1、572 2010 ASHRAEThis paper is based on findings resulting from ASHRAE Research Project RP-1235.ABSTRACTA project initiated by TC 4.4 looked at the nature, signif-icance and control of solar-driven water vapor diffusion inwall systems. The project combined experimental and simula-tion work to provide
2、an in-depth characterization of thephenomena occurring during inwards vapor diffusion in insu-lated wall assemblies. Small- and large-scale laboratory testsprovided data under controlled conditions, indicating thatporous claddings that absorb rain become the source of mois-ture when subjected to sol
3、ar radiation. The vapor permeanceof the interior finish layer is a key parameter leading to mois-ture accumulation in the gypsum board. Field studies wereperformed over a period of 2 years and occurrence of solardriven diffusion was documented for different wall assemblies.Once the capacity of compu
4、ter models to reproduce theobserved behavior was verified, a parametric study wasperformed for 18 different wall assemblies in seven locationsin USA. It was found that the design of wall assemblies shouldinclude the evaluation of behavior under conditions leading toinwards diffusion in warm and mixe
5、d climates.INTRODUCTIONThe occurrence of inwards vapor flow, leading to inter-stitial condensation or moisture accumulation in interior finishmaterials, has been identified and studied by various research-ers, e.g., Wilson 1965; TenWolde and Mei 1985; Sherwood1985; Southern 1986; Andersen 1987; Sand
6、in 1993; Straubeand Burnett 1995; 1998; Knzel 1999, 2005; Karagiozis 2002;Pressnail et al. 2003; Lawton and Brown 2003; and Wilkinsonet al. 2007. Studies have investigated occurrence of thisphenomenon both in hot and humid, and cold climate areasand for different compositions of wall assemblies. Var
7、iousstrategies have been suggested by researchers although not allstudies agree on the effectiveness of each proposed method.Most of the previous work done on inwards vapor transportaimed at reporting the occurrence of inward moisture move-ment due to high temperature gradient, and a full set of dat
8、ahad not yet been produced and analyzed. The cyclic vapor flowdriven by solar radiation and the influence of the wall compo-sition on the hygrothermal performance and durability of wallsystems subjected to such flow needed to be further lookedupon to provide a more comprehensive understanding of the
9、phenomena. To answer these needs, a project entitled “Thenature, significance and control of solar-driven diffusion inwall systems” (ASHRAE RP-1235) was initiated and super-vised by ASHRAE Technical Committee TC 4.4 “BuildingMaterials and Building Envelope Performance.” Three insti-tutions were invo
10、lved in this project, namely the BuildingEnvelope Laboratory of Concordia University, Montreal,Canada, the Building Physics Laboratory of the KatholiekeUniversiteit Leuven (KULeuven) Belgium, and the BuildingTechnology Center at Oak Ridge National Laboratory (BTC/ORNL). Drs. Dominique Derome, now wi
11、th Swiss FederalLaboratories for Materials Testing and Research (EMPA)Switzerland, Jan Carmeliet, now with Swiss Federal Instituteof Technology Zrich (ETH Zrich) and EMPA Switzerland,and Achilles Karagiozis of ORNL formed the team ofresearchers at the heart of this project.The Nature, Significance a
12、nd Control of Solar-Driven Water Vapor Diffusion inWall SystemsSynthesis ofResearch Project RP-1235Dominique Derome, PhD Achilles Karagiozis, PhD Jan Carmeliet, PhDMember ASHRAEDominique Derome is Group Leader, Wood Laboratory, Swiss Federal Laboratories for Materials Testing and Research EMPA, Dben
13、dorf,Switzerland. Achilles Karagiozis is a distinguished research and development engineer, Building Technology Center, Oak Ridge NationalLaboratory, TN, USA. Jan Carmeliet is Chair of Building Physics, Swiss Federal Institute of Technology ETH Zrich and Head of BuildingScience and Technologies Labo
14、ratory, EMPA, Dbendorf, Switzerland.OR-10-061 (RP-1235) 2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions 2010, Vol. 116, Part 1. For personal use only. Additional reproduction, distribution, or transmission in ei
15、ther print or digital form is not permitted without ASHRAEs prior written permission. ASHRAE Transactions 573The overall objective of the project was to develop a betterunderstanding of the nature and significance of solar-driveninward vapor diffusion. More specifically, the project intended to:to i
16、dentify under which conditions (climate, wall com-positions) solar driven vapor transport may lead to dura-bility problemsto develop appropriate design guidelines to predict andmanage this phenomenon as a function of climateto transfer the knowledge generated from this researchinto the ASHRAE Handbo
