling panel. Schutrum and Vouris (1954) and Awbi and Hatton (1999) in Figure 2.704(T panel − Ta )1. Schutrum and Vouris (1954) and Awbi and Hatton (1999) 15 .7 was shown to be equal to 0. (1956) Schutrum and Vouris (1954) Awbi and Hatton (1999) 50 40 30 20 10 0 315 320 325 330 335 340 345 350 355 360 365 370 Tpanel (K) Figure 2." qc = 0.5 for a panel hydraulic diameter of 1.

For a heated ceiling surface the convective heat transfer is shown to be: " q c = 0. ceiling and floor to establish constant temperature and ethylene glycol was used as the working fluid. 2004) summarizes the standard radiant panel model.9) 2. ventilating and Air-Conditioning Engineers (ASHRAE) HVAC Systems and Equipment Handbook (ASHRAE. floor and 16 .15) 4 − ( AUST + 273.3m long by 3. Serpentine and parallel copper tubing circuits were mounted on the exterior of the walls. Copper tubing was arranged such that each surface could be individually controlled to simulate a hot or cold wall.4 Currently Accepted Radiant Panel Model The 2004 American Society of Heating.24 (2. For a two surface radiation heat exchange.5 Existing Test Facilities for Radiant Heating Tasker. in the ASHVE test facility (1956). Parmelee and Schutrum (1952). the chamber was capable of testing the panels under a wide variety of surrounding temperature conditions and chamber infiltration rates.2.25 DH 0. al.8) where: AUST is the area-weighted temperature of the non heated indoor surfaces The natural convective heat transfer was calculated based on the studies from Min et.15) 4 ] (2.66m wide by 3.66m high. Measuring 7. Humphreys.20 (T panel − Ta )1. Walton’s MRT method (1980) is considered to be an acceptable method to determine the radiant heat exchange in a simple enclosure. the thermal radiation for a panel is shown to be: " q rad = 5 * 10 −8 [(T panel + 273. describe an ASHVE Environment Laboratory capable of testing radiant ceiling and floor panels in a cooling or heating mode.

so it contained additional instrumentation and capabilities that are not required for just a radiant panel test facility.6 Radiant Panel Standards The German Standardization Institute (DIN) developed a radiant cooling panel standard (DIN 4715. An ASHRAE standard for testing hydronic radiant ceiling panels has been developed. The standard states that a cooling panel is to be tested in a 4m long by 4m wide by 3m high room. with temperature controllable walls (Kochendörfer. No additional ventilation is to be added to the room and the cooling panel must cover at least 70% of the ceiling. 2005). 1994). The temperatures of the walls. ceiling and floor were measured using approximately 400 thermocouples and heat flow meters were used to measure the heat exchange between the surfaces of each wall. The radiant panel cooling load is determined by measuring the inlet and outlet fluid temperature difference. which has become widely accepted in Europe. The radiant 17 . In order to give a detailed account on variables which affect the performance of radiant panels. To simulate actual conditions. 2. 12 electric heaters are used to simulate people who require cooling. The heat flow meters accurately measured the temperature difference between opposite sides of a thermal resistance layer and were set up to measure low rates of heat flow (Huebscher et al. air temperature gradients and air motion were also measured. air temperature.ceiling. It should be noted that this facility was constructed to ultimately become a psychometric laboratory. 1996). In addition. Any combination of wall temperatures was attainable. an air guard barrier was constructed around the walls to ensure the surface temperatures would remain constant.. and is currently in a public review process (Chapman. 1952).

The inlet and outlet water temperature difference across the heating panel is measured and the output is calculated knowing the water flow rate.6 is a sketch of the proposed radiant panel test facility and location of the radiant panel relative to the chamber surfaces. Figure 2. Each surface of the enclosure is to be divided into quadrants and the temperature measured at the geometric center of each quadrant. Cold Wall Left Wall Floor Figure 2. The details of the proposed standard can be found in Appendix A.panel is to be tested in a 3.66m wide by 3.66m long by 3m high chamber.6: ASHRAE Proposed Radiant Panel Test Chamber 18 . The radiant panel is to be installed above the cold wall to offset the heat los caused by a cold window. in which three walls and the floor are maintained at a constant temperature of 20oC and the back (cold) wall at 10oC.

The radiant panel heat output (qout) can be calculated theoretically by adding the B B conductive (qcond).1 Purpose of Test Facility Radiant heating panels are directly influenced by particular environmental conditions. The test chamber was built to test a line of Sigma Convector radiant panels and to demonstrate the effectiveness of radiant perimeter heating to potential customers. Details of the test method are also summarized. The chamber can simulate a wide range of test conditions. As stated in Chapter 2. and might lead to a significant error. convective (qconv) and radiative (qrad) heat transfer. 3. radiant panels are dominated by two modes of heat transfer – convection and radiation. B B B B B B B q out = q cond + q conv + q rad (3. Therefore. can B B be determined using: 19 . Alternatively the total heat output (qout).1) Measuring each mode of heat transfer can become difficult and tedious. including losses. while giving an accurate estimate of the panel heat output. surface emissivities and chamber size. while the radiative heat transfer is dependant on surrounding surface temperatures.Chapter 3 Radiant Panel Test Chamber This chapter describes the construction of an experimental facility that can be used to rate and test the radiant panels. The convective heat transfer is dependant on the surrounding air temperature. air motion and enclosure size. the primary purpose of the test facility is to create a testing environment with known surrounding conditions.

& q out = mC p (T f ,in − T f ,out )

(3.2)

By accurately measuring the flow rate and temperature change across the panel an accurate heat output can be determined for a known air temperature and specific heat, Cp.
B B

3.2

Radiant Panel Test Chamber Construction

It was decided to construct a radiant panel test facility according to the ASHRAE Draft Standard 138P (Appendix A). The test chamber, however, was sized according to the DIN cooling standard test chamber (Kochendörfer, 1996). specifications of the test chamber (Figure 3.1) are listed below: • • • • • The chamber is to be 4m by 4m by 3m high The back wall (exterior wall) is to be maintained at 10oC +/- 0.5oC The other surrounding walls and floor are to be maintained at 20oC +/-0.5oC The radiant panel should be mounted within 2.54 cm from the back wall The emissivity of the walls should be 0.9
P P P P P P P P

The details and

Radiant Panel

Cold Wall 3m 20oC
P P

10oC
P P

20oC
P P

4m

4m

Figure 3.1: Cut Away Sketch of Radiant Panel Test Facility The construction of the test booth is shown in Figures 3.2 and 3.3.

20

Figure 3.2: Test Booth Construction – Indoor View of Walls

Figure 3.3: Test Booth Construction – Outdoor View of Walls

21

3.2.1

Temperature Controllable Walls

To maintain a constant surrounding temperature the four walls and floor must be temperature controllable and made from highly conductive material. It was decided to mount cross-linked polyethylene (PEX) tubing on the exterior of each wall and the floor, while controlling the temperature by circulating water from a constant temperature source. The wall material was made out of aluminum because of its excellent

conductivity and reasonable cost. The PEX tubing was clipped onto heat conducting rails (HCR), which were glued onto the exterior of the aluminum panels (Figure 3.4). Also shown is an insert of the cross section of the tubing mounted on the wall

Wall

HCR

Figure 3.4: Mounting PEX Tubing to Exterior Surface of Chamber Wall

22

To optimize the heat transfer and temperature controllability of the walls and floor, several improvements were suggested. For example, clipping copper tubes onto the back of each surface and building an enclosure around the chamber to maintain a constant surrounding air temperature would have been useful. With a limited budget, however, these suggestions were deemed unfeasible and PEX tubing was used instead of copper and the exterior walls were heavily insulated, minimizing heat loss or gain from the back of the wall. The walls and floor were painted with a high gloss white paint to reduce reflectivity for long wave radiation. Infiltration into the room was prevented by taping over the seams of wall and floor using Turf® tape (Figure 3.5).
P P

To determine the necessary pipe spacing, an energy balance at each wall had to be performed. Neglecting the temperature difference between the exterior and interior of the wall and assuming the back losses to be negligible, the energy balance gives: q w = q rad + q c where: qw is the heat removed by circulating water qc is the convective heat transfer
B B B B

(3.3)

To determine the amount of tubing required on the wall, the following assumptions were made: • • • • • • • Wall emissivity was equal to 0.9 because it would be painted white Interior convective heat transfer coefficient was equal to 6 W/m2K (Awbi, 1999) Interior air temperature was equal to 22oC PEX tube thickness of 5mm with a conductivity of 20 W/mK Mean back wall water temperature was 9oC Mean surrounding wall and floor water temperatures were 18oC Negligible convective and conductive back losses from the exterior surfaces
P P P P P P P P

23

5: Chamber Interior. Left and Front Wall with Door 24 .Figure 3. Back.

as documented in Appendix B.Floor Panel – Front Wall Panel – Left Wall Panel – Right Wall View Factor 0.1: View Factors of Radiant Panel Test Chamber Surfaces Surfaces (i – j) Panel – Back Wall Panel . were calculated from first principles given by: B B Fij = 1 Ai Ai A j ∫∫ cos θ i cos θ j πR 2 dAi dA j (3. 1981) and all view factors were checked to add to unity.4) where: Tsur is the surrounding surface temperature ε is the emissivity of the surface B B The tube spacing was designed for a 1. Fij. and the results are summarized in Table 3. B B B B q c = h(Ta − Tsur ) (3.257 0.2 m wide panel at a temperature of 100oC. Table 3. To verify that the equation was correct. the view factor from the panel to the back wall was compared to a published view factor equation (Siegel and Howell.If the heat removed by the water is assumed to be equal to the net incoming radiative and convective heat transfer.067 0. and the temperature of the surface (Tsur) and air (Ta). h.6) 25 . The P P view factors.169 0. then the radiant heat flux can be estimated as: " 4 4 q rad = Fεσ (T panel − Tsur ) (3.169 The convective heat transfer can be calculated knowing the heat transfer coefficient.1 below.337 0.5) where: θi and θj are the polar angles between the surface normal R is the radial distance from the point of interest to the surface of interest B B B B The view factors were calculated using Mathcad 11.

The difficulty was to determine an accurate convective heat transfer coefficient.4 and 3.7) where: dT is the temperature difference between the water and wall dx is the distance between the surface of the wall to the inner portion of the pipe Ap is the pipe contact surface area B B Approximately 1/3 of the PEX tubing surface area is in contact with the heat conducting rail mounted to the wall.7).2: Incoming Heat Flux for each surface Surface Back Wall Floor Front Wall Left/Right Wall Temperature (K) 283 293 293 293 Incoming Heat Flux (W/m2) 260 143 46 98 P P Using Fourier’s law (Equation 3. Assuming the worst case. We can therefore approximate the surface area of the pipe to equal: Ap = πDo l 3 (3. the contact surface area of tubing required to balance out the incoming energy could be calculated q out = k p A p dT dx (3. The incoming heat flux for each wall and floor is summarized in Table 3. P P With knowledge of the radiant and convective heat transfer. the amount of energy to be removed by the water can be calculated using Equations 3. floors and windows can vary between 1 and 6 W/m2K.2 below and shown in Appendix B.8) where: l is the total length in contact 26 . Awbi and Hatton (1999) reported that the convective heat transfer coefficients for walls.6. Table 3. a heat transfer P P coefficient of 6 W/m2K was used to calculate the energy transferred through convection.

Table 3.3: Pipe Spacing and Pipe Runs Wall/Floor Back Wall Floor Front Wall Left Wall Right Wall Pipe Spacing (m) 0. W= 16k PEX πDo (Twall − Tw ) 0.10) For a 0.75 3. The circulating water flow rate was calculated using: & q = mC p (T f . To ensure the wall temperature remained steady. To select a pump.in ) P P (3.44 Number of Pipe Runs 15 6 2 3 3 It was ultimately decided to use the same number of pipe runs for each surface because it was desired to be able to mount the radiant ceiling panel anywhere in the chamber and have the capability of maintaining a cold wall temperature on any surface of the enclosure.44 1. 27 . the tube spacing. In total. it was important to estimate the pressure loss across the serpentine piping circuit mounted on the walls. 5 circuits were used for 5 walls. thus minimizing the water temperature change. W.07 1.9) where: qfluxin is the incoming radiative and convective heat flux B B The tube spacing and pipe runs for each surface were calculated (Appendix B) and summarized in Table 3. a flow rate of 0.5oC water temperature increase.015q fluxin A. was determined by equating the heat flux to Equation 3.3.7 to obtain.125 L/s was calculated for the cold wall (Appendix B).out − T f . the water was circulated through the pipes at a high flow rate.27 0.Having calculated the heat flux on each wall (Table 3.wall (3.2).

81m/s2 P P The friction factor. Seventy meters of PEX tubing was mounted to each wall and ninety meters was installed for the floor.4.2 for each U-bend (Munson.12) (3. 2002): hL = f l V2 Di 2 g (3. a total of 22 pipe runs were installed per wall.13) where: KL is the loss coefficient B B Using a KL value of 0.027 was determined. a smooth surface roughness condition could be assumed. 28 .each circuit having a separate pump.. 2002). Calculating a Reynolds number of approximately 15000 (Appendix B) a friction factor value of 0. f. was determined using a Moody Chart (Appendix B) with the knowledge of the Reynolds number given by: Re = where: ρ is the fluid density µ is the dynamic viscosity Using PEX tubing.11) where: f is the friction factor V is the velocity g is the gravitational constant = 9. The pressure drop across each circuit was approximated using the head loss equation (Munson et al. To account for any additional heat transfer.. 2002): V2 hL = K L 2g ρVD µ (3. To connect the pipe runs a total of 21 U-bends were used. the pressure drop for each B B circuit was calculated (Appendix B) and summarized in Table 3. The head loss caused by each U-bend was calculated using the following equation (Munson et al.

because it would require a cold water source no matter what the surrounding test conditions were. The wall temperatures were controlled using a proportional-integral (PI) control loop. The chamber temperature conditions were maintained using two water storage tanks. 29 .Table 3. winter requires heating). thereby adjusting the amount of hot or cold water entering the system. which compared the measured wall temperature to the set point temperature.6.4: Pressure Drops for each circuit Wall/Floor Back Wall Floor Front Wall Left/Right Wall Pressure Drop (kPa) 138 193 138 138 The wall temperatures were maintained at a constant temperature using a motorized three way control valve. thus ensuring a steady wall temperature. The back wall had a separate storage tank. Using a data acquisition and control system. If the wall temperature was above the set point. The other surfaces would require heating or cooling depending on the season (summer requires cooling. The motorized three way ball valve opens and closes to control the amount of cold water being mixed with the return water. an average wall temperature was measured using 4 thermocouples located at the center of each surface quadrant. A diagram of the system is shown in Figure 3. the analog output module from the data logging system would send out a voltage signal between 2 to 10V to energize the ball valve.

Figure 3.6: Radiant Panel Test Chamber Enclosure Temperature Control System 30 .

3 kPa (Appendix B) for the pipe circuit up to the radiant panel and the maximum radiant panel pressure drop was known to be 41. The temperature of the radiant panel was controlled based on the desired inlet water temperature and using a PI loop controlling the 3-way ball valve. The maximum desired flow rate was 0.2 L/s at a pressure drop of 89. For the main line 1” copper tubing was used to minimize the pipe friction and ½” copper tubing was used for the radiant panel.027 was estimated from the Moody Chart (Appendix B). With the known flow rate. One part was made up of regular ceiling tiles heavily insulated to simulate an adiabatic surface. the pump for the radiant panel was specified based on the desired flow rate knowing the overall pressure drop.4 kPa.3. adjusting the amount of hot water allowed into the system. Similar to the surrounding walls. so a friction factor of 0. The water flow rate was set and measured using a circuit balancing valve. The second portion was the radiant ceiling panel insulated with ½” thick Batt insulation. The radiant panel control loop was designed to behave in much the same way as for the walls. the pressure drop was calculated to be 48. The panel was divided into six equal segments and panel temperature readings were taken at the center of each segment. The inlet and outlet temperature difference was measured using two thermocouples.2 The Ceiling The ceiling design was split into two parts.2 L/s. Copper tubing was used to circulate the water because of the high water temperature (100oC). 31 .2. A system schematic is shown in Figure 3. so the pressure was estimated by approximating pipe lengths. The pumps were specified before construction of the radiant panel circuit.7 kPa.7. so a pump was selected which could supply 0.

7: Radiant Panel Control Circuit 32 .Figure 3.

8). The exterior of the completed test chamber is shown in Figure 3. Tape Radiant Panel Cold Wall Figure 3.9: Overall View of Completed Radiant Panel Test Facility 33 .To minimize infiltration through the ceiling. Radiant Panel and Cold Wall Insulation Data Acquisition System Figure 3. all tiles were taped to prevent air flow into the room (Figure 3.9.8: Chamber Interior.

5. the average wall. two regular T-Type thermocouples were installed at the inlet and outlet ports of the radiant panel. a thermopile was assembled capable of measuring the temperature difference. It was planned that these thermocouples would eventually be replaced with thermistors. Left Wall Temperature (oC) Avg.8m above the center of floor. Front Wall Temperature (oC) Avg. radiant panel and air temperatures were recorded 34 . Table 3. To minimize experimental error. The variables measured during the experiment are listed in Table 3. the air temperature was measured using a shielded regular T-Type thermocouple suspended 1. The radiant panel surface temperature was measured using six T-Type thermocouples equally spaced over the radiant panel. To determine the radiant panel heat output the temperature difference between the inlet and outlet port of the radiant panel had to be measured. Floor Temperature (oC) Avg. floor.3 Procedures and Data Processing The temperature of each wall and floor was measured using four T-Type thermocouples (Omega). Exterior Wall Temperature (oC) Avg. The voltage measured from the thermopile was converted to a temperature reading. As a solution.3.5: List of Data Recorded from Experiment Measured Quantity Tin and Tout Tbackwall Tfloor Tfrontwall Tleftwall Trightwall Tair Description Radiant Panel Inlet and Outlet Temp (oC) Avg. Finally. Each surface was divided into four quadrants and the temperature was measured and recorded at each quadrant center. Right Wall Temperature (oC) Center Room Air Temperature (oC) Using a Superlogics data acquisition system communicating with Dasylab (Dasylab. 2005). It was later discovered that the data acquisition system was not sensitive enough to measure the EMF voltage from the thermopile.

Using a PI controller in DasyLab (Dasylab. floor and radiant panel had to be maintained at their desired temperatures. no differential pressure gauge was installed. a voltage signal through the output modules was sent to the valve based on the average wall or inlet water temperature. the walls. The Superlogics data acquisition system had 32 thermocouple input channels.3. The valves were installed such that the hot or cold return water from the circuits would mix with the source temperature water. a pressure differential should be read across the circuit balancing valve and then converted to a flow rate. the stabilization time took approximately 1 hour (Appendix C). 35 . The radiant panel flow rate was determined by measuring the pressure difference across the circuit balancing valve. 2006) (Appendix B). Running a steady state temperature test as shown in Figure 3. The wall temperatures and radiant panel inlet water temperature were controlled using the motorized three way ball valves.10.1 Test Procedure To determine the radiant panel temperature and heat output under known surrounding conditions. however. the surrounding surfaces and radiant panel were brought to their corresponding set point temperatures and held at constant temperatures for as long as possible. 2005). Ideally. 8 analog input channels and 8 analog output channels.into excel along with the inlet and outlet radiant panel water temperatures. To begin the experiment. Two Omega analog pressure gauges were used to measure the pressure and then converted into a flow rate using the conversion chart supplied with the valve (Armstrong Pumps. 3. due to budget limitations.

A 24 inch.2 Test Samples Two sample radiant panels were tested.00 Test Data Trightwall Tleftwall Tfrontwall Tfloor Tcold Tpanel Twater in Twater out Tair o 30.4 Test Results The test results for the 24-4 and 24-8 panels tested are summarized in Tables 3. Tests were done at 5 different inlet water temperatures ranging from 50oC to 100oC.7 below.00 90.00 0.10: Surface and Water Temperature versus time Once the walls were maintained at a steady state. 36 .3.6 and 3.00 80. 8 pass (24-8) radiant panel were tested.00 70. 3.100. temperature measurements were taken every 4 minutes for at least 30 minutes. and the seams between the panel and ceiling tiles were taped to eliminate any infiltration.00 0 2000 4000 6000 8000 10000 Time (s) 12000 14000 16000 18000 Figure 3.00 50. 3.00 40.00 10.00 Temperature ( C) 60. 4 pass (24-4) and a 24 inch.00 20. The experimental data can be found in Appendix C. Each panel was insulated.

78 61. The panel temperature versus the inlet water temperature for both panels is plotted in Figure 3.99 20.in (oC) 51.77 73.7: Experimental Results. panel temperature and air temperature were recorded for five inlet water temperatures.57 Heat Output (W/m) 207.11 and 3. The other surrounding surface temperatures were maintained between 19 and 21oC. the heat output. The surrounding conditions were maintained at a relative steady temperature.51 700.75 87. although it slightly deviated from the desired set temperature.94 20.66 Tpanel (oC) 48.71 86.27 19.46 Tair (oC) 20.6: Experimental results.98 55.08 For each panel.12.13 19.68 553.89 Tair (oC) 20.94 18.in (oC) 51.45 66.89 298. 24-8 Tf. 37 . The error bars shown in Figure 3.74 613.35 73.30 19.11 and the panel heat output versus mean water temperature is plotted in Figure 3.66 100.56 75. 24-4 Tf.91 86.4 Tpanel (oC) 46.11 99. which can be found in Appendix D.47 374.52 21. The back wall was maintained at 12oC by running city water through the back wall circuit.29 Table 3.68 Heat Output (W/m) 261.18 490.96 62.01 21.92 86.67 332.18 64.12 based on the uncertainty analysis in the test results.03 407.10 56.75 77.Table 3.

00 540.00 240.00 Heat Output (W/m) 440.00 740.00 640.12: 24-4 and 24-8 Heat Output versus Panel Inlet Water Temperature 38 .00 50 60 70 80 Inlet Water Temperature (oC) 90 100 Figure 3.00 140.00 40.11: Mean Panel Temperature versus Panel Inlet Water Temperature 840.100 90 Mean Plate Temperature ( oC) 80 70 24-4 Panel 24-8 Panel 60 50 40 50 60 70 80 Inlet Water Temperature (oC) 90 100 Figure 3.00 24-4 Panel 24-8 Panel 340.

the overall heat transfer coefficient. an analytical analysis on the radiative heat transfer was performed. calculated using a known water flow rate and an estimated inlet fluid temperature. Figure 4. Whillier. is rarely known in actual applications. Therefore. manufacturers install radiant panels based on their estimated heat output. 39 . In reality. It relates the heat output and panel temperature to panel size. The current Reverse Solar Collector (RSC) model is based on the classical fin tube solar collector model presented by Hottel. The panel temperature.Chapter 4 Heat Transfer Model The current Chapter presents the development of an analytical model for a perimeter radiant heating panel. mean water temperature and surrounding air temperature. To determine the overall heat transfer coefficient. With knowledge of the inlet flow conditions. back losses and bond conductance between the tube and panel. and a numerical analysis on the convective portion. The model was developed as an extension of the model by Hadlock (2004). a radiant panel model should be related to these known variables. The back losses were approximated by measuring the insulation thickness on the back of the panel and the effects of the bond conductance on panel performance are discussed. the model can be used to predict the mean panel temperature and heat output.1 Reverse Solar Collector Model Most radiant panel models predict the heat output based on an estimated panel temperature. Bliss (1958).1 shows the geometry used to derive this model. however. 4.

2). An energy balance on a small elemental region on the fin surface (Figure 4.δ (W − Do ) / 2 Tpanel B B Di B B • • Do B B • • • Tf B B W/2 • • W is the tube spacing (W-Do)/2 is considered the fin length δ is the fin thickness Do is the outer tube diameter Di is the inner tube diameter Tf is the average fluid temperature Tpanel is the mean panel temperature B B B B B B B B B B Figure 4.1) ∆x δ Tb B B dqc B B qx B B qx+dx B B (W-D)/2 dqback B B Figure 4.2) where: L is the panel length 40 .2: Energy Balance on Fin Element From Fourier’s law we know that: q x = − kδL dT dx x (4. yields: q x = q x + dx + dq c + dq rad + dqback dqrad B B (4.1: Radiant Panel Sheet and Tube Dimensions A radiant panel version of the model is derived in the following manner.

knowing that the radiant panels are heavily insulated in the back.7) where: tins is the thickness of the insulation B B Equation 4.. x x (4.3) The convective heat transfer rate can be expressed as: dq c = hc ∆xL(T − Ta ) (4.7 is based on the assumption that the surrounding air temperature is close to the bulk air temperature in the enclosure. 41 .and q x + dx dT = −kδL dx d 2T − kδL 2 dx dx − . as: dqback = k ins ∆xL(T − Ta ) t ins (4.4) Similarly.6) The back losses can be estimated.5) where the radiative heat transfer coefficient is given by: hr' = " q rad T panel − Ta (4. The convective. radiative and back loss coefficients can be added together to give an overall heat transfer coefficient: U L = hc + hr' + k ins t ins (4. the radiative heat transfer rate can be expressed as: ' dq rad = hrad ∆xL(T − Ta ) (4.8) Details on how the heat transfer coefficients were obtained are discussed in the later sections of this Chapter..

10) To solve the 2nd order differential equation two boundary conditions have to be specified.) = 0 dx dx dx (4.13) where.12) Solving the 2nd order differential equation (Hottel. with the corresponding boundary conditions. m= UL kδ (4.8 into Equation 4.1 a new equation can be developed: dT dT d 2T − U L ∆xL(T − Ta ) − kδL − (−kδL − kδL 2 ∆x − ..9 by ∆x and L and as ∆x approaches 0. the temperature of the fin can be expressed as: T= cosh(mx)(T panel − Ta ) cosh(m(W − D) / 2) + Ta (4.9) By dividing Equation 4. Whillier and Bliss.10 can be written as: d 2T U L = (T − Ta ) kδ dx 2 P P (4. so at x = 0: dT dx x =0 =0 (4. 42 .By substituting Equations 4.. thus: T P P x = (W − D ) / 2 = T panel (4. 1958).2 to 4. At the centerline of the fin (the ends where the fins connect) symmetry is assumed.14) and T is the temperature of the plate at some location x.11) The portion of the fin which connects to the plate just above the tubing is assumed to take the same base plate temperature. Equation 4.

which is generally known.17 predicts the heat transfer rate of a radiant panel based on known air and panel temperatures. the energy collected can be shown to be: q 'fin = (W − D) F [−U L (T panel − Ta )] (4. Ideally. it can be shown that the heat output for a radiant panel is: ' qu = ((W − Do ) F + Do )[U L (T panel − Ta )] (4. 1 hwπDi ln( Do / Di ) 2πk 1 Cb Tf B B Tp B B Tb B B Tpanel B B Figure 4. equal to: F= tanh(m(W − D) / 2) m(W − D) / 2 (4. A resistance network can be used to estimate the panel temperature based on the fluid temperature.By applying Fourier’s law at the base of the fin to both sides of the tube. the net heat output of the panel should be calculated based on the inlet fluid temperature.17) Equation 4.3: Radiant Panel Resistance Network 43 .16) To complete the heat transfer for a section of the radiant panel. as seen in Figure 4.15) where: F is the fin efficiency factor.3. By doing a similar analysis. the portion of the plate over the tube must be accounted for as well. however.

bond and panel thermal resistances. the heat transfer for the first pipe run can be expressed as: & q 'panel 1 = mC p (T f . and the conductive heat transfer through the pipe can be assumed to be negligible. From Equation 4. radiant panels are typically made from copper tubing.21 can be used to estimate the heat transfer of the panel based on the fluid temperature and air temperature for a section of the radiant panel. Having a high thermal conductivity.18) where: Cb is the bond conductance.20.in − T f . The heat transfer from the fluid to the plate can therefore be written as: q 'panel = T f − T panel 1 1 + hwπDi C b (4.21) Equation 4. The total heat transfer rate for multiple pass systems depend on the radiant panel tube configuration (serpentine or parallel header). the panel heat output can be shown to equal: q 'panel = WF '[U L (T f − Ta )] (4.19) where: b is the bond width Adding the fluid.out1 ) (4. and is equal to: B B Cb = kbb tb (4.22) 44 .20) where: F ' is: F'= 1/ U L ⎡ 1 1 1 ⎤ + + W⎢ ⎥ ⎣U L [ Do + (W − Do ) F ] C b πDi hw ⎦ (4. for a serpentine system.To maximize heat transfer. the temperature change across the pipe is minimal.

in − T f .20.in − T f .For a 4 pass system the other three passes can be written as: & q 'panel 2 = mC p (T f .23) (4.out .out1 − T f .out 4 ) (4.28) it can be shown that: 45 . For example: & mC p dT f dy = WF '[U L (T f − Ta )] (4. since each pass will have the same inlet and outlet fluid temperature.25.28) where: y is the position along the tube. the total heat transfer rate can be written as: & q 'paneltotal = mC p (T f .out 2 − T f .25) By summating Equations 4. For the serpentine circuit.26 was used for the analysis. the fluid outlet temperature can be predicted at the end of each circuit by setting Equations 4.26) For the parallel circuit system. giving: & q 'paneltotal = mC p n(T f .out 3 ) & q 'panel 4 = mC p (T f . Solving the first order differential equation (Equation 4.27) where: n is the number of passes Equation 4. since serpentine radiant panels were used in the experiment.22 to 4.out 4 ) (4.20 by the number of passes.24) (4.25 equal to Equation 4.out 2 ) & q 'panel 3 = mC p (T f . The tradeoffs between a header system and a serpentine system can be seen based on the outlet temperature. the heat transfer rate can be determined by multiplying Equation 4.n ) (4.22 to 4.out 3 − T f .

