1、346 2010 ASHRAEThis paper is based on findings resulting from ASHRAE Research Project RP-1299.ABSTRACTHigher-efficiency HVAC filters generally have a higherpressure drop and are widely assumed to increase energyconsumption in smaller air conditioning systems. To explorethe effects of filters in real
2、 buildings, we monitored 17 resi-dential and light-commercial forced air cooling systems inAustin, TX. Measurements were made once per month for oneyear at each site with filters from three different MERV rangecategories. Measured parameters included system airflow,fan power draw, outdoor unit power
3、 draw, cooling capacity,pressure drops across filters and coils, and duct leakage.Higher-efficiency (MERV 11-12) filters generally had a smallimpact on parameters related to cooling energy consumptionin the residential and light-commercial test systems whencompared to lower-efficiency (MERV 2) filte
4、rs. The medianenergy consequence of higher-efficiency filtration in the testsystems was estimated as a decrease of approximately 16 kWhper ton of nominal capacity (4.6 kWh per kW) per month ofcooling season operation, albeit with large variation, withmost of these small savings coming from fan energ
5、y reduc-tions. These results suggest a weak link between higher-effi-ciency filters and energy use in residential and light-commercial systems and that other factors should governfilter selection.INTRODUCTIONHigh-efficiency filtration in forced air heating, ventilat-ing, and air-conditioning (HVAC)
6、systems is used to protectbuilding equipment and occupants, but can also influencebuilding energy use. Filters with a high MERV (MinimumEfficiency Reporting Value, as defined by ASHRAE Standard52.2-2007) typically have a greater pressure drop than a filterwith a lower MERV. The energy consequences o
7、f a greaterpressure drop due to filtration are well known for largecommercial systems, where fan and motor controls typicallymaintain required airflow rates. A higher pressure drop filtercauses the fan motor to draw more power to overcome thepressure drop and deliver the required amount of air, thus
8、increasing energy consumption (Chimack and Sellers 2000;Fisk et al. 2002). This association between energy use andfilter pressure drop is widely assumed to hold true for smallerresidential and light-commercial systems, but operationaldifferences between small and large systems suggest verydifferent
9、energy consequences.The central difference is that increasing the pressure dropof a filter in most residential HVAC systems generally causesdiminished airflow, although evidence is limited. Parker et al.(1997) measured a 4 to 5% airflow rate reduction when replac-ing standard disposable filters with
10、 high-efficiency pleatedfilters in residential air conditioner field tests. Diminishedairflow generally decreases cooling capacity, power draw ofthe compressor, and system efficiency. Parker et al. (1997)predicted by computer simulations and laboratory tests that a5% reduction in airflow from a valu
11、e recommended by mostmanufacturers of 400 CFM ton1(193 m3h1kW1) to380 CFM ton1(184 m3 h1kW1) would decrease sensiblecooling capacity by approximately 2%. This suggests that asystem would run 2% longer to meet the same cooling load. Inlaboratory experiments, Rodriguez et al. (1996) tested 3.5-ton(12.
12、3 kW) air conditioners and reported approximately 6 toEnergy Implications of Filtration in Residential and Light-Commercial BuildingsBrent Stephens Jeffrey A. Siegel, PhD Atila Novoselac, PhDStudent Member ASHRAE Member ASHRAE Member ASHRAEBrent Stephens is a graduate student research assistant, Jef
13、frey Siegel is an associate professor, and Atila Novoselac is an assistant professorin the Department of Civil, Architectural, and Environmental Engineering at the University of Texas at Austin.OR-10-038 (RP-1299) 2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.
