BULLETIN NO. 26 Moisture and T emperoture Control in Buildings Utilizing Structural I nsuloting Boord FRANK B. ROWLEY, M.E. PROFESSOR OF MECHANICAL ENGINEERING MILLARD H. LA JOY, M.S.(M.E.) ASSOCIATE PROFESSOR OF MECHANICAL ENGINEERING .EINAR T. ERICKSON, B.M.E. RESEARCH FELLOW, ENGINEERING EXPERIMENT STATION • UNIVERSITY OF MINNESOTA INSTITUTE OF TECHNOLOGY ENGINEERING EXPERIMENT STATION VOL. L NO. 26 JULY 16, 1947 Entered at the post office in Minneapolis as semi-monthly second- class matter, Minneapolis, Minnesota. Accepted for mailing at special rate of postage provided for in Section 1103. Act of October 3, 1917, authorized July 12, 1918. I TABLE OF CONTENTS INTRODUCTION ................................................................... . ........................................................... . PAGE 1 1 2 Plan of Investigation ....................................................................................................................................... .. Previous Investigations ............ .. TYPES OF CONSTRUCTION TESTED... ............................ .......................................................... 3 Test Bungalow ............. ................................................................... ..................................... 3 Interior Construction ..................... . ... . . ................... .... ............................................................................ 5 Exterior Construction ........................ ................ .......... ...................................................................... 6 Ceiling and Attic Construction.... . .............................................................................................. 6 Interior and Exterior Construction of Test Wall Panels 9 Location of Test Walls in Bungalow ..................................... :::.:::::::::::: .::::::::::::::::: 12 Equipment and Instruments ................................ .............................................................. 12 Test Apparatus . ...................... .... ........................................................... 12 Instruments ........................................ ...................... ................................................................................. 17 TEST RESULTS ........................................................................................................... -............................................................ 18 Wall Moisture Study............................ ......................... ......................................................... 18 Results after 66-Day Test Period............................................................................................. 18 Conclusions from Wall Moisture Study...... ............................................... 22 Attic Ventilating Requirements.. . .......................... ..................................................... .................. 23 Conclusions from Attic Ventilation Study.................................................................. 27 Attic Air Temperatures................................ ............................................................................... 28 Conclusions from Attic Air Temperature Study........................... ...................... 35 SUMMARY 37 ILLUSTRATIONS FIGURE PAGE 1. VJ.ew of Test Bungalow within Cold Room .... ;>.... ........................................ 3 2. View of Entr~n(!e to Completed Bungalow...... . ............... ..................................................... 4 !3. Plan View of Test Bungalow Showing Position of Test Walls......... 4 4. Typical Wall Construction Showing Horizontal and Vertical Joints in Application of Insulating Board Lath ............. " ................................................................... 7 5. Typical VTall Construction for Application of 4-Foot by 8-Foot Insu- l~tl.ng Board Sheathing ........ . ....... ........................ ................................................... 7 6~ Typical Wall Construction for Application of 2-Foot by 8-Foot Insu- lating Board Sheathing ........................................................................... "............................................... 7 ~. Exterior Wall Pahel Fitted to Grooved Studs..................................................................... 8 8. Ceiling and Attic Constructions ...... 9. Gener:al Ctmstruction of Test Walls ..... 10. Constru~tion of Test Wall Panels ... ............................................................................ ]l() 11. Cross:-Sectional View of Test Bungalow within Cold Room..... 14. 12. S~ctional View of Air Conditioning Unit and Distribution Du~............... 15 13. Sectional View of Attic Ventilator .... 16 14. Interior View of Finished Room 2 within Test Bungalow.. 1~ 15. Effect of Ventilation on Attic Air Temperatures for Construction 4~........ 34· 16. Effect of Ventilation on Attic Air Temperatures for Consttu-c:tion 5·..... 34::. 1~4:90260 TABLES TABLE PAGE I Selection of Vapor Resistant Materials for Wall Panels in Test Bun- galow .... .................... . .................. 11 II Selection of Insulating Board Lath Coating 12 III Wall Moisture Study 19 IV Attic Ventilating Requirements to Prevent Frost Accumulation 24 V Summary of Attic Ventilating Requirements to Prevent Frost Ac- cumulation for Five Different Attic Constructions.. 27 VI Relationship between Attic Air Temperatures and Inside and Outside Air Temperatures at Different Ventilating Rates for Construction 4 30 VII Relationship between Attic Air Temperatures and Inside and Outside Air Temperatures at Different Ventilating Rates for Construction 5 32 VIII Summary of Relationship between Attic Air Temperatures and Out- side Air Temperatures at Different Ventilating Rates for Construc- tions 4 and 5 36 ACKNOWLEDGMENT The authors gratefully acknowledge the cooperation of the Insulation Board Institute and the assistance of its Tech- nical Advisory Committee in the planning, organization, and development of this research program. This committee in- cludes 0. W. Frost, chairman (1945-46), I. R. Birner, A. S. Bull, R. E. Donnelly, D. L. Gleaves, E. M. Jenkins, R. T. Miller, W. L. Scott, H. W. Stein, and P. D. Close (1945) and S.M. Van Kirk (1946), technical secretaries. They also wish to express their appreciation to Miss Fleurette Halpern for the work which she has done in editing and arranging the material. Moisture and Temperature Control in Buildings Utilizing Structural Insulating Board INTRODUCTION . The occurrence of condensation within the walls and attics of con- ventional residential structures has been under intensive study in various laboratories throughout the country and by members of the building industry during the past decade. The methods of construc- tion which will maintain a home substantially free of any condensa- tion problems have been fairly well established. With respect to con- densation within the walls, three common precautions are: (1) to reduce the relative humidity of the inside air; (2) to use sufficient insu- lation in the cold walls; and (3) to use a vapor resistant interior sur- face construction. Condensation problems in the attic may be con- trolled with the use of vapor resistant materials located on the warm side of the ceiling and with vapor seals around the attic door and fixture outlets. As an added precaution in unheated attics, provision should be made for supplying ventilation in order to remove any vapors that might accumulate in the attic because of slow transmis- sion through the vapor resistant ceiling, through the attic door, or through any openings around the door and electric outlets. Plan of Invesiigaiion This investigation, the third in a series undertaken for the Insula- tion Board Institute, was conducted in a full-scale test bungalow to observe the occurrence of moisture condensation and the distribution of temperatures in various building constructions utilizing structural insulating board. Previous tests with these materials had been con- ducted on small test panels only. By duplicating, as nearly as pos- sible, the variables in standard residential constructions, it was thought that the resulting conditions observed during operation of the test bungalow would be more conclusive than data gained heretofore from the small test panels. Four main objectives were included in this investigation, as fol- lows: 1. To establish more firmly the conclusion that an interior finish possessing a permeability rate of 1.25 grains per square foot per hour per inch of mercury vapor pressure difference is a safe material to recommend in the field. Under normal conditions previous investigations on individual test panels in a special vapor transmission test apparatus1 had shown that the rate of 1.25 was safe regardless of the type of exterior finish used. 1 F. B. Rowley and C. E. Lund, Vapor Transmission Analysis of Structural Insulating Boart'L (University of Minnesota Engineering Experiment Station Bulletin No. 22, 1944) pp. 10-16. 2 MOISTURE AND TEMPERATURE CONTROL 2. To determine what effect different rates of vapor permeability in interior and exterior surface constructions have on moisture condensation within the walls of standard residential structures. Seven different types of wall construction were included. Insulating board lath having low, medium- low, and high permeability rates were tested in combination with struc- · tural insulating board sheathing having low, medium, and high permeabil- ity rates. 3. To· provide experimental data on the volume of ventilating air required to prevent the formation of frost in the attic for different ceiling construc- tions. Variations in attic construction included a weather-stripped attic door, a paint vapor barrier, and added insulation. 