air distribution system case study

8/3/2019 Air Distribution System Case Study

1/16

CASE AIR DIST (FINAL).DOC

Air Distribution Systems

Codes and Standards Enhancement (CASE) Study

November 28, 2000

Pacific Gas and Electric Company

Patrick Eilert

Pacific Gas and Electric Co.

P.O. Box 770000, H28E

San Francisco, CA 94177

Phone: 530-757-5261

E-mail: [email protected]

This report was prepared by Pacific Gas and Electric Company and funded by California utility customers under theauspices of the California Public Utilities Commission.

Neither PG&E nor any of its employees and agents:

1. makes any written or oral warranty, expressed or implied, regarding this report, including but not limited to thoseconcerning merchantability or fitness for a particular purpose;

2. assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information,apparatus, product, process, method, or policy contained herein; or

3. represents that use of the report would not infringe any privately owned rights, including, but not limited to, patents,trademarks, or copyrights.

Copyright 2000 Pacific Gas and Electric Company. All rights reserved.

Reproduction or distribution of the whole or any part of the contents of this document without the express writtenpermission of PG&E is prohibited. Neither PG&E nor any of its employees makes any warranty, express or implied, orassumes any legal liability of responsibility for the accuracy, completeness, or usefulness of any data, information,method, policy, product or process disclosed in this document, or represents that its use will not infringe any privately-owned rights, including but not limited to patents, trademarks or copyrights.


2/16

This page intentionally left blank


3/16

COPYRIGHT 2000 PACIFIC GAS AND ELECTRIC COMPANY

CASE STUDY PAGE I

Table of Contents

ACKNOWLEDGMENTS ............................................................................................................................ II

INTRODUCTION.......................................................................................................................................... 1

TECHNOLOGY DESCRIPTION................................................................................................................1

CURRENT PRACTICE................................................................................................................................ 2

ECONOMICS ................................................................................................................................................5

KEY STAKEHOLDERS............................................................................................................................... 9

IMPLEMENTATION STRATEGIES AND RECOMMENDATIONS ...................................................9

UPDATE THE ALTERNATIVE CALCULATION METHOD (ACM) TO CREDIT DUCT SEALING ........................ 10SUPPORT AIR DISTRIBUTION SEALING THROUGH UTILITY PROGRAMS...................................................... 10

RESEARCH AIR DISTRIBUTION SYSTEM PERFORMANCE AND IMPROVE TOOLS.......................................... 10

BIBLIOGRAPHY ........................................................................................................................................ 11


4/16


PAGE II AIR DISTRIBUTION SYSTEMS

Acknowledgments

The New Buildings Institute developed this Codes and Standards Enhancement Study for thePacific Gas and Electric Company under Contract 4600010614, part of PG&Es Codes andStandards Program for 1999. Project managers for PG&E included Jennifer Barnes, PatrickEilert, Gary Fernstrom and Marshall Hunt.

Jeffrey A. Johnson, Senior Program Director of the New Buildings Institute, managed thisproject.

Subcontractors on this project were:

Eley Associates:

Charles Eley

GARD Analytics:

Roger Hedrick

Berkeley Solar Group:

Bruce Wilcox

The following individuals reviewed and advised the project: Cathy Higgins, New BuildingsInstitute; Bill Pennington and Jonathan Leber, California Energy Commission; Pete Jacobs,Architectural Energy Corporation; and Mark Modera, Lawrence Berkeley National Laboratory.

Produced by:New Buildings Institute, Inc.PO Box 653

White Salmon, WA 98672

E-mail: [email protected]

Web: www.newbuildings.org


5/16


CASE STUDY PAGE 1

Introduction

This Codes and Standards Enhancement(CASE) Study provides information tosupport including air distribution efficiencymeasures into the nonresidential energyefficiency standards (standards). This CASEstudy for air distribution systems discussesthe:

technology,

current practice,

economics,

key stakeholders, and

implementation options and

recommendations for inclusion intocodes.

Air distribution systems are used in a largemajority of nonresidential buildings to movecooling and heating energy from centralequipment to locations throughout thebuilding. Air distribution systems alsodistribute ventilation air throughout thebuilding.

