EPA-600/R-96-116 September 1996 LARGE BUILDING HVAC SIMULATION Final Report by Lixing Gu, Muthusamy V. Swami, and Vailoor Vasanth Florida Solar Energy Center 300 State Road 401 Cape Canaveral, FL 32920 DCA Contract 93-RD-66-13-00-22-009 EPA Contract 68-DO-0097, Work Assignment 3-12 EPA Project Officer: Marc Y. Memetrez National Risk Management Research Laboratory Research Triangle Park, NC 27711 Prepared for: State of Florida Department of Community Affairs 2740 Centerview Drive Tallahassee, FL 32399 and U. S. Environmental Protection Agency Office of Research and Development Washington, DC 20460 ------- FOREWORD The U. S. Environmental Protection Agency is charged by Congress with pro- tecting the Nation's land, air, and water resources. Under a mandate of national environmental laws, the Agency strives to formulate and implement actions lead- ing to a compatible balance between human activities and the ability of natural systems to support and nurture life. To meet this mandate, EPA's research program is providing data and technical support for solving environmental pro- blems today and building a science knowledge base necessary to manage our eco- logical resources wisely, understand how pollutants affect our health, and pre- vent or reduce environmental risks in the future. The National Risk Management Research Laboratory is the Agency's center for investigation of technological and management approaches for reducing risks from threats to human health and the environment. The focus of the Laboratory's research program is on methods for the prevention and control of pollution to air, land, water, and subsurface resources; protection of water quality in public water systems; remediation of contaminated sites and groundwater; and prevention and control of indoor air pollution. The goal of this research effort is to catalyze development and implementation of innovative, cost-effective environmental technologies; develop scientific and engineering information needed by EPA to support regulatory and policy decisions; and provide technical support and infor- mation transfer to ensure effective implementation of environmental regulations and strategies. This publication has been produced as part of the Laboratory's strategic long- term research plan. It is published and made available by EPA's Office of Re- search and Development to assist the user community and to link researchers with their clients. E. Timothy Oppelt, Director National Risk Management Research Laboratory EPA REVIEW NOTICE This report has been peer and administratively reviewed by the U.S. Environmental Protection Agency, and approved for publication. Mention of trade names or commercial products does not constitute endorsement or recommendation for use. This document is available to the public through the National Technical Information Service, Springfield, Virginia 22161. i a ------- ACKNOWLEDGEMENTS The authors thank the U.S. Environmental Protection Agency (EPA) and contract manager, Marc Menetrez, for funding this work. We also thank the Florida Department of Community Affairs (DCA) and planning manager, Mo Madani, for co-funding the work. The very helpful cooperation of Southern Research Institute (SRI), Bobby E. Pyle, Ashley D. Williamson and Susan E. McDonough is especially acknowledged. Thanks are due to Michael Anello of the Florida Solar Energy Center for his helpful work. ii ------- TABLE OF CONTENTS ACKNOWLEDGEMENTS ii LIST OF FIGURES iv LIST OF TABLES v EXECUTIVE SUMMARY 1 1. BACKGROUND 3 1.1 FSEC 3.0 Capabilities 3 1.2 Scope of Present Work 4 2. RADON TRANSPORT AND HVAC SYSTEM 6 2.1 Introduction 6 2.2 Radon Transport and Pressure Equation 6 2.3 HVAC System (Duct and multizone airflow and pressure) 7 2.4 Zone Radon Balance Equations 10 3 PRELIMINARY SIMULATION 11 3.1 Introduction 11 3.2 Simulation Procedure 11 3.3 Preliminary Simulation Results 11 3.4 Closure 12 4 VALIDATION 18 4.1 Introduction 18 4.2 Geometry Description of Soil and Concrete Slab 18 4.3 Simulation Results Compared to Full Airflow 18 4.4 Simulation Results in a Typical School Day 27 4.5 Closure 27 5 PARAMETRIC STUDY 30 5.1 Introduction 30 5.2 Varying Outdoor Airflow 30 5.3 Varying Ambient Radon Level 30 5.4 Varying Soil Radium Content 30 5.5 Closure 30 6 CONCLUSION 35 7 REFERENCES 36 iii ------- LIST OF FIGURES Figure 1-1. FSEC 3.0 software structure and interfaces 5 Figure 3-1. Schematic of air conditioning plan at Polk Life and Learning Center 13 Figure 4-1. Schematic of three dimension mesh configuration 19 Figure 4-2. Zone configuration of Polk Life and Learning Center 20 Figure 4-3. Indoor radon level comparison at Cafeteria 24 Figure 4-4. Indoor radon level comparison at Room 109 24 Figure 4-5. Indoor radon level comparison at Room 102 25 Figure 4-6. Indoor radon level comparison at Audiology 25 Figure 4-7. Indoor radon level comparison at Conference Room 26 Figure 4-8. Indoor radon level comparison at Conference Room in a typical school day 28 Figure 4-9. Indoor radon level comparison at Cafeteria in a typical school day 28 Figure 4-10. Indoor radon level comparison at Room 109 in a typical school day 29 Figure 4-11. Indoor radon level comparison at Room 102 in a typical school day 29 Figure 5-1. Effect of outdoor airflow on indoor radon levels (0 pCi/L ambient) 32 Figure 5-2. Effect of outdoor airflow on indoor radon levels (4 pCi/L ambient) 32 Figure 5-3. Effect of ambient radon level on indoor radon levels (1000 cfm) 33 Figure 5-4. Effect of ambient radon level on indoor radon levels (2000 cfm) 33 Figure 5-5. Effect of soil radium concentration on indoor radon levels (0 pCi/L) 34 Figure 5-6. Effect of soil radium concentration on indoor radon levels (4 pCi/L) 34 iv ------- LIST OF TABLES Table 3-1. Comparison of airflow rate between design and simulation 14 Table 3-2. Comparison of airflow rate between measurement and simulation at testing condition 16 Table 4-1. Multizone and terminal airflow rate from simulation of air distribution system 21 Table 4-2. Comparison of indoor pressures between simulation and measurement 22 Table 4-3. Comparison of indoor radon levels between simulation and measurement ... 