4. To obtain more definite data on attic air temperatures, with and without ventilation, as related to outdoor temperatures which, in turn, could be used to determine with greater accuracy heat losses through the ceiling area. Previous Investigations In 1940 and 1941, the University of Minnesota Engineering Experi- ment Station published two extensive studies on moisture and conden- sation control in building construction and operation.2 In 1941, the Insulation Board Institute entered upon a cooperative research pro- gram with the Station to study more specifically moisture control in buildings constructed with structural insulating board. Two studies have been published under the IBI cooperative pro- gram. The first investigation3 was concerned with the vapor trans- mission analysis of structural insulating board. In this investigation tests were conducted to determine the vapor permeability of various structural insulating boards: srnall test panels constructed with this material were tested for vapor permeability; interior paints and wall coverings applied to structural insulating board were tested for effec- tiveness as vapor barriers-all under temperature and humidity con- ditions commonly prevailing in residential buildings. The primary purpose of the second investigation4 was to find vapor resistant coatings for structural insulating board which would main- tain their effectiveness under conditions of high humidities-condi- tions which are prevalent in vegetable storage warehouses where it is necessary to protect both the structure and the stored product from the damages resulting from moisture condensation. 2 F. B. Rowley, A. B. Algren, and C. E. Lund, Methods of Moisture Control and Their Application to Building Construction (University of Minnesota Engineering Experiment Station Bulletin No. 17, 1940). F. B. Rowley, A. B. Algren, and C. E. Lund, Condensation of Moisture and Its Rela- tion to Building Construction and Operation (University of Minnesota Engineering Exper- iment Station Bulletin No. 18, 1941). 3 F. B. Rowley and C. E. Lund, Vapor Transmission Analysis of Structural Insulating Board (University of Minnesota Engineering Experiment Station Bulletin No. 22, 1944). 4 F. B. Rowley, M. H. LaJoy, and E. Erickson, Vapor Resistant Coatings for Structural Insulating Board (University of Minnesota Engineering Experiment Station Bulletin No. 25, 1946). ---------------------------------- TYPES OF CONSTRUCTION TESTED 3 TYPES OF CONSTRUCTION TESTED Tes:l: Bungalow The full-scale test bungalow in which the different wall and attic constructions were incorporated is shown within the large cold room in Figure 1. A view of the entrance to the completed bungalow is shown in Figure 2, while Figure 3 represents a plan view of the struc- ture. The bungalow had a floor area of 21 feet by 21 feet 8 inches and a first floor ceiling height of '8 feet 6 inches. It was identical to the one used in a previous investigation described in detail in Bulletin No. 18.5 The test bungalow was of standard frame construction and was so designed that several different types of interior and exterior con- struction could be incorporated in the walls and studied at the same time. In the present investigation 15 test wall sections were built into the structure, each of which was 4 feet wide and extended the full height of the first floor. These walls were constructed with different combinations of vapor resistant materials on the interior and exterior sides of the studding. All walls were substantially the same in the over-all coefficient of heat transmission. The bungalow contained three rooms with four double-hung win- dows and one exterior door fitted with a storm door. The attic door opened to an attic stairway. In the attic there were three gable ends with a window fitted into each. The attic window at the rear of the bungalow was used for supplying ventilating air, while the opposite 5 Rowley, Algren, and Lund, Condensation of Moisture and Its Relation to Building Construction and Operation. Figure 1. View of test bungalow within cold room Figure 2. View of entrance to completed bungalow 4-0 S-E 6-F u 3-B ROOM I ROOM 2 1-B 7'-3" x 12'-o" / l'-d'x 13~6" a-c 2-A 9-C ~II l K 15-H I 0-A ROOM 3 S-o" x 13'-s" 14-I I 1-0 J 13-F 12-G Figure 3. Plan view of test bungalow showing position of test walls TYPES OF CONSTRUCTION TESTED 5 window was opened a slight degree for the discharge of air into the low temperature room. Inferior Construction The interior finish of all walls consisted of 1/2-inch plaster applied to %-inch insulating board lath, and the ceiling was finished with 1h-inch plaster applied to l-inch insulating board lath. Two coats of plaster were used: the first, a brown coat, consisted of two and one- half parts sand to one part plaster by volume; the finish coat consisted of three parts gaging plaster to one part finishing lime. The first coat was allowed to dry for five days and then was followed by the appli- cation of the finish coat which was allowed to dry for 30 days before the start of the test. After the first few weeks of preliminary tests, the ceiling was painted with two coats of paint to serve as a vapor barrier. The composition of this coating was as follows: 1. Composition of primer (A-21) 6 by weight: pigment, 58 per cent, and vehi- cle, 42 per cent. Composition of the pigment consisted of: titanium dioxide, 10 per cent; silica, 20 per cent; metro-nite, 70 per cent. Composition of the vehicle consisted of: vegetable oils, 84.8 per cent; drier and thinner, 15.2 per cent. 2. Composition of fiat white paint (C-20) 6 by weight: pigment, 59 per cent, and vehicle, 41 per cent. Composition of the pigment consisted of: titanium calcium, 72.2 per cent, and magnesium silicate, 27.8 per cent. Composition of the vehicle consisted of: vegetable oils, 22.1 per cent; resins, 7.9 per cent; drier, 3.2 per cent; turpentine, 2.7 per cent; m1neral spirits, 64.1 per cent. On the basis of previous tests this paint coating was selected as one which would provide a reasonably good vapor barrier. Actual tests of this coating when applied to lf2-inch plaster on 1/2-inch insulating board resulted in a permeability rate of 1.61 grains per square foot per hour per inch of mercury vapor pressure difference. Since a l-inch insulating board lath was used, the permeability rate of the ceiling of the test bungalow would be slightly lower than this value because of the added resistance of the thicker lath. This treatment was taken in accordance with the permeability rate of from 1.00 to 1.50 which was originally planned for use as the ceiling barrier. The application of the insulating board lath to all of the 4-foot wide exterior wall sections as well as to the ceiling surface was done by methods used in actual practice. Figure 4 shows the lath application. The lath boards used were commercially fabricated to give a well- fitted shiplap, or V-joint. The vertical and horizontal joints of a standard wall section were duplicated because joints have always been a questionable source of vapor leakage. In any typical construction the average face dimensions of a solid wall are longer than those used in the test wall. Thus, approximately 10 per cent more joints were a Composition, code numbers, and test results of paint finishes given by Rowley and Lund, Vapor Transmission Ana~ysis of Structura~ Insu~ating Board, Bulletin No. 22, pp. 56-62. 6 MOISTURE AND TEMPERATURE CONTROL used in the test panels in this investigation than would be found in an average construction. Exterior Construction All exterior wall panels were constructed on studs spaced 16 inches on centers with structural insulating board as the sheathing and 3/4-inch by 8-inch cedar as the siding. As shown in Figure 5, wood strips, 3;4-inch wide and 11/s-inch deep, were used as nail strips for the sheathing. The outside surfaces of the fixed studs were grooved to receive these nailing strips. Figure 4 shows the stud grooves, while Figures 5 and 6 show the sheathing applied to the nailing strips which fit into the grooves. Figure 7 shows the two sections fitted together. The combined exterior section of the wall was attached to the studs by long wood screws. By this method the outer section of the wall could be fastened securely to the studs and yet could be removed easily for inspection during or after the test period. The siding was applied over the sheathing and nailed to the nailing strips with the exception of the top, center, and bottom siding boards which were applied with wood screws to facilitate removal for inspection. The exterior siding was painted with two coats of pure white lead and linseed oil paint. Application of insulating board sheathing was in accordance with standard practice. Where the schedule called for 4-foot by 8-foot sheathing, one board alone was taken to cover the complete wall panel as shown in Figure 5. Where the schedule called for 2-foot by 8-foot sheathing, these were applied as shown in Figure 6. Thus, the hori- zontal and vertical joints found in actual construction were included. Above the first floor ceiling line the construction consisted of wood sheathing, roof boards of ponderosa pine shiplap, and siding of 1J2-inch by 6-inch redwood which was painted with both one coat of primer and one coat of outside white paint. Ceiling and Attic Construction In the study of frost accumulation in the attic, five different ceil- ing and attic constructions were used. These are shown in Figure 8. The variations in these constructions included the effects of using or not using a rubber gasket around the attic door, a vapor barrier on the ceiling plaster, and added insulation. Construction 1 consisted of %-inch plaster over l-inch lath with no vapor barrier or additional insulation and with a loose~fitting attic door. Construction 2 was iden- tical to Construction 1 with the addition of a rubber gasket fitted around the attic door. Construction 3 was the same as Construction 2, plus two coats of paint applied to the ceiling plaster as a vapor bar- rier. The description of this paint is covered above under the section entitled Interior Construction. Construction 4 was identical to Con- struction 3 with the addition of %-inch insulating board applied to the tops of the ceiling or attic floor joists as added insulation. Construction Figure 4. Typical wall construction showing horizontal and vertical joints in application of insulating board lath Figure 5. Typical wall construction for application of 4-foot by 8-foot insulating board sheathing with no vertical or horizontal joints Figure 6. Typical wall construction showing horizontal and vertical joints in application of 2-foot by 8-foot insulating board sheathing Figure 7. Exterior wall panel fitted to grooved studs 1/2" PLASTER UNPAINTED CONSTRUCTION I. (LOOSE FITTING ATTIC DOOR) CONSTRUCTION 2. (ATTIC DOOR WEATHER STRIPPED) 1/2" INSULATING BOARD CONSTRUCTION 4. (ATTIC DOOR WEATHER STRIPPED) CONSTRUCTION 5. (ATTIC DOOR WEATHER STRIPPED) Figure 