1、 T. Fazli is Ph.D. candidates in the Department of Civil, Architectural and Environmental Engineering, Illinois Institute of Technology, Chicago, IL. B. Stephens is an assistant professor in the Department of Civil, Architectural and Environmental Engineering, Illinois Institute of Technology, Chica
2、go, IL. Characterizing the In-Situ Size-Resolved Removal Efficiency of Residential HVAC Filters for Fine and Ultrafine Particles Torkan Fazli Brent Stephens, PhD, PE Student Member ASHRAE Associate Member ABSTRACT The majority of human exposure to airborne particles occurs inside buildings, particul
3、arly in residences. In order to improve indoor air quality and reduce indoor particle concentrations, high efficiency particle filters in central forced air heating, ventilating, and air-conditioning (HVAC) systems are increasingly being used. ASHRAE Standard 52.2 contains the most widely used metho
4、d of evaluating the performance of HVAC filters in the U.S., but the standard only assesses the removal efficiency of particles 0.310 m in diameter. However, the vast majority of particles in any indoor or outdoor environment (by number) are less than 0.1 m in diameter, or ultrafine particles (UFPs)
5、, which have been shown to provoke alveolar inflammation and consequently intensify heart and lung disease. Further, the 52.2 test procedure is also limited only to laboratory testing and does not provide information about the performance of HVAC filters in real environments. Therefore in this proje
6、ct, an in-situ test method is used to measure the particle removal efficiency of a wide range of commercially available filters in 20 particle size ranges from 0.01-2.5 m in diameter. The test procedure involves measuring particle concentrations upstream and downstream of an HVAC filter installed in
7、 a residential air-handling unit. The size-resolved particle removal efficiency for each size bin is calculated by subtracting the average ratio of downstream-to-upstream pollutant concentrations from unity. A combination of a TSI NanoScan SMPS and a TSI Optical Particle Sizer is used with an automa
8、ted electronically actuated switching valve to measure particle concentrations. We also characterize the impacts of filter pressure drop on system airflow rates using a flow plate device. Measurements are conducted in a controlled and unoccupied residential apartment unit located in a graduate stude
9、nt residence hall on the main campus of Illinois Institute of Technology. More than a dozen filters with various minimum efficiency rating values (MERV) from a variety of manufacturers have been tested thus far and we are beginning to develop an online database of these results for engineers and hom
10、eowners to use. INTRODUCTION Long-term exposure to airborne particulate matter is connected with adverse human health effects, such as increased risks of cardiopulmonary mortality, respiratory symptoms, and lung cancer (Pope et al. 2002; Beelen et al. 2014). The majority of human exposure to airborn
11、e particles occurs inside buildings (Sundell 2004), (Klepeis et al. 2001). Indoor particles consist of both particles emitted from a variety of indoor sources (such as cooking and smoking) and particles of outdoor origin that penetrate into the indoor environment (Ozkaynak et al. 1996). As buildings
12、 are becoming more airtight in order to decrease CO2 (Smeds and Wall 2007), the balance between outdoor-infiltration and indoor-generated indoor particulate matter concentrations are shifting. Airborne particles are typically classified by their aerodynamic diameter in specific size fractions, such
13、as PM10 (respirable particles, dp 95% MERV 16 (HEPA) N/A N/A 95% 95% 95% MERV 14-15 N/A N/A 75 95% 90% 90% MERV 13 Black 2200 10 Blue 90% 90% MERV 12 Navy Blue 1900 80 90% 90% Purple 1500 8-9 Purple MERV 11 Red 1000 7 Red 65 79% 85% MERV 10 50 64% 85% MERV 8 Light Blue 600 5 Green 70 85% MERV 7 4 Gr
14、een 50 69% MERV 6 400 35 49% MERV 0 5 20 34% Given these typical test standards, there is limited information on two key aspects of residential HVAC filters. One is the removal efficiency of particles smaller than 0.3 m. The other is the removal efficiency or filters in real residential environments
15、 with real air-handling units in which differences in face velocities, airflow rates, pressure drops, and installation characteristics might not be captured accurately by a laboratory test. Therefore this work uses an in-situ test method to measure the particle removal efficiency of a wide range of
16、commercially available filters in particle size ranges from 0.01-2.5 m in diameter in an unoccupied apartment unit. METHODS/MATERIALS Experiments were performed in StudioE (the Suite for Testing Urban Dwellings and their Indoor and Outdoor Environments), an unoccupied apartment unit on the third flo
17、or of Carman Hall for graduate student housing on the main campus of Illinois Institute of Technology in Chicago, IL. The unit has a floor area of 60 m2 (646 ft2) and a volume of 150 m3 (5297 ft3). A 100% recirculating central air-handling unit is installed in the living room and connected to interi