22 Table 4-4. Radon entry rate from different zones [Bq/s] 23 v ------- EXECUTIVE SUMMARY This report represents work performed by the Florida Solar Energy Center (FSEC) for the Environmental Protection Agency (USEPA-No: 68-D0-0097) and the Florida Department of Community Affairs (DCA-No: 93-RD-66-13-00-22-009). Although individual tasks were funded separately by the two agencies, this report, for the sake of completeness, represents the combined efforts of all simulation related tasks. Project goals: The primary goal of the project was to establish the potential for using models to analyze radon levels in large buildings. This was done by applying modelling tools, developed in earlier work and integrated in the computational platform FSEC 3.0, to analyze pressures, airflows and indoor radon levels in a school building monitored by the US EPA and the SRI. Discussion of effort and results The effort of the US EPA contract is to simulate pressures, airflows, and radon levels in the Polk Life and Learning Center at Bartow, Florida, monitored by the US EPA and Southern Research Institute (SRI). First, only the air distribution system of the school building monitored by EPA was simulated to obtain and refine the distribution system parameters. This was done by trial and error while adjusting values of the distribution system parameters and comparing the results with the "test and balancing report" provided by Associated Air Balance Council. After adjustments, the differences between measured and predicted airflows were less than 5%. Next, a steady-state simulation of the soil/slab composite was carried out and the results were compared with experimental data. Because of the nature of the boundary conditions over the slab, a 3-D discretization was required to model the soil/slab composite correctly. Soil/slab parameters were adjusted by trial and error to obtain a reasonable match between predicted and measured values of pressures and airflows. Results of the steady state simulation comparison with measured indoor radon levels agreed to within 6%. Due to paucity of detailed data, it is important to note that the adjusted material properties may not necessarily represent the true values and the calibration may not necessarily translate to other cases. Keeping the adjusted parameters obtained from earlier runs constant, the next step is to compare measured and calculated indoor radon levels for a transient seven-hour period and a "typical school day" where the system was "on" for the first 12 hours and "off" for the rest of 12 hours. The figure compares histories of predicted and measured indoor radon levels, in one station, for a "typical school day". It is evident that while the agreement at the beginning and end of the "on" cycle is good, the model predicts higher radon dilution rates during the "on" cycle than shown by the experiment. However, the model and experiment compare very well during the "off" period. The disparity noted during "on" times appears consistently in all zones. This is 1 ------- a significant cause for concern and is possibly due to two factors. 1) The model assumes well mixed zones which may not be true in actuality. The ventilation efficiency may not be 100% leading to different radon levels within a zone and a single-point measurement may be insufficient. 2) The ambient radon level may be higher than assumed. Due to the unavailability of data on ambient radon levels, we assumed a constant of 3.5 pCi/L for simulation purposes. Results of other work for the FRRP (see Tyson et al., 1993) show that ambient radon levels may not only be higher than established action levels, but may also vary cyclically during a 24-hour day. Clearly, the model would predict lower rates of dilution and would approach measured values if higher ambient radon levels are used in the simulation. Undoubtedly, these two factors namely, ventilation efficiency and ambient radon levels, must be investigated further before answering the question definitively. Next, parametric analysis of the effect of varying outdoor airflow, ambient radon level and soil radium content was carried out for this specific building. Indoor radon level decreases with increasing outdoor airflow through the air distribution system, due to dilution. When ambient radon level and soil radium content are varied, there appears to be a linear relationship between indoor radon level and ambient or soil radium content occurs. This determination is specific to the building studied and is based on assumptions stated in the report and may not necessarily translate to other similar buildings. Caveats: It is crucial to note that the nature of the work performed here is an exploratory one primarily to establish the potential of using models to analyze large buildings and to identify the essential areas for experiment and simulation to compliment each other in providing an accurate, yet cost efficient strategy to study radon in large buildings. This objective was substantially achieved through a preliminary simulation of airflows and pressures in a school building monitored by the US EPA and the SRI. Since only a limited set of experimental data were available, several assumptions were made to successfully complete the simulations. The results presented in this report, should therefore, be viewed in light of the assumptions stated and applied only to the specific problem analyzed. The result should in no way be construed to represent generalizations for large-buildings. The present report concludes with a list of areas that need further attention. Polk Life & Learning Cenfer Rm. 109 (Stotion 2) 4/21-4/22 20 I 8 2 ------- 1. BACKGROUND 1.1 FSEC 3.0 Capabilities Under support from DCA, Florida Solar Energy Center (FSEC) developed and integrated radon transport in the soil and slab, HVAC system operation, multizone airflow, and zonal contaminant balance into Florida Software for Enervironmetal Computation (FSEC 3.0, 1992). FSEC 3.0 has the following capabilities: Zone thermal balance Zone moisture balance Zone contaminant balance, including radon Heat and moisture transport in envelop Multizone airflow, including air distribution system Several HVAC system models, including VAV box performance Duct heat and moisture exchange Radon transport in soil and slab Detailed air movement in space, used for investigation of ventilation effectiveness In addition to the above capabilities, FSEC 3.0 offers the following features that make it a promising computational framework for integration of the various models: Performs transient or steady 1, 2 or 3-D simulations The main computational processor is based on the Galerkin finite element methodology. This lends itself well to irregular shapes and boundary conditions Program has already been designed to accommodate up to 250 governing equations. Radon transport equations have been incorporated. Several choices for modeling combined heat and mass transport in building are available. This feature is critical to accurately predicting latent loads, indoor conditions and A/C run times in hot humid climates. Program allows the user to modify time steps, material properties and boundary conditions on a run-time basis. This is especially important when properties and boundary conditions are functions of space, time, or field variables. A building simulator performs the heat and mass balance calculations for the building zones. Subroutine slots are already available to link with other interzone airflow codes. Can be run in both PC and VAX/VMS based environments 3 ------- Many of the capabilities of FSEC 3.0 derive from the software structure itself. The general architecture of the software is given in Figure 1-1. The .Computational Processor Segment (CPS) is the heart of the software. It performs the following major operations: Computes the capacitance, stiffness and Jacobian matrices and