17、ok of Fundamentals.This paper presents a summary of the work performed inthis project. Figure 1 presents the interrelations of the differenttasks of the project. The next section reports the experimentalwork which looked at the behavior of assemblies using small-scale samples and large-scale samples
18、 in laboratory settings,and in the field. The following section reports the parametricstudy performed using simulation. Finally, guidelines andmajor conclusions are presented.EXPERIMENTAL WORKA hierarchy of experimental analysis was performed.First, the influence of temperature and temperature gradi
19、entswas studied at the small-scale level under constant and cyclicconditions. In a second step, laboratory experiments on large-scale including artificial heat/rain cycling were done. Finally,in a third step, walls were exposed to the real climate in a fieldtest building.Small-Scale Laboratory Tests
20、Two series of tests were performed at the Laboratory ofBuilding Physics of the Katholieke University of Leuven, Bel-gium. The objective was to document simultaneously thebehavior of eight assemblies under well controlled and docu-mented conditions to provide a set of data that could be used toreliab
21、ly verify the capacities of the simulation tools to calculateheat and moisture transfer at high temperature. This workresulted in the design and construction of a test set-up dedicatedto this study and the development of a novel test procedure, seeFigure 2. The testing conditions were determined bas
22、ed on pre-liminary simulations of wall behavior under the South Carolinaclimate. Then, more simulation was performed to fine-tune thetesting conditions, temperature, relative humidity, initial liquiduptake and duration. The simulation provided the test air tem-perature that would result in surface c
23、onditions equivalent tothe average ones resulting from sun radiation exposure. Underconstant conditions, the walls were exposed to 40C (104F)exterior and 19C (66F) interior temperatures, where the brickwas initially wetted to 50% of capillary saturation and coveredto prevent outward drying for 17 da
24、ys, as shown in Figure 2B.Then, the cover was removed from the brick and the test con-tinued for 20 days. The same test was repeated with daily cyclicconditions, where the exterior temperature was 40C (104F)for 8 hours and 19C (66F) for 16 hours.Eight different assemblies were tested. Figure 3 prese
25、ntsthe composition of the wood-framed walls with brick clad-ding. Three components were varied: the weather resistivemembrane (WRB: spun-bonded polyolefin SPBO and build-ing paper BP), the interior finish (vapor open paint and vaportight wall vinyl covering, VWC) and the presence or absenceof wood s
26、tud.Some tests results are presented. Figure 4 shows the simi-lar behavior of the walls with low permeance interior finish(Figure 4A) and with vaportight interior finish (Figure 4B) interms of the variations of moisture content of the cladding andthe sheathing. The vaportight interior finish leads t
27、o an impor-tant accumulation in the gypsum board. Similar results arefound for cyclic conditions, as shown in Figure 5, although themagnitude of moisture accumulation is less and the onset ofdrying is delayed.Overall, the data illustrates the clear role of the interiorfinish. For walls with vinyl wa
28、ll paper, a high moisture contentin the gypsum board and no complete drying of the wall by theend of the test was observed. For the walls with paintedgypsum, a lower moisture content in gypsum was alwaysobtained and the walls had dried out. The wetting of theoriented-strand board (OSB) exterior shea
29、thing was found todepend slightly on the type of WRB and the presence orabsence of wood studs. Finally, cyclic loading led to the sameoverall observation, although the reduction of hours underhigh temperature gradient resulted in lower moisture contentsand a delay of the onset of drying, with a redu
30、ction in the rateof drying.Large-Scale Laboratory TestsThe large-scale tests were performed at the Laboratory ofBuilding Envelope of Concordia University, Montreal,Canada. The objective was to document the behavior of fivelarge-scale assemblies under more complex, and somewhatmore realistic, conditi
31、ons than the small-scale tests, in orderto, eventually, document a size effect. This work also resultedFigure 1 Organization and interrelations of the tasks of theproject. 2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Tran
32、sactions 2010, Vol. 116, Part 1. For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission. 574 ASHRAE Transactionsin the design and construction of a test set-up dedicated to thisstudy and
33、the development of a novel test procedure, seeFigure 6. The wetting conditions were chosen to be similar tothe small-scale test ones, with 60% capillary uptake for thebrick in the wetting stage.The testing conditions in terms of surface temperaturewere determined based on preliminary simulations of