As discussed in Chapter 2. be convenient to determine an equation which can predict the heat output from the panel. the outlet water temperature is usually not known. In radiant perimeter 46 .2 Radiant Exchange Model Several methods of determining the radiant heat exchange for a panel have been presented (Walton. based on known parameters.mean = T f . It would. assume a 2D radiant heat exchange.in − Ta ⎡ A panelU L F ' ⎤ = exp ⎢− ⎥ & mC p ⎥ ⎢ ⎣ ⎦ (4.in − Ta ) and the mean plate temperature (Duffie and Beckman. They also assume low temperature differences between the panel and surrounding surface temperatures. The inlet fluid temperature and flow rate are generally known. FR. This can be achieved by using a heat removal factor. therefore.31) T p . defined as the ratio of the actual heat output of the panel to the heat B B output if the entire panel were at the fluid inlet temperature. the heat removal factor for a radiant heating panel can be shown to be: FR = ⎡ ⎛ A panel F 'U L ⎢1 − exp⎜ − ⎜ & A panelU L ⎢ mC p ⎝ ⎣ & mC p ⎞⎤ ⎟⎥ ⎟⎥ ⎠⎦ (4. 1980 and Steinman et al. however.T f . 1991): (4.. most of these models. Following a similar derivation as in Duffie and Beckman (1991).outfinal − Ta T f . 1989).out ) A panel FRU L (1 − FR ) (4. the heat output from the panel can be shown to be: q panel = A panel FRU L (T f .in − T f .in − & mC p (T f .30) Finally.29) When trying to predict the heat output from the panel.32) 4. and do not account for temperature differences between the surfaces.

The net radiant energy leaving the surface is known as the 47 . and then a three-surface radiation model.heating applications.4: Cut Away Sketch of a Seven Surface Radiant Panel Test Chamber Each surface in the enclosure has some portion of radiant energy leaving the surface and another portion incident on it. both methods will be compared to Walton’s MRT method accepted by ASHRAE (ASHRAE.4). discussed in Chapter 3. It is therefore necessary to determine which method can be used to approximate the radiative heat transfer from a radiant perimeter heating panel. Exact Radiation Heat Exchange Model For an enclosure similar to the radiant panel test chamber. 2004). and surrounding surface temperatures also vary significantly. The exact solution will be discussed first. Radiant Panel Ceiling Left Wall Cold Wall Floor Figure 4. Finally. the panel temperature is often at a much higher temperature than the surrounding conditions. there are a total of seven surfaces (Figure 4.

33.35.33) The incident radiation on a surface is known as the irradiance.35) For a seven surface enclosure the net radiation heat transfer from the panel can be determined by solving Equation 4.34.36 can be expressed in matrix form by performing a net radiation balance at each surface and bearing in mind that the surfaces cannot view themselves.radiosity. As shown in Equation 4. 2005). Gi = ∑ Fi − j J j j =1 n (4. the radiosity can be found substituting Equation 4. J. For each surface. This requires the knowledge of view factors that can be computed explicitly in this case. J i = ε iσTi + (1 − ε i )∑ Fij J j 4 N (4. This equation can be shown to be: 48 . For a given opaque surface.35.33 giving the following equation (ASHRAE. the net radiation transferred to the panel can be solved from the radiosity at each surface. G and is given by Equation 4. it can be shown that ρi = 1 – εi and hence: B B B B J i = ε i Ebi + (1 − ε i )Gi where: Ebi is the emissive power of a surface i Gi is the incident radiation on a surface i ρ is the reflectivity of the surface B B B B (4.34) q flux = ∑ Fi − j ( J i − J j ) j =1 n (4.34 into Equation 4. which includes the emitted and reflected radiative components as shown in Equation 4.36) j =1 Equation 4.

For reasons of simplicity.5) because five of the seven surrounding surfaces have equal surface temperatures and emissivities.37 can be solved by matrix inversion to obtain the radiosities at each surfaces.37) Equation 4.5: Three Surface Enclosure Approximation 49 . the radiosities can be calculated and hence the net radiative transfer from the panel can be determined as follows: " q rad = J panel − ∑ F panel − j J j n (4.⎡T14 ⎤ ⎡(1−ε ) −1 F12 F13 F14 F15 F16 0 ⎤⎡ J1 ⎤ ⎢ 4⎥ ⎢ ⎥⎢ ⎥ F23 F24 F25 F26 F27 ⎥⎢J 2 ⎥ (1−ε ) −1 ⎢T2 ⎥ ⎢ F21 −1 ⎢T 4 ⎥ ⎢ F31 F32 F34 F35 F36 F37 ⎥⎢J 3 ⎥ (1−ε ) ⎢3 ⎥ ⎢ ⎥⎢ ⎥ −1 (1−ε )⎢ F41 J 4 ⎥ = εσ⎢T4 4 ⎥ F42 F43 F45 F46 F47 ⎥⎢ (1−ε ) −1 ⎢T 4 ⎥ ⎢ F F52 F53 F54 F56 F57 ⎥⎢J 5 ⎥ (1−ε ) ⎢ 54 ⎥ ⎢ 51 ⎥⎢ ⎥ −1 J6 ⎥ F62 F63 F64 F65 F67 ⎥⎢ (1−ε ) ⎢T6 ⎥ ⎢ F61 ⎢ 4⎥ −1 ⎥⎢ ⎢ 0 ⎥ F72 F73 F74 F75 F76 (1−ε ) ⎦⎣J 7 ⎦ ⎣ ⎣T7 ⎦ (4. a three surface enclosure was assumed (Figure 4.38) j =1 where: Jpanel = total radiosity leaving the panel surface Jj = Radiosity from other surfaces in the room B B B B Simplification to the Exact Radiant Heat Exchange Model The exact radiant heat transfer can be calculated using a seven surface enclosure. With knowledge of the surface temperatures and emissivities. Radiant Panel (Surface 1) Cold Wall (Surface 2) Surroundings (Surface 3) Figure 4.

The fictitious surface would have the same overall characteristics of the actual enclosure thus minimizing the error associated with the assumption. Walton assumed that the temperature difference between surfaces was not great and that surface panel was much smaller than surface j. F13. Doing a net energy balance on each surface.F12. 2004) suggests that the radiant heat transfer from a panel can be calculated using Walton’s (1980) MRT method. can be calculated using the published view factor equation for perpendicular rectangles with a common edge. The view factor from the panel to the third surface.39) The view factor from the panel to the cold wall.39 by solving for the radiosities by matrix inversion. Walton showed that the net heat flux from radiation could be expressed as: " *3 q rad = ∑ 4σε j F panel − j Tavg (T j − T panel ) n (4. the set of radiosity equations derived from Equation 4. MRT Method The ASHRAE HVAC and Systems Handbook (ASHRAE. since the sum of all B B B B view factors in an enclosure is equal to unity. Walton assumes the radiant panel would exchange energy with a fictitious surface. F12.36 can be written matrix form as: ⎡(1 − ε ) −1 ⎢ (1 − ε ) ⎢ − F21 ⎢ − F31 ⎣ − F12 (1 − ε ) −1 − F32 ⎡T1 4 ⎤ ⎤⎡ J1 ⎤ − F13 ⎢ 4⎥ ⎥⎢ ⎥ − F23 ⎥ ⎢ J 2 ⎥ = εσ ⎢T2 ⎥ ⎢T3 4 ⎥ (1 − ε ) −1 − F33 ⎥ ⎢ J 3 ⎥ ⎦⎣ ⎦ ⎣ ⎦ B B (4. The remaining view factors can be calculated using the reciprocity equation. is simply 1 . In most radiant 50 .Considering only three surfaces.40.40) j =1 * where: Tavg is the area weighted average enclosure temperature To use Equation 4. The radiant heat transfer from the panel can be calculated using Equation 4.

00 450.00 Actual MRT Three-surface enclosure 300. Radiant perimeter ceiling panels are typically significantly warmer than the surrounding temperatures.00 400.00 500. Method Comparison The radiant heat output in an enclosure with the same properties as the Sigma radiant panel test chamber for the three models discussed are compared in Figure 4. It can be seen that the MRT method is within 8% of the actual solution and the three-surface enclosure is within 0. 550.heating perimeter ceiling cases this is not true. so some degree of error would be associated with using Equation 4. It was therefore decided to use the three-surface enclosure approximation to determine the radiant heat transfer from the panel.6 below.00 150. The solutions for each case can be found in Appendix E.6: Predicted Radiative Heat Flux versus Panel Temperature for Three Analytical Models 51 .3%. because accuracy was not compromised over computation time.00 Heat Flux (W/m2) 350.00 200.00 250.40.00 40 50 60 70 Panel Temperature (oC) 80 90 100 Figure 4.00 100.

Awbi and Hatton (1999) conducted various experiments to determine the natural convective heat transfer. has found that the radiant panel heat output can vary as much as 27% if the incorrect convective heat transfer coefficient is used.41) 4. since it is believed that the convective heat transfer has a minimal effect on panel performance. Lomas (1995). the heat transfer coefficient can be calculated by dividing it by the panel to air temperature difference. 52 . the radiative heat transfer must be expressed as a heat transfer coefficient.To use the reverse solar collector (RSC) model derived in Section 4.’s (1956) ASHVE test facility data. hr' = " q rad T panel − Ta (4. they gave limited details concerning the surrounding conditions and limitations to the developed correlations.1. however. The ASHRAE HVAC and Systems Handbook (2004) has accepted this correlation.3 Convective Heat Transfer Coefficient In current analytical radiant panel models. It was therefore decided to determine the convective heat transfer coefficient using a CFD model simulating the test chamber conditions. however. Those experiments were based on full sized ceiling panels with a uniform surrounding enclosure temperature. the convective heat transfer is based on a correlation from Min et al. The heat transfer coefficients were then obtained for various panel sizes and temperatures and the correlation used in the RSC model. Using the heat flux calculated for the three-surface enclosure assumption.

W. The ceiling is a distance. H.3. the fundamental laws of mass. and assumed to be isothermal. four of which are considered isothermal. and one surface which is considered adiabatic. No infiltration was added to the model. above the isothermal floor (BC) assumed to be at Twarm. momentum and energy conservation apply. y D dT =0 dy E B W Tpanel B A Twarm B B Tcold B B H L Twarm B B C B x Figure 4. apart. L.7: Schematic of CFD Model 4. and assuming that the fluid is incompressible. Using the Boussinesq approximation. since the radiant B B panel test chamber was sealed during tested. The end walls (AB and CD) are set a distance. the radiant panel (EA) of length.1 Numerical Model The system studied is shown in Figure 4.2 Governing Numerical Equations In natural convective heat transfer. which relates the density change to temperature change.4. The ceiling (DA) is made of two parts. and 53 .7. It consists of five surfaces.3. and the acoustic tiles (DE) considered to be adiabatic.

The added component accounted for the turbulent eddy transport of momentum.44) where: p is the pressure β is the volumetric thermal expansion coefficient The energy equation is given by: ⎛ ∂ 2T ∂ 2T ⎞ ⎛ ∂T ∂T ⎞ ⎟ = α⎜ 2 + 2 ⎟ ⎜u +v ⎜ ∂x ⎜ ∂x ∂y ⎟ ∂y ⎟ ⎠ ⎝ ⎠ ⎝ (4. a turbulence model was assumed adding a component to the governing laminar convective equation. the k-l model is: B B 54 . The turbulence viscosity. The momentum x and y-equations are: (4. can be expressed by various models.45) where: α is thermal diffusivity The viscous dissipation has been neglected in Equation 4. the air velocities could not be assumed to be negligible. The mass conservation or continuity equation is given by: du dv + =0 dx dy where: u and v are velocities in the x and y-direction respectively. which is a valid assumption when the air velocities caused by buoyancy effects are very small in natural convective heat transfer. the governing equations can be written. To take this effect into account. For example. µt. Due to some thermal instability caused by the cold wall.42) ρ⎜u ⎜ ⎛ ∂ 2u ∂ 2u ⎞ ⎛ ∂u ∂v ⎞ ∂p + v ⎟ = − + µ⎜ 2 + 2 ⎟ ⎜ ∂x ∂y ⎟ ∂x ∂y ⎟ ⎝ ∂x ⎠ ⎠ ⎝ (4.43) ρ⎜u ⎜ ⎛ ∂ 2v ∂ 2v ⎞ ⎛ ∂u ∂p ∂v ⎞ + v ⎟ = − + ( µ l + µ t )⎜ 2 + 2 ⎟ + gρβ (T − T∞ ) ⎜ ∂x ∂y ∂y ⎟ ∂y ⎟ ⎠ ⎝ ∂x ⎠ ⎝ (4.has negligible viscous dissipation.45.

3 Model Parameters A total of 10 different panel temperatures were simulated over 9 different panel sizes.005 Ns/m2. the model dimensions were 3. and therefore they were set as functions of temperature. The density was evaluated from the ideal gas law. varying minimally over the expected air temperature range and because the convective heat transfer was evaluated at the bulk mean air temperature. ρ = p/RT.46) where: Cµ is an empirical coefficient and fµ is another coefficient k and l are the model parameters B B B B In this work the STAR CD constant turbulence model was used with a turbulence viscosity of 0. while the floor (BC) and front wall (CD) were set to 293K. P P 4. For the same reasons.9624m in length (BC) and 3.3. The panel sizes simulated ranged from 0. The temperatures ranged from 318 K to 363K. 2003).25m to 2m. The specific heat and thermal conductivity of the air were evaluated at a reference air temperature of 293 K. shown below: 55 . The percentage change of this variable for the expected temperature range was 15%. typically incremented by 0. with an incremental value of 5 K. No radiative effects were included since the model was simply used to predict the natural convective heat transfer coefficient. each simulation gave a temperature profile within the room.1 µ t = f u C µ / 4 ρk 1 / 2 l (4. To simulate the test chamber constructed at Sigma. A total of 90 simulations were run. the molecular viscosity was calculated from the Sutherland equation (LMNO.25m.048m in height (AB). The cold wall (AB) was set to a temperature of 283 K.

4. however it was still evaluated using the Sutherland equation.4 K for air P P (4. Between 1000 to 4000 iteration were required to solve the governing equations at a specified tolerance of 10-5. 56 .3. while the pressure relaxation factor was set to 0. The generated blocks were merged and the boundary locations identified and fluid properties specified. A coarse grid size of 0.47) It was later decided to add a turbulence model making the laminar viscosity negligible. the 2D model of the radiant panel test chamber was created. The density and laminar viscosity relaxation factors were set to 1. The u and v velocity components were calculated with a P P relaxation factor of 0. creating a 2D block structure.3. Using the mesh command. the grid was generated for the 2D block structure.4 Numerical Simulation Method Using the STAR-CD’s Create Geometry option. The following steps were performed to create the model: • • • 26 vertex locations described the main structure of the test chamber The vertices were connected with splines.6 will discuss the grid refinement.458*10-6 kg/(msK1/2) for air S = 110. momentum and energy) are specified in STAR CD’s Properties option.25x10-5m) which expanded (x2) to the coarse grid size.7 and the temperature was calculated with a relaxation factor of 0. These integral equations were solved using the CD differencing scheme and the SIMPLE solution algorithm. while at the boundary locations a very fine grid was used (6.bTa3 / 2 µ= Ta + S where: b = 1. P P • The governing equations (mass.032m was used for the interior. Section 4. they must be integrated over the control volumes generated by the mesh.95.3. To solve these equations.

49) 1 Ta Gr = 3 gβ (T panel − Ta ) DH ν 2 .50 and 4. through convection.48) The corresponding Nu.3. The convection coefficient. hc = q "panel (T panel − Tair ) (4. 4.51) where: ν is the kinematic viscosity The temperature gradients within 1 cm from the corner of the cold wall and radiant panel were ignored.51) were calculated using the properties of air at the measured value.β = (4. This was done because at this point the temperature gradient approached infinity because of the singularity in the model. 57 .49.50) Ra = Gr Pr = 3 gβ (T panel − Ta ) DH να (4. The heat output from this unaccounted section is varies for each panel size and temperature.5 Data Analysis To determine the natural convective heat transfer coefficient it was necessary to calculate the heat flux from the panel at every cell location immediately below the panel. The temperature gradient divided by the cell size gave the heat flux which was ultimately converted into the total heat output of the panel. Gr and Ra numbers (Equations 4. Nu = hc DH dT * =+ ka dy * y *= 0 (4. was calculated with knowledge of the heat output and the temperature of B B the air at the center of the chamber. however on average 15W are not accounted for – 1% of the total panel heat output. hc.4.

such a fine mesh size was not necessary and would be very time consuming.064 0. To simulate the entire enclosure. it was necessary to compare the heat output from the panel over various coarse grid refinements and boundary grid refinements.26 19. The finest grid refinement was done at the panel and cold wall surfaces because these surfaces had the greatest influence on the air temperature distribution.032 19.0125 0.4.025 0. The following results tabulated in Table 4.05cm and had a grid expansion of two until it matched the coarse grid size.3.032 0.6 Grid Refinement To determine the grid size required to simulate the natural convection in a perimeter ceiling heated room.064 0.00625 0.0125 0. The steps taken were: • • Refine the boundary grid size start at the panel and cold wall and compare to the previous grid Refine the coarse grid and compare to previous grid The temperature gradient for the ceiling was taken within 1 cm of the exterior wall and the non-heated ceiling was assumed to be adiabatic. Table 4. The grid refinement procedure was two-fold.003125 Heat Flux (W/m2) P P 0. however.003125 0.025 0.064 0.032 0. The grid size along the front wall and floor started at 0. To determine the effects immediately below the panel it was necessary to have a very small grid size.42 58 .1 below summarize the grid refinement procedure (Appendix F).59 19.4 19.00625 0.32 19.82 19.064 0.1: Grid Refinement Results Coarse Grid Size (cm) Grid Size Start at Panel and Cold Wall (cm) 0.47 19.12 19.032 0.

The ASHVE test chamber had a uniform surrounding temperature of 293 K at each wall and floor and the entire ceiling was a radiant heating panel. a coarse grid size of 0. Gr and Ra numbers were calculated and compared to the obtained experimental data.00625cm for the cold wall and panel. For each surface. 1952) and the experimental results are accepted by ASHRAE and used to predict the natural convective heat transfer for radiant panels. the ASHVE test chamber was created in STAR-CD..From the grid refinement results. The other boundaries had a starting mesh size of 0. This model was selected because the test chamber was described in detail (Tasker et al.05cm. by positioning the thermocouple in the exact location and holding it steady. Using the same method that was used to model Sigma Convectors radiant panel test chamber. the panel width is the entire ceiling at a constant temperature and the chamber length is 7. The same solution algorithms were used as those in the above model.. Figure 4. These results were collected manually. The differences between the ASHVE test chamber and Sigma Convectors test chamber was the size and the boundary conditions. however.3. 1956) and comparing the predicted CFD results to the experimental values based on measurements done almost 50 years ago.7 Model validation The model was validated by simulating the ASHVE test chamber (Min et al.8 is similar to the model developed for the ASHVE test chamber.032cm was chosen with a starting grid size of 0. The Nu. The natural convective heat transfer coefficient was calculated by measuring the air temperature below the panel at specific locations and determining the temperature gradient. the grid expansion factor was 2 until it reached the coarse mesh size. 4.47 m and 59 .

so an attempt was made to predict 3D effects on the natural convective heat transfer coefficient.8 below. A total of 5 simulations were performed. not included in the 2D simulation. Only one panel size had to be modelled and the panel temperature was varied from 308K to 348K in 10K increments. On the other hand. 80 70 60 50 Nu 40 StarCD ASHVE 30 20 10 0 1E+11 1. The convective heat transfer effects from the left and right walls.5E+11 4E+11 Figure 4. in reality. thus supporting the 2D model used to determine the natural convective heat transfer coefficient. The ASHVE test data showed a wide range of scatter as shown in Figure 4.9. could be 60 . the predicted results from STAR-CD were within the experimental results (Appendix F). not infinitely wide. The experimental chamber was.5E+11 GrPr 3E+11 3.2. The ASHVE test facility consisted of 4 walls and a floor at 293K.8: Nusselt versus Rayleigh number for numerical model results and experimental results from the ASHVE laboratory The temperature profile for a 348 K heated ceiling in the room is shown in Figure 4.5E+11 2E+11 2.44 m high.

and the Nusselt number was calculated.10 it can be seen that the front and rear wall x-temperature gradient eventually becomes zero at a certain distance from the wall. Figure 4. The temperature gradient can then be applied to the left and right wall and a natural convective heat transfer coefficient can be calculated (Appendix F). The percentage change of the Nusselt number from cell to cell was calculated and when the difference was below 1%.9: Temperature Profile for the Numerically Simulated ASHVE Test Facility From Figure 4. the distance from the wall was recorded. 61 .11 compares the 2D results from STAR-CD with the 3D approximation and the ASHVE test results. This temperature gradient was then applied to the left and right wall. A Nusselt number was calculated for each cell below the heated ceiling. Figure 4.taken into account by assuming a temperature distribution similar to that of the front wall is generated. The distance it took for the temperature gradient to become constant was determined from the numerical model results of the ASHVE test facility.

10: Side profile and Top View of ASHVE Test Chamber and y Direction Temperature Gradient Profile 70 65 60 55 StarCd (2D) StarCd (3D) ASHVE ASHVE (Fit) Power (StarCd (2D)) Power (StarCd (3D)) Nu 50 45 40 1E+11 1.5E+11 Figure 4.11: 2D and 3D numerical model simulation results versus ASHVE experimental data 62 .y Front Wall Ceiling Rear Wall Floor x Left Wall Front Wall dT =0 dx Rear Wall z Right Wall x Figure 4.5E+11 3E+11 3.5E+11 2E+11 GrPr 2.

the 2D predicted results from STAR-CD are deemed acceptable. An example of the simulation results of the natural convective heat transfer are shown in Figure 4. As expected.12. Taking the properties of the air at the center of the room. however. some stratification under the radiant panel occurs. the cold wall gave rise to air circulation within the center of the room. since the ASHVE experimental results are almost 50 years old and measurement techniques have greatly improved. 4.8 Numerical results The effects of panel size and temperature on the natural convective heat transfer coefficient was determined simulating 9 panel sizes over 10 panel temperatures (Appendix F). having recorded the air temperature at the center of the chamber. the Nusselt.52) The total heat flux from the panel was calculated by adding up the heat output from each cell and dividing by the panel area. Grashof and Rayleigh numbers were calculated for each panel size and temperature. 63 . q panel = ∑q i =1 n celli Acelli (4.48.As a conclusion.3. For each case. the temperature gradient under the panel was measured and a heat flux for each cell was calculated using Fourier’s law: " q cell = k dT dy (4.53) A panel The heat transfer coefficient was then calculated using Equation 4.

25m and a 1.12: Temperature profile for a 0. This would indicate the effect of the convection heat transfer coefficient with panel size.5m panel at 318K 64 .13). Figure 4.The Nusselt number was compared to the aspect ratio of the panel width to the chamber size (Figure 4.

This was because the smaller panel sizes had a smaller characteristic length.3 Aspect Ratio 0.13: Nusselt Number versus Aspect Ratio As expected the Nusselt number increased as the aspect ratio increased. The Rayleigh number was compared to the aspect ratio (Figure 4. Temperature plots of the small panel size and larger panel size can be seen in Figure 4. which is proportional to the Nusselt number. Correspondingly. Therefore. the Rayleigh number increases.6 Figure 4.2 0. As expected with higher panel temperatures. This is because there is a greater effect of the cold wall on the air motion.45 43 41 39 37 Nu 35 33 31 318K 323K 328K 333K 338K 343K 348K 353K 358K 363K 29 27 25 0 0.14). At larger panel sizes. as the aspect ratio increased.5 0.12. the Rayleigh 65 .4 0. the air temperature distribution beneath the panel was not affected by the cold wall as greatly as with the smaller panel sizes.1 0. the characteristic length played a much smaller role.

15.5m level when a hotter panel 66 .number remains fairly constant over the various aspect ratios indicating that the air motion is primarily due to the temperature difference between the panel and the exterior wall. This can be explained in the following manner.15) was plotted. the surrounding air warmed up and became stratified.00E+07 0. This was seen in the minimal air temperature change at the 1.50E+08 1.1 0.00E+08 5.2 0.50E+08 2. the Nusselt number would increase as well because the convective heat transfer coefficient should increase as more air motion occured. as the radiant panels became warmer.00E+00 0 0.00E+08 3.3 Aspect Ratio 0.5 0.50E+08 3. the Nusselt number decreased. it can be seen that the expected results did not occur.50E+08 4. the Nusselt number versus the Rayleigh number (Figure 4. 5.6 Figure 4.4 0.14: Rayleigh number versus Aspect Ratio Finally.00E+08 318K 323K 328K 333K 338K 343K 348K 353K 358K 363K Ra 2. From previous experiments. it was expected that as the Rayleigh number increased.00E+08 1. As the Rayleigh number increased. From Figure 4.00E+08 4.

When stratification occurs. 45 40 35 30 25 Nu 20 15 10 0.31974624 0.376277635 0. while the Rayleigh number increases because of the less dense air. it can be seen that the convective heat transfer is dependant on the ratio of the panel length to the ceiling length.00E+08 4.190531622 0. lowering the Nusselt number.255138931 0.00E+08 2.temperature was simulated.50E+08 4. Schutrum and Vouris (1954) stated that the convective heat transfer coefficient was not dependant on panel size.134000227 0.50E+08 Figure 4. it can be concluded that the Nusselt and Rayleigh number are dependant on panel temperature and panel sizing.12.50E+08 3. From Figure 4.440884943 5 0 2.222835277 0.061317005 0.15: Nusselt versus Rayleigh number From these results. the convective heat transfer decreases. however. 67 . which is the case for full size panels occupying the entire area of the ceiling.00E+08 Ra 3. As this ratio increases. the Nusselt number flattens out showing that the panel size will have minimal effect on the convective heat transfer coefficient.

16: Convective Heat flux versus panel temperature and panel size From the plot it can be concluded that the convective heat transfer is fairly significant in smaller panel sizes. panel size and panel temperature (Appendix G).66 40 20 0 20 30 40 50 ∆T 60 70 80 Figure 4. a correlation was developed to determine the heat flux from the air temperature.61 1.27 1.44 1.17 2.94 1. because the convective heat transfer is an important component in determining the heat output from a panel. however.The combined effect of panel temperature and panel size on the convective heat transfer is plotted in Figure 4.46 0. and is minimal at larger panel sizes. 1954). 140 120 100 80 qc (W/m2) 60 0. the results from STAR-CD were correlated to the same variables 68 . This is expected because previous case studies have shown that the convective heat transfer is small for large panel sizes compared to the radiative heat transfer (Schutrum and Vouris. this is not the case.16 below. In perimeter heating cases. For comparative purposes.92 2. From the results of STAR-CD.42 2.

(1956) and Awbi and Hatton (1999).54 below and has a correlation coefficient of R2 = P P 0.17 below compares the correlated STAR-CD convective heat flux with the published equations from Min et al. while the air temperature was assumed to be 291K. Schutrum and Vouris (1954) and Awbi and Hatton (1999).17: Comparison of various Heat Flux Correlations 69 . (1956). " qc = 0.as the published equations from Min et al. The heat flux equation is shown in Equation 4. the characteristic length was 1.44m. 80 70 60 50 Heat Flux (W/m ) 2 40 30 20 Numerical Study Min et Al (No Cold Wall) Awbi (Colder Wall) Schutrum (Fully heated Ceiling) 10 0 20 30 40 50 T panel .T air (K) 60 70 80 Figure 4.9997 (Appendix G).0615 ln( DH ) + 0. In this comparison.54) where: DH is the hydraulic diameter of the panel B B Figure 4.9832 (4.0046 (T p − Ta ) 0.9937 DH 1.

201 (T panel − Tair ) 0.54. From Figures 4. For radiant panel perimeter heating applications.06 to 0. comparing the RSC model to the experimental data.0808 AR 0. P P For the purposes of the thesis. it can be seen that the convective heat transfer is dependant on the size of the radiant panel relative to the size of the ceiling. Equation 4.45m to 2. For the radiant panels which are mounted under the ceiling.16.’s (1956) correlation and Awbi and Hatton’s (1999) correlation.0511 ln( AR ) +1. Equation 4.4 Back Losses Typically radiant panels are either installed embedded in the ceiling.13 and 4.99 (Appendix G). panel temperature and air temperature (Appendix G).55) where: AR is the aspect ratio (Lpanel/Lceiling) This equation is valid for a panel temperature ranging from 318 K to 363 K and an aspect ratio of 0. Equation 4.5 and has a correlation coefficient of R2 = 0.As can be seen the STAR-CD correlation lies between Min et al. indicating that the convective heat flux equation is within the acceptable range. It should be noted that Equation 4.65m and a panel temperature range from 318K to 363K.55 will be used to predict the natural convective heat transfer coefficient.54 is valid for a hydraulic diameter range from 0. Using a similar derivation to Equation 4. or mounted hanging beneath the ceiling. " qc = 0. 4.55 predicts the convective heat flux with knowledge of the aspect ratio. 70 .8375 B B B B (4.54 predicts the convective heat transfer from a radiant panel with knowledge of the hydraulic diameter and panel temperature. it would be more useful to determine the convective heat transfer with respect to the aspect ratio.

the back losses must be taken into account.18) using the RSC model (Appendix H).00635m t=0.0127m t=0. W/m2K) 19 17 15 13 11 9 7 5 40 50 60 70 Panel Temperature (oC) 80 90 100 Figure 4. knowing the thickness and conductivity of the material. 23 21 t=0. The back loss heat transfer coefficient can be shown to be: hbackloss = k ins t ins (4. Therefore. In this case. as much as 50% of the heat transfer can be lost if the panel is not properly insulated.0254m t=0.003175m t=infinite Overall Heat Transfer Coefficient (UL.the back losses can be determining by calculating radiative and convective heat transfer coefficients above the panel.56) Plotting the effects of insulation thickness on panel performance (Figure 4. The back losses can be estimated from Equation 4. In cases where the panel is embedded in the ceiling (as in the radiant panel test chamber) it is necessary to insulate the back portion in order to minimize the heat transfer to the outside surfaces.7.18: Effect of Panel Insulation on the Overall Heat Transfer Coefficient 71 . to validate the reverse solar collector model. the back losses are determined using a conductive heat transfer model.

thus spreading the heat conducting paste over the entire contact area. each manufacturer using an alternative method.2 cm thick and had a conductivity value of 0. a 0. the panel thermal resistance should be minimized. because Sigma Convectors wanted to simulate their actual installation conditions using the insulation they provide. it was found that it was 1. the effective panel performance could be estimated knowing the radiative and convective heat transfer. It was however difficult to estimate the thickness and width.5 cm wide and 0.The proposed ASHRAE standard 138P (Chapman. Measuring the insulation used during testing. k. Sigma Convectors. The conductivity of the heat paste is known to be 1.04 W/mK. this resulted in a significantly higher panel output than actual proposed test conditions. improves the bond conductance by adding a conductive heat paste between the tube and panel.08 W/mK. The tube is then clipped into place.5W/mK. of 0. Assuming the paste distributes evenly and over the entire contact area.67 cm. the width of the paste is estimated to be 1. Equation 4. From conservation of volume.09 cm. thickness and width of the heat paste are known. Thicker insulation was not added. 4. 2005) suggests that the panel should be insulated with 2. There are several methods used to minimize the thermal resistance.3 cm thick strip of heat paste is applied to the extruded aluminum panel used for the 24-4.20 can be used to estimate the bond conductance. for example. Although. 72 . if the conductivity. the thickness can be calculated to be 0.5 Bond Conductance To optimize panel performance. During manufacturing.54 cm thick insulation with a minimum conductivity.