force vectors on an element basis, using numerical volume and surface integrations. Assembles the element matrices and force vectors. Solves the resulting linear or nonlinear algebraic equations. This portion of the software can be independently executed without interfacing with User Defined Programs (HDP). The buildings simulator is connected to the CPS through a common interface. Similarly, other UDPs can be connected to the CPS through this interface. UDPs are stand-alone software elements (subroutines); they may get some inputs from the CPS and return some outputs to the CPS. For instance, the building simulator gets surface temperatures and moisture conditions from the CPS and returns the zone air temperatures and moisture conditions to the CPS through the interface. During each iteration or time step, certain parameters can be modified through JJser Defined Routines (UDR). These modifications can be local or global (see Figure 1-1). Local modifications are performed on an element level - i.e. field variable dependent material properties and/or boundary conditions. Global modifications are performed at the beginning of an iteration or time step. Examples of global modifications are time dependent material properties or boundary conditions, variable time-step simulations, numerical solution schemes (direct iteration versus Newton type iterations) etc. 1.2 Scope of Present Work The U.S. EPA and the SRI monitored and collected data of indoor pressures and radon concentrations in a large school building at Bartow, Florida. Data under several test conditions were obtained. FSEC used the integrated computational software, FSEC 3.0, to simulate HVAC system and multizone airflows, indoor pressures, radon transport in the soil and slab and indoor radon levels in the large building. The simulation was validated by measured data. A limited parametric study shows the influence of outdoor airflow, ambient radon level and soil radium content on indoor radon levels. 4 ------- INPUT PROCESSOR DOMAIN SPECIFICATION Geometry B.C. Bags Material sat flags MESH GENERATION PROGRAM EQUATION & B.C. SPECIFICATION o From library o User defined X MATERIAL PROPERTY SPECIFICATION c From library o Uaer defined HI MASTER CONTROL o Solution atralegy o Simulation parameters o Output oontroi ir VECTOR ARRAY FOR STORAGE (STATIC PORTION) (DYNAMIC PORTION) I COMPUTATIONAL PROCESSOR SEGMENT (CPS) T J BUILDING SIMULATOR UDP*1 , USER DEFINED PROGRAM UDP #2 I GLOBAL DYNAMIC ! MODIFIER o Numerical solution control o Simulation parameter control LOCAL DYNAMIC MODIFIERS o Variable properties o Variable B.C. values o Variable source/link terms "oT =L !USER DEFINED PROGRAM UDP #3 . UDP # n , ETC. . . N T E R F A C E USER INPUT T 1 OUTPUT PROCESSOR Figure 1-1. FSEC 3.0 software structure and interfaces. 5 ------- 2. RADON TRANSPORT AND HVAC SYSTEM 2.1 Introduction The governing equations for radon transport and diffusion in soil and slab, radon balance, multizone airflows and zone pressures are presented in this Chapter. Pressure and radon transport equations in soil and slab were primarily obtained from information and sources provided by Rogers & Associates Engineering, Inc. (Rogers & Nielson, 1991). The air distribution system model was integrated from A1RNET, developed by the National Institute of Science and Technology (Walton, 1989). Since these mathematical formulations can be found in the references, only brief descriptions are given in this Chapter. 2.2 Radon Transport and Pressure Equation The pressure equation, derived from Darcy's equation for flow through a porous media, is given by (Yuan & Roberts, 1981): dP P <3t 0 = V- K VP (2-1) It should be noted that Darcy's law is valid for a Reynolds number I^K < 1 (Cheng, 1985), where ReK is Reynolds number based on air permeability and defined as pvKI,2//u. Radon concentration balance (Rogers & Nielson, 1991) including multiphase radon generation and transport in porous media may be expressed as: dC K - V-D VC ---VP-VC ~kC +RpAE (2-2) -v_ c a a a ~ c v OT (i where P Pressure fPa] P0 Reference pressure [Pa] t Time [s] K Bulk air permeability in porous media [m2l fj. Dynamic air viscosity [1.8x10"5 Pa.s] C. Radon concentration [Bq/m3] Dc Effective radon diffusion coefficient [nf/sj Kc Effective air permeability in porous media [nfj k 22Rn decay constant [2.1x10" s '] R Soil :26Ra concentration [Bq/kgJ p Bulk dry density [kg/nrj 6 ------- E,. Effective 2~Rn emanation coefficient [dimensionless) 2.3 HVAC System (Duct and multizone airflow and pressure) Mathematical formulations of several elements of the HVAC system used in the present simulation are listed below. Power law element - CRACKS Based on the power law, the airflow through a cracks is expressed as m - C (AP)"1 = C (P P)"' (2-3) ij mjv ' mvp i y y ' where m is mass flow rate fkg/sl. Cmj is the flow coefficient at the j-th crack and AP is pressure difference across the crack, "i" indicates the i-th zone where air flow enters and "j" indicates the j-th zone or specific ambient condition where air flow leaves, "i^" is exponent of flow equation at the j-th crack. For simplicity, it is assumed that there is one crack connected between i-th and j-th zones in the brief description. Therefore, the j-th crack is located between i-th and j-th zones. However, multiple cracks between two zones are allowable in the integrated FSEC 3.0 If it is assumed that J cracks exist in the i-th zone, the crack air flow in the i-th zone may be written as follows <2 - Cm2(P, -p/> (2-4) "g - c«/pi " pj)"' m , - C ,(P. PK)"J i, J m,Jv i K7 Based on mass conservation, total mass flow should be equal to zero in the steady-state condition, that is J " Y, riy = 0 (2-5) j i By substituting the air flow of each crack in the i-th zone, Eq. (2-4) into Eq. (2-5). the air flow 7 ------- of the i-th zone may be rewritten as 3.) 1 -> C ,(P P.) 2 - ... + C (P. - P.) J + ... + C ,(P \J m,2 ^ i 2' mjv i y m,Jv i (2-6) In general, the expression of the i-th zone air flow may be written as fi(P,,P2,...,Pjt...,PJ) - 0 (2-7) Duct Svstem The pressure loss in ducts due to friction is given by APf - f L pv2 D 2 (2-8) where f Friction factor L Duct length D Hydraulic diameter v Velocity The dynamic losses due to the fitting is , 2 AP. - c d o pv (2-9) where C0 Dynamic loss coefficient The total pressure loss in a duct is AP - APr + £ Ap^ (2-10) Rewriting the above equations in terms mass flow rather than velocities, one obtains m 2pA f-'Ec ID 1.7 AP 1'2 (2-11) where A is the cross section area and f can be calculated by using the non-linear Colebrook 8 ------- equation I fU2 1.44 - 2.0In 2.0 In 1+- 9.3 Re-f"2 D (2-12) and where e Surface roughness Re Reynolds number Reynolds number is defined as Re pVD mD (2-13) An Fan Element Fan performance is normally characterized by a performance curve, which relates the total pressure rise to the flow rate for a given fan speed and air density. The performance curve may be represented by cubic polynomials. AP = ao + atm + a,ifi 2 + a3m ' (2-14) where AP