8. Ceiling and attic constructions TYPES OF CONSTRUCTION TESTED 9 5 was the same as Construction 3 but for the addition of %-inch insu- lating board applied to the under side of the roof rafters and to the wall area extending above the first floor line. Interior and Exterior Construction of Test Wall Panels The materials selected for interior and exterior construction of the test wall panels possessed different rates of vapor permeability. As many .combinations as were possible within the size limitation of the test bungalow were incorporated in nine different panels. Figure 9 Figure 9. General construction of test walls ------------------------~~-- 10 MOISTURE AND TEMPERATURE CONTROL shows the general construction of all the test walls. Figure 10, a cross section of each wall panel, describes the materials used in their construction. All of the exterior walls were constructed with 4-foot by 8-foot sheathing, except Walls A and C, in which 2-foot by 8-foot sheathing was used in order to investigate the effect of horizontal and vertical joints on test results. WALL "A" ASPHALT PAPER WALL "8" WALL "E" WALL •H• 8'' WOOD SIDING Figure 10. Construction of test wall panels WALL "C" WALL •yn TYPES OF CONSTRUCTION TESTED 11 Representative vapor permeability rates selected to conform to the various classifications were as follows: Low permeability rate-below 0.50 grains per square foot per hour per inch of mercury vapor pressure difference Medium-low permeability rate-between 1.00 and 1.25 grains per square foot per hour per inch of mercury vapor pressure difference Medium permeability rate-between 3.0 to 5.0 grains per square foot per hour per inch of mercury vapor pressure difference High permeability rate-above 30.0 grains per square foot per hour per inch of mercury vapor pressure difference Table I represents a summary of the selection of vapor resistant materials. In Group 1, Walls A and B were constructed with vapor resistant asphalt paper in conjunction with plain untreated insulating board lath in order to meet the requirement of a low permeability lath. The permeability rate of 0.09 for the asphalt paper was based on actual tests in the special vapor transmission test apparatus. For the walls in Groups 2, 3, and 4 which required a lath possessing a medium-low permeability rate, it was necessary to select and try out various coatings on a plain board. A plain lath, code No. P-6,7 was selected to be coated with various vapor resistant paints. Results of these tests are shown in Table II. Four different coatings were tested. Specimen P-6-5 with one coat of A-21 primer, plus one coat of C-20 fiat white paint, gave a permeability rate of 1.14, which was within the requirements. To establish more firmly the results of this treatment, 7 Rowley and Lund, op. cit., pp. 22-25. Table I. Selection of Vapor Resistant Materials for Wall Panels in Test Bungalow- Permeability Rates Expressed in Grains per Square Foot per Hour per Inch of Mercury Vapor Pressure Difference Wall Insulating Board Group No. 3 4 5 6 7 Desig- nation (Figs. 2 and 10) 2-A and 10-A 1-B and 3-B 8-C and 9-C 4-D and 11-D 5-E and 7-E ...... 6-F and 13-F ..... 12-G ······································· 15-H 14-I * Asphalt paper barrier. t Surface treated. Insulating Board Lath Sheathing Permeability Code Permeability Code rate No. rate No. 0.09* P-2 0.44 R-1 ..................... 0.09* P-2 0.44 R-1 1.I6t P-6 0.44 R-1 I.I6t P-6 0.44 R-1 I.I6t P-6 3.70 R-6 I.I6t P-6 51.80 S-5 69.9 P-3 0.44 R-1 69.9 P-3 3.70 R-6 69.9 P-3 51.80 S-5 12 MOISTURE AND TEMPERATURE CONTROL Table II. Selection of Insulatinq Board Lath Coatinq-Vapor Transmission Test Results for Selection of Insulatinq Board Lath Coatinq with a Permeability Rate between 1.00 to 1.25 Grains per Square Foot per Hour per Inch of Mercury Vapor Pressure Difference Permeability Rate Specimen No. Coating Grains per sq ft per hr per in. Hg P-6-1 P-6-2 P-6-3 P-6-4 P-6-5* P-6-6* 1 coat A-21 primer 1 coat A-2 asphalt paint 1 coat A-21 primer 1 coat A-2 asphalt paint 1 coat A-21 primer 2 coats A-21 primer 1 coat A-21 primer 1 coat C-20 flat white paint 1 coat A-21 primer 1 coat C-20 flat white paint * Surface coverages: A-21 Primer ..................................................... 650 square feet per gallon C-20 Paint ......................................................... 572 square feet per gallon 0.55 0.59 1.74 1.04 1.14 1.17 a duplicate test was conducted as shown by sample P-6-6. The perme,. ability rate of this sample amounted to 1.17. The average of P-6-5 and P-6-6 was 1.16. This was used for all lath requiring a medium- low permeability rate. An interior finish construction having a high permeability rate was planned for the walls in Groups 5, 6, and 7. For this purpose 1/2-inch plaster without additional vapor resistant material was applied to plain 1/2-inch insulating board. A plain insulating board lath, code No. P-3, with a permeability rate of 69.9 was used as a plaster base. With respect to the sheathings it was possible to select from the regular run of commercial insulating boards those which possessed low, medium, and high permeability rates. As shown in the last two columns of Table I, commercial sheathing R-1 having a permeability rate of 0.44 was used for all walls of high vapor resistant exterior con- struction. This included the walls in Groups 1, 2, and 5. Sheathing R-6 with a medium permeability rate of 3.70 was selected for the walls in Groups 3 and 6. Plain sheathing S-5 with a high permeability rate of 51.8 was used for the walls in Groups 4 and 7. Location of Test Walls in Bungalow Figure 3 indicates the location of the 15 test walls around the test bungalow. Beginning at the left foreground of Room 1, the walls are numbered from 1 to 15, consecutively. Types of wall are designated by letter; for example, Walls 1-B and 3-B were identical in construc- tion. There were duplicate walls for all those in Groups A through F. Single tests were conducted on Walls G, H, and I. These 15 walls utilized all space available for 4-foot by 8-foot panels. Two small areas COLD ROOM ---------RETURN Figure 11. Cross-sectional view of test bungalow within cold room 14 MOISTURE AND TEMPERATURE CONTROL remained, as indicated by sections J and K, which were constructed similarly to Walls I and F, respectively. These were used as control panels and were removed occasionally for inspection to indicate the time at which the main test walls should be removed for final exami- nation. Equipment and Instruments Test apparatus-The cold room which housed the test bungalow was 30 feet square and 25 feet high with walls constructed of 2-inch by 6-inch studs spaced 16 inches on centers and covered with %-inch fiber insulating board on inside and outside wall surfaces. The stud space was insulated with mineral wool 6 inches thick, and a vapor resistant paper was nailed to the studs next to the outside surface of the wall to prevent the vapor in the warm exterior air from entering the wall. The paper was omitted from the inner or cold surface of the wall so as to allow some breathing through the cold inner surfaces. This method of construction proved satisfactory in operation and did not present any condensation problems. The room was cooled by circulating the cold room air over direct expansion coils supplied by a 25-ton ammonia compressor. By means of a 6,000-cubic foot per minute fan, the air was circulated from return duct openings at the top of the cold room through the coils into the supply duct and delivered again to the cold room by a vent directly opposite the return duct vent. The cooling coils were separated in compartments by an arrangement that allowed one coil to be defrosted while the other was in operation. A cross-sectional view of the test bungalow, cold room, and cooling unit is shown in Figure 11. Conditioned air was supplied to the test bungalow from an air con- ditioning unit located in the basement below the cold room. The unit consisted of a steam heating coil with thermostatic controls, a water spray head and eliminator, an automatic temperature and humidity control aparatus, and a 1,000-cubic foot per minute fan. A sectional view of the system appears in Figure 12. In operation, air was drawn from the test bungalow, passed through the conditioning unit, and delivered at the proper temperature and humidity to the rooms of the bungalow. The humidity control was maintained by a humidistat placed in a room of the test bungalow and connected to a solenoid water valve in the spray water line. The major part of the heat was supplied by a steam heating coil built into the air conditioning unit. This was supplemented by electric resistance heaters which were placed in the supply air ducts and controlled by thermostats in adja- cent return air ducts. This method resulted in economical and satis- factory control of both temperature and humidity in the test bunga- low. TYPES OF CONSTRUCTION TESTED 15 The attic ventilation was supplied by the ventilator shown in Figure 13. The motor and 700-cubic foot per minute fan were attached to the cold room wall, and a 4-inch duct carried the air to the plenum chamber fitted over the north or rear attic window. To assure uniform delivery of the air through the open window into the attic, a baffle was placed over the supply duct vent. A bell-shaped orifice with static and impact tubes connected to a manometer in the test bungalow de- termined the volume of air supplied. The air was carried from the cold room through the supply duct to the plenum chamber into the attic space and out through the opposite window which was open. HUMIDISTAT GLOBE VALVE AONETIC ~ !HU.