3mm 0.0mm 1.5mm.5mm 0. it is important to precisely know the dimensions of the heat paste.19) using the RSC model developed (Appendix I).67cm with a thickness of 0. By increasing the heat paste thickness by 0.3mm 45 40 45 50 55 60 65 70 75 80 85 90 95 Fluid Inlet Temperature (oC) Figure 4.4mm 0. a panel temperature difference over 2oC can be seen. Plotting the effects of bond conductance on the panel temperature (Figure 4.7mm 0. 73 .19: Effect of Heat Paste Thickness on Panel Performance With a 2oC temperature difference the heat output from the panel can vary as much as 20W/m as documented in Appendix I.1mm 1.6mm 0. the results show that the panel temperature can vary by as much as 5oC with a variation in thickness of 1mm. To minimize the uncertainty in the calculation of heat output from a panel.013cm.9mm 1.Following a similar analysis for the 24-8 panel the heat paste had a width of 1.8mm 0.2mm 1. 85 80 75 Mean panel Temperature ( oC) 70 65 60 55 50 0.

Calculate the radiative.001oC of the previous calculated temperature. Input the fluid inlet temperature 2.31 6.30 5. the solution methodology is summarized below. 74 .4. convective and conductive heat transfer coefficients 4. Calculate the outlet fluid temperature using Equation 4.32 The solution is considered to be converged once the predicted panel temperature is within 0. The steps for solving are as follows: 1. The convective and radiative heat transfer coefficients are dependant on panel temperature. Calculate the heat output using Equation 4. Solve for the panel temperature using Equation 4.6 Solution Method Having calculated the necessary variables to predict the panel heat output and panel temperature. and therefore an iterative procedure is required to determine the unknowns. Estimate the panel temperature 3.

13cm thick for the 24-8 panel.02 86.in ( C) 51. P P P P Running the model at the experimentally obtained inlet water temperatures.7mm thick with an estimated conductivity of 0.74 613.92 86. Table 5.66 100.26 64.56 o Heat Output (W/m) 207.98 55. The following parameters were therefore used to run the model: • • • • • The surrounding temperatures during the experiment were maintained at 20oC and the cold wall was at 12oC. The water flow rate was 0.2 for the 24-8 panel.1: RSC Model Results and Experimental Results for the 24-4 Panel RSC Model Tf.29 75 .67cm wide with a thickness of 0.13 B B P P o Heat Output (W/m) 216.35 73.1 Results The analytical model was used to simulate the conditions of the radiant panel test chamber as closely as possible.46 586.91 86.08W/mK.39 378.18 490.25 478. The insulation on the back of the panel was measured to be 12.71 75.56 75.4 B B P P Experiment Tpanel ( C) 46. The panel sizes were 0. The experimental results can be found in Appendix C and RSC model results in Appendix I.057L/s. 5.47 374.65 291.61m wide and had 4 and 8 tube passes connected in serpentine fashion. The heat paste was calculated to be 0. The heat conducting paste between the tube and the panel was 1.09cm and a conductivity of 1.1 for the 24-4 panel and Table 5.89 B B P P o Tpanel ( C) 46. the results are summarized in Table 5.78 61.Chapter 5 Results and Discussion This chapter compares the experimental and RSC model results for the two radiant panels tested and discusses the assumptions and errors associated with both cases.51 54.5W/mK for the 24-4 panel.89 298.18 64.

1 below for the 24-4 panel.49 67.1 56.95 398.03 77.46 66.32 610.10 314.96 62.68 332.08 5.75 87.68 553. The model is within 2% for each of the five measured panel temperatures (Appendix I).05 57.03 407.1 Mean Radiant Panel Temperature The mean radiant panel temperature is plotted against the inlet fluid temperature for RSC model and experiment in Figure 5.1.46 o Tpanel ( C) 48.42 o Heat Output (W/m) 233.71 86.02 Heat Output (W/m) 261.Table 5. 76 .2: RSC Model Results and Experimental Results for the 24-8 Panel RSC Model Tf.66 o Experiment Tpanel ( C) 48.61 700. 90 85 80 75 Mean Plate Temperature ( oC) 70 65 Experimental Predicted 60 55 50 45 40 40 50 60 70 80 90 100 110 Inlet Fluid Temperature (oC) Figure 5.35 497.11 99.75 77.76 89.77 73.in ( C) 51.1: Experimental and Analytical Model Panel Temperature versus the Inlet Water Temperature for a 24-4 Panel As can be seen the model predicts the actual mean radiant panel temperature very closely.

3 and Figure 5.The accuracy of the model is further shown in Figure 5.4 below. where experimental and predicted panel temperatures are equal. The predicted values are very close to the ideal case.2: Predicted Panel Temperature versus Actual Panel Temperature for a 24-4 Panel The experimental results for the 24-8 radiant panel tested are compared to the RSC model in Figure 5. concluding that the mean plate temperature predicted by the RSC model is a valid analytical method for radiant heating panels. 100 90 Predicted Mean Plate Temperature ( oC) 80 70 Ideal Actual 60 50 40 40 50 60 70 80 90 100 Experimental Mean Plate Temperature (oC) Figure 5.2. 77 . a plot comparing the predicted panel temperatures to the experimental measured values.

100 90 Experimental Mean Plate Temperature ( oC) 80 70 Experimental Predicted 60 50 40 50 60 70 80 Inlet Fluid Temperature (oC) 90 100 Figure 5.3: Experimental and Analytical Model Panel Temperature versus the Inlet Water Temperature for a 24-8 Panel 100 90 Predicted Mean Plate Temperature ( oC) 80 70 Ideal Actual 60 50 40 40 50 60 70 80 90 100 Experimental Mean Plate Temperature (oC) Figure 5.4: Predicted Panel Temperature versus Actual Panel Temperature for a 24-8 Panel 78 .

00 Experimental Predicted 300. the predicted model is within 2% of the measured data – the greatest deviation occurring at the higher panel temperature (Appendix I).Similar to the 24-4 panel temperature results.00 500.5 and Figure 5.00 100.00 Heat Output (W/m) 400.2 Radiant Panel Heat Output Figure 5.00 50 60 70 80 Fluid Inlet Temperature ( C) o 90 100 110 Figure 5. concluding that the RSC is a valid mean panel temperature prediction method.00 200.5: Experimental and Analytical Heat Output versus Inlet Water Temperature for a 24-4 Panel 79 .00 600.1.6 compare the RSC model panel heat output with the experimental results for both test cases. The predicted panel temperatures are still in good agreement with the experimental data. 700. 5.

6: Experimental and Analytical Heat Output versus Inlet Water Temperature for a 24-8 Panel Overall. the model is accurate within 5% of the 24-4 experimental results and 10% for the 24-8 results the greatest error occurring at the higher panel temperatures.00 600. Uninsulated. At the higher mean water temperatures. In most of the test runs. This under-prediction is expected because of the radiant panel installation and temperature measurement setup.800. This additional air flow contributed to the heat output measured from the experimental test 80 . a significant air flow was felt around the chamber.00 700. quick-disconnect hoses. which was unaccounted for by the analytical model.00 200. were installed between the radiant panel and the radiant panel water source. caused by the opening of a garage door located close to the radiant panel test chamber. the RSC under-predicts the actual heat output.3 meters in length.00 Heat Output (W/m) 500.00 300. At the higher mean water temperatures. 0.00 Experimental Predicted 400.00 50 60 70 80 Inlet Fluid Temperature (oC) 90 100 Figure 5. more heat was lost to the surroundings.

Figure 5. Running the RSC model for this scenario the results are close to the obtained experimental results as can be seen in Figure 5. 70oC mean water temperature test.7 below compares the experimentally measured air temperature for the 0. the air temperature at the geometric center of the room must be determined.3 Predicted Air Temperature Results To determine the convective heat transfer.53m Wide Panel (STAR-CD) 0. the air flow around the chamber was minimized during the 24-8.75m Wide Panel (STAR-CD) 293 288 283 315 320 325 330 335 340 345 350 355 360 365 Panel Temperature (K) Figure 5.75m wide panel.6.7: Experimental Air Temperature Measurements for a 0.53m wide panel and a 0.61m Panel and the Predicted Air Temperature from the Numerical Model at the Center of the Radiant Panel Test Chamber versus The Panel Temperature 81 . 5. Identifying that this was a source of error. The air temperature inside the room was experimentally measured using a shielded thermocouple located 1.1.chamber. 318 313 308 Air Temperature (K) 303 298 Experimental 0.83m above the floor at the center of the room.61m wide panel to the results predicted by the STAR-CD simulation for a 0.

First an attempt will be made to quantify the uncertainty associated with the measurement of the heat paste thickness. Looking at the numerical results. the wall temperatures fluctuated. Three cases in particular will be discussed.5oC higher in some cases than the predicted numerical results. the bond conductance has a significant impact on the predicted panel temperature. In addition. Finally.2. which ultimately caused the variation in air temperature. During the experiment. The air temperature results therefore deviated slightly from the expected results.5. For instance. and the walls remain at the same temperature. 5. so the effects of the side walls were not accounted for. the air temperature increases in a linear fashion versus the panel temperature. the wall temperatures were difficult to maintain at a constant value without a chilling system. varying the thickness of the heat paste by 82 . STAR-CD modelled the chamber in 2D. the air temperature measurement location and increase caused by the increased back wall temperature will be considered. Due to this temperature control method. the effects of the cold wall temperature not being at 10oC on the correlated convective heat transfer coefficient will be analysed. several assumptions and simplifications have been made to use the RSC model.2 Uncertainties and simplifications in the RSC Model As in any heat transfer model. Cold water had to be constantly added to the system in order to maintain a 20oC temperature on the wall. This would be expected because only the panel temperature is varied. and secondly.1 Thickness of the Heat Paste As discussed in Section 4.The measured air temperature was as much as 2. 5.

0. When the radiant P P panels for testing were built.5mm.3 below summarizes the uncertainty P P of each predicted panel temperature at various different inlet fluid temperatures.05cm and the thickness measured 0.19: Cb = kbb tb (5. Table 5.18 was considered: ⎛ 1 1 T panel = T f − q 'panel ⎜ ⎜ h πD + C b ⎝ w i B B ⎞ ⎟ ⎟ ⎠ (5. it was assumed that no uncertainty was associated with the heat transfer from the panel. To relate bond conductance to the temperature of the panel.5mm could result in a panel temperature increase/decrease of 2oC. Cb. the heat paste was applied though a caulking gun. the uncertainty in the panel temperature increases.2) To determine the error associated with the bond conductance. The uncertainty in the bond conductance was then determined using the RSS method (Appendix D) and shown to equal 5.3cm +/-0. which was accurate to within 0. From these results it can be concluded that at greater inlet fluid temperatures. ultimately affecting the panel temperature.5cm +/-0. 83 . the fluid temperature or the heat transfer coefficient of the circulated water. the amount used varied from panel to panel.05cm for the 24-4 panel. the predicted panel temperature is within 4oC of the calculated value (Appendix D). is calculated from Equation 4. Equation 4.1) and the bond conductance. By applying the heat paste manually.4 W/mK. the width of the heat paste measured was 0. For example. During the construction of the 24-4 and 24-8 radiant panels used for testing the initial thickness and width of the heat paste was measured. At this level of uncertainty. the heat paste temperature drop becomes a greater. This was expected because as the heat flux from the panel increases.

64 3. the test chamber was simulated in STAR-CD using the same boundary conditions as initially proposed.51 54.3: Uncertainty associated with bond conductance Tf. however. the STAR-CD model for a 0. One difference.13 Uncertainty (oC) 1.2 Convective heat transfer coefficient To determine the convective heat transfer coefficient. Figure 5.71 75.34 4.25 478.8: Convective Heat Transfer versus Panel minus Air Temperature at different back wall temperatures 84 .Tair (oC) Figure 5.51 2.8 shows a minimal change in the predicted convective heat output.Table 5.91 86.35 73.65 291.66 100.09 5.56 Tpanel (oC) 46.in (oC) 51. 70 60 50 qc (W/m2) 40 30 STAR-CD (12 C) STAR-CD (10 C) 20 10 0 20 30 40 50 60 70 80 Tpanel . To determine how the convective heat transfer was affected by this temperature difference.03 2.78 61.26 64.46 586.02 86.39 378. was the temperature of the cold wall.2.4 Heat Output (W/m) 216. Comparing the results for the 12oC and 10oC back wall temperature.61 m wide panel was run at various panel temperatures and two back wall temperatures (10oC and 12oC) (Appendix F).

Comparing the results.83m above the floor instead of 1. In addition. Comparing these predicted air temperatures to the experimental results for the 24-4 panel (Figure 5. the predicted results were below the actual measured temperature values (Figure 5. the cold wall temperature difference had a minimal effect on the convective heat transfer. 85 .5m during the experiment. a temperature increase of 0.61m panel used in Section 5.2. one being that the numerical model was in 2D and the left and right wall temperature effects were not taken into account and another cause being that the back wall during the experiment was not at the desired temperature. a 3D enclosure would need to be set up. since the air temperature was measured 1. caused by the difference in expected surrounding temperatures.7).2. some measured values are within experimental error. It was therefore decided to consider only the effects caused by a difference in the back wall temperature on the predicted air temperature. 5.2. the numerical model temperatures were taken at the experimental height. which would become a very tedious task in STARCD.With such a small change.3 Air Temperature As mentioned earlier the air temperature is dependant on the surrounding wall temperatures. When comparing the STAR-CD results to the experimental results. can be approximated considering the numerical model for the 0. To take into account the left and right wall effects in a numerical model.9). The air temperature. This can be explained by various causes.5oC at the center of the room is expected (Appendix F).

The air velocity profile plots are shown in Figure 5. which was expected. because of the buoyancy effects of the air.9: Air Temperature versus Panel Temperature for the 24-4 Panel The air velocity profiles predicted by STAR-CD for the 10oC and 12oC cold wall temperature.318 313 308 303 Tair (K) Air Temp 1. 86 .83m Experimental Air Temp 298 293 288 283 315 320 325 330 335 340 Tpanel (K) 345 350 355 360 365 Figure 5.10. A slightly higher air velocity was predicted for the lower cold wall temperature.5m Air Temp 1. the results were very close to each other.

10: Predicted Air Velocity Profile with 12oC and 10oC back wall Temperature 87 .Figure 5.

Chapter 6 Summary and Conclusions 6. The RSC model panel temperature and heat output predictions were in good agreement with the experimental results. and the panel temperature and panel heat output were within experimental uncertainty. a correlation was developed to predict the natural convective heat transfer. the radiative heat transfer could be determined. Numerically. As radiative and convective heat transfer coefficients were required to run the model. based on a flat plate solar collector model.61m wide panel with 4 passes and a 0. A 4m by 4m by 3m test chamber was constructed in which the surrounding walls and floor were maintained at a constant temperature and the heat output from an installed radiant panel was measured.1 Summary An analytical model has been developed to predict the panel temperature and heat output for perimeter radiant panel systems with a known inlet temperature and flow rate. a 0. an analytical analysis of the radiative heat transfer was performed. simple algorithm which could be used to predict panel performance. and a numerical model was developed to predict the convective heat transfer coefficient. 88 . The RSC model followed the same trends as that in the experimental results.61m wide panel with 8 passes. To validate the analytical model. Using the conventional radiative heat exchange method assuming a three-surface enclosure. The panels were tested at 5 different inlet water temperatures ranging from 50oC to 100oC. an experimental analysis was performed on radiant panels. concluding that the RSC model is a viable. Two radiant panels were tested.

2 Conclusions From the analytical and numerical model results several conclusions can be drawn. DH. The numerical model was used to determine the convective heat transfer coefficient and the air temperature in the chamber. • A correlation has been developed relating the characteristic length and temperature of the radiant panel to the convective heat transfer coefficient. • The air temperature predicted by the numerical model was close to the measured air temperature. The numerical model assumed an infinite wide chamber. • The panel temperature had a small effect on the convective heat transfer coefficient. The analytical model was used to predict the panel heat output and panel temperature. The smaller the aspect ratio. 89 . The variation was not as significant as the characteristic length. This is an important observation because radiant panels are typically installed as perimeter heaters adjacent to cold walls. B B of the panel.6. the greater the heat transfer coefficient. The conclusions drawn from the model are: • Higher convective heat transfer coefficients are observed when physically smaller sized radiant panels are installed adjacent to a colder wall. however in some cases it was not within the measured air temperature uncertainty. which are validated by the experimental results and observations. The convective heat transfer will have a greater effect on the overall heat output. explaining the temperature difference. so the left and right wall effects were not accounted for.

so a thermopile was installed. More accurate results were obtained when the air flow around the exposed radiant panel pipe was minimized. was incapable of measuring the millivolt signal generated by the thermopile. As expected. It is therefore important to accurately determine and specify the surrounding conditions. • The back losses have a significant effect on the panel performance. Almost half of the total heat output is due to the losses through the back using the regular supplied 12. • The heat output from the panel was predicted within the experimental error. The purchased data logger. air temperature and water flow rate. the reverse flat plate solar collector model was within 10% of the measured experimental results for both panels tested.3 Recommendations Currently. the overall heat output also increased. An alternative would be to 90 . however. • Approximately 85% of the effective panel heat output is through radiation. This is mainly due to the inaccuracy in the measured water temperature. the experimental test chamber predicts the heat output with an uncertainty of up to 25%. This problem was identified in the early stages of the experiment. surrounding temperatures. With knowledge of the inlet water temperature. 6.7mm thick Batt insulation. as the panel temperature increased.Conclusions that can be drawn from the analytical model validated with the experimental results are: • The analytical model used to predict the panel temperature is within experimental uncertainty.

Installing the radiant panel below the ceiling would give a better estimate of the effective heat transfer to the room. Observations during the experiment. Other studies that could be looked into are the effects of the convective heat transfer on various back wall temperatures.install a thermistor to measure the water temperature. The analytical model assumes constant surrounding temperatures when predicting the radiant heat transfer.1oC. In addition. however it would be interesting to determine at which temperature the back wall will have a significant effect on the air flow.5oC and were maintained within 1oC between quadrants. however. although the other surrounding wall temperatures were maintained at 20oC. almost 50% of the measured heat output was due to the back losses. radiant panels are insulated within the ceiling panels. to ensure that a steady temperature can be maintained. 91 . the cold wall temperature was maintained around 12oC. The city water was incapable of maintaining a back wall temperature of 10oC. or installed below the ceiling. A model should therefore be developed that takes into account the variability of the surrounding surface temperatures. indicated that the surrounding temperatures fluctuated approximately 1. During the experiment. this could not be done immediately. It was noted that a 2oC increase had minimal effects. as installed in the booth. The back wall temperature was maintained by running city water through the mounted copper and PEX tubing. and should be considered in order to get more realistic results. A small chiller should be installed in the system to cool the water down and save on water. a small chiller should be installed in this system as well. Due to budget constraints. which can be accurate up to 0. Typically. From the observations made.

References
Adlam, T.N., 1949, “Radiant Heating,” 2nd edition, The Industrial Press, New York, USA. Armstrong Pumps – Circuit Balancing Valves, 2006, “ArmFlo Venturi Style,” http://www.armstrongpumps.com/Data/performancecurves/Links/01_02_005/CBVV_composite_curves.pdf. ASHRAE, 2003, “Handbook of HVAC Applications – SI Edition,” American Society of Heating, Refrigeration and Air Conditioning Engineers, Inc., Atlanta. ASHRAE, 2004, “Handbook of HVAC Systems and Equipment – SI Edition,” American Society of Heating, Refrigeration and Air Conditioning Engineers, Inc., Atlanta. ASHRAE, 2005, “Handbook of Fundamentals – SI Edition,” American Society of Heating, Refrigeration and Air Conditioning Engineers, Inc., Atlanta. Awbi, H.B., Hatton, A., 1999, “Natural Convection from Heated Room Surfaces,” Energy and Buildings, vol. 32. Buckley, N.A., 1989, “Application of Radiant Heating Saves Energy,” ASHRAE Journal 31(9), Atlanta. Chapman, K., 2005, Personal Communication on the status of the ASHRAE Standard – Method of Testing for Rating Hydronic Radiant Ceiling Panels. Conroy, C., and Mumma, S.A., 2001, “Ceiling Radiant Cooling Panels as a Viable Distributed Parallel Sensible Cooling Technology Integrated with Dedicated Outdoor Air Systems,” ASHRAE Transactions, vol. 107, part 1. Dasylab, 2005, “Instruction Manual for Dasylab Data Acquisition System Laboratory, Version 8.” Din 4715, 1994, “Method of testing and rating cooling panels,” Halle, Germany. DOE (U.S. Department of Energy), 2005, “2005 Buildings and Energy Data Book,” Office of the Energy Efficiency and Renewable Energy. Duffie, J.A., Beckman, W.A., 1991, “Solar Engineering of Thermal Processes,” John Wiley & Sons, Inc. Toronto, Canada. Hadlock, C.J., 2004, “Heat Transfer Model for Radiant Heating,” University of Waterloo, Waterloo, Canada

92

Heid, H., Kollmar, A., 1943, “Die Strahlungsheizung,” Carl Marhold Verlagsbuchhandlung, Halle, Germnay. Hottel, H.C., Whillier, A., 1958, “Evaluation of Flat-Plate Collector Performance,” Trans. Of the Conference on the Use of Solar Energy, vol. 2. Huang, N.Y, 2005, “Thermal Performance of Double Glazed Windows with Inner-Pane Venetian Blinds,” University of Waterloo, Waterloo, Canada. Huebscher, R.G., Schutrum, L.F., Parmalee, G.V., 1952, “A Low-Inertia Low-Resistance Heat Flow Meter,” ASHVE Transactions, vol. 58. Kilkis, I.B., Sager, S.S., Uludag, M., 1994, “A simplified model for radiant heating and cooling panels,” Simulation Practice and Theory 2. Kochendörfer, C., 1996, “Standardized Testing of Cooling Panels and Their Use in System Planning,” ASHRAE Transactions, vol. 102, part 1. Kramer, J., 2005, Personal communication on the manufacturing of radiant heating panels. Laouadi, A., 2004, “Development of a radiant heating and cooling model for building energy simulation software,” Building and Environment, vol. 39, Issue 4. LMNO Engineering, Research and Software Limited, 2003, “Gas Viscosity Calculator,” http://www.lmnoeng.com/Flow/GasViscosity.htm. Lomas, K.J., 1995, “The UK applicability study: an evaluation of thermal simulation programs for passive solar house design,” Building and Environment, vol. 31. Min, T.C., Schutrum, L.F., Parmalee, G.V., Vouris, J.D., 1956, “Natural Convection and Radiation in a Panel-Heated Room,” ASHAE Research Report No. 1576. Munson, B.R., Young, D.F., Okiishi, T.H., 2002, “Fundamentals of Fluid Mechanics, 4th edition,” John Wiley & Sons, Inc., Toronto. OMEGA, 2004, “The Temperature Handbook and Encyclopedia,” 5th edition, Omega Engineering, Inc. Schutrum, L.F., Vouris, J.D., 1954, “Effects of Room Size and Non-Uniformity of Panel Temperature on Panel Performance,” ASHVE Transactions, vol. 60. Shoemaker, R.W., 1954, “Radiant Heating,” 2nd edition, McGraw-Hill, New York Siegel, R., Howell, J.R., 1981, “Thermal Radiation Heat Transfer,” McGraw-Hill, New York.

93

Sigma Convectors, 2004, “Radiant heating Panel Product Information Catalogue,” Brampton. Simmonds, P., 1996, “Practical Applications of Radiant Heating and Cooling to Maintain Comfort Conditions,” ASHRAE Transactions, vol. 1996, part 2. STAR-CD, 2001, “STAR-CD Version 3.15 Methodology,” Computational Dynamics Limited. Steinman, M., Kalisperis, L.N., Summers, L.H., 1989, “The MRT-Correction Method: A New Method of Radiant Heat Exchange,” ASHRAE Transactions, vol. 95, Part 1. Tasker, C., Humphreys, C.M., Parmalee, G.V., Schutrum, L.F., 1952, “The A.S.H.V.E Environment Laboratory,” ASHVE Transactions, vol. 58. Walton, G.N., 1980, “A New Algorithm for Radiant Interchange in Room Loads Calculations,” ASHRAE Transactions, vol. 86, Part 2. Xia, Y., Mumma, S.A., 2005, “Ceiling Radiant Cooling Panels Employing HeatConducting Rails: Deriving the Governing Heat Transfer Equations,” ASHRAE Transactions, vol. 112, Part 1. Zhang, Z., Pate, M.B., 1986, “A numerical study of heat transfer in a hydronic radiant ceiling panel,” Numerical Methods in Heat Transfer, ASME HTD 62.

94

Appendix A ASHRAE Draft Standard 138P

95

permission must be obtained to reproduce all or any part of this document from the ASHRAE Manager of Standards. Inc. Refrigerating and Air-Conditioning Engineers. NE · Atlanta GA 30329 96 . Fax: 404-321-5478. Instructions and a form for commenting are provided with this draft. Ext. INC. Atlanta. E-mail: cramspeck@ashrae. 1791 Tullie Circle. Until final approval by the ASHRAE Board of Directors. NE. GA 30329-2305. Public review of this proposed revision has been authorized by a subcommittee of the Standards Committee. Phone: 404-636-8400.BSR/ASHRAE Standard 138P This draft standard will be submitted to the American National Standards Institute Board of Standards Review (BSR) for approval. This draft has been recommended for public review by the responsible project committee. Although reproduction of drafts during the public review period is encouraged to promote additional comment. REFRIGERATING AND AIR-CONDITIONING ENGINEERS. 502. AMERICAN SOCIETY OF HEATING. ASHRAE STANDARD BSR/ASHRAE Standard 138P Method of Testing for Rating Hydronic Radiant Ceiling Panels PUBLIC REVIEW DRAFT August 1999 ©1999 American Society of Heating.org. this draft revision is subject to modification. 1791 Tullie Circle.

4) must be signed and dated. but electronic submittal is preferred). choose one of the following methods: (1) (2) sign the comment forms electronically. If you disapprove. The Project Committee will inform commenters 97 . 1999. files uploaded to an ftp site. If method (2) is used. b) Comment forms may be submitted electronically either as files (MS Word 7 preferred) attached to email (UUENCODE preferred).org Ftp server address: ftp.org. uploaded to the ftp site. 2. For signature submittal. or on 3. 1998) 1. NE Atlanta. indicate whether approval is contingent upon considering them for inclusion (1) in the current proposal or (2) in future revisions of the current proposal. please specify whether you approve or disapprove of the proposed document. Highlighting pens should not be used since highlights will not reproduce. Commenters’ signatures are required on all electronic files sent on disk. NOTE: Submit public review comments to: Manager of Standards ASHRAE 1791 Tullie Circle. Comments submitted on the Form for Commenting provided and complying with the instructions provided on the Form and in 1(a)-(b) above will be forwarded to the Project Committee for consideration.review. logon to anonymous ftp in directory: public.ashrae.NOTICE COMMENTS ON DRAFT STANDARDS (Revised December 4. INFORMATION ON REVIEW AND DISPOSITION OF COMMENTS: All comments received by ASHRAE will be acknowledged. 2) must be received by October 4. 5) may be submitted electronically on the Form for Commenting attached to the standard (submittal in paper form is acceptable. uploaded to the ftp site.review. If you provide technical comments with your approval. GA 30329-2305 E-mail: public.comment@ashrae. mail paper versions to ASHRAE address or fax: 404-321-5478. d) Supplemental background documents to support your comments may be included. 3) must be substantiated. Please do not return marked-up copies of the draft. give your reasons. INSTRUCTIONS FOR FILING COMMENTS: a) Comments on this standard: 1) must comply with these and other instructions provided in the Form for Commenting. then the box beneath the signature line on the Form for Commenting must be checked.comment. or sent via e-mail. (Alternatively.5" floppy disk.) c) In your comments. or sent via e-mail to convey non-exclusive copyright. 6) must be sent to ASHRAE Manager of Standards. or send a copy of the comment form via mail or fax with the commenter’s signature and indicate whether the original comments are on disk.

and instrumentation for determination of heating and cooling capacities for hydronic ceiling panels in a specific indoor configuration and thermal environment. 2. Purpose This standard establishes uniform methods of laboratory testing for rating the thermal performance of hydronic radiant ceiling panels used for heating and/or cooling of indoor spaces.1 This standard specifies procedures. d) Provide specific wording changes or action that would resolve commenter's concerns. 1. electronic files in wordprocessed (MS Word 7 preferred) or scanned form are preferred. If supplementary documents are provided. b) Sign and date the non-exclusive copyright release. e) Provide a brief substantiation statement that presents the rationale. including E-mail or Internet address if available. Indicate whether attachments have been provided. Submittal in electronic form is preferred. Provide an abstract of lengthy substantiations. 2. and the temperature of the surrounding space.2 Panel performance is measured as a function of hydronic working fluid (water) temperature. Submittal in paper form is acceptable.) c) Identify the specific section that is the subject of the comment. The goal is to allow rating of panels for heat transferred to or from the space to be conditioned. Use a separate form for each comment. Scope 2. and supporting documentation as well as any technical data and back up. A paper version of the comment form is included in the public review draft standard. SUPPLEMENTAL INSTRUCTIONS TO COMMENTERS Commenters must submit comments to ASHRAE Headquarters on the form provided.3 This standard covers testing of hydronic ceiling panels in the following surface temperature ranges: Heating panels: room air temperature to 150oC (300oF) Cooling panels: below room air temperature 98 . 2. panel surface characteristics. Following are supplementary instructions for some of the numbered sections in the Form for Commenting: a) Provide all of commenter's contact information. apparatus.of the proposed disposition of their comments unless the Manager of Standards informs commenters that there will be another public review. 3. (See 1b.