Air distribution system construction in lightcommercial buildings is often very similar tothat in residential buildings. Systems in bothtypes of buildings are often plagued byproblems that cause sizeable energy losses.

In recent years, poor quality ductwork hasbeen recognized as a significant source ofenergy losses. The use of ceiling insulationlocated at the suspended ceiling is commonand results in the ductwork being locatedoutside the conditioned space. Leaks inducts run in unconditioned attics result inlarge amounts of cooled and/or heated airbeing lost to the outdoors. The leaks also

increase infiltration loads.This study analyzes the savings potential ofimproving the air distribution systemefficiency in Californias nonresidentialbuildings.

Technology Description

Air distribution systems are the primarymeans of moving heating and coolingenergy to spaces within a building. Thesesystems consist of supply and return ducts orplenums through which heated or cooled airis moved.

Ideally, duct systems should be perfectlyefficient, delivering to the conditionedspaces all of the air from the centralequipment, with no change in temperature.In reality, however, duct systems fall shortof these ideals. Ductsand especiallyplenumsleak, resulting in supply air beingdiverted to and return air being drawn in

from undesired locations. In addition, ductsusually pass through locations withtemperatures different from the air inside theduct. The resulting heat transfer through theduct material changes the air temperature,usually in undesirable ways.

Nearly all nonresidential buildings utilize airdistribution systems to provide spaceconditioning. The exceptions are buildingsthat are hydronically heated without centralair conditioning systems and buildingswhere distributed space conditioning

equipment is used. Distributed equipmentincludes space heaters, packaged terminalair conditioners, window air conditioners,and similar systems.

Air distribution systems are generally builtonsite using some sort of duct material soldexplicitly for that purpose. This might belengths of flex duct, usually insulated,attached to terminal units or other ductsystem components; ducts constructed ofsheet metal with site-applied insulation; orducts built of ductboard, which includesinsulation. Any of these can be used toconnect the central air handling equipmentto some sort of terminal unit in the space.

The overall air distribution systemfrequently includes combinations of thevarious duct constructions. A central plenummade of sheet metal or ductboard is oftenused to distribute air from the central unit,


6/16


PAGE 2 AIR DISTRIBUTION SYSTEMS

with flex duct branches connecting theplenum to the various individual spaces.

In constructing the ductwork, pieces of thematerial must be connected to each other.There are numerous ways to make these

connections, depending on the duct materialused. Some methods, however, appear to beless durable than others and any methodmay be improperly installed.

Once the building has been completed, theductwork is usually out of sight, and leakscan go undetected. Improving ductworkefficiency is best accomplished by usinghigh quality, durable duct connections.Usually this means using some type ofmechanical fastenings.

For residential construction, duct efficiencytests methods exist, such as ASHRAEStandard 152P (ASHRAE 2000). These testmethods provide a means of testing theductwork after installation and confirmingthat its efficiency meets specified criteria. Inaddition, an aerosol duct sealing method canbe used to economically seal leaky ducts.This involves sealing the registers and otherdesigned air outlets, pressurizing the ducts,and spraying an aerosol sealant into the duct.As the air leaks through the unintendedopenings, the aerosol sealant graduallybuilds up on the edges of the leak until theleak is sealed.

Current Practice

Two significant studies (RLW 1999; Felts etal. 1999) show that rooftop units provide themajority of commercial building cooling inCalifornia. Figure 1 illustrates the number ofbuildings in the RLW study served bypackaged versus built-up (chiller) systems.Approximately 90% of the buildings in thestudy were served by packaged systems.

Figure 1. Market Penetration of Packaged

and Built-up Systems

Figure 2 shows a frequency distribution bycooling capacity of packaged unitarysystems, including: single packaged rooftopAC; single packaged rooftop heat pump;split system AC; split system heat pump; ordual fuel heat pump, without evaporative

condenser. These data show that themajority of the systems are 5 tons or less,with 5 tons being the most popular unit size.