fan total pressure rise = the fan total pressure at outlet minus the fan total pressure at inlet [PaJ a0 ... a3 Coefficients of the polynomial The performance of a given fan at various speed and air densities may be related to a single fan performance curve through the "FAN LAW" / \ N,P» n2P2/ (2-15) and AP, AP, ' > N N22P2; (2-16) 9 ------- where Q Volume flow rate [nrVs] N Fan rotational speed 2.4 Zone Radon Balance Equations The indoor radon balance equation at the i-th zone may be written as follows v cnttv.i + Q , (C - C ) + Y Q (C - C ) a,* a,r x/ xj iv aj a,K noz (2-17) where i-th zone j-th zone V Zone volume [m'] Ca Indoor radon concentration [Bq/nr] CaAmbient radon concentration [Bq/m3] Fei..iy Radon entry from the slab [Bq/s] Qj_i Indoor air flow from j-th zone to i-th zone [rrrVs] (Qj.;=0) Q,,f Infiltration from ambient [m3/s] noz Number of zones It should be noted that Qlllf is considered return flow to the building return plenum in the present simulation. The ambient radon concentration will be modified by combining all return flows from all zones with the outdoor air flow. Its expression is C Qqa^~ * ^2 Q»if,£a,! Q(JA Qjnf.i (2-18) where Qoa Outdoor airflow rate through the air distribution system [irr/s] Ambient radon concentration [Bq/m'] 10 ------- 3 PRELIMINARY SIMULATION 3.1 Introduction Before detailed simulation of indoor radon concentration and radon entry from the slab, the air distribution system should be simulated to calculate indoor pressures and multizone airflows. Based on the design data from the air conditioning plan by die Langbein & Bell Engineers and testing data from the testing and balancing report by the Associated Air Balance Council, the input file for the HVAC system simulation was created. By refining parameters of each component, acceptable results were obtained and compared with design and testing data. Indoor pressures, multizone airflow rates and terminal flow rates through the duct system were calculated in the HVAC system simulation. 3.2 Simulation Procedure The efforts for the preliminary simulation are described below. List and characterize all components of HVAC system Based on the air conditioning plan of the building, the HVAC system has to be discretized into a number of component elements used in the simulation. Components are composed of ducts with different cross section and lengths, VAV boxes, fans, etc. Individual VAV box or the fan is considered to be one element, and ducts with the same shape and cross section are also considered to be one element. The parameters of most elements required for the simulation were obtained from the ASHRAE handbook, the US EPA publications, or other sources. However, where the parameters of some elements are not known, a best guess was assigned for initialization and these parameters were adjusted in comparison to experimental data. Input file preparation When discretization of the HVAC system and characterization of each element are accomplished, the input file, which consists of the node number, element number, element type for different components, and nodal connectivity of each element, is created. Simulation and refinement By trial and error, adjustment of some parameters with the initialized best guess are made through test simulations. All the parameters used in the HVAC system are calibrated through refinement process to match measured data. 3.3 Preliminary Simulation Results The preliminary simulation results show the terminal airflow rate comparison between the design and simulation, and testing data and simulation, respectively. Constant inlet flow from the fan 11 ------- is assigned and indoor air pressures are set to zero gauge. The VAV boxes are assumed to be fully open. Figure 3-1 shows the schematic of the air conditioning plan of the building with terminal nodal numbers. Table 3-1 shows the comparison of design and simulation air flow rates based on the air conditioning design plan. The purpose is to create an input file with duct component parameters needed in the simulation for further refinement of component parameters. The first column indicates the node number, corresponding to each terminal listed in Figure 3-1. The second and thi'd columns show that airflow rates at each terminal, corresponding to the node number in the first column, from design and simulation, respectively. The fourth column lists the percent difference. A maximum 6.04% difference, as shown in the Table 3-1, was observed between prediction and design data. It should be noted that some design deficiencies were found and will be discussed later. The last row in Table 3-1 is total inlet air flow rates of design and simulation. Table 3-2 shows the comparison of measured and predicted airflow rates at each terminal, based on the Testing and Balancing Report. The Report presents real performance of the HVAC system for different VAV boxes and terminals. Due to the difference of HVAC system performance between testing and design, some component parameters are adjusted compared with the first simulation. The second and third columns list the measured and predicted airflow rates, respectively, corresponding to nodal number in the first column. The fourth column shows the percent relative difference. A maximum difference of 4.62% was obtained. The last row in Table 3-2 is total measured and predicted inlet airflow rates. It should be pointed out that discrepancy exists between design and testing. For example, maximum outdoor airflow rate is 1200 cfm, according to the air conditioning design plan. However, results from a recent testing and balancing report showed the maximum outdoor airflow of 3047 cfm. The performance of some VAV boxes from testing report differs from design. Therefore, corresponding adjustments are necessary. Since the testing data show the present HVAC system performance, parameters adjusted by comparison to testing data are used in subsequent simulation. 3.4 Closure Excellent comparison between testing or design and predicted airflow rates have been obtained. Airflow validation is based on the current parameters of duct, fan and cracks. Multizone airflows and indoor pressures will be used in calculation of radon entry through the slab and indoor radon level. Some changes, compared to design, of the building HVAC system are found. For example, main duct size, connected to terminals defined in Nodes 57, 59, 62, 64, 67, 69, 72 and 74 in the Cafeteria, were changed from 20" diameter to 15" diameter, (Figure 3-1), so that it is hard to achieve 500 CFM for each terminal, if the duct size connected to these terminals are the same. FSEC 3.0 can air in the design or redesign of the HVAC system. 