N (0.4 This standard does not cover: a) electric heating panels. Values shall be based on the National Institute of Standards (NIST) values. kPa (psig). designated K. designated oC (degree Fahrenheit. which.2 lbm/s) and the unit of volumetric flow rate is cubic metres per second.325 kPa (29.2 kg/m3 ( 62. 99 . The unit of length is either the millimetre.768 ft2). oF) or the kelvin. 3. in) or the metre.5 Force.6 Pressure. designated mm (inch. The unit of mass flow rate is kilograms per second. The unit of density is kilogram per cubic meter designated as kg/m3 (lbm/ft3).2248 lbf). The density of water at standard conditions. W (3. or the kilopascal. lb) and the unit of time is the second. or field testing. ft). b) heating panels that are part of the ceiling or floor structure. designated s. in turn. 3.2 Basic Units.7 Energy and Work. The unit of force is the newton. The unit of energy or work is the joule. production. 3. 3. 3. 3.1 System of Units The International System of Units (SI) has been employed in the text with inch-pound (I-P) units shown parenthetically.8 Fluid Density.3 Flow Rate. The unit of mass is the kilogram. The unit of power is the watt.2. or c) test methods for design. The unit of temperature is the degree Celsius. m3/s (35. designated m (foot. kg/s (2. The unit of pressure is the pascal.412 Btu/hr).314 ft3/s).48x10-4 Btu). m2 (10. J (9. Units of Measurement 3.3 lbm/ft3). designated kg (pound. are based on the fundamental values of the International Bureau of Weights and Measures.92 in Hg) and 20oC (68oF) is 998. 101. The unit of area is square meters.4 Area. 3. 3. Pa.

179 kJ/kg·K at 60oC ( 0.02 x 10-6 (0. Temperature 20oC (68oF) 40oC (104oF) 60oC (140oF) 80oC (176oF) 100oC(212oF) Dynamic Viscosity N s/m2 (lbm/ft hr) 1.0254) 4.47) 6. The two are related through the density of the fluid.0395) 6. For water the specific heat.67 x 10-7 (0.9 Viscosity.02 x 10-3 (2.51 x 10-4 (1. The viscosity of water decreases with increasing temperature as shown below.0140) 2. 100 .56 x 10-7 (0. 3.60 x 10-7 (0.50 x 10-4 (0.11) 3.325 kPa (29.68) Kinematic Viscosity m2/s (ft2/hr) 1.3.0114) 3. Cp.11 Reynolds Number Re. is 4.85) 2.10 Specific Heat.0181) 3.998 Btu/lboF at 140oF).60 x 10-4 (1.82 x 10-4 (0. Barometric pressure is the atmospheric pressure measured at the location of the test room and used to correct panel ratings to standard conditions of 101.92 in Hg)(1 standard atmosphere).12 Barometric Pressure.94 x 10-7 (0.57) 4. 3. The units of dynamic viscosity are N s/m2 (lbm/ft hr) and the units of kinematic viscosity are m2/s (ft2/hr). Reynolds Number is a dimensionless quantity used to indicate whether a flowing fluid is laminar or turbulent.

effective panel surface temperature: The area weighted average surface temperature of the panel under test.4 Symbols Used SYMBOL Ap C Cp m _ n pb ∆h q Re ta te ti tg tm to tp AUST ∆t v µ v B B B B B B B B B B B B B B B B B B B B B B DESCRIPTION effective panel surface area panel performance coefficient specific heat mass flow rate panel performance exponent (from q=C∆tn) barometric pressure energy loss in the hydronic circuit heat transfer rate per unit area Reynolds Number air temperature enclosure exterior wall temperature panel inlet water temperature globe temperature panel mean water temperature panel outlet water temperature panel effective surface temperature area weighted average enclosure surface temperature temperature difference air velocity dynamic viscosity kinematic viscosity P P UNIT m2 (ft2) W/m2·tn (Btu/ft2·tn ) kJ/kg·K (Btu/lbmoF) kg/s (lbm/s) dimensionless kPa (in Hg) m H2O (psig) W/m2 (Btu/hr ft2) dimensionless o C (oF) o C (oF) o C (oF) o C (oF) o C (oF) o C (oF) o C (oF) o C (oF) o C (oF) m/s (ft/s) N·s/m2 (lbm/ft hr) m2/s (ft2/hr) P P P P P P P P P P P P B PB P B B B B P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P B B P P P P 5 Definitions air temperature: Temperature of the air in the test enclosure obtained using a sensor shielded to minimize radiation effects between the sensor and the enclosure (see section 7. 101 . cooling panel: A panel specifically designed for conditioning of a space through energy transfer to the panel from the occupants or space.1.2).

or black body at the same temperature. Instruments and Methods of Measurement 6. The specified requirements correspond to two standard deviations about the mean and are based on an assumed normal distribution of the errors involved.1. mean water temperature: The arithmetic average between the inlet and outlet water temperature. hemispherical emittance: A property of a material that governs the emission of energy by radiation relative to that emitted by a perfect emitter.enclosure surface temperature (AUST): The area weighted average surface temperature of the enclosure or test chamber in which the panel to be tested has been installed. Random errors can be determined only from an adequate statistical sample. outlet water temperature: The temperature of the water at the outlet of the panel under test. inlet water temperature: The temperature of the water at the inlet to the panel under test. The specifications for instruments and methods of measurement that follow include accuracy and precision requirements.1 Accuracy and Precision. The temperature obtained is a result of thermal equilibrium between energy gained or lost by radiation and convection (see section 7. globe temperature: The equilibrium temperature obtained inside a uniformly painted 150 mm (6 in) diameter black globe. 6. heating panel: A panel specifically designed for conditioning of a space through energy transfer from the panel to the occupants or space.3). 102 . head loss: The measured energy loss in the flowing fluid across a panel under test at the specified mass flow rate.

8-19892. The resolution of an instrument is the smallest change in input that will result in a measurable change in output. Two sets of values have been shown for water temperature as it is possible to make the measurements using individual sensors or sensors connected to read differentially.4.2oC (0.3 Mass Flow Meter Calibration.3 Temperature Measurements 6. In no case shall the resolution of the measurement device be greater than the accuracy specified for the instrument.1oC (0.2oC (0.2oF) 0.4.3.4oF) 0.05oC (0.2oC (0.1 m/s (0.2 Accuracy and Precision.1-1986(RA 91)1.1 m/s (0.4. 6.1oC (0. 6.05%. including their associated readout devices. For calibration of the mass flow measurement devices. 103 . The accuracy of the liquid flow rate measurement shall be 0.3.1 Standard Practice. shall be within the following limits.2oF) 0. 6.1oF) 0.3 ft/s) Standard 41.4oF) 0.05oC (0.4 Flow Rate Measurements 6.1% of the measured quantity in mass units per unit time. time and mass measurements shall be made with an accuracy of 0.4oF) 0.1oF) 0. 6.4oF) 0.2 Accuracy. 6.Instrument errors shall be such that two standard deviations of the accumulated data from the mean (precision) do not exceed the specified values. Flow rate measurements shall be made in accordance with ASHRAE Precision 0.1 Standard Practice.2 Resolution.3 ft/s) 0. Temperature measurements shall be made in accordance with ASHRAE Standard 41. Accuracy Water Temperature (individual sensors) Water Temperature (differential sensors) Air Temperature Surface Temperature Air Velocity 6. The accuracy and precision of the instruments.2oC (0.

7. 7.) by 3.1.) by 3. The temperature of each surface of the enclosure shall be 104 .1 Test Chamber. All interior surfaces shall have a smooth. The uniformity of the interior surface temperature will be determined by the heating or cooling method used to maintain the temperatures as well as the control system.9 or higher and shall be temperaturecontrolled. designated the exterior wall.4. This condition shall be deemed to have been met if the dew point of the air in the chamber is more than 2oC (3.1 AUST Temperature Uniformity. As a minimum.) high. Tests shall be performed in a chamber with inner dimensions of not less than 3. flat-painted finish with a surface emittance of 0. The chamber shall be constructed so as to minimize infiltration or exfiltration.6. The control of interior surface temperature shall be such that one wall of the enclosure. the chamber shall be tested for air tightness in accordance with ANSI/ASHRAE 119-1988 (RA 94)4 and shall have a leakage class of A or B. For cooling tests the humidity shall be controlled so as to prevent condensation on a test panel or any of the enclosure surfaces.05 m (10 ft. Prior to being used to test panels. Energy transferred to or from a panel under test may be directly measured by thermal energy measurement techniques if conducted using a meter calibrated in accordance with ANSI/ASHRAE Standard 125-19923. The surface shall be divided into quadrants and a temperature sensor located at the geometric center of each quadrant. Equipment and Setup 7. can be maintained at a temperature different from the remaining walls.66m (12ft.4 Thermal Energy Measurements.6oF) below the panel inlet water temperature.66m (12 ft. the temperature of each surface of the enclosure shall be measured at four locations using a temperature sensor specifically designed for surface measurements.

) diameter black globe thermometer. Globe temperature shall be measured at a location as near as practical to the air temperature measurement location. The head loss across the panel(s) under test shall be measured as the pressure difference between the panel inlet and panel outlet.1.4 Globe Temperature.1. 7.1. 7.2 Test Panel Installation. Air velocity in the vicinity of the panel(s) under test shall be measured using a temperature compensated hot wire or hot film anemometer.5 ft) above the floor of the enclosure.1.8°F). The sensor shall be placed at the geometric center of the floor. An alternate method. 7. The test panel effective surface temperature shall be determined as the arithmetic average of surface temperature measurements from at least six locations on the panel. Globe temperature in the test chamber shall be measured using a 150 mm (6 in.75 m (2.1. Air temperature in the chamber shall be measured using a shielded temperature sensor located at a height of 0. designated Installation Method B.) of the face of the panel under test. 7. 105 . has been included in this standard for cases where a standard installation is inappropriate. The measurements should be made as close as practical to the inlet and outlet water temperature measurement locations.controlled so that the variation in average surface temperature between surfaces (other than the exterior wall) is less than 1°C (1. The measurement shall be made within 300 mm (12 in.2 Panel Effective Surface Temperature. Installation Method A shall be used if it does not conflict with the manufacturer's design application for the panel under test.5 Air Velocity. A specific installation is detailed below and referred to as Installation Method A. 7. 7.3 Air Temperature.6 Head Loss.

In all installations the test panel supply and return water lines shall be insulated when within the chamber. the individual panels shall be plumbed in series. Complete details shall include both a written 106 .0 ft.030 W/mK (0.2.44 m. If Method B is used.04 W/mK (0. A standard T-grid acoustical ceiling system shall be installed such that the lower surface of the ceiling is 2. All other spaces in the T-grid shall be filled with acoustical tiles.0 ft) above the floor surface of the enclosure. The edge of the panel(s) under test shall be placed within 25 mm (1 in.2. An example might be a cooling panel intended to hang from a ceiling so that air is free to circulate around the panel or a panel (heating or cooling) that is intended to be installed in a vertical position. Inlet and outlet water temperatures shall be measured within 300 mm (12 in) of the connection to the test panel(s). 7.) of the exterior wall of the chamber.1 Installation Method A.).2 Installation Method B. 7. The insulation shall have a density of 13 kg/m3 (0.75 lbm/ft3) and a thermal conductivity of 0. In no case shall the total tubing length of the test panel(s) exceed 61 m (200 ft). Method B allows the test panel to be installed in configurations different from that specified in Method A.44 m (8. complete details of the installation must be included with the test results. The insulation shall be a minimum of 6 mm (0. The panel(s) to be tested shall be installed such that the long axis is parallel to the exterior wall of the enclosure. The back of the test panel(s) shall be insulated using 25 mm (1 in) glass fibre insulation without a foil backing. In the event that the test panel consists of more than one individual panel. 8. The test panel(s) shall be horizontal and shall be placed such that the lower surface is flush with the bottom surface of the acoustical tiles (2.25 in) thick and have a thermal conductivity of 0.023 Btu/h·ft·°F).017 Btu/h·ft·°F) or lower.

five sequential readings of all required measurements. each spaced in time by five minutes. model. 8. As a minimum. In the case of a heating panel the area of a panel to be tested shall not be less than 4.1 General Test Requirements.5oC (55oF). 107 . trial observations shall be made until steady readings are obtained. Equilibrium conditions shall be established before each determination. The description of the test panel shall be recorded. (See Section 8. For heating panels the test shall include a minimum of five panel mean water temperatures equally spaced over the operating range specified by the manufacturer.1. In all cases the panel location relative to the enclosure surfaces shall be described as is required for Method A. 12.3) 8. To test for equilibrium.1. 15oC (60oF) and 18oC (65oF). The number of determinations required to establish the performance of a panel depends on the purpose of the panel.1 Test Panel. tube sizing and placement as well as surface finish. heating or cooling.and a graphical description of the installation.3 m2 (100 ft2).2 Data to be Recorded. 8. physical dimensions.65 m2 (50 ft2). The description shall include panel manufacturer.1 Test Points. For cooling panels the area of test panels combined shall be not less than 9. For cooling panels the test shall include four panel mean water temperatures. Method of Conducting a Test 8.2. 10oC (50oF).2 Equilibrium. 8. shall be at the required value and remain within test tolerances. 8.

kg/s ( lbm/s) _ d = inside diameter of the panel tubes. exterior wall temperature. The instruments and apparatus used in the rating of a panel shall be recorded. the following readings will be made at each mean water temperature: Inlet water temperature. Readings of all relevant data shall be made simultaneously wherever possible. fluid mass B B B B flow rate.3 Instruments. _ B B B B AUST. globe temperature. serial numbers and calibration information shall be recorded. 8.1. Flow rate through the panel(s) under test shall be adjusted to achieve a Reynolds Number greater than 10000 to ensure turbulent flow within the tubes carrying water. ta. tg. as defined in section 8. The description of the test setup including panel placement shall be recorded. Data shall be recorded at a minimum of 5 minute intervals until equilibrium. model numbers. m (ft) µ = dynamic viscosity of water. 8. enclosure surface temperature. Reynolds number can be calculated using equation 1.2. m. This is particularly important in the event that Method B has been used in the rating of a panel. Test data for each determination shall be recorded. is reached. to.2. head loss across the panel. N·s/m2 (lbm/ft hr) B B P P B B Regardless of whether Method A or Method B is used.4 Test Data. outlet water temperature.2 Test Setup. Re = & 4m πdµ where m = mass flow rate of water.2. ∆h. te. It is not necessary to include this information in the test report but it must be kept on file in the event of a question with respect to the overall accuracy of the test results.2.8. below. Local barometric pressure shall be measured at the test location or obtained from a local weather station. enclosure air temperature. and air velocity in the B B vicinity of the panel. ti. 108 . Names.

9. 15oC (60oF) and 18oC (65oF).1 Panel Output.8.3.5oC (85oF ±0. For cooling panels the test shall include four panel mean water temperatures.9oF).5oC (75oF ±0.1 Heating Panels. 8.9oF).3 Test Conditions 8.9oF). Heating Panels 109 . Air temperature in the chamber shall remain between 21oC and 24oC (70oF to 75oF) during steady state operation and humidity in the chamber shall be maintained so that the dew point of the air in the chamber is at least 2oC (3. 12. All other test chamber surfaces shall be maintained at 23oC ±0.5oC (50oF ±0. The surface temperature of the exterior wall of the chamber shall be 30oC ±0.5oC (68oF ±0.3. 10oC (50oF). All other test chamber surfaces shall be maintained at 20oC ±0. Calculation of Results 9. The surface temperature of the exterior wall of the chamber shall be 10oC ±0.5oC (55oF). Air temperature in the chamber shall remain between 18oC and 21oC (65oF to 70oF) during steady state operation. The output of the panel shall be calculated for all test conditions listed.9oF). The mean water temperatures shall include the highest and lowest operating temperatures specified by the manufacturer. Tests shall be conducted at a minimum of five mean panel water temperatures equally spaced over the manufacturer’s specified operating range.6oF) below the panel inlet water temperature. The output shall be calculated using the measured data and either equation 2 (heating panels) or equation 3 (cooling panels) and shall be normalized with the area of the panel tested.2 Cooling Panels.

The characteristic performance relates the panel output to 110 . The mean panel water temperature shall be calculated at each test condition as the arithmetic average of the panel inlet and panel outlet temperature as indicated in equation 5.2 Average Chamber Surface Temperature. The average chamber surface temperature shall be calculated as the area weighted average temperature of the six surfaces. Appendix A. which make up the chamber as indicated in equation 4. AUST. to determine the coefficients C and n shown in equation 6 below.4 Characteristic Performance. excluding the panel under testing. A characteristic performance equation for the panel tested shall be determined using the method of least squares. tm = ti + to 2 9.ti ) Ap Cooling Panels 9.q= & m C p ( ti .3 Mean Water Temperature.to ) Ap q= & m C p ( to . AUST = t 1 A1 + t 2 A2 + t 2 A3 + t 4 A4 + t 5 A5 + t 6 A6 A1 + A2 + A3 + A4 + A5 + A6 9.

q = C∆ t n where ∆t = tm .2 Characteristic Performance. q.AUST (for heating panels) B B ∆t = AUST . Testing Facility Name of Testing Organization Location of Facility Person Conducting Test Panel Information Manufacturer Model Information (trade name.1 Report. The report shall also identify the name of the person who conducted the test and the laboratory at which the test was performed. ∆T.) 111 . etc. shown on the ordinate axis. model number. The characteristic performance of a panel shall be identified on a plot with the test temperature difference. Test points shall be identified on the graph using a series of circled points. 10.the temperature difference between the panel surface and the average surface temperature of the enclosure. Report and Results of a Test 10. uncertainty in results. shown as the abscissa and panel output. The list that follows will serve as a check list for the preparation of performance reports. make. The report of a laboratory test of a radiant heating or cooling panel shall include the object. results. test data. descriptions of the panel tested and any deviations from the standard.tm (for cooling panels) B B 10.

ANSI/ASHRAE 119-1988 (RA 94). ANSI/ASHRAE Standard 125-1992. American Society of Heating. “Air Leakage Performance for Detached Single-Family Residential Buildings”. “Standard Method for Temperature Measurement.Panel Description Dimensions (external. Inc. matte. Atlanta. Georgia. American Society of Heating. Refrigerating and AirConditioning Engineers. References 1. ASHRAE Standard 41. Refrigerating and Air-Conditioning Engineers. Atlanta. surface properties) Test Conditions Exact panel installation Detailed description and figures if Method B used Raw test results (temperatures. Inc. Inc. Georgia. Atlanta. 4. gloss. tube size) panel material panel surface finish (paint. Georgia. “Method of Testing Thermal Energy Meters for Liquid Streams in HVAC Systems”. Atlanta. Inc. drop etc. 1791 Tullie Circle NE. Refrigerating and Air-Conditioning Engineers.) Test Results Panel output at each test condition Uncertainty in panel output at each condition Characteristic performance (coefficients of fitted equation) Plot of derived characteristic performance and actual test points 11. pressure. Georgia. American Society of Heating. 1791 Tullie Circle NE.8-1989. flow rates. 3. Refrigerating and Air-Conditioning Engineers. 2. 1791 Tullie Circle NE. 112 . “Standard Methods of Measurement of Flow of Liquids in Pipes Using Orifice Flowmeters”. ASHRAE Standard 41. 1791 Tullie Circle NE. American Society of Heating.1-1986(RA 91).

(This appendix is part of this standard and is included as a normative appendix.) Appendix A Curve Fitting by Least Squares The following method of least squares shall be used to determine the relationship between panel temperature difference and energy output as defined in Section 9.4.i ) 2 i=1 n (A-3) In this case the residual to be minimized is 2 2 ε r = ∑ ( yi . Since the characteristic equation is a power function it must be linearized by taking the log of each term in the equation.y fitted.(A + Bxi ) ) i=1 p (A-4) where yi = log(q) xi = log(∆t) A = log(C) B=n p = number of data points B B B B to do this we must generate the appropriate sums 113 . Thus the equation q = C∆ t n (A-1) is linearized as log( q ) = log( C ) + n log( ∆t ) (A-2) and the standard method of minimizing the residual 2 ε r ≡ ∑ ( yi .

Σy = Σlog(q) Σx = Σlog(∆t) Σx2 = Σ(log(q))2 P P P P and Σxy = Σlog(q)log(∆t) which are then used to solve for the coefficients A and B in the linear fit. C = exp(A). A= ∑ yi -B ∑ xi p (A-5) B= p ∑ xiyi . Since the original power function was linearized by taking the log of each term the coefficient C must be determined by taking the exponential of A as indicated in (A-7). (A-7) 114 .∑ xi ∑ yi p ∑ x2-( ∑ xi )2 (A-6) In this case p is the number of data points used in the fit.

the calculation of panel output on the basis of the measured quantities m and ∆t. Starting with equation A1. & q = m C p ∆t (B-1) it can be seen that the individual uncertainties in mass flow rate and temperature difference will influence the result. εq = [ 2 ∂q ∂q ∂q 2 2 2 εm ] +[ εCp ] + [ ε ∆t ] & & ∂m ∂C p ∂ ∆t (B-2) Taking the derivative of equation B-1 and squaring 2 2 & & ε q = C p ∆t 2 ε m + m2 ∆t 2 ε C + m2 C p ε ∆t & 2 2 2 2 P (B-3) Dividing the equation by 115 .(This appendix is not part of this standard but is included for information purposes only.) Appendix B Uncertainty Analysis of Panel Output The rating of a heating or cooling panel depends on a number of measurements that each contain systematic and random errors. This appendix gives an example of how an uncertainty analysis on panel output could be carried out. Since the equation is purely multiplicative. the following method can be used to estimate the uncertainty in the derived result. Without careful calibrations these errors can result in very large uncertainties associated with calculated panel output.

equation B-5 reduces to εq ⎛ε & ⎞ ⎛ε ⎞ = ⎜ m ⎟ + ⎜ ∆t ⎟ & q ⎝ m ⎠ ⎝ ∆t ⎠ 2 2 (B-6) 116 .2 q = m2 C p ∆t & 2 2 (B-4) gives the fractional error in panel output as a function of the fractional error in the remaining quantities as shown in equation B-5. 2 2 ⎛ ε q ⎞ ⎛ ε m ⎞ ⎛ ε C p ⎞ ⎛ ε ∆t ⎞ & ⎟ +⎜ ⎟ ⎜ ⎟ = ⎜ ⎟ +⎜ ⎜q⎟ ⎜ ⎟ & ⎝ ⎠ ⎝ m ⎠ ⎝ C p ⎠ ⎝ ∆t ⎠ 2 2 (B-5) Given that it can be assumed that there is no uncertainty in the specific heat.

) Appendix C Using Test Results to Estimate Panel Capacities at Other Than Test Conditions When design conditions are different than the standard test conditions.87 qp (C-1) In this equation. corrects for mean temperature difference between the panel and surroundings at conditions other than standard.02 and the positive sign (+) is used. P P 1. Cs is the performance constant at design conditions and C is the performance B B constant derived from the test results at standard conditions using equation 6.(This appendix is not part of this standard but is included for information purposes only. α. Radiative Output Adjustment. radiative and convective components of the total output need to be separately adjusted. [ta-AUST] is the difference between the air temperature and the B B 117 . The ratio qr/qp is B B B B the average ratio of radiant energy transfer to total energy transfer determined in the performance test and are defined in Chapter 6 of ASHRAE Handbook. α= 1 ± k[ t a . The first adjustment. the actual performance of the panel changes. Under these circumstances. The first term in equation C-1 represents the adjustment for the radiative heat transfer component of the panel output. The standard test result may be adjusted to design conditions using the following relationships. 19961.1 and the negative(-) sign is used and for ceiling heating k is 0.AUST ] s 1 ± k[ t a . HVAC Systems and Equipment. C s = C[( α qr qr α F c ) + abc(1 )] q p 0.AUST] (C-2) For ceiling cooling k is 0.

AUST, maintained and recorded during the standard test. [ta-AUST]s is the anticipated
B B B B

temperature difference at a 50% design load condition at the location where the panels are to be installed. The second adjustment is made to account for differences in radiation angle factor, Fc,
B B

Fc =

1 1/ F p - r + [(1/ ε p ) - 1] + A p / Ar [(1/ ε r ) - 1]

(C-3)

(ASHRAE Handbook, HVAC Systems and Equipment, 1996, Chapter 61).
P P

When Fc differs from the value of 0.87 derived for the test chamber, a correction involving three
B B

factors is required. i) If the panels are to be installed in a different geometry with respect to the walls and floor. Fp-r is the radiation angle factor. Fp-r approaches unity if the panels are flat, horizontal
B B B B

and flush with the finished ceiling. For other selected configurations, Fp-r may be
B B

calculated from related figures and tables in the ASHRAE Handbook of Fundamentals, Chapter 3. 19972.
P P

ii)

If the thermal emittance of various surfaces deviate from 0.9. εp and εr are the thermal
B B B B

emittance of the panel and inside room surfaces respectively. They may be obtained from Table 3, 2ASHRAE Handbook, Fundamentals, 1997, Chapter 36.
P P

iii)

If the term Ap/Ar · [(1/εr)-1] is not close to zero. Ap and Ar are the surface area of the
B B B B B B B B B B

panel agglomerate, and the total surface of the unconditioned surfaces within the conditioned space respectively. Using actual design data for Ap, Ar and εr, this term may
B B B B B B

be adjusted in equation C-3 through Fc.
B B

118

2. Convective Output Adjustment. The second term in equation C-1 represents an adjustment for the convective heat transfer component from the panel surface. It consists of three separate effects: i) Size effect - the equivalent diameter, Des, of the panel agglomerate at the intended installation. De is the equivalent diameter of the panel under test. The equivalent diameter is four times the panel (or panel agglomerate) area divided by the length of the perimeter. Pressure (altitude) effect - barometric pressure at the intended installation location, ps. pb is the barometric pressure at test conditions. Air Velocity effect - the anticipated air velocity, vs, in the vicinity of the panel surfaces.

ii) iii)

Table C-1 shows the adjustment coefficients for each term in the correction equation. Table C-1 Convective Panel Output Adjustments for Conditions Differing From Standard Coefficient System of Units SI a b1 c (De/Des)0.08 (ps/pb)n (vs/0.15)0.8 IP (De/Des)0.08 (ps/pb)n (vs/0.5)0.8

Notes with respect to Table C-1: Des = [4 area/perimeter] of panel agglomerate, m (ft) ps = barometric pressure at actual location, kPa (in Hg) pb = barometric pressure at test conditions, kPa (in Hg) vs = anticipated air velocity at design conditions, m/s (ft/s) n = 0.62 for ceiling panel cooling and 0.50 for ceiling panel heating

119

1) Correction for altitude above sea level may also be made using the following relationships. b = {(1 - 2.22 x 10-5 hs)/(1 - 2.22 x 10-5 h)}n h, hs in meters, b = {(1 - 0.67 x 10-5 hs)/(1 - 0.67 x 10-5 h)}n h, hs in feet. where hs is the altitude of the actual location and h is the altitude of the test location.
References

1. ASHRAE Handbook, HVAC Systems and Equipment, 1996,American Society of Heating, Refrigerating and Air Conditioning Engineers, 1791 Tullie Circle NE, Atlanta, Georgia. 2. ASHRAE Handbook, Fundamentals, 1997, American Society of Heating, Refrigerating and Air Conditioning Engineers, 1791 Tullie Circle NE, Atlanta, Georgia.

120

Appendix B Radiant Panel Test Booth Calculations

121

z y x Figure B.B. The calculations were done in MathCad 11 and are shown below. This was done to determine the approximate the radiant incoming heat flux to each surface and determine how much energy must be removed by the circulating water on the exterior of the wall.1 View Factor Calculations This section summarizes the view factor calculations for the largest sized panel that would be tested at the Sigma Radiant Panel Test booth.1) A diagram of the radiant panel test booth is shown below.1) Fij = 1 Ai Ai A j ∫∫ cos θ i cos θ j πR 2 dAi dA j (B. The first principle view factor equation is given below (Equation B.1: Co-ordinate system to determine the radiant panel to surface view factors Surface i is the radiant panel and surface j is other surface. The co-ordinate system used to calculate the view factors is shown on the diagram. For example consider surface j to be the floor. 122 .