7/16


CASE STUDY PAGE 3

0%

5%

10%

15%

20%

25%

30%

1 2 3 4 5 6 7 10 12 15 20 25 30 40 50 75 100

Unit size (ton)

EstimatedNRNCUnitarySystemMarketShare

Figure 2. Market Penetration of Packaged Unitary AC and Heat Pumps by Cooling

Capacity

From these data, it is evident that lightcommercial air distribution systemsrepresent a significant portion of airdistribution systems in California.

Most research on distribution systemefficiency has been performed on residential

buildings. Air leakage for residentialsystems is generally in the range of 10% to20% of the fan airflow, on both the supplyand return side of the fan (Modera 1999).Additional inefficiencies result from heattransfer between the air in the ducts and inthe attic, particularly during cooling whenthe attic can be substantially hotter than theoutdoors due to solar radiation.

In light commercial buildings, duct systemsare typically run in the ceiling cavity

between the dropped ceiling and the roof(Delp et al. 1997). This cavity is frequentlyoutside both the air barrier and thermalbarriers; in other words, it is outside theconditioned space. This situation closelyparallels that in most residential buildings.In addition, the systems in these lightcommercial buildings usually use a cyclingfan operation (Delp et al. 1997), further

extending the parallel. Delp found that theleakage area for light commercial ductsystems was 2.6 times that of a typicalresidential system.

Problems caused by air distribution systeminefficiencies include:

Air leakage and heat transfer reduce theefficiency of the overall spaceconditioning system, becauseconditioning energy is not delivered tothe desired space. Extra energy must beconsumed to make up the losses.

The effective heating or coolingcapacity of the system is substantiallyreduced, often resulting in oversizedequipment being installed tocompensate. This can result in further

reductions in system efficiency.

When supply air leakage is unevenacross the building, some spaces may beunderconditioned or overconditionedrelative to other spaces, resulting inoccupant discomfort.

Unbalanced supply duct leakage candepressurize the building. If natural


8/16



draft combustion appliances areinstalled (such as gas water heaters) thedepressurization increases the risk ofappliance backdrafting. The resultingpresence of NO2 and water in the space

can degrade indoor air quality. If the

appliance also has a malfunctioningburner, CO can be introduced to thespace.

Leakage that is unbalanced between thesupply and return sides of the systemcan increase the overall buildinginfiltration rate. This can causeincreased energy consumption to heatand cool the infiltration air. (On theother hand, this infiltration air may bethe primary source of ventilation for the

building, improving the indoor airquality. Properly controlled ventilationsystems would undoubtedly be a bettermeans of providing this benefit.)

For larger commercial buildings,improvements in air distribution systemefficiency can result in significant savings,primarily from fan power savings. The fanpower consumption of these larger buildingsis on the same order of magnitude as theircooling energy. Since the fans runcontinuously, reducing the required airflow

reduces the fan power.

A recent study (Xu et al. 1999a) found thatthermal losses through leakage andconduction were 30% of the supply energy.Theoretically, eliminating these 30% losseswould reduce the fan power by 70%.Perhaps half of these savings are actuallyachievable, as it is not possible tocompletely eliminate the conduction losses.Also, constant air volume systems must beretrofitted to reduce the system airflowbefore any savings would be achieved.

There are several sources of leakageproblems for conventional air distributionsystems. These include:

Use of inadequate connection methodsbetween pieces of the duct system. Mostnotorious is the use of poor quality ducttape to seal connections. These

connections then fail either immediately(the tape didnt stick because thesurfaces were dusty or dirty) or in ashort span of time (the tape adhesivewas weak or wasnt able to handle thetemperature cycles encountered).

Improper installation of the ducts,particularly not adequately connectingdifferent pieces of the system. In otherwords, the installer did a shoddy job andsimply did not properly connect theducts.

Use of building cavities as airdistribution plenums. In particular,drawing return air through the ceilingplenum can result in significant leakagefrom the outdoors.