12 ------- Oj AC-I V-8 r-1 V-18 3 ao V-7 l?4 6? v-et 94 45 V-13 *vo, [V-S -4 SO 71 It 74 40 V-3 V-J 41 Figure 3-1. Schematic of air conditioning plan at Polk Life and Learning Center. ------- Table 3-1. Comparison of airflow rate between design and simulation Node No. Design cfm Simu. cfm % diff. 8 175 173 -1.13 9 365 359 -1.38 13 140 139 -0.25 15 150 150 0.44 17 175 174 -0.40 18 275 276 0.42 23 115 115 0.07 27 200 202 1.46 28 230 236 2.61 33 230 228 -0.46 35 220 218 -0.47 36 70 70 0.08 40 225 220 -1.88 41 260 255 -1.80 45 130 132 1.84 49 75 77 3.56 50 50 51 2.68 54 120 123 2.72 57 500 509 1.87 59 500 509 1.87 62 500 500 0.12 64 500 500 0.12 67 500 520 4.20 69 500 520 4.20 17 500 510 2.06 (continued) 14 ------- Table 3-1 (continued) Node No. Design cfm Simu. cfm % diff. 74 500 510 2.06 78 145 148 2.62 80 85 86 2.14 84 210 209 -0.47 86 240 245 2.36 89 210 207 -1.39 91 240 230 -3.77 95 70 71 2.61 96 75 76 2.17 101 165 174 5.75 102 100 96 -3.01 104 100 93 -6.04 105 250 262 5.13 110 325 309 -4.69 113 325 315 -3.02 115 325 315 -3.05 119 210 210 0.34 120 200 199 -0.07 124 100 98 -1.06 126 50 48 -3.07 131 250 236 -5.44 132 250 242 -2.90 160 325 315 -2.96 Total 11455 11460 0.044 15 ------- Table 3-2. Comparison of airflow rate between measurement and simulation at testing condition Node No. Meas cfm Simu cfm % diff 8 80 81 1.87 9 175 178 1.93 13 150 151 0.73 17 130 131 1.35 18 220 223 1.44 23 105 107 1.91 27 125 127 2.23 28 130 133 2.40 33 140 143 2.31 35 95 97 2.31 36 35 35 2.55 40 165 162 -1.57 41 180 177 -1.57 45 120 120 0.64 49 65 65 0.46 50 40 40 0.69 54 100 97 -2.14 57 320 314 -1.67 59 320 314 -1.67 62 500 484 -3.12 64 490 484 -1.14 67 460 453 -1.45 69 480 471 -1.82 72 490 487 -0.56 74 502 -1.47 (continued) 16 ------- Table 3-2 (continued) Node No. Meas cfm Simu cfm % diff 78 80 80 0.80 80 60 60 1.35 84 140 143 2.19 86 140 143 2.49 89 135 138 2.63 91 135 138 2.59 95 50 51 2.39 96 50 51 2.50 101 100 99 -0.59 102 55 54 -0.40 104 55 55 1.53 105 150 151 1.12 110 225 229 2.16 113 225 233 3.93 115 225 233 3.79 119 150 156 4.14 120 140 143 2.76 124 110 114 3.73 126 50 52 4.62 131 145 144 -0.21 132 150 148 -1.00 160 225 233 3.96 Total 8421 8424 0.036 17 ------- 4 VALIDATION 4.1 Introduction Following the preliminary simulation of airflow in the HVAC system of Polk Life and Learning Center in Bartow. Florida, multizone airflows, zone pressures, indoor radon concentrations, and radon entry rates from the slab were simulated and compared with experimental data during certain time periods, using the integrated FSEC 3.0 software. This Chapter describes the comparison. It should be noted that calibration of parameters in the last Chapter is for the HVAC system only. The parameter calibration in this Chapter relates to radon transport in the soil and slab. After calibration, the simulation is compared to measured data for one typical school day. 4.2 Geometry Description of Soil and Concrete Slab Soil and Slab In order to correctly model radon entry, a 3-D soil and slab discretization becomes necessary to obtain radon entry and indoor radon levels for different indoor pressures at different zones. The schematic of 3-D mesh is shown in Figure 4-1. It should be noted that the large elements are used in the present simulation to reduce computational time. Since cracks are not discretized separately, weighted-average properties of air and concrete are used for the element. These properties will be adjusted based on the crack size at each element. Zone According to the experimental layout and observation from Polk Life and Learning Center, seven (7) zones are used in the present simulation, as shown in Figure 4-2. These zones are labeled Room 102 for Zone 1, Conference room for Zone 2, Cafeteria for Zone 3, Rm 105 for Zone 4, Audiology Room for Zone 5, Room 109 for Zone 6 and Corner room for Zone 7, respectively. It should be noted that measurement data are available in Zones 1, 2, 3, 5 and 6. 4.3 Simulation Results Compared to Full Airflow A special experiment is set up in order to control VAV box performance and establish the calibration for VAV boxes. Thermostats in all zones are set to cPF lower than the normal setting, and fan airflow is set to the maximum, so that VAV boxes in the duct system can be assumed to be fully open. Since the VAV box is controlled by temperature differences between zones and thermostats, the special setting was required to avoid adjusting parameters related to VAV box performance in the simulation. Time period for the experiment was from 12:30 PM to 7:00 PM on April 4, 1993. Figures 4-3 to 4-7 shows comparisons of simulations to measurements during this time period for the Cafeteria, Room 109, Room 102, the Audiology Room and the Conference Room, respectively. From the observation of measured data, mentioned by Marc Menetrez (EPA), and suggested by Bobby Pyle (SRI), the lowest 18 ------- ^ Concrete Slab f¦ f ' i l ! ! i i i i - r - i 1 - ^ i : ' ! III! .- i i I . r i M i j >i i . _ -I ! i i i = 1 1 i ___L-Kr 11 ^XSoil Figure 4-1. Schematic of three dimension mesh configuration. indoor radon concentration are comparable to the ambient radon level. In this case, outdoor radon concentration is set to 5 pCi/L. Further, during this validation, the radon entry rate can be calculated based on precalculated indoor pressures obtained by computing airflow rates through the air conditioning system. Two steps of the building simulation, steady-state and transient conditions, are used to refine the material properties of soil and concrete slab used in the input file. Validation at Steadv-State Condition The purpose of steady-state simulation is to adjust material properties of soil and slab, such as diffusion coefficient, emanation coefficient, air permeability, moisture content, etc. In other words, radon entry rate from the slab will be calibrated, comparing simulated indoor radon level with experimental data. As mentioned before, some material properties of the concrete slab, especially for radon diffusion coefficient and air permeability, represent the properties of combined concrete and crack by estimating crack size for each element of the slab. Therefore, as long as a reasonable comparison is obtained, the material property estimation is considered 19 ------- Polk County Life and Learning Center L_J ' © HVAC i Rm 109 \ CD Audio (5) J 1 Conf. 2 Cafe f5' Rm 105 a ! Rm 102 | 1 figure 4-2 Zone configuration of Polk Life and Learning Center. reasonable. The radon concentration and pressure distribution in the soil and slab will be used as the initial conditions for transient simulation during the seven (7) hour time period. It should be noted that the measured data of fan and outdoor airflows at 7:00 PM are used as steady-state inputs. It was observed from experimental data that fan and outdoor airflows change very little during the seven-hour period, so that these flows are essentially constant. Table 4-1 lists all airflows from the present simulation for the air distribution system. The first column gives nodal connectivity for each element used in the simulation. These elements represent those from terminals to zones, zone to zone, and zone to return plenum. Nodal connectivity shows element connection between two nodes, where the first node indicates the air inlet and the second the air exit. In the simple terminology, inlet node is called "from" and exit node is called "to". The negative sign indicates the airflow direction is opposite to the direction of nodal connectivity. The other three columns are airflow rates, expressed in different units. It is assumed that air density is 1.2 kg/m3. 