9624 – panellength.2 and B.3 and B.2.3 into Equation B. R.048 B B B B To describe the location of the floor x-co-ordinates: y-co-ordinates: z-co-ordinates: xj1 = 0. zj2 = 0 B B B B The radial distance.2) The polar angles can be expressed in terms of the x.1.4 show the view factor calculations 123 .9624 B B B B zi1 = 3.3) Substituting Equations B. yi2 = 3.9624 B B B B yj1 = 0. Expressing the polar angles B B B B in terms of the x. zi2 = 3. the view factor can be solved. yj2 = 3.y and z coordinates: cos θ = zi R (B.048. xj2 = 3.9624 B B B B zj1 = 0. xi2 = 3.To describe the location of the radiant panel x-co-ordinates: y-co-ordinates: z-co-ordinates: xi1 = 0. is the distance between any two points from surface i to j. The surface normals are the same for the floor and radiant panel. The radial distance can be calculated with the following equation: R 2 = ( xi − x j ) 2 + ( y i − y j ) 2 + ( z i − z j ) 2 (B. Figures B. A polar angle is defined as the angle it makes with surface normal (Siegel and Howell.9624 B B B B yi1 = 3. 1981). B. so θi = θj.y and z co-ordinates.

Figure B.2: MathCad Worksheet to Calculate the View Factors 124 .

Figure B.3: MathCad Worksheet to Calculate the View Factors 125 .

337 0.1.Figure B.1: Summary of View Factors View Factor 0.169 The published view factor equation by Siegel and Howell (1981) is shown in Figure B.169 0.4: MathCad Worksheet to Calculate the View Factors The view factors calculated can be summarized in Table B.067 0.5 on the next page and verifies the view factor equation. 126 .257 0.Floor Panel – Front Wall Panel – Left Wall Panel – Right Wall Table B. Surfaces (i – j) Panel – Back Wall Panel .

5: MathCad Worksheet using a Published View Factor Equation to Validate View Factor Calculations 127 .Figure B.

a coefficient of 6 W/m2K P P was used with the following equation (B.4) where: F is the calculated view factor from above. The calculations are summarized in Figure B.2 Energy Balance on Each Surface With knowledge of the view factors. except for the cold wall which was at 283 K.5) The panel temperature was assumed to be 363K. " q c = h(Ta − Tsur ) (B.5). 1999). P P P P The radiative heat transfer was calculated with the following equation (B.4) " 4 4 q rad = Fεσ (T panel − Tsur ) (B.B. Assuming the worst case from published values (Awbi and Hatton. The total incoming heat flux can be estimated by assuming the convective heat transfer coefficient is 6 W/m2K and the air temperature inside the chamber is 22oC. the radiant heat flux incoming to each wall can be calculated. The convective heat transfer can be calculated knowing a heat transfer coefficient. while the surrounding surfaces were at 293 K.6 on the following page: 128 .

6: MathCad Worksheet to Calculate the Incoming Heat Flux to Each Wall 129 .Figure B.

7 and Figure B. since each pipe-run will be 4m long.B. Heat conduction rail effects have been ignored as well as any bond conductance.7: MathCad Worksheet to Calculate the Tube Spacing 130 . Figure B. Figure B.3 Pipe Spacing The derivation for the pipe spacing is shown below.8 show the calculation. Solving for the pipe length required the number of pipe-runs can be calculated. It is derived from the Fourier’s law and assuming a 1/3 of the pipe is in contact with the wall. From the number of piperuns the tube spacing can be calculated dividing the length of the chamber of the total number of pipe-runs.

8: MathCad Worksheet to Calculate the Tube Spacing 131 .Figure B.

B. Equations B. (2002).8) 132 . hL = f l V2 Di 2 g V2 2g (B.9).6) hL = K L (B. knowing the Reynolds number and assuming a smooth surface roughness condition for the PEX tubing.6 and B.7 are the head loss equations used to determine the pipe circuit pressure drop.4 Pressure Drop Calculations The pressure drop from each circuit can be estimated from the head loss equation for pipe friction and the loss coefficient for tube u-bends given by Munson et al. The Reynolds is given by: Re = ρVD µ (B.7) The friction factor for the tubing can be determined from the Moody chart (Figure B.

9: Moody Chart 133 .Figure B.

11) Figure B.10.Specifying the desired flow rate to obtain a minimal pressure drop. Figure B. the pressure drop for each circuit is summarized in the work sheets below (Figure B.10: MathCad Worksheet to Calculate the Pressure Drop for Each Cicuit 134 .

11: MathCad Worksheet to Calculate the Pressure Drop for Each Circuit B.5 Circuit Balancing Valve The Circuit Balacing Valve Conversion chart to determine the flow rate from a pressure differential reading is shown in Figure B.Figure B. 135 .12.

12: Circuit Balancing Valve Conversion Chart 136 .Figure B.

Appendix C Experimental Test Data 137 .

C. inlet and outlet panel water temperatures. air temperature and flow rate were recorded. Using Equation C. panel temperature. The surface temperatures.1) The results are in imperial units because Sigma Convectors requested that all the experimental data should be this way. Each panel was tested at five different mean water temperatures. & q out = mC p (T panel − Ta ) (C. 138 .1 Experimental Data The following tables are the experimental results for the 24-4 and 24-8 panels tested.1 the heat output for each panel was calculated.

66 73.63 139.38 71.90 190.72 68.12 70.87 64.81 126.05 69.10 65.49 61.42 137.02 55.33 100.90 140.90 16.38 135.31 62.90 353.68 249.00 134.03 78.23 74.15 55.41 62.02 70.87 63.00 69.17 68.60 69.16 302.44 69.77 70.61 69.35 122.38 0.93 69.88 84.31 83.50 252.00 82.44 121.49 63.81 127.61 54.07 54.89 92.71 72.90 Tair (F) 60.72 124.90 3.40 117.25 0.52 57.24 Tcold (F) 61.06 73.09 87.40 75.75 111.14 262.32 381.88 0.15 9.22 511.95 73.96 71.21 69.83 69.24 68.50 70.36 129.15 274.06 109.50 77.90 16.41 68.59 62.65 66.13 0.67 84.22 62.23 75.96 134.00 0.99 68.83 55.06 123.43 116.63 0.83 -513.28 69.32 54.88 74.70 72.15 69.69 75.55 70.66 73.99 70.60 69.54 72.90 289.45 66.29 54.80 62.96 65.31 80.90 151.15 233.56 70.13 133.50 0.90 126.90 264.57 69.00 0. MWT = 49oC Apanel (ft ) 2 24.50 0.08 66.82 Heat Flux Flow Rate Heat Flux 2 MWT (F) (BTU/hrft) (GPM) (BTU/hrft 0.03 5.16 66.21 281.97 74.11 131.78 79.46 71.75 62.08 56.97 139 .75 0.59 78.58 70.10 70.60 68.39 302.50 6.39 67.94 56.64 68.25 0.19 67.85 68.63 71.20 70.01 128.53 69.22 69.88 0.49 68.05 67.70 69.83 73.50 70.90 8.18 68.03 85.56 121.58 579.00 125.96 67.90 99.88 79.52 70.94 54.87 68.90 131.67 72.49 58.81 95.63 0.10 80.04 88.13 65.22 68.61 71.32 71.35 68.86 68.64 121.23 56.14 123.07 71.90 151.90 255.36 56.84 2.13 69.30 90.25 0.58 74.46 67.90 3.06 706.27 72.87 69.56 63.00 Tfloor (F) 64.73 66.13 131.90 124.87 282.65 68.90 141.61 69.70 70.04 124.28 118.44 71.78 62.96 66.00 72.23 Tpanel (F) Twater in (F) Twater out (F) 60.12 69.90 137.30 55.17 67.26 91.92 68.66 69.74 61.51 69.43 7.97 68.00 0.25 0.82 78.73 64.46 32.85 70.62 65.45 68.00 0.64 104.18 69.69 75.21 69.06 68.38 61.33 62.19 68.33 67.14 64.53 98.90 297.90 1.Experimental Data for 24-4 Panel.07 75.05 71.23 72.05 67.95 70.48 69.74 72.71 72.19 85.28 61.06 56.63 63.05 61.08 66.56 72.02 71.63 128.91 68.76 54.44 199.90 116.69 67.32 67.90 138.90 2.13 130.49 57.90 151.38 0.42 67.01 56.65 62.74 68.87 64.48 71.38 0.85 67.13 0.43 91.40 594.32 87.75 0.84 119.61 68.62 63.92 68.19 138.63 69.42 61.70 68.98 74.59 276.88 124.69 73.99 118.39 54.83 66.46 126.49 528.63 127.87 55.90 4.38 68.99 66.75 0.53 62.63 0.29 118.47 118.56 88.21 69.27 302.55 69.64 126.13 0.10 65.79 113.47 55.75 Trightwall (F) Tleftwall (F) Tfrontwall (F) 61.91 68.48 119.94 64.86 130.02 71.25 0.90 -256.30 65.44 92.51 92.89 69.67 99.32 72.75 61.04 69.

31 68.13 0.64 123.32 130.77 121.88 69.38 68.36 68.90 96.39 54.22 68.28 69.90 84.94 69.90 113.53 123.80 68.78 69.99 119.19 0.30 54.39 54.78 69.91 67.11 69.65 126.Trightwall (F) Tleftwall (F) Tfrontwall (F) Tfloor (F) Tcold (F) Tpanel (F) Twater in (F) Twater out (F) 67.88 239.35 54.15 116.63 69.01 68.90 109.71 117.42 68.23 68.67 69.25 124.49 70.90 113.50 0.79 140.26 117.83 68.07 54.90 106.58 68.23 141.66 69.70 68.32 54.88 229.63 -145.88 0.55 69.49 70.12 125.06 0.04 68.12 68.89 69.93 132.69 68.96 119.35 68.50 68.34 118.06 0.63 69.32 129.90 232.77 126.13 140 .23 54.90 116.75 0.72 69.30 69.81 68.40 68.90 113.15 54.75 0.38 214.58 131.22 68.66 69.73 122.69 138.20 121.25 0.72 118.21 54.90 -72.90 114.44 68.62 120.50 226.74 70.40 68.63 168.85 67.31 54.27 68.97 68.17 54.18 118.93 68.50 120.95 69.18 118.70 131.35 121.84 54.75 0.11 118.25 0.90 106.35 70.00 213.85 68.83 68.21 125.33 70.13 232.88 226.76 68.85 54.76 124.17 68.63 68.90 115.94 0.42 70.48 138.27 68.95 69.08 54.81 68.25 0.11 70.00 213.84 54.09 Tair (F) 68.14 117.88 229.75 0.35 68.55 114.24 119.11 54.69 121.72 118.59 114.16 119.81 69.65 70.90 119.39 70.16 68.18 70.50 226.74 70.22 124.81 118.06 54.99 116.00 0.43 68.00 67.84 122.38 68.23 54.95 141.76 123.35 69.94 68.91 67.90 122.62 114.93 69.69 0.02 124.50 260.93 70.02 70.38 116.50 68.43 120.50 70.72 70.25 465.73 69.31 68.14 135.43 68.17 115.90 113.65 124.47 70.05 113.45 68.56 125.64 134.58 68.16 Heat Flux (BTU/hrft) 287.31 143.31 0.90 107.40 114.79 69.13 227.53 118.19 69.15 134.39 68.44 0.74 120.63 121.36 54.56 MWT (F) 136.22 69.17 68.90 574.47 116.48 68.18 149.67 68.50 193.73 137.71 68.25 0.90 146.92 69.17 70.44 0.40 70.69 70.47 68.50 218.14 125.30 69.14 70.54 124.16 120.90 126.63 316.17 146.93 114.36 54.38 226.06 0.89 125.90 113.25 54.77 123.57 69.08 69.13 0.37 130.81 0.41 54.83 68.72 68.48 70.50 176.27 125.78 120.13 0.46 126.32 117.42 68.91 113.13 222.22 68.97 119.81 69.38 223.22 132.38 132.81 0.36 54.97 68.81 0.76 69.26 54.11 70.49 121.66 70.50 340.29 114.27 68.97 70.61 124.39 69.44 68.12 68.25 167.90 111.44 135.24 68.84 123.70 68.15 114.90 88.75 292.87 68.59 70.40 68.02 68.85 68.87 120.90 158.47 123.31 69.37 125.32 68.31 70.20 68.23 54.97 117.75 118.68 68.90 130.95 125.81 0.69 70.42 121.90 143.75 69.58 114.36 122.37 68.90 170.90 100.84 69.39 68.76 113.90 70.95 68.34 119.30 68.21 54.33 68.91 114.30 68.04 54.79 127.50 247.90 111.29 70.37 70.38 0.50 127.96 128.36 70.39 121.66 69.25 222.79 69.97 122.85 68.81 69.72 115.30 54.67 70.90 111.18 112.69 0.08 113.28 54.23 54.96 68.90 123.75 68.77 69.95 122.44 0.63 1148.80 116.50 68.93 122.05 54.48 68.99 116.02 129.63 244.65 69.75 201.97 125.13 Flow Rate Heat Flux 2 (GPM) (BTU/hrft ) 0.20 123.04 115.60 69.90 83.05 113.56 68.80 122.90 114.33 69.18 121.46 126.92 118.94 120.

44 225.53 119.21 69.91 69.25 118.31 69.38 0.00 0.90 107.12 124.90 96.69 121.46 55.27 69.50 122.05 69.Trightwall (F) Tleftwall (F) Tfrontwall (F) Tfloor (F) Tcold (F) 68.22 212.50 0.44 125.11 69.76 70.28 124.45 218.90 102.66 70.38 Note: Shaded Test Data was the experimental data used 141 .23 70.64 70.42 70.90 68.90 102.75 0.90 96.30 107.51 69.10 116.32 70.08 70.90 109.89 70.00 0.33 69.25 69.97 Tair (F) 68.31 121.24 69.90 109.94 219.62 70.01 125.47 116.10 122.38 0.99 69.13 55.44 207.71 68.89 220.25 122.90 58.26 69.29 119.97 69.45 70.68 116.63 0.43 55.70 108.81 70.89 126.90 112.31 70.44 55.75 0.90 111.65 70.69 122.81 70.97 209.49 57.19 121.53 115.90 109.08 55.62 70.01 119.77 70.63 70.29 218.90 96.19 70.28 55.37 116.66 71.63 0.17 Tpanel (F) Twater in (F) Twater out (F) 116.38 0.07 69.88 0.13 192.15 69.95 55.49 55.49 55.28 117.25 147.91 216.14 115.71 70.26 214.44 218.33 69.09 69.75 105.70 124.14 69.43 70.25 0.00 0.25 121.13 69.69 121.00 121.67 115.20 125.44 69.31 121.19 104.57 70.90 110.68 68.21 70.27 219.90 109.78 222.31 69.50 0.14 101.97 119.06 68.29 69.58 115.72 193.36 68.35 119.90 112.40 225.51 69.00 0.29 69.17 55.49 69.04 55.90 106.26 70.95 120.21 69.02 117.36 54.39 119.90 108.54 70.91 100.63 69.50 0.53 70.38 0.05 124.03 69.97 68.25 98.03 119.72 69.30 70.69 122.04 117.29 69.39 70.00 122.38 118.18 70.15 70.35 69.81 70.61 70.47 70.46 68.20 120.85 70.49 68.52 119.85 116.44 55.26 70.40 118.31 69.36 55.83 125.60 70.25 122.03 70.87 70.43 119.93 68.42 69.82 119.21 125.99 112.35 69.88 122.37 69.86 68.86 116.50 0.09 69.34 55.74 70.39 69.86 125.56 207.11 69.63 0.79 118.90 104.23 69.97 70.36 124.29 Flow Rate Heat Flux Heat Flux 2 MWT (F) (GPM) (BTU/hrft (BTU/hrft) 0.50 124.34 70.67 69.67 70.61 70.63 122.98 54.09 115.90 73.56 70.71 70.40 125.15 116.33 124.38 0.79 124.04 71.28 117.90 103.07 70.76 69.27 55.83 117.90 113.13 69.91 55.69 70.32 69.31 122.06 115.35 69.38 118.76 125.58 114.54 192.91 71.35 125.54 205.56 103.18 119.13 55.90 103.19 125.29 70.74 126.99 226.37 119.79 119.94 122.31 122.98 204.60 70.31 69.63 0.38 54.15 68.80 68.90 109.

29 66.06 66.38 0.51 139.31 55.58 130.50 0.69 139.17 128.08 68.77 294.39 111.49 133.11 65.94 139.07 279.12 142.71 67.90 141.51 66.06 133.60 66.03 66.49 128.54 66.70 66.57 65.00 0.90 139.50 0.75 0.79 -6.06 130.54 131.22 67.88 0.06 66.90 67.87 141.25 Trightwall (F) Tleftwall (F) Tfrontwall (F) Tfloor (F) 66.38 133.60 55.25 0.67 55.49 131.64 67.38 135.89 68.69 144.49 132.08 291.11 124.18 351.90 158.00 Tcold (F) 54.19 134.81 68.67 131.87 136.63 137.82 67.40 55.93 141.30 68.87 55.69 135.25 130.07 67.86 67.10 68.90 145.07 68.98 67.69 131.42 55.79 68.00 68.13 66.11 68.87 67.90 158.09 68.00 141.04 141.68 68.44 54.90 161.29 54.90 66.13 55.90 147.01 66.12 Tair (F) 66.36 68.83 129.08 129.85 66.71 140.75 0.45 250.72 66.41 132.90 125.90 1091.75 321.76 130.90 64.71 128.78 130.09 68.69 136.13 322.06 66.88 136.29 66.75 131.90 159.90 175.86 67.89 Note: Shaded Test Data was the experimental data 142 .38 0.87 68.48 68.32 133.61 66.66 66.24 130.31 145.28 67.77 67.85 67.71 316.51 67.43 141.48 67.22 2182.90 167.66 66.16 137.66 140.17 131.17 55.84 130.06 67.77 66.36 66.14 67.02 66.35 133.11 Heat Flux Flow Rate Heat Flux 2 MWT (F) (BTU/hrft) (GPM) (BTU/hrft 0.Experimental Data for 24-4 Panel.69 136.28 67.48 66.13 0.06 159.00 54.25 138.50 67.83 67.20 130.61 67.67 67.90 102.42 188.45 66.24 66.79 138.35 317.48 67.21 334.48 Tpanel (F) Twater in (F) Twater out (F) 138.10 330.10 66.79 54.94 137.25 135.11 126.75 0.46 133.99 127.34 66.27 68.73 128.90 55.19 67.90 160.12 138.63 55.90 165.37 66. MWT = 57oC Apanel (ft ) 2 24.96 68.68 54.11 65.13 0.67 67.68 319.83 66.92 67.20 66.79 67.73 66.87 67.79 135.32 65.62 204.46 66.48 55.96 138.86 142.89 127.69 68.06 54.51 66.09 66.17 67.88 54.38 0.90 -3.92 67.01 68.13 0.34 130.83 133.19 131.39 139.95 125.38 0.90 158.64 141.38 0.95 316.50 0.45 66.40 68.20 136.

99 69.28 68.59 151.14 155.63 66.05 68.91 558.88 0.32 162.01 Heat Flux Flow Rate Heat Flux MWT (F) 2 (BTU/hrft) (GPM) (BTU/hrft ) 0.50 158.66 67.87 70.69 160.66 66.41 68.19 168.62 101.09 169.38 0.31 69.79 164.54 381.25 159.42 167.19 54.90 274.94 67.34 68.90 143.32 68.90 271.63 0.49 165.38 Tpanel (F) Twater in (F) Twater out (F) 98.81 68.97 65.21 152.92 68.75 159.90 1264.20 65.92 68.62 68.75 0.61 59.96 70.14 67.40 70.56 67.38 Tcold (F) 61.91 657.16 65.69 146.17 421.76 379.86 66.40 68.47 154.38 67.29 68.13 0.95 67.20 541.36 162.90 69.88 175.65 66.28 53.75 70.63 68.63 168.36 68.90 168.38 0.75 0.88 0.13 0.86 147.55 68.90 505.19 70.38 53.89 66.32 66.90 190.40 67.34 54.90 246.75 153.90 147.38 Trightwall (F) Tleftwall (F) Tfrontwall (F) 65.28 67.23 68.05 159.90 328.05 65.36 168.66 155.55 155.75 67.15 69.76 147.50 0.90 194.95 165.90 235.19 160.71 54.34 68.90 17.35 149.94 65.13 0.02 68.38 68.78 70.03 286.33 69.61 400.41 67.57 68.82 67.05 164.51 68.90 270.81 162.69 67.31 173.73 66.08 471.44 1010.95 68.55 154.59 69.40 56.69 165.21 97.70 153.45 Tair (F) 66.13 55.00 Tfloor (F) 66.90 191.06 157.90 91.54 147.99 66.66 54.56 157.12 54.27 68.91 184.83 54.77 66.88 0.69 154.88 134.23 148.07 549.06 160.00 0.96 70.47 53.71 155.87 67.38 66.01 148.50 155.51 186.97 153.99 58.01 148.07 67.82 164.30 164.06 160.08 154.69 163.88 150.14 163.90 200.20 68.69 69.56 68.22 69.90 189.72 147.52 68.81 148.47 147.50 0.46 153.07 67.74 66.35 68.90 210.95 68.50 66.19 35.90 279. MWT = 71oC Apanel (ft ) 2 24.26 378.54 158.68 172.44 67.91 153.50 161.12 70.94 161.75 0.04 64.78 68.43 53.80 67.00 0.Experimental Data for 24-4 Panel.59 337.93 67.39 168.18 66.34 66.13 0.90 48.66 160.57 68.48 68.90 190.68 53.38 159.90 189.62 0.91 153.00 53.88 150.32 388.80 67.33 Note: Shaded Test Data was the experimental data 143 .78 65.94 159.60 382.75 0.20 65.67 68.50 160.17 70.96 67.06 69.08 64.14 70.48 492.69 159.08 156.42 53.64 164.33 68.79 161.79 68.83 149.59 53.71 68.86 54.80 57.13 69.28 54.95 543.69 159.63 66.64 147.82 67.94 167.84 167.38 160.93 69.96 167.27 155.38 0.78 2529.60 64.17 68.87 70.38 0.72 67.84 67.21 154.56 381.69 68.58 147.72 70.10 183.

95 188.90 254.50 55.06 169.93 201.90 262.55 66.00 0.11 65.90 1285.38 0.32 66.72 70.19 180.81 203.73 170.02 68.97 169.79 174.75 522.78 558.90 259.02 505.04 169.94 181.02 63.56 181.44 196.25 Trightwall (F) Tleftwall (F) Tfrontwall (F) 63.74 65.15 66.48 66.62 70.71 66.88 163.05 201.18 185.98 169.30 55.71 68.28 204.68 71.13 0.15 66.07 517.12 57.46 64.16 170.33 174.90 349.50 0.57 65.81 65.00 196.25 66.90 289.63 180.16 71.43 66.90 252.00 71.19 67.06 181.90 251.28 66.44 181.43 55.16 568.48 67.04 187.51 173.47 183.90 245.86 68.02 509.10 187.09 188.07 66.37 0.22 55.41 67.95 55.17 54.50 67.41 66.50 0.90 264.14 66.31 183.75 0.27 189.08 56.36 174.47 65.27 65.75 0.40 66.05 67.14 66.88 0.97 57.41 66.74 204.45 70.97 69.25 0.37 0.73 Flow Rate Heat Flux Heat Flux 2 MWT (F) (GPM) (BTU/hrft (BTU/hrft) 0.65 71.36 70.00 0.51 55.29 67.03 55.38 70.Experimental Data for 24-4 Panel.96 68.18 68.11 71.16 179.90 507.61 56.82 66.50 0.70 521.90 256.25 186.90 66.95 525.67 170.22 207.88 180.63 68.18 66.24 66.88 0.85 71.31 188.00 Tfloor (F) 66.16 173.88 66.75 0.90 276.97 187.90 254.99 70.18 66.57 66.58 144 .43 170.90 258.00 66.81 187.99 174.00 67.88 519.90 253.43 65.36 186.85 67.63 0.69 180.47 66.93 Tcold (F) 55.48 66.07 509.09 588.01 66.18 579.73 169.90 294.91 197.90 254.40 170.04 66.28 172.78 508.10 71.76 2571.75 Tair (F) 61.21 188.90 258.95 174.77 65.04 63.40 66.19 170.90 261.37 66.25 196.45 67.43 66.12 64.49 66.53 170.70 65.75 191.53 55.61 66.34 56.90 279.89 65.76 189.22 173.75 0.28 177.61 62.89 55.91 528.57 67.50 181.83 200.38 55.15 56.39 56.99 174.89 66.13 0.82 67.02 55.62 67.66 65.21 61.31 66.67 173.90 290.70 65.90 260.00 193.11 71.29 55.95 174.75 183.86 174.45 184.63 428.85 55.86 187.22 67.90 281.96 65.78 562.95 187.54 67.12 186.06 581.81 181.95 67.00 0. MWT = 82oC Apanel (ft ) 2 24.55 66.95 184.50 71.67 71.60 495.71 187.93 181.65 129.48 71.81 187.60 66.07 182.63 0.13 174.63 0.14 173.20 513.81 Tpanel (F) Twater in (F) Twater out (F) 109.58 189.34 192.80 198.78 67.17 55.33 66.50 65.27 69.59 65.51 699.88 200.82 490.96 170.20 170.50 180.97 65.56 188.09 66.86 67.09 169.77 169.72 553.17 66.57 67.40 66.54 65.00 0.90 214.18 66.37 502.16 66.19 181.19 67.78 66.60 66.79 67.06 187.61 65.69 67.68 67.22 516.18 65.09 71.55 178.48 66.45 65.81 181.34 174.88 0.46 55.17 71.83 66.34 189.88 179.25 180.17 67.00 181.90 284.01 66.37 67.56 56.90 247.26 56.15 66.00 0.27 173.69 192.66 188.91 65.63 194.38 0.75 65.91 174.

81 169.02 67.20 188.90 246.78 66.06 66.90 243.23 53.44 486.07 67.19 181.12 67.27 67.36 490.82 173.13 66.25 180.64 65.59 568.04 68.40 67.00 68.75 66.63 0.90 257.16 55.90 264.97 Tair (F) 65.48 481.62 174.00 181.90 259.96 66.11 65.09 173.52 65.80 55.01 68.64 66.32 Tfloor (F) 67.63 181.35 166.25 0.84 169.36 66.63 180.93 54.24 66.90 269.02 186.13 0.23 67.49 173.75 0.96 66.16 Flow Rate Heat Flux Heat Flux 2 MWT (F) (GPM) (BTU/hrft (BTU/hrft) 0.45 187.07 67.92 189.72 169.96 67.71 189.62 0.01 174.67 67.64 538.90 245.50 181.53 174.22 66.90 240.93 168.39 489.83 175.41 563.29 66.61 66.40 170.78 53.48 168.90 255.25 0.03 174.70 166.38 0.50 67.94 65.03 511.56 67.35 66.67 55.34 68.90 281.25 0.78 173.61 66.90 67.77 66.46 189.16 66.48 67.88 67.77 187.12 65.00 182.53 66.59 68.06 182.73 67.53 55.94 182.59 67.05 68.00 145 .90 239.33 54.13 180.38 180.13 0.18 65.80 55.71 166.88 0.27 Tpanel (F) Twater in (F) Twater out (F) 169.25 0.08 176.38 0.63 182.75 168.63 0.28 492.88 479.33 55.75 180.85 67.00 0.90 246.68 65.05 68.72 165.04 67.81 180.21 175.25 0.43 67.73 66.23 167.27 187.55 519.04 Tcold (F) 55.51 167.01 189.96 67.27 66.14 55.92 65.07 66.85 67.00 0.23 168.94 169.12 66.62 55.52 173.08 67.72 67.67 493.38 66.17 189.34 529.88 0.32 187.50 0.64 67.54 66.90 240.90 241.07 187.24 534.93 188.80 67.69 67.42 66.89 67.03 66.90 284.01 529.70 66.43 66.38 66.89 189.95 173.02 67.75 0.74 545.05 67.19 480.02 66.90 247.28 174.27 539.07 68.27 67.90 240.70 174.90 246.81 181.13 174.87 174.62 67.90 269.04 68.55 66.81 187.57 66.61 66.31 66.87 167.05 174.88 66.14 67.49 66.94 175.12 175.48 188.43 66.90 254.23 67.58 68.84 186.16 67.56 55.38 180.90 167.90 256.50 181.46 173.72 66.72 65.82 66.60 67.95 482.88 67.00 0.82 187.40 66.50 189.88 174.42 169.81 181.11 54.36 187.90 267.92 66.20 54.43 66.25 166.16 67.90 272.39 187.59 55.32 55.00 181.06 180.33 67.50 0.38 186.81 180.41 509.55 169.25 188.20 65.63 0.38 0.13 0.12 169.69 68.06 66.19 66.60 68.70 167.57 67.86 527.03 65.98 513.49 169.75 66.62 55.06 54.94 181.86 167.61 174.49 168.62 65.68 53.12 0.18 66.95 68.73 173.48 54.90 229.48 67.91 187.90 68.83 54.42 54.24 67.72 173.31 55.85 188.89 175.31 66.20 67.90 187.50 168.48 67.56 182.28 67.78 169.45 492.62 68.62 181.01 68.90 257.07 65.91 515.13 55.90 263.88 180.58 515.78 67.85 53.30 481.31 66.69 182.89 187.38 67.28 68.32 68.92 66.74 67.29 175.49 66.75 180.00 0.56 180.10 67.91 65.29 67.90 258.81 55.78 187.18 66.99 68.69 180.43 66.47 175.90 264.42 67.12 68.51 54.88 173.50 0.32 65.67 68.75 0.04 66.59 186.80 66.85 516.90 495.35 67.65 459.83 66.90 244.17 67.69 168.Trightwall (F) Tleftwall (F) Tfrontwall (F) 66.