-OfF VN..VE ~AIN Figure 12. Sectional view of air conditioning unit and distribution ducts Ins:l:rumen:l:s-The measurement and control of temperatures and relative humidities were considered very important throughout this research project. All temperatures were taken with copper-constantan thermo- couples. Thermocouples were placed in the cold room, in the attic, PLENUM CHAMBER ___ ...,. BAFFLE PLATE IMPACT a STATIC TUBES 700 Ci FM FAN Figure 13. Sectional view of attic ventilator ATTIC SPACE TYPES OF CONSTRUCTION TESTED 17 in the bungalow rooms, and in the test walls. Figure 14 shows the thermocouple stations within Room 2 of the test bungalow. Four thermocouples were located in each test wall, as follows: one on the interior surface of the plaster, one on the cold side of the insulating board lath, one on the warm side of the sheathing, and one on the cold side of the sheathing. They were placed midway between the studs at approximately one-half the wall height. Permanent leads for 60 thermocouples were connected between stations and an instrument panel in the operator's room. By an arrangement of switches 48 of these could be ·connected to automatic recorders; thus, all temperatures could be taken by a manually-oper- ated potentiometer. To measure the relative humidities of the air within the bungalow and attic, a wet- and dry-bulb apparatus was used. It was comprised of two thermocouples, each mounted in a glass tube with the beaded end extending into a solid brass tip of the same diameter as the glass tube. The brass tip of the wet-bulb thermometer served to give a good Figure 14. Interior view of finished Room 2 within test bungalow showing thermocouple stations, heating duct, and double windows 18 MOISTURE AND TEMPERATURE CONTROL thermal contact between the thermocouple and wicking material. Water was fed to the wick of the wet-bulb thermometer from a con- stant level water reservoir. An electric fan was placed behind the apparatus to obtain sufficient air velocity for accurate wet-bulb tem- peratures. The moisture content of the test wall sections and other parts of the bungalow was obtained by the use of a commercial type of electric moisture meter which measured the comparative electrical conductivi- ty of a wood or fiber material, which is an index of moisture content. The instrument was calibrated for use on insulating board. The rate of air flow was determined by a calibrated orifice with the aid of a standard Pitot tube and draft gage. TEST RESULTS All of the test results will be discussed under three separate divi- sions, as follows: (1) results pertaining to moisture conditions within the walls; (2) results pertaining to attic ventilation required to pre- vent condensation of moisture on the attic roof boards; and (3) results pertaining to attic air temperatures as related to outdoor temperatures. Test results of the first two studies included in this section were obtained during the 66-day period of operation. During this time the test bungalow was subjected to continuous operation under the fol- lowing conditions: outside air was maintained between 0 F and -10 F; inside air was maintained at 70 F and 40 per cent relative humidity. Results of the third study in this section, as mentioned above, were obtained from the conductance of a special series of tests. Wall Moisture Study Two general methods were used to determine the amount of mois- ture accumulated throughout the various parts of the structures. (1) Exposed surfaces and interior sections were inspected visually to note the accumulation of frost and ice; (2) the gain in moisture content by different members of the structure was taken with an electric mois- ture gage. Results after 66-day test period-After continuous operation for a period of 66 days all exterior walls were removed from the test bunga- low and examined. No changes in operating air temperatures either inside or outside the test bungalow were made during this inspection period. A comprehensive summary covering results of tests on mois- ture content of the various materials in each wall and notations on frost accumulation on the siding is contained in Table III. In this table column 1 refers to the group and code numbers of the wall panels. The letters identify the type of wall as shown in Figure 10, and the number prefix refers to its location in the test bungalow as shown in Figure 3. The moisture contents given in columns 2 through Table III. Wall Moisture Study-Moisture Content and Frost Conditions at End of 66-Day Test Period-Continued Wall No. Moisture Content (Per Cent by Weight-Moisture Meter Reading) Lath Sheathing Siding Cold side Cold side Warm side Top board Middle board Bottom board Group 4 Walls with treated lath having ·a MEDIUM-LOW permeability rate of 1.16 and sheathing having a HIGH permeability rate of 51.8 6-F ................................................................. 13.3 6.5 5.4 5.4 9.6 8.4 9.5 8.9 8.8 8.2 13-F ................................................................. 14.9 Group 5 Wall with plain lath having a HIGH permeability rate of 69.9 and sheathing having a LOW permeability rate of 0.44 12-G .............................................................. .. Group 6 15-H ................................................................ . Group 7 14-I ................................................................. .. 6.6 6.6 7.6 10.1 6.4 7.5 Wall with plain lath having a HIGH permeability rate of 69.9 and sheathing having a MEDIUM permeability rate of 3.70 5.9 9.2 6.9 16.5 8.9 6.4 Wall with plain lath having a HIGH permeability rate of 69.9 and sheathing having a HIGH permeability rate of 51.8 6.4 12.3 7.9 17.0 20.0 16.1 Frost Accumulation on Siding Inner Surface No frost accumulation No frost accumulation Heavy frost accumulation on top board along joint at top of wall section. Middle and bottom boards clear Heavy frost accumulation on upper surface of top board. Middle and bottom boards clear Heavy frost accumulation of approximately 3/32-inch thickness on surface of top board. Only slight frost accumulation on middle and bottom boards Table III. Wall Moisture Study-Moisture Content and Frost Conditions at End of 66-Day Test Period Wall No. Moisture Content (Per Cent by Weight-Moisture Meter Reading) Lath Sheathing Siding Cold side Cold side Warm side Top board Middle board Bottom board Group 1 Walls with plain lath and paper barrier having a LOW permeability rate of 0.09 and with sheathing having a LOW permeability rate of 0.44 2-A ........................................................ 12.6 5.4 7.9 9.0 7.9 8.2 8.1 8.0 8.5 7.4 7.9 6.7 6.4 8.4 10-A ....................................................... . 1-B ........................................................... 13.9 6.1 6.2 5.4 5.4 3-B .................... 16.1 Group 2 8-C 9-C ·.·.·:.·::.·.·.·:.·.::::::·::.·:.···.·:.·:.·.·:.·:.·::::::·.·::::::::::~·.-.·.·.····· 4-D .......................................................... . 11-D ..................................................... . Group 3 5-E ..................................................... . 7-E ...................................................... . Walls with treated lath having a MEDIUM-LOW permeability rate of 1.16 and sheathing having a LOW permeability rate of 0.44 14.9 7.2 5.4 9.6 9.5 7.1 16.2 6.5 5.4 10.2 7.9 7.2 11.4 6.7 5.7 10.6 8.3 7.0 15.9 6.0 5.6 9.5 6.7 6.6 Walls with treated lath having a MEDIUM-LOW permeability rate of 1.16 and sheathing having a MEDIUM permeability rate of 3.70 17.9 6.4 5.4 8.3 7.8 7.9 13.5 7.0 6.5 13.0 11.6 9.5 Frost Accumulation on Siding Inner Surface No frost accumulation No frost accumulation No frost accumulation No frost accumulation No frost accumulation No frost accumulation Slight frost accumulation on top board for half of length. Middle and bottom boards clear No frost accumulation No frost accumulation Slight frost accumulation on top board along joint at top of wall section. Middle and bottom boards clear I TEST RESULTS 21 7 were obtained with an electric moisture gage. The figures in column 3 represent an average of these readings taken from top to bottom on the cold side of the sheathing. The moisture content readings of the cold side of the lath and the warm side of the sheathing were taken at the mid-height level of the wall section. The last column consists of notes based on visual inspection of frost accumulation on the inner surface of the siding. These notations on frost were limited to the inner surface of the siding because the temperature gradient through- out all of the walls in the test bungalow was such that the surfaces in the inner stud space were at temperatures above freezing. The lath and sheathing facing this inner stud space were examined for moisture content as shown in the table in columns 2 and 4, respectively. The walls listed in Table III are divided into seven groups in accordance with the different combinations of wall construction used. Group 1 included four walls (2-A, 10-A, 1-B, and 3-B) containing interior and exterior materials both of low permeability rates. With re- spect to these four walls there was no frost accumulation on the inner surface of the siding boards. The moisture content of the siding ranged from 6 to 9 per cent which is safe for this type of exterior construc- tion. The moisture content of the lath ranged from 12.6 to 16.1 per cent, a condition which would be expected in these walls inasmuch as a very high vapor resistant material in the form of an asphalt coated paper was located on the cold side of the lath. Moisture from the inside passed through the lath and was stopped by this barrier, which tended to build up a high equilibrium moisture content within the lath. - Walls constructed with lath having a medium-low vapor perme- ability rate of 1.16 and sheathing having a low permeability rate of 0.44 are shown in Group 2 of Table III. These included Walls 8-C, 9-C, 4-D, and 11-D. The vapor barrier on the lath consisted of one coat of A-21 primer, plus one coat of C-20 fiat white paint. As is apparent from the results, there was no frost accumulation on the inner surface of the siding for three of these four walls. Wall 4-D showed a slight frost accumulation on the top siding board for a small portion of its length, while all other boards were clear. However, this frost accumu- lation was not appreciable and should not result in any serious con- densation problem. . Group 3 consisted of two walls, 5-E and 7-E. These were constructed with treated lath having a medium-low permeability rate of 1.16 and sheathing having a medium permeability rate of 3.70. It should be noted that there was no evidence of frost accumulation on the inner surface of the siding of Wall5-E and only a slight line of frost appeared on the top siding board of Wall 7-E. The moisture content of the siding was comparatively low, whereas the moisture content of the lath was high as would be expected when a good vapor barrier is located on the cold side of the lath. 22 MOISTURE AND TEMPERATURE CONTROL Group 4 included Walls 6-F and 13-F constructed with treated lath having a medium-low permeability rate of 1.16 and sheathing having a high permeability rate of 