14、(www.ashrae.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. ASHRAE Transactions 3477% reductions in efficiency and t
15、otal capacity associated witha 10% reduction from the recommended airflow rate. Palaniet al. (1992) measured the impacts of low airflow on a 3-ton(10.5 kW) air conditioner in a series of laboratory tests as welland found similar reductions in capacity for comparablereductions in airflow. The same st
16、udies showed that moredrastic energy consequences occur when flow reductions areextreme. Parker et al. (1997) reported that system coolingenergy consumption could increase by 20% if flow diminishesapproximately 40% from 400 CFM ton1(193 m3h1kW1).The previous investigations show that if the presence
17、of ahigher-efficiency filter diminishes airflow, sensible coolingcapacity will decrease, suggesting an increase in energyconsumption due to increased system runtime. However, fanand compressor power draw also generally decrease, poten-tially limiting negative energy impacts. In addition, a changein
18、filter pressure drop can affect duct leakage by changing thepressure around duct leaks. Although we know of no directresearch of the implications of filtration on duct leakage, thereis extensive literature on the energy consequences of ductleakage in residential and light-commercial systems (e.g.,Mo
19、dera 1989; Modera 1993; Parker et al. 1993; Jump et al.1996; Walker et al. 1998; Withers and Cummings 1998; Siegelet al. 2000; Francisco et al. 2006).One of the central challenges of associating energy conse-quences with filtration is the complexity of these interactingeffects. The magnitudes, and e
20、ven the signs, of many of theseeffects are not well characterized, but are likely very system-dependent and are affected by such parameters as the fractionof the system pressure drop associated with the filter, the fan-speed setting, and the intersection point of the fan and the ductcurves. To explo
21、re these effects in real systems, we monitoredresidential and light-commercial forced air cooling systems atmultiple sites in Austin, Texas. Measured parameters includedsystem airflow rate, power draw, cooling capacity, pressuredrops across filters and coils, and duct leakage. Periodicmeasurements w
22、ere made over the course of a year at each sitewith readily available filters with different MERV categories,as rated by the filter manufacturer. The purpose of thisresearch was to assess how filter MERV and the correspondingmeasured pressure drop impact energy use in smaller air-conditioning system
23、s. The specific goal is to allow systemdesigners and users to evaluate the consequences associatedwith higher-efficiency filtration.METHODOLOGYSite Selection and DescriptionsSeventeen systems were selected as a sample of conve-nience based on the willingness of the building owners andresidents to ha
24、ve monitoring equipment installed and frequentvisits from the field personnel. Table 1 summarizes the 17 testsites. The first eight sites were residential buildings and theremaining nine were light-commercial buildings. The light-commercial buildings were all office spaces with some alsoserving a li
25、mited retail function. Each system served less than2000 ft2(186 m2) of floor area and rated air conditioner cool-ing capacities ranged from 1.5 to 5.0 tons (5.3 to 17.6 kW).Sites 1 to 15 had typical permanent split capacitor (PSC) fansand Sites 16 and 17 had electronically commutated motor(ECM) fans
26、. Most ductwork was located in unconditionedattics, with a few systems with ducts in other locations. Filterswere located in return grilles or at the air handler. All of the testsystems relied on infiltration for fresh air, rather than dedi-cated outdoor air ventilation.Test MethodologyThe test site
27、s were visited once a month for a year, duringwhich time three categories of filtration efficiency typicallyused in residential and light-commercial systems wereinstalled: low (MERV 2), medium (MERV 6-8), and high(MERV 11-12). Each MERV category filter was left in placefor three months and monitored
28、 four times: initially on the dayof installation and after one, two, and three months of usage.The final three-month period was used to repeat an installationof one of the MERV categories to assess variation in themeasurements. Unlike the other sites, Site 12 had only high-efficiency filters install
29、ed over the duration of the projectbecause of a request by the building owner.During each monthly visit, measurements were made inthe fan-only mode by activating the switch at the thermostat.Pressure measurements were made using an Energy Conser-vatory DG-700 handheld digital manometer (uncertainty
30、1%of reading), including the pressure drop across the filter(s) andcooling coil and the pressure differential between the occupiedspace and the supply and return plenums. A custom-built data-logging box was then launched to record the pressure dropacross the filter(s) and cooling coil and the power
31、draw of theair handler fan in the fan-only mode for approximately 15 min-utes at 10-second intervals. The data-logging box consisted ofa Continental Control Systems (CCS) Wattnode AC truepower meter (uncertainty 0.45% of reading and 0.05% offull-scale), two Setra pressure transducers (uncertainty 1%