Each of these problems must be addressed atdifferent points in the buildings design andconstruction process. For example, duringthe design of the HVAC system, which maybe done either by an engineer or an HVACcontractor, codes can require that if buildingcavities are used for air distribution, theymust be adequately sealed. Alternatively,codes can prohibit this practice. The codecan also specify acceptable means of ductconstruction by the mechanical contractor,which can be verified by a buildinginspector. Finally, a duct leakage test can beused to ensure that the air distributionsystem has been tightly constructed. Recentchanges to the standards for low-riseresidential buildings prohibit the use ofbuilding cavities as air distribution plenumsand require that ducts be diagnosticallytested to assure low leakage.

An alternative to contractor-installed ductsealing approaches is the use of an aerosol-based sealing technology. Once the

distribution system has been constructed andthere are no gross leaks (such as ducts thatare not connected), an aerosol sealant issprayed into the ducts while they arepressurized by a fan. The sealant adheres tothe edges of any leaks, building up to sealthe leak. This is followed by a leakage test.An alternative to extensive requirements onduct system construction is to build the


9/16


CASE STUDY PAGE 5

building with the air and thermal barriercollocated at the roof and extending downthe outside walls of the ceiling plenum.Ductwork installed in the plenum is theninside the conditioned space, and any losseshave less impact on the buildings energy

consumption. (This is a requirement ofASHRAE/IESNA Standard 90.1-1999,which prohibits use of a suspended ceilingas the air or thermal barrier.)

If this approach is used, the standards mustspecify adequate means of ensuring that theceiling plenum is in fact inside air andthermal barrier. This is true, however, onlyif the ceiling plenum is not used as a returnplenum. If the supply ducts are inside thereturn air plenum, duct losses create a shortcircuit and the air never makes it to theoccupied space. This has a negative impacton energy consumption, although it is muchless severe than when the ducts are in anunconditioned space.

The Standards previously prohibited the useof a suspended ceiling as the thermal or airbarrier. The prohibition was removed,however, because of building ownersconcerns about the additional cost to insulatethe wall areas above the dropped ceiling andthe greater difficulty in insulating the roof,

as opposed to the ceiling. The potentialbenefits related to the distribution systemwere apparently not included in the decisionto eliminate the prohibition.

Research needs in this area are limited.Investigation of the in-service life of variousqualities of duct tape, and the factors whichaffect their life, would be desirable. A testprocedure to certify proprietary ductconnection methods would also be useful.

Economics

Improved air distribution systems cansignificantly reduce the energy consumptionof buildings. Consider, for example, asingle-story building with an unsealeddistribution system in California climatezone 12. Using data from the 1998 CECresidential Alternative Calculation Methods

(ACM) manual, this system will haveseasonal average efficiencies of 75% inheating and 67% in cooling. After ductsealing, these efficiencies increase to 82% inheating and 76% in cooling, saving 9% ofthe heating energy and 12% of the cooling

energy.

Researchers (Delp et al. 1997), however,found that leakage in light commercialbuildings was 2.6 times that in residentialbuildings, indicating that the aboveestimates may be very conservative. On theother hand, ducts in commercial buildingsare somewhat more likely to be inside theconditioned space, reducing the potentialsavings.

Analysis using the DOE-2.1E energy

simulation program and ASHRAE Standard152P algorithms was conducted under thisstudy. We examined a number of prototypebuildings, and made simulation studies ofthree buildings: a medium-size office, astore in a strip mall and a full-servicerestaurant.

Other buildings were not simulated forseveral reasons. Some of the buildings didnot use ducts for the primary systems (suchas packaged terminal air conditioners usedin hotels and schools). Some buildings hadsystems with minimal ductwork outside theconditioned space (warehouse-style retailstores, for example). And for some buildingswe were unable to do adequate DOE-2modeling (for a large office building whereonly fan power effects are expected, DOE-2cannot simulate return side leakage).

The three modeled buildings were simulatedwith three cases: 30% leakage above theceiling insulation (base case), 10% leakageabove the ceiling insulation, and 30%

leakage with the insulation extended up thewalls and roof of the attic/plenum. The airchange rate of the attic/plenum wasunchanged for the three cases. Simulationswere done for four California locations:Fresno, Mt. Shasta, Oakland and PalmSprings.