04 68.72 54.06 187.26 69.90 256.93 175.95 68.00 67.16 562.80 166.50 68.64 169.00 68.37 54.13 66.01 186.00 Note: Shaded Test Data was the experimental data and used for the steady state plot 146 .59 Tair (F) 68.73 67.69 173.58 65.84 166.55 54.18 67.29 167.83 68.13 181.Trightwall (F) Tleftwall (F) Tfrontwall (F) 67.42 67.37 0.80 188.83 54.93 68.17 167.32 174.06 179.18 512.33 67.50 67.38 Tcold (F) 54.96 68.50 0.80 67.35 511.99 472.44 69.16 187.94 66.69 181.66 53.80 188.03 488.31 181.50 182.00 0.16 67.97 68.34 173.89 174.13 0.93 67.82 68.30 188.27 68.12 167.86 67.83 188.04 67.39 67.49 Flow Rate Heat Flux Heat Flux 2 MWT (F) (GPM) (BTU/hrft (BTU/hrft) 0.17 68.71 65.18 543.33 67.48 Tpanel (F) Twater in (F) Twater out (F) 168.13 0.89 173.05 65.21 67.38 0.02 Tfloor (F) 67.75 174.06 181.38 67.27 175.09 65.13 0.90 242.42 190.28 53.95 69.35 484.38 54.89 67.77 53.66 68.28 489.30 67.69 182.13 0.38 0.48 187.90 265.52 173.00 68.03 54.53 173.24 168.43 513.06 180.90 271.25 0.84 168.95 54.50 180.93 67.34 68.56 65.82 67.09 68.90 242.40 531.98 531.25 180.63 0.43 67.56 180.39 173.90 265.90 244.66 175.88 66.54 187.65 67.90 236.29 186.18 68.13 167.79 65.43 68.65 68.90 244.47 67.90 281.89 66.65 189.69 182.19 54.01 166.10 167.20 66.90 256.39 67.85 484.90 255.44 65.86 65.

45 541.73 194.83 64.07 633.80 65.60 191.67 185.95 67.48 197.23 185.69 199.61 206.16 175.29 539.92 578.02 70.Experimental Data for 24-4 Panel.95 55.17 70.08 172.39 68.00 203.04 68.90 309.90 267.39 66.00 0.93 180.22 70.31 196.82 184.12 69.35 190.92 54.60 207.51 211.53 69.28 68.42 534.26 212.70 66.15 67.90 277.85 211.12 67.83 193.95 213.28 64.13 68.96 68.74 69.31 195.91 555.41 67.85 54.90 319.50 201.51 187.44 638.19 68.69 67.90 284.88 62.52 195.83 55.69 204.63 189.79 65.22 68.90 318.13 0.90 429.46 66.05 564.97 176.90 295.24 639.09 65.37 0.67 201.92 66.39 66.37 0.58 196.80 67.27 70.90 270.00 0.46 70.56 191.00 66.12 61.73 67.05 54.83 65.11 70.17 64.81 190.47 68.31 195.91 65.03 67.50 193.14 54.87 69.40 66.55 67.29 184.93 625.57 205.81 67.59 66.90 66.69 203.52 69.54 68.90 316.00 Tfloor (F) 66.50 205.13 0.23 67.90 298.63 0.57 203.16 66.90 318.28 67.78 194.94 202.56 204.68 69.22 69.56 65.00 206.70 590.75 66.26 181.01 55.57 68.77 Tcold (F) 56.56 206.52 Note: Shaded Test Data was the experimental data 147 .71 630.88 0.91 637.00 0.90 322.90 187.90 319.35 67.34 196.16 68.52 182.08 66.90 307.91 66.00 0.96 67.41 67.11 212.36 70.26 175.56 204.90 67.02 71.25 197.66 199.90 269.44 192.44 54.00 0.98 180.53 66.45 189.90 315.82 189.29 54.05 212.77 188.15 66.31 67.15 66.97 67.11 186.56 202.17 68.90 321.95 177.39 637.22 67.63 204.39 546.33 67.31 202.63 54.80 66.41 Tair (F) 58.83 69.79 619.62 54.42 55.25 0.25 0.18 67.38 190.82 54.79 55.20 188.85 61.30 55.54 67.62 67.07 65.78 197.37 209.09 181.03 70.93 214.74 54.97 68.90 300.13 0.21 70.38 0.25 55.76 214.75 0.50 0.09 67.77 54.56 188.00 202.57 196.72 Tpanel (F) Twater in (F) Twater out (F) 160.90 282.38 0.59 195.06 185.23 210.10 183.90 289.63 194.30 65.74 69.40 173.79 65.98 569.90 321.00 0.51 56.08 65.62 194.07 69.93 196.13 Trightwall (F) Tleftwall (F) Tfrontwall (F) 68.03 615.29 642.60 70.41 597.93 67.97 67.36 643.88 213.22 70.96 70.00 66.93 178.00 0.21 70.90 312.95 645.13 0.59 67.55 213.75 66.10 624.23 69.42 858.85 66.00 185.87 71.16 65.71 592.21 67.63 0.19 188.13 204.84 67.77 59.65 54.34 189.89 196.00 192.77 67.82 65.25 0.32 54.17 54.74 189.33 65.62 Flow Rate Heat Flux Heat Flux MWT (F) (GPM) (BTU/hrft2 (BTU/hrft) 0.21 69.00 198.25 69.71 65.05 70.40 67.86 195.39 188.41 67. MWT = 95oC Apanel (ft ) 2 24.42 211.60 600.55 198.32 190.90 296.49 54.32 54.99 54.06 186.14 174.02 187.75 70.86 67.78 68.00 0.59 67.72 197.98 182.79 69.21 187.90 273.11 70.84 70.73 66.59 70.23 65.90 312.00 0.55 67.76 67.

31 121.42 117.56 125.10 285.90 133.82 262.66 113.90 136.63 0.15 117.23 67.71 122.72 67.78 68.52 132.63 119.38 0.79 69.90 137.38 Heat Flux Flow Rate Heat Flux 2 MWT (F) (BTU/hrft) (GPM) (BTU/hrft 0.90 125.22 68.24 126.74 67.37 67.53 56.27 72.84 120.63 68.46 67.50 0.46 66.76 66.01 274.25 122.01 69.90 88.42 68.13 132.86 65.88 67.52 69.00 Tcold (F) 65.00 0.92 273.91 67.13 66.66 282.22 2161.82 118.97 118.03 68.52 69.24 66.53 71.33 69.00 0.11 59.41 67.40 121.92 55.02 119.90 66.90 131.74 277.03 66.Experimental Data for 24-8 Panel.65 116.35 118.79 56.92 70.63 274.41 69.90 577.13 119.00 73.29 70.59 71.19 128.90 137.59 131.32 118.83 72.76 69.58 69.90 118.70 116.30 56.75 127.54 129.00 121.77 56.81 66.90 90.44 125.23 70.31 118.01 66.83 67.72 70.61 72.13 67.79 260.90 69.50 137.31 121.96 118.42 67.78 68.76 70.57 66.90 187.84 117.99 116.33 66.50 123.18 72.25 0.50 0.10 67.49 66.50 66.18 117.85 108.63 66.63 56.76 68.30 69.67 125.12 124.42 120.54 66.32 57.75 102.25 122.61 67.42 67.15 56.55 137. MWT = 49oC Apanel (ft ) 2 24.66 67.76 126.50 0.90 130.90 138.50 0.63 0.00 71.75 68.28 73.09 56.90 142.80 67.79 66.68 118.46 117.80 71.08 181.27 117.88 122.91 126.19 68.35 177.25 130.98 267.66 118.91 69.37 71.80 66.99 68.28 117.90 130.24 66.64 118.66 56.75 260.25 0.15 1155.53 374.91 69.47 71.40 129.19 125.59 Tpanel (F) Twater in (F) Twater out (F) 76.90 118.75 124.00 67.50 120.58 56.41 119.06 56.04 118.13 70.96 72.52 275.91 67.07 72.25 0.82 116.00 0.90 1080.42 125.00 0.99 66.90 137.63 0.13 Trightwall (F) Tleftwall (F) Tfrontwall (F) Tfloor (F) 73.72 71.75 121.81 121.84 57.09 69.18 72.52 71.90 141.32 125.97 Tair (F) 71.31 119.12 68.69 122.23 121.80 116.59 66.45 69.52 Note: Shaded Test Data was the experimental data 148 .54 126.60 118.75 119.50 0.89 56.46 69.43 73.04 152.33 71.19 69.58 71.40 119.90 59.06 121.50 121.17 56.39 55.50 0.80 236.35 70.58 71.89 67.00 132.95 123.29 119.22 125.66 69.29 56.75 0.26 125.77 251.07 68.79 61.15 68.38 0.25 71.05 70.73 65.

88 0.13 0.31 138.75 0.63 138.13 67.Experimental Data for 24-8 Panel.31 190.09 68.16 143.98 399.38 133.90 88. MWT = 60oC Apanel (ft ) 2 24.47 70.36 137.15 66.69 140.51 67.71 68.47 136.04 69.60 56.25 68.88 68.94 140.90 176.90 -48.99 1159.37 130.30 66.39 Tpanel (F) Twater in (F) Twater out (F) 153.59 132.18 68.41 66.13 67.00 Tcold (F) 57.51 286.38 151.53 157.75 56.79 68.48 132.55 159.90 178.77 69.13 145.90 168.90 579.32 68.78 68.36 66.65 133.37 145.90 169.90 175.84 138.86 264.09 144.58 133.12 127.80 66.31 68.50 0.31 135.57 339.56 130.04 55.92 67.22 144.90 203.88 0.93 66.45 67.06 56.25 Trightwall (F) Tleftwall (F) Tfrontwall (F) 67.26 68.20 67.90 169.90 143.32 140.24 133.70 416.06 133.76 67.70 56.38 Flow Rate Heat Flux Heat Flux 2 MWT (F) (GPM) (BTU/hrft (BTU/hrft) 0.73 67.64 133.48 146.27 56.04 135.75 0.99 68.88 139.13 0.74 67.89 68.75 0.13 68.59 56.14 139.90 132.25 57.66 67.99 67.02 56.83 67.25 0.90 199.21 68.78 66.63 339.26 155.63 0.16 66.15 66.88 141.64 67.57 -97.41 68.16 69.60 357.38 56.69 129.83 136.49 68.25 0.27 68.34 67.00 69.62 133.24 68.87 68.54 135.65 68.32 68.24 68.82 68.10 68.69 129.33 142.36 141.04 69.75 0.29 67.14 135.01 176.66 69.90 169.29 136.02 134.54 67.05 337.88 57.94 156.88 134.69 139.59 68.11 56.05 133.73 68.70 68.55 144.23 67.10 134.25 141.81 145.07 68.53 212.90 66.26 67.80 65.18 69.38 0.51 56.39 147.63 0.29 67.57 128.90 208.56 174.30 56.25 146.24 56.31 Note: Shaded Test Data was the experimental data 149 .04 66.95 68.75 68.14 350.44 155.22 68.13 Tair (F) 66.64 136.92 134.50 140.09 68.75 69.33 145.00 0.60 69.51 56.13 141.38 0.87 148.06 406.79 69.81 339.00 Tfloor (F) 68.52 68.31 144.34 68.72 159.85 67.62 67.90 79.66 137.46 68.29 152.11 68.41 68.90 106.84 353.87 68.

88 0.90 222.52 55.56 157.73 153.15 68.10 55.94 66.53 56.09 69.57 67.83 69.14 152.11 164.97 154.88 517.75 0.13 67.99 153.74 153.38 158.75 Trightwall (F) Tleftwall (F) Tfrontwall (F) 66.94 158.69 150.13 0.03 154.94 445.50 68.85 152.05 421.90 153.27 166.77 66.88 0.06 164.24 478.81 54.32 68.26 164.51 65.88 0.25 160.93 69.38 159.31 152.62 68.22 153.65 427.01 67.25 159.25 68.25 151.19 157.81 160.90 232.31 159.70 166.95 165.90 225.90 216.90 206.00 164.60 56.36 67.38 66.31 159.87 470.50 399.68 69.32 69.50 153.52 69.46 432.43 55.40 68.90 208.97 67.94 158.67 164.10 67.94 158.56 159.43 67.78 163.67 70.81 67.47 68.99 69.90 206.27 68.90 150.50 0.00 0.19 163.88 0.16 445.65 68.32 165.16 68.75 67.84 68.89 454.09 67.84 67.04 Heat Flux Flow Rate Heat Flux 2 MWT (F) (BTU/hrft) (GPM) (BTU/hrft 0.24 55.32 409.25 153.89 68.52 68.21 66.25 66.75 0.28 164.47 67.01 164.90 235.75 70.05 68.80 154.68 66.72 68.00 Tfloor (F) 67.68 55.77 164.63 68.90 239.90 199.07 55.59 161.15 67.24 68.94 53.05 Tcold (F) 56.61 165.50 0.86 69. MWT = 70oC Apanel (ft ) 2 24.75 158.98 152.50 0.42 165.58 65.79 55.63 417.83 67.72 68.90 210.70 69.64 67.04 66.48 56.48 68.26 152.05 165.51 68.67 67.86 66.02 67.50 465.07 152.68 67.17 69.42 55.22 152.01 67.62 165.59 152.06 152.49 66.82 Tpanel (F) Twater in (F) Twater out (F) Tair (F) 155.89 56.59 55.10 69.90 205.82 67.80 69.66 66.90 258.75 68.23 164.48 68.26 68.29 66.62 66.08 56.63 0.00 0.65 425.63 0.71 152.77 68.90 213.11 66.22 166.95 410.58 54.26 450.74 152.14 165.36 68.10 165.75 0.81 159.72 67.04 68.17 152.07 68.90 216.90 185.63 65.72 66.37 426.50 0.55 151.35 67.61 412.70 153.40 65.71 64.50 68.37 68.66 67.38 0.67 55.98 68.67 68.35 164.11 152.83 153.86 152.87 164.76 151.08 68.22 68.13 0.00 154.64 153.50 159.50 67.76 68.71 153.93 151.47 432.30 67.50 158.52 68.00 153.23 68.00 0.69 159.71 67.38 0.30 153.63 419.72 68.90 227.19 55.13 0.71 56.34 68.27 55.90 216.43 421.20 68.59 154.51 432.48 68.46 66.00 158.93 153.94 158.43 417.90 210.63 0.45 67.30 67.50 65.36 Note: Shaded Test Data was the experimental data 150 .62 153.Experimental Data for 24-8 Panel.42 152.68 153.90 222.88 53.00 150.75 66.49 68.61 66.47 153.78 66.86 66.38 158.35 67.60 69.92 67.25 158.82 163.12 412.86 154.50 0.76 65.10 68.90 209.75 0.88 159.65 370.94 152.65 68.24 67.80 153.83 68.22 68.37 68.90 204.12 67.50 56.86 67.62 0.65 153.14 66.38 159.69 67.90 208.42 55.21 68.06 160.01 300.15 66.90 213.43 69.23 154.84 152.92 54.90 212.92 151.25 157.01 67.11 68.92 152.01 68.

13 0.82 194.94 68.69 56.41 68.42 70.62 70.93 175.97 784.02 68.02 67.16 670.60 564.21 621.98 188.90 349.14 172.40 172.89 66.23 56.01 68.05 68.63 179.98 69.54 175.96 67.95 56.02 71.38 0.64 56.67 69.14 56.34 56.34 68.75 491.96 67.12 187.01 171.33 56.52 69.55 172.56 172.00 0.96 70.79 68.21 172.60 67.86 196.50 185.23 56.88 180.88 68.44 185.00 68.76 Flow Rate Heat Flux Heat Flux MWT (F) (GPM) (BTU/hrft2 (BTU/hrft) 0.90 310.70 67.96 170.39 68.00 0.48 56.63 171.32 67.64 172.32 67.98 556.54 69.12 185.29 70.94 185.50 180.30 185.12 65.51 56.49 596.31 175.64 68.73 67.69 67.09 172.45 66.18 186.28 188.90 196.31 66.54 621.94 187.63 170.07 56.72 70.90 245.10 175. MWT = 82oC Apanel (ft ) 2 24.82 637.42 68.19 188.88 178.69 67.26 178.45 631.05 70.32 196.10 68.63 0.22 68.10 193.19 171.53 172.69 178.04 68.68 186.19 182.50 178.90 271.03 179.42 68.23 67.38 70.90 335.70 174.60 56.90 275.02 172.20 186.50 68.84 69.87 0.47 579.78 66.90 282.68 173.31 178.56 68.56 179.63 179.06 178.58 178.04 178.10 182.31 179.65 172.81 180.55 68.03 70.86 67.90 238.67 67.57 69.74 68.19 171.51 67.26 71.21 70.94 188.99 68.12 67.44 68.56 67.80 69.89 70.90 268.Experimental Data for 24-8 Panel.29 573.63 606.25 0.63 0.12 0.90 298.37 69.42 69.97 56.42 698.04 67.74 66.19 56.69 644.23 70.57 69.86 68.90 318.13 Trightwall (F) Tleftwall (F) Tfrontwall (F) 65.85 69.00 0.00 179.83 67.75 0.81 179.37 0.68 180.07 70.24 69.26 173.88 543.52 66.90 310.17 171.60 68.79 536.31 189.04 67.63 0.50 184.21 66.78 68.69 188.44 191.21 182.06 179.38 179.90 345.36 623.88 0.03 66.58 66.90 242.28 56.27 69.00 0.90 392.64 171.69 68.69 68.43 198.11 69.63 0.32 477.21 69.42 68.69 68.03 67.31 187.67 68.97 69.10 67.59 690.77 67.92 67.93 177.74 170.66 56.44 187.90 286.42 172.63 0.21 177.23 597.66 484.88 0.71 192.38 173.42 66.85 68.50 68.75 0.69 67.14 68.70 68.90 291.63 190.75 71.10 67.66 68.66 68.41 67.24 172.44 171.43 67.47 68.67 185.90 322.72 68.37 550.03 Tair (F) 68.31 180.52 171.04 172.09 67.38 0.99 67.90 278.00 Tfloor (F) 67.13 70.90 315.76 56.84 171.73 70.37 69.28 67.87 151 .90 298.82 172.73 200.20 56.99 69.94 67.51 Tcold (F) 55.13 69.89 171.30 56.62 56.49 193.35 56.88 69.16 171.57 66.10 441.44 179.85 68.00 0.72 66.30 56.12 188.45 67.29 67.56 69.90 311.39 187.23 Tpanel (F) Twater in (F) Twater out (F) 173.52 188.49 66.82 66.81 56.90 289.33 56.69 183.63 0.08 56.38 0.08 171.07 393.84 176.56 186.14 70.79 56.13 0.33 66.42 174.13 68.90 287.96 177.67 68.75 0.33 172.88 0.90 220.85 575.22 71.90 303.50 179.94 67.19 69.41 583.80 186.18 172.

06 190.33 72.08 645.43 56.57 67.00 0.06 67.90 268.81 191.34 180.25 182.90 360.70 68.44 179.08 68.22 68.88 178.54 68.95 187.50 68.75 0.50 Note: Shaded Test Data was the experimental data Experimental Data for 24-8 Panel.35 59.75 200.27 60.22 69.33 71.30 69.59 71.66 181.76 71.35 56.88 198.15 172.27 67.13 68.65 59.65 186.85 671.41 Tair (F) 69.34 171.84 171.38 0.96 68.25 0.18 68.13 205.39 68.42 69.47 171.36 206.53 60.00 0.61 68.83 60.31 171.43 73.47 188.48 69.08 67.69 Tcold (F) 56.94 188.22 68.36 70.90 321.90 Tfloor (F) 69.76 67.51 73.72 187.08 567.07 737.82 71.90 326.90 834.41 70.73 60.50 68.63 0.38 0.89 60.90 384.14 69.53 70.57 56.31 196.50 68.30 70.89 68.46 67.59 Note: Shaded Test Data was the experimental data 152 .58 192.90 309.11 560.84 70.23 70.13 68.95 71.67 619.46 72.01 720.38 202.75 177.19 71.80 699.41 67.54 68.44 69.16 69.72 68.72 68.92 171.88 60.32 711.77 67.81 187.13 207.85 181.35 204.89 184.35 675.48 72.69 179.06 67.75 0. MWT = 95oC Apanel (ft ) 2 24.44 Tpanel (F) Twater in (F) Twater out (F) 171.68 187.12 768.90 335.79 68.75 0.61 67.88 180.63 60.93 69.33 Tcold (F) 61.34 Tpanel (F) Twater in (F) Twater out (F) 154.06 200.15 69.66 175.63 0.67 70.20 721.35 185.83 71.43 67.59 191.90 340.32 172.72 187.Trightwall (F) Tleftwall (F) Tfrontwall (F) 67.19 194.97 172.99 71.83 190.31 179.88 196.04 Flow Rate Heat Flux Heat Flux MWT (F) (GPM) (BTU/hrft2 (BTU/hrft) 0.46 188.59 68.88 0.10 68.88 0.38 0.90 368.64 171.29 658.19 69.24 194.37 67.75 0.11 70.78 68.50 193.42 68.96 69.49 67.84 68.50 0.94 68.66 69.00 179.11 74.13 0.51 182.27 170.31 60.89 594.35 70.77 72.19 186.26 70.63 203.30 67.80 202.45 59.67 69.28 56.63 0.02 67.75 0.67 70.64 69.48 Heat Flux Flow Rate Heat Flux 2 MWT (F) (BTU/hrft) (GPM) (BTU/hrft 0.32 72.77 Tair (F) 74.90 214.75 155.57 69.27 70.00 Tfloor (F) 75.63 68.02 652.02 199.11 184.36 197.38 56.90 374.90 280.81 209.25 Trightwall (F) Tleftwall (F) Tfrontwall (F) 70.90 360.31 60.91 69.75 71.25 74.93 690.88 200.90 355.51 56.40 193.85 68.63 204.63 171.87 69.61 187.75 179.50 0.42 69.76 749.66 68.44 643.90 322.38 68.78 70.97 186.93 170.44 68.47 179.51 171.88 1669.10 71.62 212.90 345.58 185.46 67.44 537.50 184.84 70.28 189.08 186.90 349.62 175.21 187.90 337.90 297.27 69.90 329.29 680.82 211.51 60.89 190.87 179.95 68.14 196.90 283.

Appendix D Uncertainty Analysis for Experimental Data and RSC Model 153 .

2005).2oC. D. Since the radiant panel surface temperature is averaged over six thermocouples the uncertainty in temperature measurement can be shown to equal +/-0. The following portion will discuss the uncertainty associated with the data obtained running the experiment.1).5oC (OMEGA. 2005).In any experiment there is a certain amount of error associated with obtaining data to compare to theoretical models. 154 . for brown thermocouple grade. shows that they are accurate within +/-0. the error with each measurement is +/-0. Doing an uncertainty analysis on the wall temperatures based on an average of 4 thermocouples. From the Omega handbook (Omega.1 Uncertainty in Measured Surface Temperatures Omega T-Type thermocouples were used to measure the surface temperatures.25oC. shown and derived in the MathCad worksheet below (Figure D.

1: MathCad Worksheet to Determine Temperature Uncertainties 155 .Figure D.

2 Uncertainty in Calculated Heat Output The panel heat output is determined by measuring the mass flow rate of the water and the temperature difference between the radiant panel inlet and outlet water.1) Figure D.2: MathCad Worksheet to Derive the Equation for Panel Heat Output Uncertainty 156 .2).D.62Q(Tin − Tout ) Pwidth (D. it is shown that the heat output equation is: q 'panel = 51. From the attached MathCad worksheet (Figure D.

which is accurate within 2% of the measured reading. The uncertainty of the heat output can be estimated to be between 2. Ultimately. it can be shown that the uncertainty of the heat P P output is between 2 to 2. For a 0.1.5%. however due to the resolution of the DA system.2) The water flow rate was constant for each test at a value of 0. Using two thermistors to measure the inlet and outlet P P temperature.05oC) (Huang. The water flow rate was measured using a circuit balancing valve. 2005) or a thermistor P P (+/-0. Using the root sum of the squares method (RSS) the uncertainty of the heat output of the panel can be estimated. From Equation D.057L/s.5 to 5% (Table D. A thermopile was installed for a more accurate measurement. 157 . the circuit balancing valve is accurate within 0. Equation D. Another source is the accuracy of the temperature measurement. the thermopile was ineffective. the uncertainty for the heat output of the panel can be shown to be +/-50W/m (Table D. P P The pressure difference across the venturi tube is converted into a flow rate.2 can be derived estimating the uncertainty with the heat output from the panel.14oC. The error associated with each temperature measurement is +/-0.0011 L/s.The water inlet and outlet temperature was measured using T-Type thermocouples.5oC.3. The error could easily be reduced if the thermopile would work (+/-0.1). δq panel ' ⎡⎛ δ∆T ⎞ 2 ⎛ δV ⎞ 2 ⎤ = ⎢⎜ ⎟ +⎜ ⎟ ⎥ q panel ' ⎢⎝ ∆T ⎠ ⎝ V ⎠ ⎥ ⎣ ⎦ 1/ 2 (D.1oC).057 L/s reading. The water temperature difference changed for each different panel temperature. The panel width is made from standard sized panels and minimal error is associated with it. the uncertainty of the temperature difference can be shown to equal +/-0. If the thermopile had been used. The derivation can be found in Figure D.1).

Figure D.3: MathCad Worksheet to Determine Heat Output Uncertainty 158 .

2888872 7.688 % Error 4.68823 .8675928 48.966 3.1: Panel Heat Output Uncertainty due to Temperature Difference Method Summary of Results from the 24-4 Test Based on an average steady state test run of 30 minutes .484 % Error 4.7102 3.584122 6.192435 8.07153 62.37209 2.584122 5.0331246 16.762 7.2037 10.619147 2.Error 12.35 298.Error 5.1273 49.319 12.3625 73.965 11.762007 2.6117509 10.Error 6.96507 14.4825 Thermistor + Error .1977 2.91 374.3363827 55.Table D.46123 2.02150913 60.938207 2.379 17.71 407.8522 17.461 12.852 15.294695 6.20535 100.31285635 49.99479 12.86759 73.4715319 16.1977 2.47602 Summary of the 24-8 Results Based on an average steady state test run of 30 minutes .4825 Thermistor + Error .6191 8.1273 53.966 3.849011 2.801 14.4 613.02151 99.32374 11.so 8 readings Inlet Fluid Temp Heat Output Regular Thermocouple o (W/m) % Error + Error .72083 49.36250295 54.Error ( C) 51.31911 13.84181 53.036 13.96 261.11988 86.22468 Thermopile % Error + Error 2.074 11.324 10.1811396 13.51791 86.51791338 53.7424858 10.8634 2.03638 15.8634 2.068883 14.995 12.7102 3.11 553.07152876 62.37261 48.476019 48.37929 159 .203677 2.48388 .72083 62.0820049 7.131715 11.3721 6.6840682 13.225 15.1198832 49.33638 Thermopile % Error + Error 2.05199 15.294695 7.37261 54.131715 10.9382 11.904272 55.78 207.904272 48.66 700.052 14.77 332.84181 60.20534571 53.8884267 23.80142 2.so 8 readings Fluid Inlet Temp Panel Heat Regular Thermocouple o Output (W/m) % Error + Error .31286 61.6752184 23.Error 10.07391 11.Error ( C) 51.192435 8.66 490.849 8.068883 12.

The effect of bond conductance on panel temperature can be seen in Equation 4.18 (Equation D. The following MathCad worksheets summarize the calculations (Figures D.3 have no uncertainty.3) The uncertainty in the panel temperature can be quantified if the uncertainty in the bond conductance is known based on the measured values.3 Uncertainty in Bond Conductance due to heat paste thickness Using the RSS method as described the above.4) Assuming all other variables in Equation D. ⎛ 1 1 ⎞ T panel = T f − q 'panel ⎜ ⎜ h πD + C ⎟ ⎟ b ⎠ ⎝ w i (D.D.6). 160 . an attempt was made to measure the uncertainty associated with the bond conductance due to the measurement uncertainty in the heat paste thickness and width.4 to D. the panel temperature uncertainty due to bond thickness measurement error can be quantified. The bond conductance equation is shown below: Cb = kbb tb (D.3).

Figure D.4: MathCad Worksheet to Determine Panel Temperature Uncertainty Associated with Bond Conductance 161 .

Figure D.5: MathCad Worksheet to Determine Panel Temperature Uncertainty Associated with Bond Conductance 162 .

6: MathCad Worksheet to Determine Panel Temperature Uncertainty Associated with Bond Conductance 163 .Figure D.

Appendix E Solution to the Radiosity Equations 164 .

the view factors for a 2 ft wide radiant panel was calculated first.1 View Factor Calculations for Radiant Panel Test Chamber To solve the set of radiosity equations. Figure E.1: MathCad Worksheet to Calculate the View Factors for the Radiant Panel Test Chamber 165 . the view factor calculations are summarized on the following MathCad worksheets (Figures E.E.3).1 to E. Using the same method as described in Appendix B.

2: MathCad Worksheet to Calculate the View Factors for the Radiant Panel Test Chamber 166 .Figure E.

the remaining view factors can be calculated as shown in the next few MathCad worksheets (Figure E.4 to E.3: MathCad Worksheet to Calculate the View Factors for the Radiant Panel Test Chamber Using reciprocity and tabulated parallel and perpendicular view factor equations.7) and excel (Table E.Figure E. 167 .1).

4: MathCad Worksheet to Calculate the View Factors for the Radiant Panel Test Chamber 168 .Figure E.

Figure E.5: MathCad Worksheet to Calculate the View Factors for the Radiant Panel Test Chamber 169 .

6: MathCad Worksheet to Calculate the View Factors for the Radiant Panel Test Chamber 170 .Figure E.

Figure E.7: MathCad Worksheet to Calculate the View Factors for the Radiant Panel Test Chamber 171 .