51.8. These walls were constructed in accordance with the principle of a medium-low vapor permeability interior finish and a vapor porous exterior finish. As noted in Table III, no frost accumulation appeared on the inner surface of the siding of either of the two walls, which indicated that the moisture was effec- tively retarded by the interior vapor barrier. Walls 12-G in Group 5, 15-H in Group 6, and 14-I in Group 7 were all constructed with lath having a high permeability rate of 69.9 and with sheathing having permeability rates varying as follows: 0.44 for Wall 12-G, 3.70 for Wall 15-H, and 51.8 for Wall 14-I. For all walls of these last three groups there was a heavy accumulation of frost on the top siding board. This occurred along the joint at the top of the wall section and was independent of the type of sheathing used. The effect of the type of sheathing in combination with a vapor porous lath is indicated by the results on moisture content shown in columns 5, 6, and 7 of Table III. For Wall 12-G in which a sheathing of high vapor resistance was used, the siding moisture content ranged from a low of 6.4 to a high of 10.1 per cent. Using a sheathing of medium perme- ability, such as for Wall 15-H in Group 6, the moisture content of the top siding board increased to 16.5 per cent. When a vapor porous sheathing was used, as for Wall 14-I in Group 7·, the siding took on moisture ranging to a high of 20 per cent. For the walls in these last three groups the amount of moisture entering into the inner stud space was the same in each case. With a vapor resistant sheathing the moisture was retarded in passing through and was in part absorbed by the structural members themselves, or it passed through any open- ings occurring at the joints of the structure. In walls in which the sheathing was less vapor resistant, there was the opportunity for the vapor to flow through the material until it reached a cold surface where frost and condensation occurred. Irrespective of whether the moisture is taken up by the structural members, such as the studdings and plates, or by the exterior facing, such as the siding, both may be considered serious conditions, and construction methods should be such as to reduce the chance of vapor entering into the stud space. Conclusions from wall moisture study-Results have shown that when the vapor resistant lath on the inside surface of a wall possesses a permeability rate of 1.16 grains or less, the vapor passage is suffi- ciently reduced so that no serious condensation problem may be ex- pected within the wall. This is true for exterior wall constructions having either high or low permeability as used in this investigation and is in accordance with conclusions reached from smaller scale tests which were reported previously in Bulletin No. 22.8 On the basis of s Rowley and Lund, Vapor Transmission Analysis of Structural Insulating Board. TEST RESULTS 23 the results of these tests, together with previous experience, the vapor permeability of an acceptable interior barrier to be used in standard residential construction should be 1.25 grains, or lower, per square foot per hour per inch of mercury vapor pressure difference. A:t:tic Ventilating Requirements Simultaneous with the wall moisture study, investigations were conducted in the attic to determine the volume of ventilating air re- quired to prevent the formation of frost on the cold attic surfaces. Outside air was metered through a calibrated orifice, supplied through one of the attic windows, and exhausted by gravity through an open window located in the opposite gable. These studies were conducted under five different attic constructions. The principal objective was to determine the volume of ventilating air in cubic feet per hour re- quired to prevent any serious accumulation of moisture or frost on the various building rna terials used in the attic construction. The results of this study over the extended 66-day test period are summarized in Table IV. Column 1 contains brief descriptions of the five different ceiling and attic constructions investigated. Columns 2 through 7 show the test operating conditions regarding attic ventilation and inside and outside air temperatures, while the last column con- tains the principal results of the test; namely, notations concerning visual observations of frost accumulation on the various materials used in ceiling and roof constructions. Construction 1 consisted of a ceiling finish of 1/2-inch plaster over l-inch structural insulating board lath with no vapor resistant treat- ment, a loose-fitting attic door, and a conventional uninsulated roof. This and the four other constructions are shown in Figure 8. Two dif- ferent operating conditions were involved in tests under this construc- tion. The first consisted of no ventilation being supplied to the attic. Under this condition frost accumulated on the roof board as would be expected with such a construction and an unventilated attic. The second operating condition consisted of supplying ventilation at the rate of 4,130 cubic feet per hour, which was equivalent to approximately 13;4 air changes per hour. This was considered a comparatively high rate of ventilation and of a magnitude which would be difficult to attain in actual structures where the ventilation is obtained entirely by natural means through louvres located in the gable ends of the attic. At this ventilating rate it was found that the frost accumulation from the previous run remained on the roof boards. Although selected areas were scraped free of frost at the start of each test run, obser- vations revealed that frost continued to accumulate on these areas. It was not possible to establish a safe operating ventilating rate under this construction. The capacity of the ventilating unit was limited to 1%, air changes per hour; however, this was considered high and constituted a value which in actual practice would be difficult to attain Table IV. Attic Ventilating Requirements To Prevent Frost Accumulation for Different Types of Ceiling and Attic Construction Type of Construction Cubic feet per hour Construction 1-No vapor barrier, 0 no added insulation, loose-fit- 4,130 ting attic door Construction 2-No vapor barrier, 4,090 no added insulation, rubber gasket around attic door 3,720 3,570 3,450 2,700 Construction 3-Two coats of 2,615 paint on ceiling plaster as vapor barrier, no added insu- 2,090 lation, rubber gasket around 1.565 attic door Attic Ventilation Air changes per hour 0.00 1.72 1.70 1.54 1.48 1.44 1.12 1.09 0.87 0.65 Cubic feet per hour per sq ft attic floor area 0.0 10.6 10.4 9.6 9.0 8.8 6.9 6.2 5.1 3.8 Inside Air Temperature Dry- bulb 66.0 68.5 69.0 68.0 69.0 68.5 68.0 68.5 69.0 69.0 Dew- point 45.4 45.7 44.2 44.6 45.2 46.7 45.1 45.4 42.5 42.8 Outside Air Tem- perature (DegF) 0.0 -8.0 -9.0 -7.0 -5.0 -8.0 -8.0 -5.5 -4.0 -5.5 Notations on Frost Accumulation · Frost accumulation on roof boards Frost accumulation remained on roof boards after a 9-day run at this ventilating rate Necessary ventilating rate to prevent frost ac- cumulation was not established No frost accumulation on observed knots, nails, or roof boards* No frost accumulation No frost accumulation No frost accumulation Frost accumulation on knots and nails Necessary ventilating rate is between 2,700 and 3.450 cfh No frost accumulation on knots, nails, or roof boards No frost accumulation Frost accumulation on nails only Necessary ventilating rate is between 1.565 and 2,090 cfh • Selected areas in roof boards, including knots and nail shanks extending through roof boards, were scraped clear of frost after each test. These specific areas were used for observation. Table IV. Attic Ventilatinq Requirements To Prevent Frost Accumulation for Different Types of Ceilinq and Attic Construction-Continued Type of Construction Construction 4-Two coats of paint on ceiling plaster as vapor barrier, 1!2-inch insulat- ing board on floor joists, rub- ber gasket around attic door Construction 5-Two coats of paint on ceiling plaster as vapor barrier, 1!2-inch insulat- ing board on under side of roof rafters, rubber gasket around attic door Cubic feet per hour 1.610 1,790 2,090 1,220 1.945 3,020 Attic Ventilation Air Cubic feet per changes hour per per sq ft attic hour floor area 0.67 4.5 0.75 4.6 0.87 5.4 0.51 3.1 0.81 4.9 1.26 7.7 Inside Air Temperature -------- Outside AirTem- Notations on perature Frost Accumulation Dry- Dew- (DegF) bulb point 69.0 45.4 -7.0 Frost accumulation on nails only 69.0 44.3 -6.0 Frost accumulation on nails only 69.0 44.9 -5.5 No frost accumulation Necessary ventilating rate is between 1,790 and 2,090 cfh 68.0 44.4 -4.0 Frost accumulation on roof boards at eaves 68.0 46.6 -5.0 Frost accumulation on roof boards at eaves 68.0 46.4 -4.0 No frost accumulation at eaves Necessary ventilating rate is between 1,945 and 3,020 cfh 26 MOISTURE AND TEMPERATURE CONTROL by natural ventilation. Consequently, if a particular construction were found to require such a high ventilating rate, then corrections in the construction should be recommended as a more practical and economi- cal means of reducing vapor problems in the attic. Construction 1, therefore, was not considered a satisfactory type from the standpoint of vapor transmission into the attic, and condensation problems could be expected in a home having this type of construction even though precautions had been taken to provide louvres in the gable ends of the attic for ventilating purposes. Construction 2 was identical to Construction 1, except that a rub- ber gasket was provided around the attic door in order to reduce the passage of vapor through the cracks. Observations were made on selected areas in the roof boards, such as on the knots and nail shanks extending through the boards, the particular areas of which were colder than the other surfaces in the attic construction. As shown under Construction 2 in Table IV, the ventilation was gradually re- duced from the high rate attained under Construction 1 until the rate of 2,700 cfh was reached. At this rate frost began to accumulate on the knots and nail shanks. However, the next higher rate of 3,450 cfh was sufficient to prevent the formation of frost. No attempts were made to obtain an exact limiting rate but merely to obtain, by this trial and error method, an approximate value. Under this construction the desirable ventilating rate was found to be