20 ------- Table 4-1. Multizone and terminal airflow rate from simulation of air distribution system Nodal Connectivity Airflow Rate (fcg/s) Volume Flow Rate (cfm) Volume Flow Rate (m3/s) 173-179 -0.1871 -336 -0.1586 174-179 -0.2021 -364 -0.1717 175-179 -0.3299 -592 -0.2795 176-179 -0.0141 -25 -0.0119 177-179 -0.2712 -487 -0.2298 178-179 -0.0543 -98 -0.0460 173-180 0.7996 1,436 0.6776 174-180 0.7949 1,427 0.6736 175-180 0.7432 1,334 0.6298 176-180 0.0725 130 0.0614 177-180 0.7699 1,382 0.6525 178-180 0.7960 1,429 0.6746 179-180 0.8304 1,491 0.7037 173-178 0.0543 98 0.0460 173-174 0.0631 113 0.0535 177-178 -0.1668 -300 -0.1414 Table 4-2 lists the gauge pressures (relative to ambient) at different zones obtained from simulation and measurement. RAP and Amb indicate the return air plenum and ambient, respectively. All pressures are relative to the ambient pressure. Very good agreement between prediction and measurement has been achieved. In other words, the simulation results correctly reflect the HVAC system performance in the building. When the air handing unit is on, the zones are pressurized, as shown in Table 4-2. Evidently, advection of radon through cracks carried by air flow may be negligible with positive pressure in the building. The diffusion of radon through the slab is the main factor that affects indoor radon level compared to advection. 21 ------- Table 4-2. Comparison of indoor pressures between simulation and measurement Node Number Relativity Measured (Pa) 1.50 Simulated (Pa) 1.50 180 RAP-Arab. 172 Cafe.-Amb. 2.11 2.11 171 Rm 109-Amb. 2.19 2.19 169 Audio.-Amb. 1.95 1.81 167 Rm 102-Amb. 1.68 2.12 166 Conf.-Amb. 2.12 2.12 Table 4-3 shows good agreement between simulated and measured values of indoor radon levels. Although seven zones are used in the simulation, results of simulation show only indoor radon concentrations in five zones, because only five zones are measured in the experiment. Table 4-3. Comparison of indoor radon levels between simulation and measurement Zone Number Measured (pCi/L) Simulated (pCi/L) Cafe 5.8 6.10 Rm. 109 5.4 5.33 Audio 6.9 6.65 Rm. 102 6.3 6.26 Conf. 5.3 5.37 Validation at Transient Condition During the validation period, the transient simulation is from 12:30 PM to 7:00 PM on 4/4/93. It should be noted that indoor pressures, fan flow, and outdoor air airflows are assumed to be constant during the validation time period. These values are the same as those at the steady-state condition. The simulation results of radon and pressure distribution in the soil and slab at the steady-state condition are used as initial conditions in the transient simulation. From Table 4-4, the radon entry rate from the slab into different zones remain fairly constant, even though indoor radon levels vary due to outdoor air dilution. In other words, radon entry rate from the slab is affected only slightly by the indoor radon level. It should be noted that radon entry rate in the individual zone is equal to radon flux multiplied by the individual zone area. 22 ------- Table 4-4. Radon entry rate from different zones [Bq/s] 1 Time Zone 1 Zone 2 Zone 3 Zone 4 Zone 5 Zone 6 Zone 7 j 13.0 41.1394 31.8613 52..2105 24.3544 0.4623 35.3662 12.6480 13.5 41.3059 31.9482 52.3159 24.3976 0.4627 35.4351 12.6701 14.0 41.3919 31.9948 52.3778 24.4254 0.4630 35.4761 12.6817 14.5 41.4320 32.0189 52.4119 24.4436 0.4631 35.4996 12.6878 15.0 41.4472 32.0305 52.4292 24.4554 0.4632 35.5124 12.6909 15.5 41.4495 32.0351 52.4366 24.4631 0.4632 35.5186 12.6923 16.0 41.4456 32.0359 52.4381 24.4678 0.4632 35.5210 12.6926 16 5 41.4398 32.0350 52.4372 24.4708 0.4633 35.5212 12.6926 17.0 41.4336 32.0333 52.4348 24.4725 0.4633 35.5204 12.6922 17.5 41.4278 32.0312 52.4319 24.4734 0.4632 35.5189 12.6918 18.0 41.4226 32.0291 52.4286 24.4735 0.4632 35.5172 12.6912 18.5 41.4187 32.0272 52.4259 24.4734 0.4632 35.5156 12.6908 19.0 41.4155 32.0255 52.4235 24.4731 0.4632 35.5140 12.6904 23 ------- Polk Life & Learning Center, 4/4/93 Cafeteria 30 20 12.5 13 13.5 14 14.5 15 15.5 16 16.5 17 17.5 18 18.5 19 Tlrr* (hoirs) ¦ Measured Str*Jaf«d Figure 4-3. Indoor radon level comparison at Cafeteria. Polk Life & Learning Center, 4/4/93 Rm 109 30 25- 20- -1 15- a 10- 12.513 13.51414.51515.5 16 16 J1717.51B 18.5 19 Tim* (hart) ¦ Sfrr*ict«d Figure 4-4. Indoor radon level comparison at Room 109. 24 ------- Polk Life & Learning Center, 4/4/93 Rm 102 30 20- - 15- 12.513 13.51414.5 15 15.5 16 16.517 17.518 18.5 19 Tlrr* (hour?) ¦ Measured SrmJat»d rigure 4-5. Indoor radon level comparison at Room 102. Polk Life & Learning Center, 4/4/93 Audio 40 35 10- 12.513 13.51414.5 1515.5 1616.517 (7.518 18.519 Tfcm (hcxn) ¦ M#o*red Smdatod Figure 4-6. Indoor radon level comparison at Audiology. 25 ------- Polk Life & Learning Center, 4/4/93 Conference Room 40 m 25- 2 20- ¦ 10- 12.513 13.514 14.51515.5 16 16.517 17.518 18.519 Tim# (hours) Measured -* Simulated Figure 4-7. Indoor radon level comparison at Conference Room. Figures 4-3 to 4-7 show comparisons of indoor radon levels between simulation and measurement for five (5) zones. Measured indoor radon concentrations at 12:30 PM are used as initial conditions. The airflow rates and indoor pressures remain the same as those at the steady-state condition. In order to be consistent in all simulations, the material properties in the soil and slab are kept the same. All the pressure and radon concentration distributions at the steady-state condition are also used as initial conditions for the pressure and radon transport equations. Simulation results show that dilution rate of indoor radon level due to outdoor air is faster than the measured results, although the indoor radon levels at the final hour are closer to the measured data. The explanation, first of all, is that the lumped zone air model is used in the simulation assuming 100% mixing. In reality, ventilation effectiveness is not 100%, so that indoor radon levels do not decrease as fast as indicated by the simulation. Since ventilation efficiency directly affects the simulated results, investigation of ventilation efficiency is necessary for further refinement. A correction factor for ventilation efficiency should be included in the simulation. However, these factors will be a function of flow rate, register location, zone size, etc. Detailed fluid dynamics simulation can be used to determine these factors. Secondly, the ambient radon level may be higher than assumed. Due to the unavailability of data on ambient radon levels, a constant ambient radon concentration was assumed. Finally, another more likely reason, suggested by R. Mosley (US EPA), is that the passive radon monitor being used has a slow response time and can not follow the rapid rates of change that occur. This validation is used to calibrate radon entry rate from the slab by adjusting the material 26 ------- properties of combined crack and concrete. Once the material properties are refined, they will not be changed for subsequent simulation. 