1、914 2009 ASHRAEABSTRACTWater vapor is one of the primary products of combustion. Since unvented gas fireplaces release all combustion products into the home this means that a substantial amount of moisture is added to the indoor air during fireplace operation. An anal-ysis of the indoor moisture lev
2、els in 30 homes using unvented gas fireplaces was performed using measurements from multi-ple locations in each home. Several different metrics were considered, including relative humidity (commonly used in assessments of comfort), vapor pressure (a temperature-inde-pendent metric), and dew-point (i
3、mportant for potential prob-lems at surfaces). There was a median increase in vapor pressure of about 0.1 kPa (0.015 psi) for the sample of homes. Vapor pressures were typically fairly uniform within each home, with the most distant rooms often showing a slightly lower vapor pressure. The direction
4、and magnitude of changes in relative humidity depended on the proximity to the fireplace, with locations further from the fireplace having higher relative humidity levels because of a lesser temperature influence. Dew-point levels rarely exceeded 50F (10C), which is approximately the dew-point requi
5、red for condensation on a double-pane window when the indoor temperature is 70F (21C)and the outdoor temperature is 10F (12C).INTRODUCTIONUnvented gas heating appliances are similar to other space heating and hearth products, with one exception they dont have a chimney, flue or vent. For builders an
6、d homeown-ers this is appealing because no hole in the roof or walls is required, allowing for substantial installation savings and preventing potential leaks at the point of venting. Since the installation of unvented gas fireplaces does not require a chim-ney or any other vents, they can be easily
7、 installed almost anywhere. However, since unvented gas fireplaces release combustion products with potentially adverse health effects to the living space, they have remained controversial among the building science community (e.g. Energy Design Update 2001). Moisture has been one of the greatest co
8、ncerns because of the potential effects of elevated moisture content on both health and building durability.Very few field measurement studies have been performed to assess the in-situ concentrations of combustion products in homes using unvented hearth appliances. One study looked at CO and NO2emis
9、sions in two homes in Boulder, Colorado at various times between 1997 and 2000 (Dutton et al. 2001). Results showed significant indoor pollutant accumulation when the fireplaces were used for extended periods of time. The study was careful to point out that, owing to the high elevation of Boulder, t
10、he tests were performed in conditions that are relatively oxygen-deprived, facilitating the production of CO.A modeling study which looked at moisture buildup in homes using unvented gas hearth appliances using stochastic methods, came to the conclusion that under most conditions, when the appliance
11、 is used for less than 4 consecutive hours, the indoor air relative humidity remains below the level required for fungal growth (Whitmyre and Pandian 2004). However, this paper did not address vapor pressure or dewpoint, which is important when determining the potential for mold growth on a surface.
12、From 2005-2008 the authors performed a multi-faceted research study on unvented gas fireplaces, the central focus of which was field testing in 30 homes over two winters that utilized these appliances. Testing at each home lasted for 3-4 Indoor Moisture in 30 Homes Using Unvented Gas FireplacesPaul
13、W. Francisco Jeffrey R. Gordon William B. RoseMember ASHRAE Member ASHRAEPaul W. Francisco and Jeffrey R. Gordon are research specialists and William B. Rose is a research architect at the Building Research Coun-cil, University of Illinois at Urbana-Champaign, Champaign, IL.LO-09-087 2009, American
14、Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions 2009, vol. 115, part 2. For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior writt
15、en permission.ASHRAE Transactions 915days, and included measurements of carbon monoxide (CO), carbon dioxide (CO2), oxides of nitrogen (NOx) and its components nitric oxide (NO) and nitrogen dioxide (NO2), oxygen (O2) depletion, and water vapor (H2O) in a location approximately 6 feet away from the
16、fireplace. Water vapor and CO were also recorded using portable passive sensors in 5 locations throughout the home to assess the distribution of combustion products throughout the home. Data were recorded every minute. Residents were asked to use their fire-places in a manner consistent with their n
17、ormal usage patterns.Assessing water vapor is less straightforward than assess-ing the other gaseous products of combustion. There are many interior sources of moisture; the humidity levels outdoors can change significantly in relatively short periods of time; and household materials and furnishings