10/16



For the strip store, ASHRAE Standard 152Pcalculations were done for buildings inOakland and Fresno. The 152P calculationwas not done for Palm Springs and Mt.Shasta because the spreadsheet used did notinclude weather data for these locations, and

a particular seasonal temperature value isrequired which was not available.

The strip store is a 9600-ft_ retail storelocated at the end of a strip mall. The entirestore is served by a single packaged VAVsystem. The building is modeled with theroof insulation at the suspended ceiling.

The medium-size office is a three-storyoffice building totaling 49,000 ft_. Thebuilding has three central VAV systems, oneserving each floor, with all central

equipment located in a penthouse. The ductleakage model addressed only the systemserving the top floor, assuming that theductwork for this system is located in thepenthouse, with the roof insulation locatedbelow the ductwork. Ductwork for the othertwo systems is assumed to drop from thepenthouse through a central shaft, withhorizontal runs between floors, inconditioned space.

The full-service restaurant is a single-story9060-ft_ restaurant/lounge. The kitchen isserved by a packaged single-zone systemwith the rest of the building served by apackaged multizone system. The ductwork

for the multizone system is run through theattic. The kitchen system ductwork also runsthrough the attic, but was not included in theduct loss analysis. DOE-2.1E cannot modellosses from more than one system into asingle zone.

The simulation results are shown in thefollowing tables. Tables 1 through 3 showannual electricity consumption in kWh andthe annual peak demand in kW, for themedium office, restaurant, and strip store,respectively. Tables 4 through 6 show kWhand kW savings for the 10% leakage and30% leakage with extended insulation cases.Tables 7 through 9 show the savings inpercentage terms.

For all three buildings and all four locations,

energy and demand savings of duct sealingare dramatic. Reductions in cooling energyrange from 9% to 18% for the single-storybuildings, and 3% to 4% for the three-storyoffice. Reductions in demand are evengreater, ranging from 12% to 27% for thestore and restaurant, 3% to 5% for the office.

Savings for the extended insulation withoutduct sealing are not nearly so dramatic, butare still substantial for the single-storybuildings (except for the Mt. Shasta cases).Given that extended insulation can beexpected to be at least as expensive as ductsealing, it is clear that duct sealing should bethe primary strategy.