18、or rigid sheet metal ductwork, but it is not connected to a heating or cooling system (the system is used to mimic a typical residential air handler and distribution system). Both a TSI Nanoscan SMPS and a TSI Optical Particle Sizer (OPS) were used to measure size-resolved particle concentrations, a
19、lternating upstream and downstream of the filter, from 10 nm to 10 m. Particles in the size range of 10 nm to 2.5 m were generated through a combination of burning incense and emission of NaCl particles from a TSI particle generator (Model 8026). In order to increase generated particle concentration
20、 and prevent particle dilution in the large volume of StudioE apartment, a small chamber with dimentions of 1004060 cm (391624 inches) was installed in front of the return duct. Each test was conducted for approximately 1 hour. Upstream and downstream concentrations were measured through an automate
21、d electronically actuated sampling system with TSI conductive tubing. Each sampling period was set for 4 minutes, for a total of 8 minutes per upstream/downstream combination, or about 7 complete upstream/downstream cycles within 1 hour of testing. Removal efficiency in each particle size bin (appro
22、ximately 20 bins total) for each 8-minute sample period was calculated by subtracting the ratio of the average downstream concentration to the average upstream concentration from unity. The average size-resolved removal efficiency was then reported as the average removal efficiency across all 7 samp
23、le periods for a given test. In both upstream and downstream sampling periods, we ignored the first minute of data collected to ensure that the sampling lines were cleared from the previous measurement. Size-resolved removal efficiency for each test filter is reported as an average and standard devi
24、ation across the 7 combined upstream/downstream sample periods. Figure 1 shows the schematic section and plan view of the air filter testing system in StudioE. According to ASHRAE 52.2 the inlet nozzles of upstream and downstream sample probes should have appropriate entrance diameter to maintain is
25、okinetic sampling (within 10%) at the test airflow rate. Initial measurements with a Fluke air velocity meter (Model #975) without a filter installed averaged approximately 3.3 m/s (649 ft/min) in the location of sampling probes and the calculated velocity through the sampling probe (sample flow rat
26、e divided by cross-sectional area of tubing) is approximately 3.5 m/s (688 ft/min) which is the reasonable value. Figure 1 Schematic of the air filter testing (not to scale) For each filter test, airflow rates and filter pressure drop were also measured. Filter pressure drop was measured using an En
27、ergy Conservatory DG-700 differential pressure gauge connected to ambient indoor air on one end and to a pressure tap just a few inches downstream of the filter (in the return plenum) on the other end. The airflow rate was estimated by measuring the pressure in the supply plenum and relating back to
28、 TrueFlow plate measurements. Uncertainty was estimated using the manufacturers reported value (7%). Each filter had its manufacturer-reported make, model, and rated efficiency (MERV, FPR, MPR) recorded during each test. Temperature and relative humidity in the room was also measured throughout each
29、 test at either 1 or 5 minute intervals using an Onset HOBO U12. RESULTS AND DISCUSSION Characteristics of Tested Filters Table 2 provides a list of 15 tested filters with various rated efficiencies and mean values of measured filter pressure drop, HVAC airflow rates, indoor temperature, and relativ
30、e humidity during each of the experimental conditions in the apartment unit. Filters were tested across approximately similar indoor environmental conditions, with indoor temperatures ranging 2128C (7082F) and RH ranging 4469%. The installation of filters from different manufacturers and with differ
31、ent rated efficiencies introduced various filter pressure drops, which altered airflow rates by as much as about 11%. In general, by increasing in rated removal efficiency of the filter, the filter pressure drop increased and consequently airflow rates decreased relative to the no filter condition (
32、1048 m3/hr). The minimum airflow rate decrease was with MERV 1-4, the most inefficient filters, which was only about 2%. The maximum airflow rate decrease occurred with a 1” (2.5 cm) MERV 13 filter (16%). Typically, filters with extended depths and the same rated efficiency as a thinner filter were
33、able to maintain higher airflow rates. Size Resolved Removal Efficiency of Filters Figure 2a, b and c show the mean and standard deviation of size-resolved removal efficiency measured during the filter test conditions for three different efficiency rating systems (MERV, MPR and FPR). All filters hav