between 2,700 and 3,450 cfh or, in nominal figures, approximately 11/4 air changes per hour. Construction 3 was the same as Construction 2, except that two coats of paint were applied to the ceiling plaster. Under this con- struction it was possible to reduce further the ventilating require- ments to a point where the ventilating rate required to prevent frost was established between 1,565 and 2,090 cfh. No frost accumulated at 2,090 cfh, but it began to accumulate when the rate was reduced to 1,565 cfh. Thus, by applying a vapor barrier in the form of an oil paint to the ceiling plaster, a ventilating requirement of approximately three- fourths air changes per hour resulted. Construction 4 was identical to Construction 3 but for the applica- tion of 1/2-inch insulating board to the tops of the ceiling joists; i.e., providing an attic floor of insulating boards. This application did not appreciably increase the vapor resistance of the ceiling, but had the effect of lowering the coefficient of heat transmission through the ceil- ing. This, in effect, tended to result in lower temperatures within the attic. Under this construction there was very little difference in the ventilating requirements compared to Construction 3. Construction 5 differed from Construction 4 only in the location of the added insulation within the attic. Under this last construction the 1/2-inch insulating board was removed from the tops of the ceiling or attic floor joists and applied to the under side of the roof rafters. In cases where the attic space is planned for future living quarters, a con- TEST RESULTS 27 struction of this type would be used. The ventilating requirement under this last construction was established within the range of from 34 to 11/4 air changes per hour. Conclusions from a:t:l:ic ventilation study-A summary of the amount of attic ventilation required to prevent the accumulation of frost under the five different attic constructions investigated is pre- sented in Table V. The method of measurement was that of visual observation of frost accumulation within the attic at the different trial Table V. Summary of Attic Ventilating Requirements To Prevent Frost Accumulation for Five Different Attic Constructions-Inside Air 70 F and 40 Per Cent Relative Humidity; Outside Air -5 F Attic and Ceiling Construction Construction 1-No vapor barrier, no added in- sulation, loose-fitting attic door Construction 2-No vapor bariier, no added in- sulation, rubber gasket around attic door Construction 3-Two coats of paint on ceiling plaster as vapor barrier, no added insulation, rubber gasket around attic door Construction 4--Two coats of paint on ceiling plaster as vapor barrier, Vz-inch insulating board on floor joists, rubber gasket around attic door Construction 5-Two coats of paint on ceiling plaster as vapor barrier, Vz-inch insulating board on under side of roof rafters, rubber gasket around attic door Cubic feet per hour 2,700 to 3.450 1,565 to 2,090 1,790 to 2,090 1,945 to 3,020 Safe Ventilating Rate Attic air Cubic feet per changes hour per sq ft of (No. per hour) ceiling area Rates not established 1.12 to 1.44 6.9 to 8.8 0.65 to 0.87 3.8 to 5.1 0.75 to 0.87 4.6 to 5.4 0.81 to 1.26 4.9 to 7.7 rates of attic ventilation specified. Inasmuch as this essentially was a trial and error method, it should again be noted that no exact limit- ing values were found. The results are shown in the form of two fig- ures, both approximate, the lower being the ventilating rate at which frost accumulation did appear and the higher being the ventilating rate which resulted in the elimination of frost accumulation. With respect to a construction similar to Construction 1 in which the ceiling was untreated and the attic door loose-fitting, it was con- cluded that it would be difficult to prevent moisture condensation and frost accumulation by ventilation alone. From the results of the tests it was shown that a ventilating rate of 13/4 changes of air per hour was not sufficient for such purposes. A loose-fitting attic door was found to be a serious source of vapor leakage into the attic. When the attic door was fitted with a rubber gasket to form a tight seal against vapor leakage, it was then possible to determine the necessary rate of attic ventilation of approximately 11/4 air changes per hour as a value which did prevent frost accumu- 28 MOISTURE AND TEMPERATURE CONTROL lation. This applied as well to a construction, such as that of Con- struction 2, in which no vapor resistant treatment was applied to the ceiling plaster. If the construction consisted of a ceiling painted with some type of a vapor resistant coating, such as a standard interior finish oil paint, and if the attic door were weather stripped (i.e., fitted with a rubber gasket) as in Constructions 3, 4, and 5, it was concluded that a venti- lating range of from approximately three-fourths to one air change per hour should be sufficient to prevent any frost from accumulating within the attic during extended cold periods. Attic Air Temperatures The last principal factor of this investigation was concerned with the relationship between attic ventilation and attic temperatures. In structures where the attic is ventilated with outside air, the question has arisen as to what method should be used in calculating heat loss through the ceiling area. The methods of calculation for unventilated attics have been well established, but such is not the case with respect to ventilated attics. Most residential structures being built today in the northern temperature zone are supplied with louvres in the gable ends of the attic to obtain ventilation. This is generally regarded as good construction and serves as a precaution against problems of mois- ture condensation within the attic. For these structures it is important to know what method of heat loss calculation can be applied to deter ... mine the transfer of heat through the ceiling. Some investigators have suggested that in cases where the attic is ventilated, the attic air temperature may be assumed to be equal to the outside air temperature. The ceiling heat loss would then be calculated by multiplying the ceiling area by the coefficient of heat transmission through the ceiling and by the temperature difference between the inside and the outside air. This method would not give credit for any insulating value of the roof. Consequently, the results of such a calculation would show a heat loss value which would be higher than the actual case. But, the argument has been advanced that it is more accurate to use this method than to use the standard method of calculation for an unventilated attic where full credit is allowed for the roof construction. Very little is yet known of the relationship between attic air temperatures and outside air tempera- tures at specific rates of attic ventilation. It is hoped that this investi- gation will contribute experimental data on this point. A special series of tests were conducted for both Construction 4 and Construction 5, which have been previously described and are shown in Figure 8. In these tests the inside air temperature was maintained at approximately 70 F with relative humidity at 40 per cent, while the outside air. temperature was maintained at approxi- mately -6.5 F. Tests were conducted at no ventilation, 1,500, 3,000, and 4,200 cubic feet of air per hour. TEST RESULTS 29 Thermocouples were located within the cold room at levels of 2 feet above the bungalow roof and 5 feet above the floor at the east and west sides of the bungalow. In the open attic space three thermocouples were located at different elevations above each of the three rooms. One was located 2 inches above the attic floor joists, one at mid-height, and one at 7 feet above the attic floor. One thermocouple station was also provided in the northwest corner of the attic at a level of 1 foot above the floor joists, which with the others resulted in a total of 10 set ther- mocouple stations for temperature measurements within the attic. In addition, two thermocouples were provided for exploring different parts of the attic in order to determine the temperature variations that might occur at points other than the set thermocouple stations. Ther- mocouples were also established in each of the three rooms below. A detailed record of these tests is shown in Table VI for Construc- tion 4 and in Table VII for Construction 5, respectively. The test for each different ventilating rate was continued for a sufficient length of time to insure equilibrium temperature conditions. In Construction 4 the temperature variation from room to room at corresponding elevations was not considered great. At the mid-height level the temperatures were found to be substantially the same throughout the attic. The greatest variation occurred at the stations located 2 inches above the attic floor joists where a maximum varia- tion of approximately 4.5 F between Rooms 1 and 2 was obtained. In Construction 5 the attic temperatures were approximately 12 F higher than those in Construction 4, due to the location of the added insulation. At the mid-height level a maximum variation of approxi- mately 2.5 F occurred between the three rooms. Again, the greatest variation appeared at the stations located 2 inches above the attic floor joists where a maximum variation of approximately 8 F between Rooms 1 and 2 was obtained. This large variation was due mostly to the fact that the thermocouple station above Room 2 was located near the entrance of outside ventilating air. Results shown in column 9 of Table VII represent the temperatures at the low point in the attic directly in the path of this ventilating air. Exploration at other points within the attic, such as in the corner areas and near the eaves, did not reveal any greater temperature differ- ences than existed from room to room as measured at the set thermo- couple stations. To show clearly the attic air temperatures as affected by different rates of attic ventilation, the results of the detailed tables previously shown (Tables VI and VII) have been averaged at equilibrium condi- tions, and the results of these averages have been plotted in the form of curves which are presented in Figures 15 and 16 for Constructions 4 and 5, respectively. In Figure 15 for Construction 4 the inside air temperature was maintained at 70 F, with maximum fluctuations of 0.5 F, for the four different ventilating conditions. Likewise, for these Table VI. Detailed Record of Tests Showinq Relationship between Attic Air Temperatures and Inside and Outside Air Temperatures at Different Rates of Ventilation for Construction 4 (Rubber Gasket around Attic Door, Ceilinq Plaster Painted, %-Inch Insulatinq Board on Attic Floor Joists) Time Cold Room Attic Air Temperature, Deg F Inside Air Air Temperature, Deg F Room 1 Room2 Room 3 Temperature, Deg F -- --- Q) ai~ Ul ;:: Ul ;:: Ul ;:: Hours > i~ .....;-0> Siil 01 (jj Siil 01 (jj Siil 01 (jj of 0 Ol"tj ~:~ 'Qj .