18、 can buffer water vapor, resulting in a muted response relative to outdoors.Further, whereas there are clear thresholds provided by various agencies on the other products of combustion there is no clear threshold for water vapor. In the past, ASHRAE (Standard 55:1992) showed a maximum RH of 60% in o
19、rder to provide comfort; however, that threshold has been replaced with a variable threshold that depends on other factors such as temperature and which can be as high as 85% (ASHRAE Stan-dard 55:2004). This threshold was not intended to address either health or building durability, rather, it was s
20、imply a value for general indoor air relative humidity to provide comfort. Because building surfaces can be much colder than the room air it is possible to get much higher surface relative humidity (and possibly condensation) on windows and other building surfaces even if the room air has what would
21、 be perceived as “acceptable” relative humidity based on comfort criteria.In addition to the lack of a clear threshold, water vapor is also different from the other gases in that more is not neces-sarily worse. When the air is very dry people tend to perceive it to be uncomfortable. Whereas an incre
22、ase in nitrogen diox-ide can be perceived as undesirable, an increase in humidity levels is more ambiguous unless the levels get high enough to result in degradation of building materials or mold growth. There is also evidence that dampness in buildings, even in the absence of mold, has a correlatio
23、n with illness, especially respiratory problems (Institute of Medicine 2004), though it is not clear what level of humidity corresponds to sufficient “dampness” to be of concern.Finally, the impact of temperature on relative humidity makes it so that relative humidity is not necessarily a good indi-
24、cator of actual change in moisture content in the air. When an unvented fireplace is operating and warming up the room it can reduce the relative humidity near the fireplace even though it is adding moisture to the air, which increases the dew-point and vapor pressure. Locations further away from th
25、e fireplace may at the same time experience an increase in relative humidity.This paper presents the results of the moisture measure-ments from the 30 homes in the study. Since nearly all previous published work on moisture from unvented fireplaces (as well as ASHRAE standards) have focused on relat
26、ive humidity as the metric of interest, the results expressed in relative humidity are presented. However, temperature-independent metrics such as dew-point are important for assessing the potential for problems at building surfaces that could lead to durability problems. Therefore, even while resul
27、ts are presented in terms of relative humidity the limitations of such a characterization of moisture content are discussed and results are also presented in terms of dew-point and vapor pressure.METHODOLOGYTest homes were obtained as a sample of convenience, recruited through an e-mail newsletter.
28、House age ranged from less than 2 years old to over 100 years old. Blower door tests were performed on each home. The average airtightness for the 30 homes was 11.7 air changes per hour at 50 Pascals depressurization (ACH50), with a median of 10.7 ACH50, a minimum of 5.6 ACH50, and a maximum of 27.8
29、 ACH50. Fireplace operational times were controlled by the occupants, whether by manually turning the fireplace off and on or by using a thermostat. Individual on-time periods ranged from less than 5 minutes to greater about 9 hours, with an average of about 44 minutes and median of about 20 minutes
30、.The primary humidity measurement was obtained using a capacitance-type probe with a rated accuracy of 1%. This probe was mounted at the same location as the sample gas inlet, about 6 ft (2 m) away from the fireplace. This probe also measured the concurrent temperature, and the relative humid-ity me
31、asurements were converted to dew-point and vapor pres-sure for analysis.Portable passive temperature/relative humidity sensors were placed in multiple locations throughout the homes. One sensor was always placed on the fireplace mantel. Two others were placed in convenient locations in the same room
32、 as the fireplace was located, one approximately in the middle of the room and the other at the far end of the room. Another sensor was placed in an adjacent room, which was frequently a kitchen and was always open to the fireplace room. The final sensor was placed in a distant room, usually a bedro
33、om, which was usually able to be closed to the rest of the house by the occupants if so desired. Measurements from these sensors were also converted to dew-point and vapor pressure.In order to fairly evaluate the moisture contribution of the fireplace it is necessary to use a humidity metric that is