181 0.0306 0.401 0.181 0.2359994 0 0.181 0.236 0.181 0.14163636 0.166 0.20418182 0.0306 0.762768 0.236 0.1: Excel Worksheet Calculating View Factors using Reciprocity Equations Surface 1 Surface 2 Surface 3 Surface 4 Surface 5 Surface 6 Surface 7 F11 F12 F13 F14 F15 F16 F17 Panel Back Wall Front Wall Floor Left Wall Right Wall Ceiling 0 0.28036473 0.181538 0.205401 F71 F72 F73 F74 F75 F76 F77 0 0.181538 0.181538 0 0.237232 F51 F52 F53 F54 F55 F56 F57 0.166 0 0.181 0.1558 0.0114 0.8442 F31 F32 F33 F34 F35 F36 F37 0.181 0.181 0.2054006 0.794599 F61 F62 F63 F64 F65 F66 F67 0.036615 0.7754 F41 F42 F43 F44 F45 F46 F47 0.2246 0.01831133 172 .238 0.166 0 0.2054006 0.235999 0.057 0.153 0 A1 (ft ) A2 (ft ) A3 (ft ) A4 (ft ) A5 (ft ) A6 (ft ) A7 (ft ) 2 2 2 2 2 2 2 26 130 130 169 130 130 143 F21 F22 F23 F24 F25 F26 F27 0.181 0.181538 0.Table E.153 0.166 0.0802 0 0.18672782 0 1.

2 Radiosity Equations (Exact Solution) With knowledge of the view factors.E. Figure E.8: MathCad Worksheet for Radiosity Calculations (Exact) 173 . the radiosity equations could be solved as shown in attached MathCad worksheets (Figure E.8 and Figure E.9).

Figure E.9: MathCad Worksheet for Radiosity Calculations (Exact) 174 .

The attached MathCad worksheet (Figure E.10: MathCad Worksheet for Radiosity Calculations (Three Surface) 175 .3 Radiosity Equations (Three Surface Solution) For the assumed three-surface enclosure the view factors can be calculated using reciprocity equations and published view factor equations for perpendicular plates (shown above).11) solves the radiosity equations for the known three-surface enclosure view factors.10 and Figure E. Figure E.E.

11: MathCad Worksheet for Radiosity Calculations (Three Surface) 176 .Figure E.

78 293 307.02 5.50 3.06 185.26 0.17 3.25 197.12 2 %error 6.19 293 232.2: Comparison of Radiative heat Exchange Methods Tpanel (K) 318 323 328 333 338 343 348 353 358 363 Tcold (K) 283 283 283 283 283 283 283 283 283 283 Tfront (K) 293 293 293 293 293 293 293 293 293 293 Tfloor (K) 293 293 293 293 293 293 293 293 293 293 Tleft (K) 293 293 293 293 293 293 293 293 293 293 Tright (K) 293 293 293 293 293 293 293 293 293 293 Tceiling (K) Exact qrad (W/m ) MRT Method 293 163.43 -0.16 219.The following table (Table E.11 -0.88 4. the MRT method and the threesurface enclosure method.75 432.72 509.31 177 .29 477.19 0.31 417. Table E.07 0.91 3.90 6.55 293 347.10 % error -0.34 3 .12 347.42 268.93 307.04 388.27 293 433.90 255.49 374.25 -0.27 293 197.24 0.38 4.33 293.2) summarizes the total radiative heat exchange for the exact solution.77 525.01 0.55 232.26 293 526.04 462.69 3.33 293 479.Surface 164.14 0.51 333.49 4.03 293 389.55 152.68 293 268.

Model Validation and Numerical Results 178 .Appendix F Grid Refinement.

F.1

Grid Refinement

The following excel worksheets summarize the grid refinement procedure.
Cavity: Panel Size: Coarse Grid: Panel Boundary: Cold Wall Bound. Wall Boundary x-dir fill cell: y-dir fill cell: Gridrefinement: Cell No 4492 4493 4494 4495 4496 4497 4498 4499 4500 4501 4502 4503 4511 4568 4569 4570 3.9624m x 3.048m 0.88315 0.064m 0.025cm start x 2expansion 0.025cm start x 2expansion 0.05cm start x 2 expansion 0.05915m 0.040975m 2 times Heat Flux (W/m ) 11.821 13.459 11.709 11.302 11.123 11.048 11.051 11.125 11.278 11.555 12.264 15.155 28.280 62.968 131.987 258.441
2

Cavity: 3.9624m x 3.048m Panel Size: 0.883025 Coarse Grid: 0.064m Panel Boundary: 0.0125cm start x 2expansion Cold Wall Bound. 0.0125cm start x 2expansion Wall Boundary 0.05cm start x 2 expansion x-dir fill cell: 0.059025m y-dir fill cell: 0.040625m Gridrefinement: 2 times Heat (W) 0.757 0.861 0.749 0.723 0.712 0.707 0.707 0.712 0.722 0.740 0.785 0.970 1.673 2.015 2.112 2.068 19.263 Cell No 4611 4612 4613 4614 4615 4616 4617 4618 4619 4620 4621 4622 4631 4704 4705 4706 Heat Flux (W/m ) 11.843 13.469 11.714 11.303 11.121 11.039 11.034 11.115 11.276 11.553 12.263 15.155 28.443 63.768 135.258 271.003
2

Width 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.059 0.032 0.016 0.008 0.883

Width 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.059 0.032 0.016 0.008 0.883

Heat (W) 0.758 0.862 0.750 0.723 0.712 0.707 0.706 0.711 0.722 0.739 0.785 0.970 1.679 2.041 2.164 2.168 19.474

Cavity: Panel Size: Coarse Grid: Panel Boundary: Cold Wall Bound. Wall Boundary x-dir fill cell: y-dir fill cell: Gridrefinement: Cell No 4730 4731 4732 4733 4734 4735 4736 4737 4738 4739 4740 4741 4751 4842 4843 4844

3.9624m x 3.048m 0.8829625 0.064m 0.00625cm start x 2expansion 0.00625cm start x 2expansion 0.05cm start x 2 expansion 0.058965m 0.0405625m 2 times Heat Flux (W/m ) 11.859 13.463 11.685 11.286 11.105 11.028 11.028 11.105 11.260 11.537 12.265 15.208 28.530 64.048 136.461 276.635
2

Cavity: 3.9624m x 3.048m Panel Size: 0.88293125 Coarse Grid: 0.064m Panel Boundary: 0.003125cm start x 2expansion Cold Wall Bound. 0.003125cm start x 2expansion Wall Boundary 0.05cm start x 2 expansion x-dir fill cell: 0.058965m y-dir fill cell: 0.0405625m Gridrefinement: 2 times Heat (W) 0.759 0.862 0.748 0.722 0.711 0.706 0.706 0.711 0.721 0.785 0.973 0.973 1.682 2.050 2.183 2.213 19.824 Cell No 4849 4850 4851 4852 4853 4854 4855 4856 4857 4858 4859 4860 4871 4982 4983 4984 Heat Flux (W/m ) 11.840 13.450 11.672 11.273 11.092 11.002 11.002 11.092 11.260 11.543 12.252 15.189 28.549 64.203 137.140 279.498
2

Width 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.059 0.032 0.016 0.008 0.883

Width 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.064 0.059 0.032 0.016 0.008 0.883

Heat (W) 0.758 0.861 0.747 0.721 0.710 0.704 0.704 0.710 0.721 0.739 0.784 0.972 1.682 2.054 2.194 2.236 19.591

179

Cavity: Panel Size: Coarse Grid: Panel Boundary: Cold Wall Bound. Wall Boundary x-dir fill cell: y-dir fill cell: Gridrefinement: Cell No 14342 14343 14344 14345 14346 14347 14348 14349 14350 14351 14352 14353 14354 14355 14356 14357 14358 14359 14360 14361 14362 14363 14364 14365 14380 14381 14438 14439 14440

3.9624m x 3.048m 0.88315 0.032m 0.025cm start x 2expansion 0.025cm start x 2expansion 0.05cm start x 2 expansion 0.029575 0.0204875 2 times Heat Flux (W/m ) 5.890 15.443 16.475 13.233 12.369 11.959 11.716 11.555 11.446 11.370 11.319 11.291 11.284 11.298 11.331 11.383 11.454 11.549 11.675 11.848 12.115 12.586 13.522 15.554 19.877 29.832 61.588 131.572 258.260
2

Cavity: 3.9624m x 3.048m Panel Size: 0.883025 Coarse Grid: 0.032m Panel Boundary: 0.0125cm start x 2expansion Cold Wall Bound.0.05cm start x 2 expansion Wall Boundary 0.05cm start x 2 expansion x-dir fill cell: 0.0295125 y-dir fill cell: 0.0203125 Gridrefinement: 2 times Heat (W) 0.188 0.494 0.527 0.423 0.396 0.383 0.375 0.370 0.366 0.364 0.362 0.361 0.361 0.362 0.363 0.364 0.367 0.370 0.374 0.379 0.388 0.403 0.433 0.498 0.588 0.882 1.971 2.105 2.066 19.115 Cell No 14566 14567 14568 14569 14570 14571 14572 14573 14574 14575 14576 14577 14578 14579 14580 14581 14582 14583 14584 14585 14586 14587 14588 14589 14606 14607 14680 14681 14682 Heat Flux (W/m ) 5.933 15.539 16.511 13.215 12.355 11.946 11.703 11.543 11.435 11.363 11.313 11.284 11.276 11.289 11.324 11.377 11.447 11.540 11.669 11.846 12.113 12.585 13.532 15.587 19.953 30.008 62.311 134.605 270.499
2

Width 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.030 0.030 0.032 0.016 0.008 0.883

Width 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.030 0.030 0.032 0.016 0.008 0.883

Heat (W) 0.190 0.497 0.528 0.423 0.395 0.382 0.374 0.369 0.366 0.364 0.362 0.361 0.361 0.361 0.362 0.364 0.366 0.369 0.373 0.379 0.388 0.403 0.433 0.499 0.589 0.886 1.994 2.154 2.164 19.317

180

Cavity: Panel Size: Coarse Grid: Panel Boundary: Cold Wall Bound. Wall Boundary x-dir fill cell: y-dir fill cell: Gridrefinement: Cell No 14790 14791 14792 14793 14794 14795 14796 14797 14798 14799 14800 14801 14802 14803 14804 14805 14806 14807 14808 14809 14810 14811 14812 14813 14832 14833 14924 14925 14926

3.9624m x 3.048m 0.8829625 0.032m 0.00625cm start x 2expansion 0.00625cm start x 2expansion 0.05cm start x 2 expansion 0.02948125 0.02028125 2 times Heat Flux (W/m ) 5.955 15.576 16.510 13.189 12.332 11.920 11.682 11.527 11.418 11.344 11.289 11.263 11.266 11.276 11.305 11.366 11.434 11.524 11.653 11.830 12.097 12.577 13.534 15.588 19.969 30.069 62.637 136.048 276.454
2

Cavity: Panel Size: Coarse Grid: Panel Boundary: Cold Wall Bound. Wall Boundary x-dir fill cell: y-dir fill cell: Gridrefinement: Heat (W) 0.191 0.498 0.528 0.422 0.395 0.381 0.374 0.369 0.365 0.361 0.360 0.360 0.361 0.361 0.362 0.364 0.366 0.369 0.373 0.379 0.387 0.402 0.433 0.499 0.589 0.886 2.004 2.177 2.212 19.398 Cell No 15014 15015 15015 15017 15018 15019 15020 15021 15022 15023 15024 15025 15026 15027 15028 15029 15030 15031 15032 15033 15034 15035 15036 15037 15058 15059 15170 15171 15172

3.9624m x 3.048m 0.88293125 0.032 0.003125cm start x 2expansion 0.003125cm start x 2expansion 0.05cm start x 2 expansion 0.0294825 0.02028125 2 times Heat Flux (W/m ) 5.926 15.537 16.465 13.141 12.303 11.904 11.646 11.492 11.389 11.311 11.260 11.234 11.234 11.260 11.298 11.337 11.402 11.505 11.633 11.814 12.084 12.548 13.489 15.563 19.956 30.069 62.766 136.715 279.318
2

Width 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.029 0.029 0.032 0.016 0.008 0.883

Width 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.032 0.029 0.029 0.032 0.016 0.008 0.883

Heat (W) 0.190 0.497 0.527 0.421 0.394 0.381 0.373 0.368 0.364 0.362 0.360 0.359 0.359 0.360 0.362 0.363 0.365 0.368 0.372 0.378 0.387 0.402 0.432 0.498 0.588 0.886 2.009 2.187 2.235 19.419

181

F.2

2D and 3D Model Validation

The data shown below was obtained from STAR-CD simulating the ASHVE test chamber. From the data obtained, the Nusselt, Grashof and Prandtl numbers were calculated and compared to the ASHVE test facility to validate the 2D test chamber model. The same data below was used to approximate the actual 3D case. As explained in Chapter 4, the panel width was calculated when the Nusselt number change was below 1%. The areas highlighted are where the boundary was measured. The overall Nusselt number was calculated by taking the area weighted average. Included in this Appendix are the numerical model results for a ceiling temperature at 318K. The University of Waterloo, Solar Lab can be contacted if the other panel temperature numerical model results are requested. The following Excel Worksheet summarizes model validation for a panel temperature of 318K. Table F.1 summarizes the results from every panel temperature simulated.
Panel Temp = 318K
Tair (K)
B B

297.81 318 307.90

Tpanel (K)
B B

Tavg (K)
B B

Properties of Air @ Tavg 1.1351 ρ (kg/m3) β (1/K) 0.0032 k (W/mK) 0.0269 µ (Ns/m2) 1.88E-05
B B P P P P

h (W/m2K)
P P

0.2539 46.3706 2.767E+11 0.7056

Nu Gr Pr

Cp (J/kgK) L (m)
B B

1007.3161 4.9102

Air Temp measured at 5ft from bottom surface - cell # 13926 Cell # 21080 21081 21090 q" (W/m2) 197.4264 96.2879 46.8428
P P

Width (m) 0.0080 0.0160 0.0219

q (W) 1.5794 1.5406 1.0241

182

0320 0.0320 0.4341 0.1171 0.7361 3.1187 0.7619 3.6974 3.8327 3.0646 4.4124 4.1210 0.9422 3.2166 0.1204 0.7812 3.0066 3.1412 0.1206 0.6717 3.1439 0.1323 0.7361 3.1187 0.0320 0.0320 0.0320 0.0320 0.0320 0.0320 0.1274 0.4962 4.1262 0.7425 3.8971 3.1336 0.1208 0.6588 3.1233 0.0320 0.0320 0.0320 0.0320 0.6459 0.1268 0.1214 0.0320 0.6781 3.0320 0.0320 0.8198 3.0320 0.0320 0.1290 0.9164 3.1198 0.1367 0.0320 0.1177 0.8520 3.7866 0.0320 0.0320 0.8247 4.1239 0.8005 3.1196 0.5659 8.2192 4.0320 0.7103 3.1480 0.0320 0.1822 0.1175 0.8842 3.1179 0.7700 5.0320 0.1222 0.1177 0.0320 0.1196 0.0320 0.0320 0.8714 3.0320 0.0320 0.6781 3.9809 3.0320 0.1191 0.9615 3.7103 3.6846 3.1243 0.1226 0.0320 0.1253 0.0320 0.1301 0.0320 0.1741 4.1389 0.6588 3.0320 0.0320 0.1033 4.6588 3.0320 0.0320 0.1282 0.7232 3.0320 0.7747 3.0320 0.0320 0.1544 0.1183 0.1171 0.1171 0.7941 3.5809 13.0320 0.9408 6.6250 4.7683 3.1355 4.2861 0.1216 0.0320 0.2707 4.0320 0.0320 0.0320 0.0320 0.1645 0.0320 0.0320 0.6943 5.1313 0.0324 4.1350 0.0320 0.0320 0.1167 183 .3416 4.1247 0.0320 0.22931 22932 22933 22934 22935 22936 22937 22938 22939 22940 22941 22942 22943 22944 22945 22946 22947 22948 22949 22950 22951 22952 22953 22954 22955 22956 22957 22958 22959 22960 22961 22962 22963 22964 22965 22966 22967 22968 22969 22970 22971 22972 22973 22974 22975 22976 22977 22978 22979 22980 24.1403 4.

1125 0.0320 0.1123 0.5171 3.1146 0.5428 3.0320 0.1123 0.1119 0.5171 3.5042 3.0320 0.5750 3.0320 0.1150 0.1150 0.1134 0.6266 3.0320 0.1136 0.0320 0.0320 0.1163 0.1142 0.0320 0.1121 0.0320 0.5235 3.0320 0.0320 0.0320 0.5686 3.5042 3.0320 0.1138 0.0320 0.1125 0.0320 0.0320 0.5815 3.0320 0.1134 0.4913 0.0320 0.22981 22982 22983 22984 22985 22986 22987 22988 22989 22990 22991 22992 22993 22994 22995 22996 22997 22998 22999 23000 23001 23002 23003 23004 23005 23006 23007 23008 23009 23010 23011 23012 23013 23014 23015 23016 23017 23018 23019 23020 23021 23022 23023 23024 23025 23026 23027 23028 23029 23030 3.5428 3.5042 3.5042 3.5557 3.0320 0.0320 0.5106 3.1148 0.6330 3.0320 0.5300 3.5879 3.4977 3.1125 0.1161 0.6330 3.6266 3.5106 3.0320 0.1117 184 .5557 3.0320 0.0320 0.0320 0.0320 0.5171 3.5944 3.5171 3.5557 3.1125 0.5493 3.1130 0.0320 0.1134 0.1144 0.5428 3.1158 0.1163 0.0320 0.0320 0.5171 3.0320 0.0320 0.5364 3.1142 0.5750 3.1161 0.0320 0.1125 0.1121 0.5944 3.1144 0.1128 0.1144 0.0320 0.5106 3.5686 3.0320 0.5428 3.0320 0.1156 0.1121 0.5300 3.5428 3.1123 0.0320 0.1121 0.0320 0.0320 0.0320 0.6008 3.5042 3.0320 0.1138 0.0320 0.0320 0.1130 0.0320 0.6266 3.5171 3.1125 0.5042 3.1134 0.1130 0.5300 3.1138 0.1134 0.0320 0.1140 0.0320 0.6137 3.1121 0.5750 3.1152 0.0320 0.0320 0.1161 0.6201 3.1121 0.5300 3.0320 0.5622 3.1130 0.0320 0.1132 0.0320 0.

0320 0.1121 0.4913 3.0320 0.1121 0.1119 0.0320 0.0320 0.0320 0.5235 3.1117 0.1117 0.5042 3.1119 0.4977 3.0320 0.4977 3.1121 0.0320 0.4913 3.5171 3.0320 0.0320 0.0320 0.1117 0.1117 0.4913 3.0320 0.1117 0.1119 0.5042 3.1119 0.1117 0.0320 0.4913 3.4977 3.1117 0.0320 0.1121 0.0320 0.1123 0.0320 0.0320 0.1119 0.0320 0.1119 0.5042 3.0320 0.1117 0.0320 0.1130 185 .4977 3.0320 0.0320 0.5042 3.1117 0.0320 0.0320 0.1128 0.1117 0.0320 0.5042 3.0320 0.0320 0.4977 3.1128 0.0320 0.1117 0.1130 0.1117 0.5235 3.4977 3.4913 3.1128 0.0320 0.0320 0.1128 0.0320 0.4913 3.4977 3.0320 0.4913 3.1119 0.4977 3.0320 0.4977 3.1121 0.4913 3.1117 0.0320 0.1119 0.1119 0.0320 0.5235 3.4913 3.0320 0.1125 0.0320 0.0320 0.4913 3.1119 0.1117 0.4977 3.0320 0.1123 0.1119 0.1119 0.0320 0.4977 3.1119 0.1123 0.5300 0.0320 0.4977 3.1119 0.4977 3.4977 3.5106 3.5235 3.4913 3.0320 0.5171 3.4913 3.4913 3.1117 0.1117 0.4913 3.0320 0.0320 0.1125 0.1123 0.5106 3.4977 3.0320 0.0320 0.4913 3.1123 0.5235 3.5106 3.5106 3.1119 0.0320 0.4913 3.0320 0.0320 0.1128 0.5300 3.23031 23032 23033 23034 23035 23036 23037 23038 23039 23040 23041 23042 23043 23044 23045 23046 23047 23048 23049 23050 23051 23052 23053 23054 23055 23056 23057 23058 23059 23060 23061 23062 23063 23064 23065 23066 23067 23068 23069 23070 23071 23072 23073 23074 23075 23076 23077 23078 23079 23080 3.5106 3.0320 0.0320 0.

1130 0.6201 3.0320 0.0320 0.1208 0.0320 0.0320 0.0320 0.5557 3.1158 0.1148 0.0320 0.0320 0.7876 3.6588 3.5622 3.5944 3.0320 0.5428 3.1171 0.6073 3.0320 0.0320 0.0320 0.1134 0.0320 0.0320 0.0320 0.5622 3.6523 3.5750 3.0320 0.0320 0.0320 0.1175 0.6459 3.0320 0.1134 0.0320 0.6266 3.0320 0.6008 3.6395 3.0320 0.1146 0.1177 0.1142 0.0320 0.5879 3.1156 0.5750 3.0320 0.0320 0.7747 3.0320 0.8069 0.5300 3.1167 0.7425 3.5686 3.7490 3.1189 0.6717 3.0320 0.1200 0.0320 0.1179 0.1130 0.1148 0.1138 0.0320 0.6910 3.5300 3.1132 0.1167 0.0320 0.0320 0.6073 3.1212 0.1136 0.7554 3.0320 0.1187 0.5815 3.6846 3.1161 0.0320 0.1140 0.7296 3.1152 0.1140 0.23081 23082 23083 23084 23085 23086 23087 23088 23089 23090 23091 23092 23093 23094 23095 23096 23097 23098 23099 23100 23101 23102 23103 23104 23105 23106 23107 23108 23109 23110 23111 23112 23113 23114 23115 23116 23117 23118 23119 23120 23121 23122 23123 23124 23125 23126 23127 23128 23129 23130 3.0320 0.7683 3.0320 0.1202 0.0320 0.6781 3.0320 0.5493 3.6201 3.0320 0.0320 0.0320 0.6459 3.5622 3.7168 3.0320 0.0320 0.5557 3.1134 0.5300 3.1198 0.0320 0.7039 3.1185 0.1150 0.1158 0.6781 3.5428 3.0320 0.7103 3.5879 3.1177 0.0320 0.0320 0.1218 186 .6137 3.1154 0.1181 0.5364 3.0320 0.1165 0.5428 3.1206 0.1193 0.1144 0.1144 0.0320 0.0320 0.0320 0.1138 0.1140 0.6201 3.1169 0.1146 0.0320 0.5815 3.1130 0.1154 0.1158 0.

2707 4.1542 0.2168 0.1339 5.9036 3.3351 4.9809 4.9422 3.0131 4.0320 0.8557 96.1480 0.9615 3.4124 4.1311 0.5723 24.9229 3.0320 0.1301 0.1243 0.9537 13.8842 3.1321 0.0320 0.0160 0.4343 0.0320 0.5797 5.0646 4.0320 0.1249 0.1269 187 .6943 6.0320 0.0320 0.0320 0.0320 0.1441 0.5873 46.5026 4.1255 0.0320 0.0320 0.6250 4.0320 0.0320 0.0320 0.1387 0.4585 0.1292 0.1241 0.1367 0.2192 4.1262 0.1229 0.1237 0.0320 0.1822 0.0320 0.1336 0.0320 0.0320 0.4517 0.0244 1.3073 197.7765 8.1350 0.0320 0.8183 5.1268 0.8069 3.0320 0.1274 0.7868 1.8778 3.0320 0.5409 1.0388 4.8134 3.0080 7.0968 4.2865 0.1220 0.0320 0.0320 0.0320 0.0320 0.1290 4.8392 3.0219 0.8649 3.0320 0.0320 0.23131 23132 23133 23134 23135 23136 23137 23138 23139 23140 23141 23142 23143 23144 23145 23146 23147 23148 23149 23150 23151 23152 23153 23154 23155 23156 23157 23158 23159 23160 23169 23242 23243 3.1284 0.0320 0.1412 0.1741 4.1643 0.0320 0.0320 0.1218 0.

459 11.708819 51.380 11.449 64.514 Nu 63.266E+11 log(Nu) 1.291 11.666 1.459 11.575 63.400E+11 2.8843655 66.754E+11 3.805 1.810 1.71 1.388E+11 Nu 45.66 1.407 GrPr 1.266E+11 log(Nu) 1.988 46.864 48.685 log(GrPr) 11.242E+11 1.242E+11 1.275 11.952E+11 2.5654229 48.2559586 64.81 1.687 1.784 62.952E+11 2.2861384 60.9778819 58.6E+11 2.311E+11 3.380 11.82E+11 1.663 1.365 11.876E+11 3.52 11.832 GrPr 1.689 1.645 48.53 Nu 62.799 1.291 11.876E+11 3.41 11.094 11.82 log(GrPr) 11.69 1.317E+11 2.514 188 .57E+11 2.884E+11 2.805 log(GrPr) 11.944 65.0693448 GrPr 1.44 11.78 1.415 11.77 1.400E+11 2.816 1.368 48.Table F.3734835 45.795 1.26 11.094 11.1: Model Validation Results Star CD Simulation (2D Approx) Panel Temp (K) 308 318 328 338 348 (3D Approx) Panel Temp (K) 308 318 328 338 348 ASHVE Data log(Nu) 1.

60623 0.200 71.665 159.032 0.032 0.275 2.911 1.981 3.571 49.358 54.803 400.366 36.2429625 323 Heat Flux (W/m ) 12.446 1.62942 Tair (K) 290.029 0.032 0.3 Numerical Model Results The numerical model results for the 0.980 48.912 0.016 0.016 2 Width 0.859 86.296 1.008 0.181 53. Panel Size = 0.029 0.057 318.016 0.547 55.342 67.008 0.159 97.401 2 Width 0.147 1.2429625 318 Heat Flux (W/m ) 10.032 26.197 3.181 87.407 1.032 0.794 42.299 Cell No 14810 14811 14812 14813 14832 14833 14924 14925 14926 290.032 0.485 2.657 1.032 0.029 0.032 0.856 0.029 0.010 1.630 138.840 199.799 28.346 0.243 Heat (W) 0.758 28.25m panel size have been included.441 59.029 0.032 0.626 3.529 78.029 0. Temperature range from 318K to 363K Panel Size: Tpanel (K) Cell No 14810 14811 14812 14813 14832 14833 14924 14925 14926 0. Solar Lab.032 0.597 1.032 0.807 2.51501 0.870 46.545 2.645 Tair (K) Panel Size: Tpanel (K) Cell No 14810 14811 14812 14813 14832 14833 14924 14925 14926 290.084 2 Tair (K) Panel Size: Tpanel (K) Width 0.805 277.F.131 3.594 1.606 43.243 Heat (W) 0.008 0.032 0.029 0.032 0.407 1.55 0.164 2.25m. Results for the other panel sizes simulated can be obtained from the University of Waterloo.029 0.032 0.324 1.076 1.032 0.115 41.032 0.156 107.724 1.223 48.166 1.032 0.993 0.634 36.524 3.179 118.741 Cell No 14810 14811 14812 14813 14832 14833 14924 14925 14926 290.032 0.118 240.008 0.034 Cell No 14810 14811 14812 14813 14832 14833 14924 14925 14926 0.016 0.032 0.032 0.016 0.531 1.2429625 343 Heat Flux (W/m ) 20.510 31.540 1.101 2.008 0.221 2.526 Tair (K) 290.803 3.029 0.450 2.029 0.469 1.032 0.032 0.780 3.970 55.008 0.574 1.377 359.275 441.160 67.241 1.032 0.440 38.960 220.355 2 Width 0.542 54.2429625 333 Heat Flux (W/m ) 16.032 0.032 0.334 49.387 43.243 Heat (W) 0.721 179.853 3.873 63.032 0.570 33.356 1.643 1.032 0.2429625 328 Heat Flux (W/m ) 14.2429625 338 Heat Flux (W/m ) 18.000 1.854 1.884 2 Panel Size: Tpanel (K) Width 0.029 0.483 1.109 2 Tair (K) Panel Size: Tpanel (K) Width 0.011 Tair (K) Panel Size: Tpanel (K) Cell No 14810 14811 14812 14813 14832 14833 14924 14925 14926 290.016 0.651 38.248 77.032 0.833 482.668 31.029 0.742 1.243 Heat (W) 0.032 0.455 3.032 0.032 0.6503 189 .931 43.920 49.016 0.870 2.5801 0.980 37.243 Heat (W) 0.719 33.243 Heat (W) 0.173 1.707 51.039 61.

080 607.952 83.2429625 353 Heat Flux (W/m ) 24.099 1.473 2 Width 0.008 0.032 0.320 2 Tair (K) Panel Size: Tpanel (K) Width 0.259 128.032 0.815 2 Panel Size: Tpanel (K) Width 0.138 2.032 0.256 2.688 72.191 94.707 5.587 64.243 91.824 159.785 2.7379 190 .032 0.583 79.315 324.241 Cell No 14810 14811 14812 14813 14832 14833 14924 14925 14926 290.032 0.243 Heat (W) 0.766 4.032 0.306 2.032 0.371 Cell No 14810 14811 14812 14813 14832 14833 14924 14925 14926 0.184 5.549 70.6936 0.806 1.392 138.029 0.2429625 358 Heat Flux (W/m ) 26.915 2.029 0.721 1.032 0.516 4.254 Tair (K) 290.Panel Size: Tpanel (K) Cell No 14810 14811 14812 14813 14832 14833 14924 14925 14926 0.678 2.938 303.160 4.243 Heat (W) 0.032 0.032 0.032 0.859 110.346 73.032 0.589 148.595 66.032 0.6712 0.499 77.430 2.242 565.032 0.243 Heat (W) 0.032 0.068 75.533 65.739 65.850 2.032 0.008 0.436 4.441 59.341 4.904 2.029 0.562 61.243 Heat (W) 0.912 2.2429625 363 Heat Flux (W/m ) 28.032 0.032 0.029 0.903 85.2429625 348 Heat Flux (W/m ) 22.008 0.601 56.029 0.016 0.218 2.618 282.71652 Tair (K) 290.061 2.098 5.029 0.532 59.092 71.491 523.016 0.032 0.016 0.581 75.184 4.750 2.290 1.483 2.029 0.524 102.195 118.274 Tair (K) Panel Size: Tpanel (K) Cell No 14810 14811 14812 14813 14832 14833 14924 14925 14926 290.338 261.008 0.016 0.332 2 Width 0.971 1.523 4.107 4.812 69.083 2.002 649.849 4.032 0.029 0.