§:[ 'Qj ~:~ 'Qj .... N (") ~·til ~!! >~ ... > > > Test ,.Q ~:S~ ~(jj ~ ~(jj ~ ~(jj ~ s s s o_ -u; .... . ... ::::g ~~ -Ul _,..,.. s::o ..... > s::o ..... > s::o ..... > 0 0 0 "';'t! '+j'tOQ ..... 0 ..... 0 ..... 0 "';' 0 0 0 Nl-4 11)0> ..... s::u N:;::: ~~ ~ N:;::: ~~ ~ N:;::: ~~ r-.. JXi JXi JXi No Ventilation -3.0 -4.0 -3.5 14.5 15.5 14.0 15.0 14.5 14.0 14.5 15.0 13.5 13.5 70.5 70.5 70.5 2 -4.0 -4.0 -4.0 14.0 15.0 13.5 14.0 13.5 13.5 14.0 14.5 13.5 13.5 70.0 70.0 70.5 3 -4.0 -5.0 -4.0 13.0 15.0 13.5 14.0 13.5 13.5 13.5 13.5 12.5 13.0 70.0 71.0 70.5 4 -5.0 -5.0 -5.0 12.5 14.0 12.5 13.5 13.5 13.0 13.5 13.5 13.0 13.0 70.0 70.0 70.5 5 -5.0 -5.5 -5.5 13.0 13.5 12.5 13.0 13.0 12.5 13.0 13.0 12.0 12.0 71.0 70.0 70.5 6 -5.0 -6.0 -5.0 13.0 13.5 12.5 13.0 12.5 12.5 13.0 13.0 12.0 12.5 69.0 70.0 70.5 7 -5.0 -6.0 -5.0 13.0 13.0 12.0 13.0 12.0 12.0 12.5 13.0 12.0 12.0 69.0 70.0 70.5 8 -6.0 -6.0 -6.0 12.0 12.5 ll.S 12.5 12.0 12.0 12.5 12.0 11.0 11.5 69.0 70.0 70.0 9 -6.0 -6.0 -6.5 12.0 13.0 12.0 12.5 12.0 12.0 12.5 12.0 11.0 12.0 71.0 72.0 70.5 10 -6.0 -6.0 -5.5 12.0 13.0 12.0 12.5 12.0 12.0 12.5 12.5 u.s ll.S 69.0 70.0 69.0 ll -6.0 -6.5 -5.5 12.0 13.0 12.0 12.0 12.0 12.0 12.0 12.5 11.5 12.0 69.5 70.0 69.0 1.500 cfh Attic Ventilation -6.0 -6.5 -5.5 12.0 13.5 11.0 12.5 10.0 ll.S 12.0 12.5 11.0 11.5 69.5 70.0 69.5 2 -6.0 -6.0 -6.0 11.0 13.0 11.0 11.0 9.5 11.5 ll.5 12.0 10.5 11.0 70.5 70.5 69.0 3 -6.0 -6.0 -6.0 11.0 12.5 11.0 11.0 9.0 10.5 ll.S 10.5 10.0 10.5 70.0 70.0 69.0 4 -6.0 -6.0 -6.0 10.5 12.5 10.5 10.5 8.5 10.5 ll.S 10.5 9.5 10.0 69.5 70.5 69.0 5 -6.0 -6.5 -6.0 10.0 12.0 10.0 11.0 9.0 10.0 11.0 10.0 10.0 10.0 70.0 70.0 70.0 6 -5.0 -6.0 -6.0 10.5 12.0 10.0 10.5 8.0 10.5 10.5 10.0 10.0 10.0 70.5 69.0 70.5 7 -5.5 -6.5 -5.5 11.0 12.0 11.0 11.0 9.5 ll.S 11.0 10.5 10.5 11.0 69.0 70.0 70.0 8 -6.0 -6.5 -6.0 11.0 12.5 10.0 10.5 8.0 10.5 11.0 10.0 10.0 10.0 70.5 70.5 69.5 Table VI. Detailed Record of Tests Showinq Relationship between Attic Air Temperatures and Inside and Outside Air Temperatures at Different Rates of Ventilation for Construction 4 (Rubber Gasket around Attic Door, Ceilinq Plaster Painted, l/2-lnch Insulatinq Board on Attic Floor Joists)-Continued Attic Air Temperature, Deg F Time Cold Room Inside Air Air Temperature, Deg F Room I Room2 Room3 Temperature, Deg F Q) ~~ til 1:: til 1:: til 1:: Hours I> 'i~ .....''"' 1>· .... I>~ ... I> ~·: I> I> Test Q)tll ~= ~=~ ~Qj ~ ~Qj ~ ~Qj ~ El El El tl_ ..... 1ii -o ~~ .,.til .......... ~=:o ..... :> ~=:o ..... :> ~=:o ..... :> 0 0 0 -o ";'tl '+;400 ..... 0 ..... 0 ..... 0 0 0 0 (';! ... LOQ) ..... ~=:u (';!:;::: :s~ ~ (';!:;::: :s~ ~ (';!:;::: :s~ ~ p:; p:; p:; LSOO cfh Attic Ventilation-Continued 9 -6.0 -7.0 -6.0 10.5 12.0 9.5 10.0 8.0 9.5 10.0 10.5 9.5 9.5 69.5 70.0 70.5 10 -6.0 -7.0 -6.0 10.0 12.0 9.0 10.0 8.0 9.5 10.5 9.5 9.0 9.0 70.0 69.5 70.0 11 -6.0 -7.0 -6.0 10.0 11.0 9.5 10.0 8.0 10.0 10.5 10.5 9.5 9.5 70.0 70.0 70.0 12 -6.5 -7.0 -6.0 9.5 11.5 10.0 10.0 8.0 10.0 10.0 10.0 9.5 10.0 69.5 70.0 70.5 3,000 cfh Attic Ventilation -6.5 -7.0 -6.0 9.0 12.0 9.0 9.5 6.5 9.0 10.0 10.5 9.5 9.0 71.0 70.0 70.0 2 -6.5 -7.5 -6.0 9.5 10.5 9.0 9.5 6.0 9.0 10.0 10.0 9.0 9.0 69.5 70.0 70.0 3 -6.0 -7.5 -6.0 9.0 11.0 9.0 9.0 6.0 9.0 9.5 9.5 9.0 8.5 70.0 70.0 70.5 4 -6.5 -8.0 -6.5 8.5 10.0 8.5 9.0 6.0 9.0 9.0 10.0 9.0 9.0 69.5 70.0 70.5 5 -6.0 -7.5 -6.0 9.0 10.0 9.0 9.0 6.0 9.0 10.0 10.0 9.0 9.5 69.5 70.0 71.0 6 -6.5 -7.5 -6.0 9.0 10.5 9.0 9.0 6.0 9.0 9.5 10.5 9.5 9.0 70.0 69.0 70.0 7 -6.0 -8.0 -6.0 9.0 10.5 9.0 9.0 6.0 9.0 9.5 10.0 9.0 9.0 69.0 69.5 69.5 8 -6.0 -7.5 -5.5 9.0 10.5 8.5 9.0 7.0 9.5 10.0 10.0 9.0 9.5 69.0 69.5 70.0 4,200 cfh Attic Ventilation 1 -5.5 -7.0 -6.0 9.0 11.0 8.5 9.0 5.0 9.0 9.0 8.5 9.0 9.0 70.5 70.0 70.0 2 -5.0 -7.0 -6.0 8.5 10.0 9.0 9.0 5.5 9.0 9.5 9.0 9.0 9.0 70.0 71.0 70.0 3 -5.5 -7.0 -5.5 9.0 9.5 8.5 9.0 5.0 9.0 9.0 9.0 9.0 9.0 70.0 70.0 70.5 4 -5.5 -7.0 -5.5 8.5 10.5 9.0 9.0 6.0 8.5 9.0 9.0 8.5 9.0 70.5 70.5 70.5 5 -5.5 -7.0 -5.5 9.0 10.0 9.0 9.0 5.5 9.0 9.5 8.5 9.0 9.0 69.0 70.5 70.0 6 -6.0 -7.0 -5.5 9.5 11.0 9.0 10.0 9.5 9.5 10.5 10.0 9.0 9.5 70.0 69.5 69.5 Table VII. Detailed Record of Tests Showing Relationship between Attic Air Temperatures and Inside and Outside Air Temperatures at Different Rates of Ventilation for Construction 5 (Rubber Gasket around Attic Door, Ceiling Plaster Painted, ¥2-Inch Insulating Board on Under Side of Roof Rafters) Time Cold Room Attic Air Temperature, Deg F Inside Air Air Temperature, Deg F Room 1 Room 2 Room 3 Temperature, Deg F -- Q) a;~ Ill ~ Ill ~ rll ~ Hours > af~ ....;Q) E!Ui 0" a; E!Ui 0" a; E!Ui 0" a; of 0 Cll'O ]:§, ·a; ~:~ ·a; ~:[ ·a; .... N (') >'lil >· ... >iS: ... > > > Test ..Q 21;) ~:! ~.;::~ ~a; ~ ~a; ~ ~a; ~ Ei Ei Ei o_ .... .... ... rn =0 jjO =0 0 0 0 +<0 ~~ ......... .... 0 ..... > ..... 0 ..... > .... 0 ..... > 0 0 0 -o ";<0 '+j'OQ ~~ r:-:. ~~ r:-:. ~~ r:-:. N,_, LJ)Q) -=u N:;::: N:;::: N:;::: r:I; r:I; r:I; No Ventilation -1.5 -3.0 -2.0 30.5 31.0 30.0 31.0 28.5 30.0 31.0 29.5 30.0 30.0 70.0 72.0 73.0 2 -2.5 -3.0 -3.0 29.0 31.0 29.5 30.0 28.5 29.5 30.5 29.0 29.5 29.5 71.0 72.0 71.0 3 -3.0 -4.0 -3.0 28.5 31.0 29.0 29.5 28.0 29.0 29.5 28.5 29.0 29.0 69.0 71.0 71.0 4 -3.5 -4.5 -3.5 28.0 29.5 28.0 29.0 27.0 28.0 29.0 27.0 29.0 28.0 69.0 70.5 70.0 5 -3.5 -5.0 -4.0 27.5 29.5 29.0 28.5 26.5 27.5 28.5 27.5 28.0 28.0 70.0 70.5 71.0 6 -4.5 -5.0 -5.0 27.0 29.0 28.0 28.0 26.0 27.0 28.0 27.0 28.0 27.0 70.0 70.0 70.0 7 -5.0 -5.5 -5.0 26.5 28.0 27.0 27.5 26.0 27.0 27.5 26.0 27.0 27.0 70.0 70.0 70.0 8 -5.5 -6.0 -5.5 26.0 28.0 26.5 27.0 25.0 26.0 27.0 25.0 27.0 26.0 70.0 69.0 69.0 9 -5.5 -6.5 -6.0 25.0 28.0 26.0 27.0 25.0 26.0 27.0 28.0 26.0 27.0 70.0 69.0 69.0 10 -4.5 --5.0 -5.0 26.0 27.0 26.0 26.5 25.0 26.0 26.5 25.0 26.0 26.0 69.0 70.0 70.0 11 -5.5 -6.0 -5.5 25.0 28.0 26.0 26.5 25.0 26.0 26.5 25.0 27.0 26.0 70.0 70.0 70.0 12 -6.0 -7.0 -6.0 25.0 28.0 26.0 26.0 24.0 25.0 26.0 25.0 27.0 25.5 69.5 70.0 70.0 13 -7.0 -7.0 -7.0 25.0 27.0 25.5 26.0 24.0 25.0 26.0 24.0 27.0 25.0 70.0 70.0 70.0 14 -7.0 -7.5 -6.5 24.5 27.0 25.0 25.5 24.0 25.0 25.5 24.0 27.0 25.0 70.0 70.0 70.5 15 -7.0 -7.0 ·-6.5 24.5 27.0 25.0 25.5 23.0 25.0 25.5 24.5 27.0 25.0 69.5 69.5 70.0 1.500 cfh Attic Ventilation -6.0 -7.0 -6.0 20.0 22.0 20.0 21.0 16.5 20.0 20.5 19.0 22.5 20.0 69.0 70.0 71.0 2 -6.0 -7.5 -6.0 20.5 22.5 21.0 21.5 17.0 20.5 21.0 19.0 22.0 20.5 69.5 70.0 70.5 3 -5.5 -7.0 -6.0 20.5 23.0 21.5 21.5 17.5 21.0 21.5 19.0 22.5 21.0 69.0 69.0 70.0 4 -6.0 -8.0 -6.0 20.5 23.0 21.0 21.5 17.0 21.0 21.5 19.0 22.0 21.0 69.0 69.0 70.0 Table VII. Detailed Record of Tests Showinq Relationship between Attic Air Temperatures and Inside and Outside Air Temperatures at Different Rates of Ventilation for Construction 5 (Rubber Gasket around Attic Door, Ceilinq Plaster Painted, ¥2-lnch Insulatinq Board on Under Side of Roof Rafters)-Continued Attic Air Temperature, Deg F Cold Room Inside Air Time Air Temperature, Deg F Room 1 Room2 Room3 Temperature, Deg F Q) -Ui Ul .:E Ul .:E Ul .:E Hours I> i~ ..... El"lii IJI a; El"lii IJI a; El"lii IJI (i) ---0 Q)'tl .§:2- ·a; o·s ·a; .§:[ ·a; .... N C') of :>'"' >· ... ~.~ I> I> I> Test ..Q Q)U! QlUI ~a; ~ .:::·- ~a; ~ ~a; ~ El El El o_ -Ui -- --= .~ g ..... . .... ~g UI _ .... J::O J::O 0 0 0 ~; ";'ts :zg .... :> .... 0 .... :> .... 0 .... :> ";' 0 0 0 C'IM Ll)Q) N::;:: ::s~ .:.:. N= ::s~ .:.:. N::;:: ::s~ 1:'- p:; p:; p:; 3.000 cfh Attic Ventilation -6.5 -8.0 -6.5 19.0 22.0 19.0 19.5 14.0 18.5 19.0 19.0 21.5 19.0 70.0 70.0 70.0 2 -6.0 -7.5 -6.5 19.0 21.5 20.0 20.0 14.5 19.0 20.0 19.0 22.0 19.5 70.0 70.0 70.5 3 -6.0 -7.5 -6.0 19.0 21.5 20.0 20.5 15.0 19.5 20.0 19.0 22.5 19.5 70.0 70.0 70.5 4 -6.0 -7.5 -6.0 19.0 23.0 20.5 20.0 15.0 19.5 20.0 19.0 22.5 19.5 70.0 70.5 70.5 4,200 cfh Attic VentUation --6.0 -6.5 -6.0 24.0 27.0 25.0 25.0 17.0 23.0 23.0 22.0 27.0 24.0 69.5 70.0 70.0 2 -6.0 -6.5 -6.0 21.0 24.5 22.0 22.0 16.5 21.5 22.0 21.0 25.0 21.0 70.0 69.5 70.0 3 -5.0 -6.0 -5.0 21.0 24.0 22.0 22.0 16.0 21.0 21.0 21.0 24.5 21.0 69.5 70.0 70.0 4 -5.5 -6.5 -6.0 21.0 24.0 21.0 22.0 15.5 21.0 21.0 21.0 24.0 21.0 69.5 70.0 69.5 5 -5.0 -6.0 -5.5 20.0 23.0 20.0 21.0 15.0 20.0 20.5 20.0 23.0 20.0 70.0 70.0 70.0 6 -5.5 -6.0 -5.5 20.0 23.0 20.0 21.0 15.0 20.0 20.5 20.0 23.0 20.0 70.0 70.5 70.0 7 -5.5 -6.5 -5.5 20.0 23.0 20.0 20.5 15.0 20.0 20.0 20.0 23.0 19.0 70.0 70.0 70.0 8 -5.5 -6.5 -6.0 19.0 23.0 20.0 20.0 14.0 19.5 20.0 20.0 23.0 19.0 70.0 70.0 70.0 9 -6.0 -7.0 -6.0 19.0 22.0 20.0 20.0 14.0 19.0 19.0 20.0 22.0 19.0 69.5 69.5 69.0 10 -5.5 -7.0 -6.0 19.0 22.0 20.0 20.0 14.0 19.0 19.0 19.0 22.0 19.0 70.0 70.0 70.0 11 -5.5 -7.0 -5.5 19.0 22.0 20.0 20.0 14.0 19.0 19.0 19.0 23.0 19.0 70.0 70.0 70.0 12 -6.0 -7.0 -6.0 19.0 22.0 19.0 20.0 14.0 19.0 19.0 19.0 22.0 19.0 70.0 70.5 70.5 13 -6.0 -7.5 -6.0 19.0 22.0 19.0 20.0 14.0 19.0 19.0 19.0 22.0 18.0 70.0 69.5 69.5 14 -7.0 -8.0 -7.0 18.0 20.0 19.0 19.0 14.0 18.0 18.0 18.0 21.0 18.0 69.5 69.5 70.5 15 -6.0 -7.5 -6.0 18.5 21.0 19.0 19.5 14.0 19.0 19.0 19.0 22.0 18.5 70.0 70.0 70.0 16 -5.5 -8.0 -6.5 18.5 21.5 19.0 20.0 13.5 18.0 18.0 18.5 21.0 18.0 69.0 69.5 69.0 17 -6.0 -7.5 -6.0 18.5 21.5 19.0 20.0 14.0 19.0 19.0 19.0 21.5 18.5 69.5 70.5 70.0 34 MOISTURE AND TEMPERATURE CONTROL same four test runs the cold room temperature was maintained at -6.5 F, with maximum fluctuations of 0.5 F. Since the inside air and the outside air temperatures were substantially constant during all of the test runs, the curve shows directly the effect of ventilation upon the attic air temperatures. At zero or no ventilation the attic air temperature was 12.1 F. When the ventilating rate was set at 1,500 cfh, the attic air temperature dropped to 9.8 F. This drop of 2.3 F was encountered because of the amount of ventilation supplied. A further increase in the ventilating rate to 3,000 cfh resulted in a lower attic air temperature of 9.1 F, which was equivalent to an over-all drop of 3 F. When the ventilating rate was increased again to 4,200 cfh, the resulting drop in attic air temperature was only 3.1 F. In Construction 4 an interior finish of 1/2-inch plaster over l-inch insulating board lath was applied to the ceiling, and the tops of the ceiling or attic floor joists were covered with 1/2-inch insulating board. The effect of ventilation upon reducing attic air temperatures was con- sidered small. It was found that the roof remained effective in provid- ing thermal resistance against heat loss; hence, it would be an error not to allow some value for the roof in heat loss calculations with re- spect to this type of construction. 