4.4 Simulation Results in a Typical School Day Following previous validation for short time periods to calibrate radon entry rate through the slab, a typical day is chosen to continue to validate simulation results of the building under study. The typical day is a normal school day starting from 6:00 AM, 4/21/93 (Wednesday) to 6:00 AM, 4/22/93 (Thursday), as suggested by Bobby Pyle, SRI. The A/C was on in the first twelve (12) hours and off in the next twelve (12) hours. When the A/C is on, certain amount of outdoor air is brought through duct system to dilute the indoor radon concentration. When the A/C is off, no outdoor air enters the zone, and indoor radon concentration increases due to radon entry from the slab. The indoor radon level increases linearly with time, based on the magnitude of radon entry rate from the slab. Figures 4-8 to 4-11 show comparisons of simulation results to measured data. The indoor radon concentration decreases during A/C on-time period and increases linearly when the A/C was off, as expected. It should be noted that since the ventilation efficiency factor is not included in the present simulation, the indoor radon level decreases faster than measured data. From observation of measurement and suggestion from SRI, ambient radon level is set to 3.5, pCi/L when A'C was on, because the minimum indoor radon level is 3.6 pCi/L. It is assumed that when A/C is on for a long time, it brings enough outdoor air throughout the building to reach the minimum indoor radon level, which is equivalent to that of the ambient condition. 4.5 Closure It can be seen that from the figures that reasonable comparisons between prediction and measurement has been obtained. Material properties were not changed between the seven-hour calibration and the one day validation, showing the material properties used in the input file are a good approximation after adjustment. Radon entry from the slab varies slightly but may be considered to be constant during A/C on and off period. From the observation of experimental data, pressure differences between on and off periods is approximately within 1 Pa. However, there are some unknown effects causing discrepancy with measured data. The possible explanation may involve ventilation efficiency, leakage area, or possibly instrument response times. It should be pointed out that indoor positive pressures in the building are measured when A/C was on, so that advection term of radon entry from the slab is relatively small compared to indoor negative pressures. Since ambient radon level was unavailable during this period, a constant ambient radon level was assumed. Results of other work for the FRRP (see Tyson et al., 1993) show that ambient radon levels may not only be higher than established action levels, but may also vary cyclically during a 24-hour day. Clearly, the model would predict lower rates of dilution and would approach measured values if higher ambient radon levels are used in the simulation. Undoubtedly, these two factors namely, ventilation efficiency and ambient radon levels, must be investigated further before answering the question definitively. 27 ------- Polk Life & Learning Center Conference (Station 5) 4/214/22 20 Measured Simulated o. o 8 10 12 14 16 18 20 22 24 26 28 30 6 Time (hours) Figure 4-8. Indoor radon level comparison at Conference Room in a typical school day. Polk Life & Learning Center Cafeteria (Station 6) 4/214/22 22 Measured o CL Simulated m m -o a oc 8 10 12 14 16 18 20 22 24 26 28 30 6 Time (hours) Figure 4-9. Indoor radon level comparison at Cafeteria in a typical school day. 28 ------- Polk Life & Learning Center Rm. 109 (Station 2) 4/21-4/22 20 Measured Simulated D. O O "O o oc Time (hours) Figure 4-10. Indoor radon level comparison at Room 109 in a typical school day. 1 8 16 <£ 14 c o 1 2 ^10- c 0) o c o <_> c o "D O SXL 8 - 4 - Polk Life & Learning Center Rm. 102 (Station 4) 4/214/22 Measured Simulated 8 10 12 14 16 18 20 22 24 26 28 30 Time (hours) Figure 4-11. Indoor radon level comparison at Room 102 in a typical school day. 29 ------- 5 PARAMETRIC STUDY 5.1 Introduction Following the validation simulation of the large building, parametric studies are presented in this Chapter, using the building configuration of Polk Life and Learning Center. It should be noted that airflow rates of supply and return are the same as those used in the seven-hour simulations. 5.2 Varying Outdoor Airflow Figures 5-1 and 5-2 show indoor radon levels as a function of outdoor airflow for different ambient radon levels when the A/C is on. Indoor radon levels decrease with increasing outdoor airflow through the air distribution system. When small amounts of outdoor airflow are introduced, indoor radon levels increase dramatically because of less dilution. However, when a large amount of outdoor airflow is introduced, for instance, above 1500 cfm for this building, there is little effect to reduce indoor radon levels. The optimal outdoor airflow can be determined from the present simulation, based on building configuration, air conditioning system and radon levels of ambient and soii conditions. On other hand, as long as the ambient radon level is lower than the indoor level, adding more outdoor air can dilute indoor radon. However, when the ambient radon level is higher than indoor levels, outdoor airflow will have the opposite effect; that is, the indoor radon level will increase. This is an important consideration in determining action levels for indoor radon. 5.3 Varying Ambient Radon Level Figures 5-3 and 5-4 show the indoor radon level varying with ambient conditions for different amounts of outdoor airflow through the air distribution system. Indoor radon levels at different zones tend to increase linearly with increased ambient radon levels. Consequently, even though a large amount of airflow is introduced, the indoor radon level may remain high when the ambient radon level is high because fresh air dilution is not effective. 5.4 Varying Soil Radium Content Figures 5-5 and 5-6 show the effect of soil radium concentration at different outdoor airflow rates and ambient radon levels. The indoor radon level increases when radium concentration in the soil increases, and vice versa. From this investigation, the relationship between indoor radon and soil radium content seems linear for a certain amount of airflow. The audiology room has the highest indoor radon level in the building based on the simulation results. 5.5 Closure Through limited parametric studies, it is clear that outdoor airflow is the main factor in reducing indoor radon level by dilution. Since bringing in more outdoor air will lead to a penalty of 30 ------- higher energy demand, any radon reduction strategies should be evaluated to optimize both good indoor air quality and energy consumption. It is worth noting that since no experimental data are available to validate the parametric studies for pressure difference between indoor and outdoor, the advection effect in parametric studies is not shown in the present report. 