Appendix G Methodology to Determine the Convective Heat Transfer Coefficient 191 .

the convective heat flux versus temperature difference follows a fairly linear equation. 140 120 100 80 qc (W/m2) 60 0.1.1) For the smallest panel size.6821 (G.94 1.2) 192 . a linear fit gives the following equation.61 1.27 1.92 2.66 40 20 0 20 30 40 50 ∆T 60 70 80 Figure G.17 2.42 2.46 0. with an r-squared value of 0.5674(T p − Ta ) + 4.G.44 1.1 Deriving the Correlation to Predict the Convective heat Flux versus Panel Temperature.9999: q c = 1.1: Convective Heat Flux versus Panel minus Air Temperature for Various Panel Sizes A correlation for each panel size could be written of the form: q c = A(T p − Ta ) + B (G. Air Temperature and Panel Size As shown in Figure G.

0249 1.3706 B 0. there is a minor change in coefficient “B”.5153 0.6109 1.1603 1.For the largest panel size.0311 1. From these equations it can be seen that the A and B coefficients vary significantly with panel size.0362 1.1 could be correlated using an equation of the form: q c = A(T p − Ta ) B (G.0148 1.8476 (G.9334 0. To simplify.9364 1.4559 0.4579 0. the linear fit gives the following equation with an r-squared value of 0.1: Coefficients for power law fit to each panel size simulated Characteristic Length (m) 0.4509(T p − Ta ) − 0.9999: q c = 0.3) The A. the heat flux plots in Figure I.2683 1.0078 1.1 in excel. showing that the fit accurately predicts the simulated heat flux values.0709 0. Coefficient “A” and “B” would be correlated with the characteristic length for each panel to ensure an accurate fit is obtained. Table I.1667 2.407 0. Table G.7876 0. With two coefficients dependant on panel size.6887 0. since it is an exponent to the temperature difference. B and r-squared values were obtained from applying a linear fit to each curve in Figure G.4) Applying the power fit to each panel size.4441 1. the correlated equation can be very large and complex to use.4248 2.6609 A 2. The r-squared value varies from 0. it will also be correlated to the effective panel diameter.0399 193 .9787 0.9997 to 1 for each equation.9200 2.6155 0.1 below summarizes the coefficients for the power fit for each panel size.999 1. Although coefficient B does not vary greatly over the panel sizes.

Equation G. However. to show coefficient A can be approximated with an equation of the form: A = FDe G (G.2).9937x 2 R = 0.5 -1.3. coefficient A was plotted against the effective panel diameter (Figure G. for each panel size a different correlation is required.5) 2.9999 0. knowing the panel to air temperature difference.0046 0 0 0. Length (m) 2 2. coefficient B can be correlated of the form: B = H ln( De ) + K (G. From power fit trend line has been added to the plot.4 predicts the convective heat flux. From a logarithmic fit to the curve.5 3 Figure G.5 A 1 y = 0.6) 194 .5 1 1.5 Char.2: Coefficient A versus Characteristic Length Coefficient B was similarly plotted versus the panel characteristic length and is shown in Figure G. To include the effects of panel size in the heat flux correlation.5 2 1.

qcbar B B (qcorr .4.06 y = 0.qcbar)2 was done for each value B B B B P P r 2 ∑ (q = 1− ∑ (q corr − qc ) 2 c − q cbar ) 2 195 .9966 1.96 0.9832 2 R = 0. Length (m) 2 2.5 1 1.92 0 0.1.3: Coefficient B versus Panel Characteristic Length Combining the Equations G.02 1 B 0. r. The r-value was found as follows: • • • • The numerical model convective heat fluxes was averaged.98 0.5 3 Figure G.04 1.5 Char.9937 De 1.7) The correlation coefficient.9997. is derived in excel below.qc)2 was done for each correlated value B B B B P P (qc . G.0615 ln( De ) +0.9832 (G.0046 (T p − Ta ) 0.0615Ln(x) + 0.94 0.6. The r-value was found to equal 0.5 and G. the convective heat flux can be approximated with the following equation: qc = 0.

5344097 21.35762358 26.9245505 78.2552171 24.48378437 27.570528 57.76912914 18.10772274 0. qc (W/m2) 47.26817036 68.0343448 27.36136737 0.417309 670.0122242 21.8540735 Corr.7559929 46.73048641 25.7408514 46.68178 17.07220042 0.174579 556.8144712 101.08990315 23.200233 296.82707112 63.qc (W/m2) 48.58106 19.13889459 0.382872 334.979157 372.3114105 22.60025245 24.22002639 0.5838354 48.0958423 26.24096713 32.4449196 17.507982 187.5287303 24.043431 0.01356447 0.7715147 12.31884243 15.08674339 0.8642779 18.04256115 36.65524774 12.8048258 20.00345106 1.65006804 87.02777983 0.03051223 0.1312888 86.11885107 0.359786 320.040319 510.05718325 0.73359756 18.34798151 36.70026 80.9088992 0.16875572 0.1766117 20.8407772 18.058181 241.82706237 20.03649949 0.231176 619.99858 272.08333494 0.07206423 0.2093E-05 0.5920565 63.4649819 248.04413159 0.4643155 55.628784 487.025585 96.00062297 0.13348493 0.24293878 0.1618075 25.8382168 29.90178272 16.10774903 0.2313369 123.3510059 36.809253 377.0519872 20.6705121 21.8063976 33.20751909 0.1594247 117.484024 385.79303613 41.47642044 33.2838043 14.6641007 50.10888495 0.7438274 71.3982283 17.19260445 1.850894 322.249656 410.12331728 0.01359374 0.72737 191.1837016 40.5320697 19.qc)2 0.203438 11.75727848 20.728265 673.283743 30.09029999 0.09307576 (qc .42741794 28.15983432 0.645184 32.044505 16.30213806 16.31534406 0.36341775 79.2809515 23.17889702 22.30154145 16.491745 26.01554065 0.52999092 24.47646354 11.7206672 36.5697556 17.0449067 13.7502964 19.01351802 7.76821881 36.54899025 (qcorr .29422712 0.583892 1712.738663 271.09513818 0.4074076 19.67254824 31.0914663 15.5264597 51.7654064 13.6109448 0.88187297 13.654105 467.05059116 0.589E-06 0.7524857 27.95471754 23.5691786 32.9803475 145.01529514 30.49408413 26.338016 564.97992157 18.00433838 0.8290705 17.2840369 13.89493683 17.60395809 29.26913191 1.3205241 0.58090772 55.14167776 0.83350212 40.231252 178.4163693 26.294454 254.05523186 0.0114066 41.8558679 22.12263971 0.42842477 20.26877287 14.29188163 0.90534012 17.60393292 0.0885752 71.7186353 15.qcbar)2 114.073786 211.0007488 0.64555263 21.693036 2417.754846 343.228902 176.02036965 0.00513345 0.9322096 29.088967 1132.25199903 196 .2991491 36.00022594 0.936484 444.4679728 11.8254461 20.

1643415 32.1106558 65.82674617 50.33572047 0.52894123 0.03964713 0.424748 107.08873289 43.04791439 37.1936339 64.9829759 29.3367425 29.13038851 0.00028965 0.270982 66.2420266 34.3560685 0.03682888 0.62213208 40.78492175 7.52299769 32.02563989 0.95805511 47.19616918 0.46833807 38.25087 816.99949712 0.642505 24.8625784 6544.450968 234.341024 162.2770316 31.84563406 32.7543672 41.25418392 103.01354619 0.8465324 69.32015 374.780443 172.18538141 35.2508068 9.916762 56.05325 98.31972344 29.03675924 25.301616 41.4225516 26.59450037 0.9296746 52.03438237 0.01181888 0.4126359 34.792428 53.0456855 0.41024745 1.13609268 0.9406909 114.60477687 0.254403 70.8344135 24.61483806 44.1791144 118.66971 349.16840069 29.99974853 33.2328154 22.75585232 0.15732318 0.5133683 35.33.9246E-07 0.7406751 43.1833396 0.19506831 0.596214 45.6670041 47.34043 557.15200063 51.13188561 33.4535676 37.27402 60.62945524 22.840055 42.20365687 0.7632676 0.32771001 0.90915997 55.9762858 48.8397279 29.219773 224.06687355 0.888144 110.9139146 15.62970864 29.01526179 0.4978055 27.241258 65.6519179 44.22500166 0.13331194 34.38743062 48.31717736 0.5391198 12.3646512 62.35061675 1.8621101 109.34675046 0.8954314 38.1182223 26.70192437 46.32239465 28.46473476 0.368508 4214.3372174 60.03498057 0.914666 127.18240875 0.9213543 114.9057423 32.22378163 32.5611322 27.21234439 0.3649537 94.608456 198.09121403 25.52902397 34.04410809 8.56985856 111.733523 3250.2349453 37.50260865 0.710873 51.0829405 40.2986896 26.4937691 9.4019156 197 .9473168 95.42160687 26.4907539 37.52297619 55.670012 40.89365851 118.29948123 0.08466014 36.6630323 24.2187284 40.03857414 29.51057329 0.00319987 10.16126123 43.00283555 0.33914676 1.577444 5312.7238972 34.54059 1126.6136783 102.8701247 67.78168018 0.10757538 0.20304578 0.96854181 4.4292078 47.058807 44.0813323 31.19710845 0.5879681 29.3707132 56.6855095 31.

2228 0.2: Coefficients for power law fit to each aspect ratio simulated Aspect Ratio 0.5) versus the aspect ratios and correlated to give the following equations: A= 0.201 AR 0.3763 0.9334 0.2 Deriving the Correlation to Predict the Convective heat Flux versus Panel Temperature. The coefficients are summarized in Table G. coefficients A and B were determined from Figure G.2551 0.9787 0.3706 B 0.4409 0.999 1.5055 A 2.6155 0.0399 Coefficients A and B are then plotted (Figure G.0249 1.0709 0. Assuming the power fit curve similar to the previous derivation.1905 0.0808 198 . Air Temperature and Panel Size To develop an equation used to predict the natural convective heat flux with respect to the radiant panel to ceiling ratio.1 for various aspect ratios simulated.8) (G.1340 0.0148 1.7 can be done.7876 0.8375 (G.0311 1.0362 1.0078 1.3197 0. Table G. a similar derivation to Equation G.407 0.1603 1.9) B = 0.4 and G.G.4559 0.5153 0.0613 0.0511ln( AR) + 1.6887 0.2 below.

5 0.0511Ln(x) + 1.4 0.4: Coefficient A versus AR 1.5 -0.9978 0.8375 0 0 0.3 Aspect Ratio 0.96 0.2 0.5 A 1 y = 0.2.06 y = 0.1 0.04 1.6 Figure G.5 0.5: Coefficient B versus Aspect Ratio 199 .98 0.0808 2 R = 0.2 0.02 1 B 0.94 0.6 Figure G.5 2 1.92 0 0.3 Aspect Ratio 0.201x 2 R = 0.9881 1.1 0.4 0.

9.8375 (G.10) The correlation coefficient for Equation G. G.0808 AR 0.Combining Equations G. the convective heat flux can be approximated using the following correlation: qc = 0. 200 .7.0511 ln( AR ) +1.4.8 and G.201 (T panel − Tair ) 0. since similar coefficients were used for the correlation derivation.10 is equal to the one found for Equation G.

Appendix H Effects of Panel Insulation on Panel Performance 201 .

0254m Panel Temp (oC) 90 85 75 65 55 45 90 85 75 65 55 45 90 85 75 65 55 45 90 85 75 65 55 45 90 85 75 65 55 45 Enclosure Wall Temp (oC) 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Enclosure Cold Wall Temp (oC) 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 Enclosure Air Temp (oC) 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 hr (W/m K) 7.820 6.288 9.598 12.107 8.052 6.000 0.896 0.575 1.000 0.150 3.000 0.052 6.04W/mK t=0.900 0.896 0.150 3.820 6.299 6.820 6.847 13.658 14.083 11.889 0.820 6.713 7.04W/mK t=0.889 0.598 12.325 7.598 0.473 7.893 0.789 9.299 12.608 7.147 20.598 12.575 1.150 3.608 7.889 0.900 2 hbl (W/m K) 1.893 0.548 7.04W/mK t=0.000 0.325 7.093 10.575 1.888 0.812 20.698 10.900 0.242 14.000 2 UL (W/m2K) 9.652 6.575 1.888 0.598 12.598 12.363 11.893 0.The Excel sheet below summarizes the effects of insulation thickness on the overall heat transfer coefficient Insulation Detail k=0.513 14.896 0.013 13.959 20.473 7.652 6.893 0.052 6.891 0.652 6.575 3.889 0.900 0.150 3.299 6.003175m k=0.891 0.511 11.891 0.123 9.893 0.541 20.889 0.325 7.888 0.150 6.660 14.900 0.473 7.214 7.325 7.508 k=0.936 9.299 6.608 2 hc (W/m K) 0.943 7.518 9.04W/mK t=0.473 7.808 20.299 6.652 6.608 7.325 7.891 0.312 20.891 0.863 10.299 6.888 0.052 6.608 7.00635m k=0.150 3.04W/mK t=infinite 202 .820 6.361 8.652 6.0127m k=0.888 0.052 6.896 0.575 1.000 0.896 0.473 7.

Appendix I Reverse Solar Collector Model Panel Temperature and Panel Heat Output Results 203 .

306 0.585 221 (13 ft) (2.05664 4 0.1 Reverse Solar Collector Model Results The Reverse Solar Collector Model was run for at the experimental inlet fluid temperatures for the 24-4 panel and 24-8 panel. RSC Model for 24-4 Panel Radiant Panel Variables: Enclosure Dimensions: Length (m) Width (m) Height (m) Emmisivity Radiant Panel Dimensions Length (m) Width (m) Thickness (m) Char Length (m) Radiant Panel Details # of Tubes Flow Rate (USGal/min.1.885 0.057 0.9624 3.015875 0.1524 0.056782 0.08 0.401 3.625 0.238 0.014859 0.9624 0. m) kpanel (W/mK) Panel Insulation k (W/mK) Thickness (m) Bond Conductance k (W/mK) Thickness (m) Width (m) View Factors Panel to Floor Panel to Side Walls Panel to Front Wall Panel to Back Wall 3.9624 3.I.048 0. The complete results are summarized in Table I. m) Innner Tube Diameter (in.0167 204 . The results are shown below.002 1. kg/s) Tube Spacing (in.6096 0.0009 0.0 ft) 0.9 6 0. m) Outer Tube Diameter (in.0127 1/2in 1.5 0.

565 3079.888722 13.6704 78.8819 0.297 68.565 3079.896306 0.833 0.67 0.88874 0.393897 6.67 0.833 27.890844 0.297 27.833 27.8814 0.936662 378.57 19.78 51.50828834 46.828 0.888048 0.92007913 75.6717 78.337622 6.26912 7.833 27.67 hfi (W/m K) 3079.53314 13.67 0.96243 13.833 27.45705 14.565 3079.297 27.125581 86.67 0.260387 54.8768 0.38867 7.3 19.01727837 86.297 205 .94 18.08672 14.9539 0.6054 90.260389 70 64.9521 0.896714 6.017279 90 86.833 0.67 0.297 68.833 0.12558053 86.9521 0.695154 6.66 86.70591434 64.35 73.827 0.26038885 64.833 27.510455 46.9528 0.297 68.93 20.91 73.1622 48.706675 64.9533 0.4843 56.35 61.67 0.900194 0.12564071 86.565 3079.955 13822 68.556235 586.Fluid Inlet o Temp ( C) 51.01724317 75.65128 0.565 3079.4677 56.833 27.518953 586.67 0.67 0.67 0.695154 6.565 3079.297 68.236165 54.695141 6.458926 478.9539 0.565 3079.900193 0.6054 FR qpanel (W/m) Panel Temp o ( C) (Calculated) 46.8794 0.45413 14.8819 0.890864 0.269122 0.8877 13.78 61.67 0.604 78.8814 0.26 19.337618 6.297 68.828 0.3 19.833 0.833 27.562587 0.50828834 54.833 0.833 27.565 3079.896662 7.827 0.565 3079.558698 586.8777 0.896306 0.508288 57 54.565 3079.57 19.57 6.26 19.9533 0.8846 0.96383 13.833 27.827 0.297 68.565 3079.58852 13.9521 13822 13822 13822 13822 13822 13822 13822 13822 13822 13822 13822 13822 13822 13822 13822 68.8842 0.900194 0.61 20.8819 0.771054 6.125641 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 19.890844 0.53267 13.508288 46.8805 0.53314 14.8805 0.297 27.833 27.829 216.6704 78.93 20.26 19.67 0.91 86.9537 0.9517 0.833 27.9537 0.565 3079.1637 48.4 100.6704 90.955 13822 68.833 F ReD NuD kwater (W/mK) 0.028813 86.253755 482.829 216.66 86.1622 56.35 61.89456 13.829 216.640372 64.78 51.920079 75.039643 6.26 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12.458718 586.383246 478.222605 378.297 68.67 0.89456 13.253751 378.898651 0.9533 0.89455 13.297 68.896306 0.833 27.8777 0.27 20.67 0.27 20.4 Panel Enclosure Enclosure Enclosure hr hc UL Cb o Temp ( C) Wall Temp Cold Wall Air Temp 2 2 2 (W/m K) (W/m K) (W/m K) (W/mK) o o o (Estimate) ( C) Temp ( C) ( C) 50 46.890843 0.893945 6.9537 0.829 216.67 0.955 13822 68.565 3079.4843 56.8777 0.705905 64.771054 7.66 86.5942 67.67 0.460394 478.9548 13822 68.828 0.93 20.57593 14.35 61.67 0.833 0.94 18.61 12.556296 27.565 3079.705914 80 74.607 90.771071 6.08363 13.827 0.94 20.565 3079.896661 6.829 216.695154 6.888722 0.8814 0.1622 48.565 3079.893551 0.955 13822 68.67 0.297 68.37778 291.4 100.9533 0.833 27.08402 14.94 18.70590539 64.96382 13.9537 0.769648 6.892203 0.893551 0.12802344 86.93 18.889821 0.337618 6.61 12.9539 0.8814 0.827 0.02881296 86.236165 54.565 3079.565 3079.565 2 F' Outlet Fluid Temp o ( C) 48.888722 0.386217 291.3 19.8805 0.833 27.823 292.8846 0.3 19.689833 6.892219 6.297 27.823 0.51045493 46.5937 67.53314 13.08677 14.9533 0.67 0.7066752 64.882 0.5392 90.297 0.6511 0.45713 14.893551 0.128023 86.823 0.5489 67.27 20.67 0.78 51.4843 67.12563922 0.70591424 74.89456 13.828 0.508293 46.67 0.833 27.35 61.91 73.895409 0.8846 0.565 3079.45706 27.337142 6.65128 0.5937 78.386146 291.96382 14.297 68.9521 0.297 27.377936 291.67 0.8819 0.297 68.67 0.386147 380.4844 56.826 0.458954 590.22868 14.01912792 75.297 68.900194 0.297 68.565 3079.565 3079.833 27.8805 0.833 0.822 0.833 27.9539 0.833 0.893568 0.8846 0.66 86.896315 0.8805 0.266181 7.6053 90.94 18.9539 13822 13822 13822 13822 13822 68.61 12.565 3079.64037238 64.91 73.57 19.27 20.832 0.61 12.93 20.4 100.254117 378.833 0.67 0.01727906 75.565 3079.57 19.297 68.5937 67.4 100.08672 14.297 68.67 0.019128 75.823 0.651279 0.2605932 54.833 0.297 68.260593 54.26 19.78 51.8777 0.50828835 46.91 73.66 100.27 19.437496 478.833 27.297 68.3 19.26038886 54.269194 7.50829274 46.297 68.565 3079.017243 75.1622 48.26038713 54.

057 0.9624 3. m) Innner Tube Diameter (in.056782 0.048 0.0013 0.014859 0.05664 8 0.306 0.885 0.6096 0.0762 0.08 0.9624 3.9 3 0.9624 0.0 ft) 0.24-8 RSC Model Radiant Panel Variables: Enclosure Dimensions: Length (m) Width (m) Height (m) Emmisivity Radiant Panel Dimensions Length (m) Width (m) Thickness (m) Char Length (m) Radiant Panel Details # of Tubes Flow Rate (USGal/min. m) Outer Tube Diameter (in.5 0.0167 206 .401 3. kg/s) Tube Spacing (in.015875 0.0127 1/2in 1.625 0. m) kpanel (W/mK) Panel Insulation k (W/mK) Thickness (m) Bond Conductance k (W/mK) Thickness (m) Width (m) View Factors Panel to Floor Panel to Side Walls Panel to Front Wall Panel to Back Wall 3.238 0.585 221 (13 ft) (2.002 1.

31844784 89.486476 70 66.52 20.8743 398.9903 0.42079082 89.565 3079.9382406 14.9907 13822 6.8772 0.269231 0.9337 67.89062346 0.316 615.021 610.13 20.89558797 13.9907 0.67 hfi (W/m K) 3079.89480901 13.58372209 0.11 99.0512818 14.05807 0.0303 0.9907 0.05807 0.67 0.565 3079.77 62.8738 400.9907 13822 6.6958759 13.99 13822 68.89913732 0.52 20.8765 6.6958759 13.317 497.297 207 .565 0.297 68.565 3079.47528 89.8687 0.66 99.8687 0.96 51.939 609.8743 398.0259377 67.89301256 0.486475 57.021 19.67 0.04973865 57.85906401 7.67 0.67 0.99 13822 68.67 0.01 21.80555 77.76480496 77.354 3079.419029 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 20.52 20.5703417 14.37472 89.269231 0.0512819 14.486566 57.9904 0.8743 398.9899 13822 68.06754 48.297 68.01 21.9355 0.47096592 57.66 99.201 497.8772 0.68 6.8718 0.297 68.049742 48.049739 60 57.52 20.297 68.58501242 0.58209803 7.89300529 0.88756945 0.269231 19.269231 19.89559297 13.9302 0.67 0.418997 89.269231 19.9907 13822 13822 13822 13822 13822 68.11 86.99770586 67.269231 19.269231 0.89913867 0.9356 0.99 13822 68.96 51.99 20.286 233.102 233.80553 77.269231 19.71 73.80555 89.8687 503.023 610.297 68.67 0.934 67.269231 19.96 51.778256 19.75145856 7.297 68.02619643 67.026196 67.38193795 7.51066108 6.763491 95 89.8703 0.8772 0.895588 13.9905 0.017 3079.565 0.297 19.049739 48.96 51.764805 77.269231 0.68 21.947 314.06754 57.565 3079.68 21.565 3079.67 0.565 0.8718 0.67 0.04973864 48.71 73.48647579 66.934 67.269231 0.77 62.9270421 6.8885301 0.663535 77.67 0.51084 57.102 233.297 68.51084 57.9356 0.88851239 14.99 19.63020968 0.04662274 48.49558 57.9905 0.77 62.66 99.269231 0.269231 19.8687 0.67 0.77 62.046623 48.3745774 14.297 68.9356 0.934 67.9907 0.13 20.917 610.76349067 89.02593768 77.9907 13822 6.9326 0.025935 67.318448 89.269231 0.06447 48.9907 0.41902852 89.297 0.9356 0.18215049 7.354 3079.942 314.8765 0.67 0.89913733 0.67 0.297 19.420791 89.01 21.9907 13822 6.58345046 0.297 19.269231 0.6958759 19.1185179 14.52 21.9309 77.297 68.96 62.41899706 89.565 3079.565 3079.9309 0.9903 0.7727365 14.050666 14.565 3079.67 0.269231 19.99 19.89913733 13.9903 13822 13822 13822 13822 13822 13822 13822 13822 13822 13822 3079.67 0.9352 0.565 3079.565 3079.85844084 6.3720076 14.470966 57.297 68.67 0.297 68.269231 0.565 2 F' Outlet Fluid Temp (oC) 48.8771 0.565 3079.269231 F ReD NuD kwater (W/mK) 0.71 73.89226318 0.0259353 67.7727882 14.7727374 19.297 68.77 73.565 3079.11 86.8718 0.58501151 7.297 19.9905 0.06754 48.67 0.88851208 0.71 86.89062318 0.49752596 0.9903 0.9326 0.8765 0.89300522 0.49752596 6.565 3079.89300529 0.565 0.297 68.66 99.67 0.88851239 0.18474087 7.269231 0.71 73.8765 0.947 Panel Temp (oC) (Calculated) 48.9326 0.565 3079.85906395 6.48656632 57.297 68.9309 0.05833 0.7785227 19.269231 19.4864758 57.04973865 48.297 68.67 0.76347338 77.239 497.8242313 19.85906972 6.269231 19.6958616 13.947 314.297 68.41902796 0.565 0.89832346 0.58506351 7.47351 89.763473 77.06754 48.66353458 77.47354 0.9906 13822 6.565 3079.339 3079.70728 77.04974227 48.565 0.297 68.88919112 0.11 86.316 497.48647527 57.8743 398.11 86.354 3079.297 68.9309 0.9352 0.0512875 14.5837221 0.269231 19.99 13822 68.565 3079.49751029 6.9326 0.7785227 19.9352 0.67 0.102 315.3746112 14.51093 57.269231 0.9352 0.99 19.89062346 0.935 0.269231 19.8718 0.Enclosure Enclosure Enclosure Panel Inlet Fluid 2 2 2 Temp (oC) Wall Temp Cold Wall Air Temp hr (W/m K) hc (W/m K) UL (W/m K) Cb (W/mK) o Temp ( C) (oC) Temp (oC) (oC) (Estimate) 51.67 0.05807 0.80684 77.51084 FR qpanel (W/m) 233.01 21.3745769 14.7698407 14.895588 13.47355 89.8674 0.68 21.862 314.565 3079.8761 0.18477541 7.68 21.9318 0.9905 0.67 0.269231 19.934 67.7785243 19.66 50 48.565 3079.89064451 0.01 21.102 233.67 0.67 0.13 20.13 19.13 20.58372369 0.9901 0.997706 67.7634909 77.18474132 7.99 19.297 68.7081971 13.025938 85 77.8772 0.49752598 6.

25 75.98 207.71 86.215171189 2.66 Experiment Mean Plate Heat Output Temp (oC) (W/m) 48.08 RSC Model Mean Plate Temp Heat Output (oC) (W/m) 48.56 374.46 700.03 66.288524819 10.89 613.46 86.29 RSC Model Mean Plate Panel Heat Temperature Output (W/m) o ( C) 46.91 86.32 89.35 77.68 77.56 Temperature Error 1.10 57.18 75.66 100.49 314.413389777 -0.503050477 4.71 378.89 55.017351347 -2.088407553 2.51 216.05 233.68 56.77 73.004069085 1.92 490.16877502 12.61 87.18 298.02 Temperature Error 0.Table I.75 553.76 497.4 Experimental Mean Plate Panel Heat Temperature Output (W/m) o ( C) 46.11 99.74 86.35 73.65 54.239913061 Heat Output Error 10.95 67.91932009 5.358890588 24-8 Inlet Fluid Temp (oC) 51.1 261.02 478.46 332.86432854 208 .145780823 2.104493461 -1.03 398.13 586.879687854 Heat Output Error -4.818058438 -0.42 610.666566056 -0.226013377 1.26 291.96 62.1: Summary of the 24-4 and 24-8 RSC Model Results 24-4 Fluid Inlet Temperature (oC) 51.78 61.47 64.39 64.189043239 0.373889823 -1.75 407.

33 1.41 0.60 63.08 70. the two simulations run with a bond thickness of 0.68 71.3mm 46.03 53.91 61.2.24 60.25 77.00 60.53 52.1mm 44.14 61.2 Effect of Bond Thickness on Panel Performance To determine the effect of bond conductance on the mean radiant panel temperature and heat output.74 1.00 90. 209 .69 52.00 0.10 62.60 77.29 71.93 1.3 at 90oC.39 68.03 79. Table I.91 54.86 52.35 62.6mm 45.5mm 45.00 80.55 53.38 61.15 0.20 53.8mm and 1.50 69.2mm 44.66 68.3mm is shown in Table I. The results are summarized in Table I.91 76.30 78.9mm 45.2: Effect of Bond Thickness on panel Temperature Bond Thickness (mm) Fluid Inlet Temperature (oC) 50.16 To see how the heat output is affected by the bond conductance.73 54.7mm 45.62 62.56 0.98 71.40 79.09 54.37 52.4mm 45.I.86 62.3mm 44.29 0.00 70.78 80.37 53.58 76.38 70.25 76.46 60.22 69. the RSC model was run at various bond thicknesses.54 1.85 0.79 69.98 0.0mm 44.8mm 45.94 68.69 60.95 78.66 78.

94 507.27 19.16 19.27 19.61 19.26 19.84 76.90 78.57 19.98 488.61 12.22 14.22 31.31 518.17 19.16 76.22 14.58 14.3: Effect of Bond Thickness on Panel Heat Output Fluid Inlet Temperature (oC) Bond Thickness (mm): 90 90 90 90 90 Bond Thickness (mm): 90 90 90 90 90 Panel Temp Enclosure Wall Enclosure Cold Enclosure Air (oC) (Estimate) Temp (oC) Wall Temp (oC) Temp (oC) 0.31 31.31 507.31 31.3 90 75.26 12.58 14.61 12.61 12.61 12.16 76.26 19.57 19.57 14.93 507.57 19.26 19.15 1.27 19.15 78.57 19.31 31.61 12.31 31.93 UL (W/m2K) Cb (W/mK) qpanel (W/m) 210 .8 90 77.79 488.26 12.61 12.57 19.55 488.17 14.57 19.22 14.79 19.61 12.17 14.57 19.27 499.73 507.26 19.57 19.Table I.26 19.15 78.61 19.57 14.26 19.61 12.26 19.80 488.16 14.27 19.26 19.