34、e the lowest measured removal efficiency for particle sizes of about 300 nm and the highest efficiency for particle sizes near 2.5 m, consistent with filtration theory and previous measurements. Table 2. Charactrestics of Fifteen Tested Filters and Relevant Parameters Manufacturer Model Depth, cm (i
35、n) Rated efficiency Filter P, Pa (in. w.c.) Airflow rate, m3/hr (cfm) Temp, C (F) RH, % FLANDERS EZ-Flow II 2.5 (1) MERV 1-4 11.9 (0.047) 1027 (604) - - WEB-ECO Filter Plus 2.5 (1) MERV 8 37.8 (0.152) 1005 (591) 27 (81) 44 AIRGUARD Dp-4-40 Max 10.2 (4) MERV 8 28.9 (0.116) 1005 (591) 28 (82) 45 FLAND
36、ERS Pre-pleat-40 5.1 (2) MERV 11 36.4 (0.146) 982 (578) - - CULUS Pleated Panel 5.1 (2) MERV 11 34.8 (0.140) 990 (582) 22 (72) 47 AIRGUARD PWG-STD4-402 10.2 (4) MERV 11 37.2 (0.149) 981 (577) 23 (73) 44 AIRGUARD Dp-13-STD1-102 2.5 (1) MERV 13 56.8 (0.228) 900 (530) 28 (82) 45 FLANDERS Pre-Pleated 40
37、 5.1 (2) MERV 13 38.0 (0.152) 974 (573) 26 (79) 52 3M Filtrete Clean Living 2.5 (1) MPR 600 52.0 (0.209) 982 (578) - - 3M Micro Allergen Defense 2.5 (1) MPR 1000 45.8 (0.184) 964 (567) 22 (72) 68 3M Filtrete Helathy Living 2.5 (1) MPR 2200 51.0 (0.205) 963 (567) 26 (79) 48 RHEEM Dust for particle si
38、zes 300 nm to 1 m, the mean absolute standard deviation was 4%; and for particle sizes 1 m to 2.5 m, the mean absolute standard deviation was only 2%. The highest uncertainty across all filters and size bins was 15%, so all tests were considered valid. Furthermore, to address concerns about the repe
39、atability of the test method, a minimum of 10% (1 out of 10) filters was tested in triplicate (filters drawn at random). The average removal efficiency for all of the particle size bins was shown to deviate less than 6% on an absolute basis in these replicate tests (shown in Figure 3). For the MERV
40、11 filter, the mean absolute standard deviation across tests was only 2% and for the MPR 1000 filter, the mean absolute standard deviation was only 3% (Figure 3). These low uncertainties demonstrate good repeatability of the methods used herein. Figure 2 Mean and standard deviat ion of size-resolved
41、 removal efficiency for three different efficiency rating systems; (a) MERV, (b) MPR and (c) FPR. Figure 3 Results of triplicate filter test for two random selected filter CONCLUSION An in-situ upstream/downstream filter test method was used to measure the size-resolved fine and ultrafine particle r
42、emoval efficiency of 15 commercially available residential HVAC filters in an unoccupied apartment. The removal efficiency for ultrafine particles generally ranged from 35% to 80% for MERV 8 filters and MERV 13 filters, respectively. For the particle sizes between 300 nm to 1 m, the removal efficien
43、cy of the filters ranged from a minimum of 16% for MERV 1-4 filters to a maximum of 80% for MERV 13 filters. Similarly, for particle sizes 1 m to 2.5 m removal efficiencies ranged from 30% for MERV 1-4 to 95% for MERV 13. These results demonstrate that in order to achieve substantial removal of both
44、 fine and ultrafine particles by central HVAC filters in residential environments, higher efficiency filters than what is typically recommended in standards such as ASHRAE 62.2 are required. ACKNOWLEDGEMENTS This work was supported in part by an ASHRAE Graduate Student Grant-in-Aid Award to Torkan F
45、azli and in part by an ASHRAE New Investigator Award to Brent Stephens. REFERENCES ANSI/AHRI. 2009. “ANSI/AHRI Standard 680 (I-P) Standard for Performance Rating of Residential Air Filter Equipment.” Arlington, VA. ASHRAE. 2007. “Standard 52.2: Method of Testing General Ventilation Air-Cleaning Devi
46、ces for Removal Efficiency by Particle Size.” American Society of Heating, Refrigerating and Air-Conditioning Engineers. Beelen, Rob, Ole Raaschou-Nielsen, Massimo Stafoggia, Zorana Jovanovic Andersen, Gudrun Weinmayr, Barbara Hoffmann, Kathrin Wolf, et al. 2014. “Effects of Long-Term Exposure to Ai
47、r Pollution on Natural-Cause Mortality: An Analysis of 22 European Cohorts within the Multicentre ESCAPE Project.” The Lancet 383 (9919): 78595. doi:10.1016/S0140-6736(13)62158-3. EPA, December. 2009. “Integrated Science Assessment for Particulate Matter.” US Environmental Protection Agency Washingt
48、on, DC. Home Depot. 2015. “The Home Depot, Air Filters Buying Guide.” http:/ Klepeis, N E, W C Nelson, W R Ott, J P Robinson, A M Tsang, P Switzer, J V Behar, S C Hern, and W H Engelmann. 2001. “The National Human Activity Pattern Survey (NHAPS): A Resource for Assessing Exposure to Environmental Po
49、llutants.” Journal of Exposure Analysis and Environmental Epidemiology 11 (3): 23152. doi:10.1038/sj.jea.7500165. Monarca, Silvano, Donatella Feretti, Alberto Zanardini, Massimo Moretti, Milena Villarini, Berthold Spiegelhalder, Ilaria Zerbini, Umberto Gelatti, and Eugenio Lebbolo. 2001. “Monitoring Airborne Genotoxicants in the Rubber Industry Using Genotoxicity Tests and Chemical Analyses.” Mutation Research/Genetic Toxicology and Environmental Mutagenesis