32、of full-scale) connected to an Onset Flexsmart (uncertainty1% of full-scale), and an Onset HOBO Energy Logger Pro.The box was connected to pressure taps, voltage taps, and 0 to20 Amp CCS current transducers (uncertainty 1% of read-ing) that remained installed for the duration of the one-yeartest per
33、iod.During each monthly visit in the cooling season, measure-ments were made with each system in fan-only mode, then theequipment was left to monitor and log for approximately 24hours with the thermostat operated normally by the buildingoccupants. Also, during the cooling season visits, additionalco
34、ntinuous measurements were made of the power draw of theoutdoor unit using the same instrumentation as describedabove and Onset HOBO U12 dataloggers for temperature andrelative humidity measurements (uncertainty 0.4C (0.7F)and 2.5% from 10% to 90% RH; 6-minute response time).Temperature and relative
35、 humidity measurements were taken 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 either print or digital
36、form is not permitted without ASHRAEs prior written permission. 348 ASHRAE Transactionsoutdoors, in the zone that contained the majority of the ductwork (often the attic), inside the return plenum, inside thesupply plenum, and at a single supply register.Duct leakage and airflow measurement tests we
37、re con-ducted during one visit over the course of the year long mon-itoring period. Duct leakage was measured using an EnergyConservatory Duct Blaster (uncertainty 3% of flow reading)and Model 3 blower door in accordance with ASHRAE Stan-dard 152-2004. The Duct Blaster alone was used to make total(i
38、nterior + exterior) leakage measurements and the blowerdoor was added to make exterior leakage measurements. Thetests were repeated with the return side of the system sealed offto separate supply and return leakage. Monthly estimates ofduct leakage were assessed by correcting for changes in thesuppl
39、y plenum operating pressure observed during each visitand using a power-law flow-leakage approximation followingthe procedure in the Duct Blaster manual. System airflow rateswere measured with an Energy Conservatory TrueFlow meter-ing plate and DG-700 digital manometer (uncertainty 7% ofreading). Mo
40、nthly corrections were made based on changes inthe supply plenum pressure measured during each visit fol-lowing the calculation procedure in the instrument manual.Table 2 summarizes the equipment used in the field tests andthe manufacturer-reported accuracies of each device.Calculation of Energy Con
41、sequencesPreviously, similar studies have relied on the metrics ofcapacity (sensible and latent) and the coefficient of perfor-mance (i.e., efficiency) in attempts to address the complicatedrelationship between flow changes, system runtime, and over-all energy consumption. We used the same metrics t
42、o describethe cooling performance of the systems using the measureddata. The total capacity, qt(Btu/h, W), calculation is shown inEquation (1). The first term defines sensible capacity and thesecond term defines latent capacity. qt= Qfan(CT + Whfg)(1)whereQfan= volumetric flow rate of air (ft3/h, m3
43、/s) flowing through the cooling coil; = air density, assumed constant (0.075 lbm/ft3, 1.2 kg/m3);C = specific heat of air, assumed constant (0.24 Btu/(lbmF), 1.005 kJ/(kgK);T = temperature difference across the cooling coil (F, K);W = humidity ratio difference across the cooling coil (lbm/lbm, kg/kg
44、); andhfg= latent heat of vaporization for water, assumed constant (970 Btu/lb, 2257 kJ/kg).The coefficient of performance, COP, calculation isshown in Equation (2).(2)whereWou= power draw of outdoor unit, including the compressor (W); andWfan= fan power draw (W).Table 1. Test Site CharacteristicsSi
45、te Building UseFloor Area,ft2(m2)Rated Cooling Capacity1, tons (kW)DuctworkLocationAir Handler LocationNumber ofFiltersFilter Location21 Residential 1830 (170) 4.0 (14) Attic Closet 1 Slot2 Residential 1430 (133) 3.0 (11) Attic Garage 1 Slot3 Residential 1080 (100) 2.5 (9) Between floors Closet 3 Gr
46、illes4 Residential 320 (30) 1.5 (5) Attic Attic 1 Grille5 Residential 1140 (106) 2.5 (9) Attic Attic 1 Grille6 Residential 1500 (139) 3.0 (11) Attic Attic 1 Grille7 Residential 1200 (111) 3.0 (11) Between floors Closet 1 Slot8 Residential 1350 (125) 3.0 (11) Attic Garage 1 Slot9 Commercial 1300 (121
47、) 5.0 (18) Attic Attic 2 Grilles10 Commercial 1300 (121) 3.5 (12) Attic Attic 2 Grilles11 Commercial 1320 (123) 3.5 (12) Attic Attic 2 Grilles12 Commercial 1860 (173) 5.0 (18) Attic Attic 3 Grilles13 Commercial 1430 (133) 3.5 (12) Attic Closet 1 Slot14 Commercial 980 (91) 3.0 (11) Attic Closet 1 Slo
48、t15 Commercial 1000 (93) 2.5 (9) Attic Closet 2 Slot16 Commercial 760 (71) 1.5 (5) Outdoor Outdoor closet 1 Grille17 Commercial 280 (26) 1.5 (5) Conditioned space Closet 1 Slot1Cooling capacity corresponds to the nominal capacity of the outdoor unit.2Slot = Filter slot at the air handling unit, Gril
49、le(s) = Return grille(s).COPqtWouWfan+-= 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 either print or digital form is not permitted without ASHRAEs prior written permission. ASHRAE Transactions 349The cooling capacities were calculated during eachrecorded cycle only when the systems reached a period ofsteady-state operation. Measu