31 ------- Ambient Radon level 0 pCl/L Outdoor airflow (CFM) Rm 102 Audio Rm 109 3000 Figure 5-1. Effect of outdoor airflow on indoor radon levels (0 pCi/L ambient), 18 16 - s- \ o 1 4 - LL > _a> 12 - c o -o D a: 10 - L. 8 8 - 'w C 6 - Ambient Radon level 4 pCi/L 500 1000 1500 2000 Outdoor airflow (CFM) 2500 Rm 102 Conf Cafe Audio Rm 109 3000 Figure 5-2. Effect of outdoor airflow on indoor radon levels (4 pCi/L ambient). 32 ------- 14 Outdoor airflow 1000 CFM Rm 102 o a. 1 o Audio Rm 109 3 4 5 6 7 Ambient Radon level (pCl/L) Figure 5-3. Effect of ambient radon level on indoor radon levels. (1000 cfm). Outdoor airflow 2000 CFM 1 2 Rm 102 1 0 Conf a. B Cafe Audio 6 Rm 109 4 2 0 Ambient Radon level (pCt/L) Figure 5-4. Effect of ambient radon level on indoor radon levels (2000 cfm). 33 ------- 0 pCi/L ambient and 1000 CFM OA Rm 102 2.5 - Conf o Q. Cafe Audio Rm 109 tx 0.5 200 250 50 100 150 Ra226 soil concentration (Bq/kg) Figure 5-5 Effect of soil radium concentration on indoor radon levels (0 pCi/L). 4 pCi/L ambient and 1000 CFM OA Rm 102 Conf tx Cafe Audio o 5.5 - Di Rm 109 5 - 4.5 150 200 100 250 50 Ra226 soil concentration (Bq/kg) Figure 5-6. Effect of soil radium concentration on indoor radon level f (4 pCi/L). 34 ------- 6 CONCLUSION Multizone airflow, indoor pressure and radon concentration, and radon entry rate from the slab are simulated in a large building, Polk Life and Learning Center at Bartow, Florida. Excellent comparison between the testing or design and predicted airflow rates at the terminals have been obtained in the HVAC system simulation. Reasonable comparison of the indoor radon level between simulation and measurement is obtained for both cases, seven-hour calibration and one typical day validation. Following the validation, parametric studies show that outdoor air flow rate is main factor affecting the indoor radon concentration. However, ambient radon level and soil radium content affect indoor radon level directly. linear relationship is shown between indoor and outdoor radon levels. One can conclude that the best strategy for the present problem to reduce indoor radon concentration is to increase the rate of outdoor airflow. In order to reduce the indoor radon level, the amount of outdoor airflow can play an important role in radon reduction strategy. However, a penalty of increasing energy demand will occur in order to cool more outdoor air. Therefore, an optimal condition should be determined to use minimum energy while maintaining good indoor air quality. Topics for further investigation Ventilation efficiency Less energy consumption by introducing more fresh air Exhaust fan effect Other indoor pollutant Zone energy and moisture simulation Cost analysis Soli depressurization system analysis Pressure difference between indoor and outdoor Caveats: It is crucial to note that the nature of the work performed here is an exploratory one primarily to establish the potential of using models to analyze large buildings and to identify the essential areas for experiment and simulation to compliment each other in providing an accurate, yet cost efficient strategy to study radon in large buildings. This objective was substantially achieved through a preliminary simulation of airflows and pressures in a school building monitored by the US EFA and the SRI. Since only a limited set of experimental data were available, several assumptions were made to successfully complete the simulations. The results presented in this report, should therefore, be viewed in light of the assumptions stated and applied only to the specific problem analyzed. The result should in no way be construed to represent generalizations for large-buildings. The present report concludes with a list of areas that need further attention. 35 ------- 7 REFERENCES Cheng. P., "Geothermal Heat Transfer," Handbook of Heat Transfer Applications. 2nd Ed., Edited by Rohsenow, W. M., Hartnett, J. P. & Ganic, E. N., McGraw-Hill Inc., New York, 1985 FSEC 3.0, Florida Software for Enervironment Computation - User's Manual. Version 3.0. FSEC-GP-47-92, Florida Solar Energy Center, Cape Canaveral, Florida, 1992 Rogers, V. C. & Nielson, K. K., "Multiphase Radon generation and Transport in Porous Materials," Health Physics. Vol. 60, No. 6, pp. 807-815, 1991 Tyson, J. L., Fairey, P. W. & Withers, C. R., "Elevated Radon Levels in Ambient Air.'' Indoor Air Quality and Climate Helsinki, Finland, June 27 - July 2, 1993 Walton, G., AIRNET User Manual. NISTIR 89-4072, US Department of Commerce, Washington DC, 1989 Yuan, Y. C. & Roberts, C. J., "Numerical Investigation of Radon Transport through a Porous Medium." Transaction of American Nuclear Society. Vol. 38, pp. 108-110, 1981 36 ------- TECHNICAL REPORT OATA (Pkase read Instructions on the reverse before completi l. REPORT NO. 2. EPA-600/R-96-116 3. f 4. TITLE ANO SUBTITLE Large Building HVAC Simulation S. REPORT DATE September 1996 6. PERFORMING ORGANIZATION CODE 7. AUTHOR(S) Lixing Gu, Muthusamy V. Swami, and Vailoor Vasanth 8. PERFORMING ORGANIZATION REPORT NO. FSEOCR-616-93 9. PERFORMING ORGANIZATION NAME AND ADDRESS Florida Solar Energy Center 300 State Road 401 Cape Canaveral, Florida 32920 10. PROGRAM ELEMENT NO. 11. CONTRACT/GRANT NO. 68-D0-0097 Work Assignment 3-12 12. SPONSORING AGENCY NAME ANO ADORESS EPA, Office of Research and Development Air Pollution Prevention and Control Division Research Triangle Park, NC 27711 13. TYPE OF REPORT AND PERIOD COVEREO Final report; 3/92-4/93 14. SPONSORING AGENCY CODE EPA/600A3 ^.supplementarynotesAppCD oject officer is MarcY. Menetrez, Mail Drop 54. 919/ 541-7981. 16. abstractrep0r£ discusses the monitoring and collection of data relating to indoor pressures and radon concentrations under several test conditions in a large school building in Bartow, Florida. The Florida Solar Energy Center (FSEC) used an integrated computational software, FSEC 3.0, to simulate heating, ventilation, and air-conditioning system and multizone airflows, indoor pressures, radon transport in the soil, and slab and indoor radon levels in the large building. The simulation was validated by measured data. A limited parametric study shows the influence of out- door airflow, ambient radon level, and soil radium content on indoor radon levels. 17. KEY WORDS ANO DOCUMENT ANALYSIS a. DESCRIPTORS b.IDENTIFIERS/OPEN ENDED TERMS c. COSATI Field/Group Pollution Soils Radon Slabs Heating Radium Ventilation Air Conditioning School Buildings Pollution Control Stationary Sources Heating, Ventilation, and Air-conditioning Systems 13 R 08G.08M 07 B 13 C 13H.13A 13M.05I 18. DISTRIBUTION statement Release to Public 19. SECURITY CLASS (This Report) Unclassified 21. NO. OF PAGES 42 20. SECURITY CLASS (This page) Unclassified 22. PRICE EPA Form 2220-1 (9-73) ------- |