14 11.. ~12 0:: ::) ti ffi 10 a.. ~ LLI 1-8 26 18 vATTI 1 C AIR 1 INSIDE AIR 7cfF I" ~ OUTSIDE AIR -6.~ t--r--..... ..... t--. r--.... J/v -~ ~---- ~ 0 400 800 1200 1600 2000 2400 2800 S!OO 3600 4000 44100 ATTIC VENTILATION (CFH) Figure 15. Effect of ventilation on attic air temperatures for Construction 4 '" "'~ / ATTIC AIR INSIDE AIR 70°F I-- 'r--... / OUTSIDE AIR-6.S'F 1'-. _,I .............. " re-r--~ r-r--t-- .9 'f 0 400 800 1200 1600 2000 2400 2800 3200 3600 4000 4400 ATTIC VENTILATION (CFH) Figure 16. Effect of ventilation on attic air temperatures for Construction 5 TEST RESULTS 35 Figure 16 shows similar results concerning Construction 5. For the four different ventilating conditions in this test the inside air was maintained at 70 F, with maximum fluctuations of 0.5 F, and the out- side air was maintained at -6.5 F, likewise, with maximum fluctuations of 0.5 F. Inasmuch as the inside and outside air temperatures were substantially constant during all of the test runs, the curve may again be used to show directly the effect of ventilation upon the attic air temperature. In this construction the insulating board was removed from the tops of the ceiling or attic floor joists and applied to the under side of the roof rafters. This resulted in the ceiling having lower ther- mal resistance than in Construction 4, but with the roof having a comparatively high thermal resistance. As would be expected, the attic air temperatures in Construction 5 were found to be considerably higher than those in Construction 4. For the no-ventilation-condition the attic air temperature was 25.3 F. When the ventilating rate was set at 1,500 cfh, the attic air temperature was found to be 20.8 F, which was equivalent to a drop of 4.5 F. When the ventilation was increased further to 3,000 cfh, the attic air temperature dropped to 19.7 F. Com- pared to the no-ventilation temperature, this represented a decrease of 5.6 F. At the highest ventilating rate of 4,200 cfh, the attic air tem- perature was 18.8 F, or 6.5 F lower than that for no ventilation. Even at this high ventilating rate of 4,200 cfh, which is equivalent to approxi- mately 13,4 air changes per hour, the attic air temperature was con- siderably higher than the outside air temperatures. Again, for this type of construction it would be a serious error not to G~-llow some value for the roof in heat loss calculations. Conclusions from attic air temperature study-:-The principal signi~ ficance of these experimental data on attic air temperatures lies in their use in calculating heat loss through the ceiling of a ventilated attic. Therefore, Table VIII has been prepared to show the comparison of actual attic air temperatures with outside air temperatures when used in such calculations. Since the heat transfer by conduction through a given construction is directly proportional to the design temperature difference, the ratio between temperature differences would be equivalent to the ratios between actual heat losses. In this table the last column shows the ratio between temperature differences as obtained from columns 5 and 6. Column 5 shows the design tempera- ture difference under actual conditions, while Column 6 shows· the design temperature difference between inside air and outside air, which has been proposed by some investigators as the temperature difference to be used in calculations on heat loss through the ceiling of a ventilated attic. Under Construction 4, in terms of percentage, it can be seen that if the attic air temperature is assumed to be equivalent to the outside air temperature, the heat loss calculations would be approximately 36 MOISTURE AND TEMPERATURE CONTROL 25 per cent too high in an attic ventilated at the rate of 13/4 air changes per hour. Under Construction 5 this error would be considerably greater, amounting to approximately 50 per cent at the corresponding ven- tilating rate. It was recognized that a construction such as Construction 5 may not be used where louvres will provide attic ventilation. How- ever, many structures similar to this type are built with the thought of planning future living quarters within the attic. In these structures windows usually replace the louvres, and there may be some reluc- tance on the part of the residents to open the windows for ventilating purposes for fear of causing serious heat loss. A significant fact appar- ent from these results is that with constructions of this type and with a large amount of ventilation in the attic, the roof still offers substan- tial thermal resistance. It was concluded that if the full insulating value for the roof were allowed in the calculations in the case of a ventilated attic, the magni- tude of the error would be relatively small and would depend upon the type of construction and the amount of ventilation provided. In Table VIII. Summary of Relationship between Attic Air Temperatures and Outside Air Temperatures at Different Rates of Ventilation for Constructions 4 and 5- Comparison of Attic Air Temperatures with Outside Air Temperatures Ventilat- ing Rate (cfh) 0 1,500 3,000 4,200 0 1.500 3,000 4,200 Inside Air Tempera- ture When Used in Calculations on Heat Loss through the Ceiling Area Outside Air Tempera- ture Actual Attic Air Tempera- ture Design Temperature Difference for Heat Loss Calculation through Ceiling Inside air minus Inside air minus attic air outside air (A) (B) Construction 4-lf2-lnch Insulating Board on Floor Joists 70.0 -6.0 12.1 57.9 69.9 -6.4 9.8 60.1 76.3 69.5 -6.6 9.1 60.4 76.1 69.5 -6.0 9.0 60.5 75.5 Construction S-1/2-lnch Insulating Board on Under Side of Roof Rafters 69.9 -6.9 25.3 44.6 69.3 -6.4 20.8 48.5 75.7 70.2 -6.5 19.7 50.5 76.7 69.7 -6.6 18.8 50.9 76.3 Ratio BtoA 1.27 1.26 1.25 I.5a· 1.52 1.50 a construction like that of Construction 4 of this investigation, this error would be small. For example, at a ventilating rate of 13;4 air changes per hour, the calculations, allowing full value for the roof, would be approximately 5 per cent too low. However, if no valuci were allowed for the roof, the calculations would result in being. approximately 25 per cent too high. In the case of Construction 5, the error would be somewhat larger. At a ventilating rate of 13;4 air SUMMARY 37 changes· per hour, the calculations, again allowing full value for the roof, would be approximately 14 per cent too low. If no value were allowed for the roof, the resulting calculations would be approximately 50 per cent too high. Thus, for the above two constructions it would be considerably more accurate to allow full value for the roof in the heat loss calculations than to neglect it entirely. SUMMARY This investigation was conducted on a full scale test bungalow of conventional frame construction under simulated winter conditions. The inside air temperature was maintained at 70 F and 40 per cent relative humidity, while the outside air temperature was maintained between 0 F and -10 F. As previously noted, the four objectives of this research program were: (1) to establish more firmly whether an interior surface construction of 1.25 grains per square foot per hour per inch of mercury vapor pressure difference possessed sufficiently low vapor permeability to safeguard a home against wall condensa- tion problems; (2) to determine the effect of different combinations of vapor resistant interior and exterior surface constructions upon mois- ture condensation within the walls of a standard residential struc- ture; (3) to obtain experimental data on the volume of ventilating air required to prevent the formation of frost within the attic, utilizing different ceiling constructions; and (4) to determine the relationship between attic air temperatures and outside air temperatures at differ- ent rates of attic ventilation. Based on the results obtained over an extended test period and on the specific constructions, operating temperatures, and humidity in- volved, the following may be concluded: 1. Conventional frame walls constructed with an interior finish having a vapor permeability rate equal to, or lower than, 1.25 grains per square foot per hour per inch of mercury vapor pressure difference may be con- sidered safe with respect to effectively reducing moisture condensation in residential structures. 2. The degree of vapor resistance of the exterior finish of an exposed wall in a residential construction is not as critical a factor as the vapor resistance of the interior finish. If the interior finish is of sufficiently high vapor re- sistance, the exterior construction may possess either a high or low perme ... ability rate. 3. To minimize effectively the formation of frost within the attic, the ventilat- ing requirements amount to approximately one air change per hour, pro- vided the attic door is made to fit tightly with the aid of a gasket and the ceiling is made substantially vapor resistant, such as by the application of two coats of an interior oil paint. The ventilation may be provided me- chanically or by natural circulation through sufficient louvre area. "For 38 MOISTURE AND TEMPERATURE CONTROL inside air conditions of 70 F and relative humidities of more than 25 per cent in combination with -10 F outside air, attic ventilation should be used. For these conditions and with a building 25 feet square, openings of 1/4 square inch per square foot of ceiling area on each exposed wall will be sufficient." (Bulletin No. 18) 4. Attic air temperatures were found to be substantially higher than the out- side air temperatures, even during operation under the high ventilating rates. In calculating the heat loss through the ceiling where the attic space above is ventilated in an amount sufficient to safeguard against con- densation, the heat loss due to such ventilation may be neglected without serious error.