34、wallbehavior in South Carolina climate. After wetting, the wallswere exposed to a radiation lamp array that led the surface toreach 60C (140F), while the interior conditions were 17C(63F) and 50% RH. After eight hours, the lamps were turnedoff for sixteen hours and the surface temperature decreased
35、tothe ambient laboratory conditions, approximately 2022C(7075F). The daily wetting-heating procedure was repeateduntil sufficient data was acquired.Five different wood-framed assemblies were tested. Theparameters included the type of cladding (brick cladding,cement stucco), the interior finish (vapo
36、r open paint and vaportight VWC), the type of exterior sheathing (OSB, extrudedpolystyrene board XPS) and the presence or absence of venti-lation in the air space behind the brick.Some tests results are presented. For example, Figure 8Ashows the moisture content in the brick cladding and in thegypsu
37、m together with the changes in moisture of the wholebackwall in the brick wall with vapor tight interior finish.Figure 8B shows the vapor pressure in three planes across thesame wall assemblies, where a vapor pressure gradienttowards the inside is present throughout the wetting phase andwell into th
38、e drying phase. Figure 9 compares the moisturecontents attained in the gypsum board of the 5 different wallassemblies. The moisture content increases rapidly in the twowalls with vaportight interior finish.The data confirmed the overall behavior observed with thesmall-scale tests, documenting the pr
39、esence of considerablevapor flow and stressing once again the role of the interior fin-ish. In addition, the large-scale tests permitted to highlight sev-eral phenomena due the size of the specimens, i.e., the presenceof vertical gradients in terms of temperature and moisture con-tent in the gypsum
40、board, the role of the air space as a capillarybreak (brick vs. cement stucco) and allowing ventilation (ventedvs. non-vented air space). Finally, the occurrence of noticeablevapor flow in the assembly with the XPS sheathing may indicatethe necessity to study materials at higher temperature.Figure 2
41、 (A) Overall view of the test setup developed for the small-scale tests. (B) View of the eight small-scale specimens lyinghorizontally. Above, exterior conditions are maintained; below, laboratory conditions reproduce interior air-conditioned conditions.(A) (B)Figure 3 Schematic representation of th
42、e small-scale spec-imens, indicating composition, geometry andoverall dimensions. 2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions 2010, Vol. 116, Part 1. For personal use only. Additional reproduction, distribut
43、ion, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission. ASHRAE Transactions 575Field MonitoringTo complement the laboratory studies, eight wall assem-blies were monitored under field conditions. The test facilitywas situated in Charleston (Holl
44、ywood), SC, where theclimate is categorized as a hot and humid climate.Table 1 lists the composition of the 8 walls. Tempera-ture, relative humidity and moisture content were recordedat several heights and depths through out the wall speci-mens as shown in Figure 9. The recorded climatic condi-tions
45、, i.e., air temperature and relative humidity, rain andwind speed, were found to be near the 30-year average. Theinterior conditions were controlled to be constant at 23C(73F) and 40% RH in summer and 21C (70F) and 60%RH in winter, representative of conditions monitored forthe climatic area.The moni
46、toring campaign provided a wealth of data. Theanalysis of the data aimed at identifying the conditions acrossFigure 4 (A) Moisture content in the brick cladding, OSB sheathing and gypsum board, with paint interior finish, of the small-scale sample exposed to constant conditions. (B) Moisture content
47、 in the brick cladding, OSB sheathing and gypsumboard, with vinyl wall covering as interior finish, of the small-scale sample exposed to constant conditions.(A) (B)Figure 5 (A) Moisture content in the gypsum board in the small-scale samples with paint as interior finish. Comparison ofconstant versus
48、 cycling testing conditions. (B) Moisture content in the gypsum board in the small-scale samples withvinyl wall covering as interior finish. Comparison of constant versus cycling testing conditions.(A) (B) 2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ash
49、rae.org). Published in ASHRAE Transactions 2010, Vol. 116, Part 1. For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission. 576 ASHRAE TransactionsFigure 6 (A) Schematic representation of the large-scale testing setup. (B) Side view of large-scale setup. (C) View of setupduring simulation of the solar radiation.(A) (B) (C)Figure 7 (A) Moisture content of the different components of the brick wall with vinyl wall covering specimen versus time.(B) Va