34、 inde-pendent of temperature, unlike relative humidity. Vapor pres-sure was used for this temperature-independent metric. Vapor pressure, measured in kilopascals (kPa), is the metric of choice for IEA Annex 14 (International Energy Agency 1991).In the absence of dehumidification, average vapor pres-
35、sure is typically higher indoors than outdoors due to interior moisture sources. Ideally, there would be a clear and steady increment between indoor and outdoor vapor pressures when the fireplace was off, and a clear increase in indoor vapor pres-sure when the fireplace was on. That would allow for
36、a direct assessment of the impact of the fireplace on indoor humidity 916 ASHRAE Transactionslevels. However, Figure 1 shows that there can be times when the outdoor vapor pressure temporarily rises above indoor conditions and that, on the time scale of a fireplace event, the variability of indoor a
37、nd outdoor vapor pressures is such that this type of evaluation is not practical. As a result, other means of assessing the contribution of the fireplace are necessary.Two methods of evaluating the contributions of the fire-places to indoor vapor pressure were employed. The first was to average the
38、indoor vapor pressure during fireplace on-times and separately average the indoor vapor pressure during fire-place off-times. The difference between the two is an indica-tion of the average contribution of the fireplace. It is certainly not exact, since there will still be elevated moisture levels a
39、t the beginning of the off-time and there will be a ramp-up period at the beginning of the on-time. Further, the changes due to factors other than the fireplace are unlikely to be the same in both on- and off-times. However, the results are still instructive. Off-times are often long enough that the
40、 impacts of the elevated moisture levels at the beginning are usually small. The average values for the on-times will typically be more impacted by the ramp-up period because on-times are usually shorter, so the difference between on-times and off-times would then be underestimated using this method
41、. Finally, while the other moisture contributors are unlikely to be consistent, when considered over the full sample of homes the likelihood is that the bias associated with on-time and off-time selection will be small.The second method of evaluating the contributions of the fireplaces to indoor hum
42、idity was to simply take the difference between the indoor vapor pressure when the fireplace shut off and the indoor vapor pressure when the fireplace had come on. In other words, the vapor pressure rise for each cycle was calculated. The results were normalized by runtime to avoid the appearance of
43、 fireplaces that ran for long periods of time being naturally greater contributors of indoor moisture. This method does not make use of the off-time data at all, but instead uses the vapor pressure at the beginning of the on-time as a baseline.As with the previous evaluation technique, this method i
44、s not perfect. It does not account for changes in vapor pressure that would have occurred without the fireplace operating due to internal sources or responses to changes in outdoor moisture levels. Also, normalizing by runtime is not accurate since the increase in moisture is not linear. However, in
45、dividual cycle runtimes were typically short enough that the impact of other sources of moisture was likely to be small in most cases, such that the average over all cases would not be heavily affected. Also, because of the cycle runtimes being short the exponen-tial nature of the curve often was no
46、t well developed such that a linear approximation was deemed reasonable.Figure 1 Example of indoor and outdoor vapor pressure (1 kPa = 0.14504 psi; F = 1.8 C + 32).ASHRAE Transactions 917RESULTSIncrease in Indoor Moisture Due to Fireplace OperationFigure 2 shows the results of the first analysis met
47、hod described in the Methodology section using box plots. The values used for the box plots are the average indoor vapor pres-sure readings for each home. In these box plots, the upper and lower bounds of the boxes correspond to the 25th and 75th percentiles of the vapor pressure readings. The lines
48、 within the boxes are the median values, and the whiskers above and below the boxes are the minimum and maximum values. The right-most box shows the difference between on- and off-time average indoor vapor pressures. Table 1 provides the quanti-tative values shown in the box plots plus the mean valu
49、e. In Table 1 the headings P25 and P75 refer to 25th percentile (1st quartile) and 75th percentile (3rd quartile), respectively.This characterization of moisture contribution from the fireplace shows a clear difference between on- and off-times. When the fireplace is off the median vapor pressure average is about 0.80 kPa (0.12 psi), and with the fireplace on the median vapor pressure average is about 0.88 kPa (0.13 psi). The median difference is about 0.1 kPa (0.015 psi), which is about 12.5% of the median off-time vapor pressure average. There is one home that shows a very