11/16


CASE STUDY PAGE 7

Table 1. DOE-2 Predicted Medium Office

Electricity Consumption

City CaseAnnual

kWhPeakkW

Fresno 30% 497,886 180.310% 478,067 171.3

Ext Insul 495,787 179.2Mt. 30% 475,245 170.5Shasta 10% 457,521 163.5

Ext Insul 473,941 169.9

Oakland 30% 474,120 168.010% 457,650 161.6

Ext Insul 473,150 167.2

Palm 30% 503,126 184.0Springs 10% 484,357 177.6

Ext Insul 500,955 183.4

Table 2. DOE-2 Predicted Restaurant


City CaseAnnual

kWhPeakkW

Fresno 30% 620,227 174.610% 515,849 146.7

Ext Insul 599,815 156.1

Mt. 30% 486,577 127.2Shasta 10% 413,132 108.3

Ext Insul 493,030 125.7Oakland 30% 472,154 118.4

10% 402,259 104.0Ext Insul 466,880 106.5

Palm 30% 724,000 190.2Springs 10% 593,287 153.8

Ext Insul 714,587 181.1

Table 3. DOE-2 Predicted Retail Store


City CaseAnnual

kWhPeakkW

Fresno 30% 250,092 105.210% 214,965 79.9

Ext Insul 242,939 100.3Mt. 30% 179,202 63.1Shasta 10% 160,554 52.4

Ext Insul 180,068 63.3Oakland 30% 192,904 75.9

10% 175,585 61.0

Ext Insul 190,945 72.5Palm 30% 310,159 122.9Springs 10% 254,804 89.9

Ext Insul 299,176 115.5

Table 4. DOE-2 Predicted Medium Office

Electricity Savings

City CaseKWh

SavingsPeak kWSavings

Fresno 30% - -10% 19,819 9.0

Ext Insul 2,099 1.1Mt. 30% - -Shasta 10% 17,724 7.0

Ext Insul 1,304 0.6

Oakland 30% - -10% 16,470 6.4

Ext Insul 970 0.8

Palm 30% - -Springs 10% 18,769 6.4

Ext Insul 2,171 0.6


Electricity Savings

City CaseKWh


Fresno 30% - -10% 104,378 27.9

Ext Insul 20,412 18.5

Mt. 30% - -Shasta 10% 73,445 18.9

Ext Insul -6,453 1.5Oakland 30% - -

10% 69,895 14.4Ext Insul 5,274 11.9

Palm 30% - -Springs 10% 130,713 36.4

Ext Insul 9,413 9.1


Electricity Savings

City CaseKWh


Fresno 30% - -10% 35,127 25.3

Ext Insul 7,153 4.9Mt. 30% - -Shasta 10% 18,648 10.7

Ext Insul -866 -0.2Oakland 30% - -

10% 17,319 14.9

Ext Insul 1,959 3.4Palm 30% - -Springs 10% 55,355 33.0

Ext Insul 10,983 7.4


12/16



Table 7. DOE-2Predicted Medium Office

Electricity Savings

City CaseKWh


Fresno 30% - -10% 4% 5%

Ext Insul 0% 1%Mt. 30% - -Shasta 10% 4% 4%

Ext Insul 0% 0%

Oakland 30% - -10% 3% 4%

Ext Insul 0% 0%

Palm 30% - -Springs 10% 4% 3%

Ext Insul 0% 0%


Electricity Savings

City CaseKWh


Fresno 30% - -10% 17% 16%

Ext Insul 3% 11%

Mt. 30% - -Shasta 10% 15% 15%

Ext Insul -1% 1%Oakland 30% - -

10% 15% 12%Ext Insul 1% 10%

Palm 30% - -Springs 10% 18% 19%

Ext Insul 1% 5%


Electricity Savings

City CaseKWh


Fresno 30% - -10% 14% 24%

Ext Insul 3% 5%Mt. 30% - -Shasta 10% 10% 17%

Ext Insul 0% 0%Oakland 30% - -

10% 9% 20%

Ext Insul 1% 4%Palm 30% - -Springs 10% 18% 27%

Ext Insul 4% 6%

These savings are dramatic, particularly thedemand reductions achieved in the retailstore. For the small systems modeled, thefan and compressor will be single-speedequipment and the instantaneous demandwill not be decreased. Over an averaging

period that is longer than the cycle time ofthe system, however, the equipment will runfor a shorter time, and demand will bereduced proportionally.

As an independent check on these results,the ASHRAE Standard 152P algorithmswere used to check the impact ondistribution system efficiency of thereduction in leakage from 30% to 10% forthe strip store in Oakland and Fresno. Theseresults are shown in Tables 10 and 11.

Table 10. ASHRAE 152P Calculated

Distribution System Efficiencies - Retail

Store

Oakland FresnoLeakage 30% 10% 30% 10%Heating, design 44% 57% 42% 57%Heating, seasonal 52% 63% 53% 65%Cooling, design 42% 55% 33% 49%Cooling, seasonal 75% 80% 64% 72%

Table 11. Savings Based on ASHRAE

152P Calculation of Distribution System

Efficiency and Comparison to DOE-2

Results Retail Store

152PSavings

DOE2Savings

City Oak. Fres Oak. FresCooling, design 23% 33% 20% 24%Cooling, seasonal 7% 11% 9% 14%

The DOE-2 and 152P results are consistent,despite some significant problems in makingthe comparison. For example, in making the152P calculation, we assumed that two-thirds of the leakage was on the supply sideand one-third on the return side, whereas inthe DOE-2 calculation, all the leakage isfrom the supply side. Also, the 152Pcalculation assumes a cycling fan operation,whereas DOE-2 models a continuous fanoperation.

Costs for duct sealing are relatively low.Post-construction aerosol duct sealing is


13/16


CASE STUDY PAGE 9

expected to be about $250 per system(Modera 2000a). This includes sealing theducts and performing a leakage test.Systems in light commercial buildings serveabout 360 square feet per ton of cooling(Delp et al. 1997). If an average system is 4

tons, this works out to about $0.18 persquare foot ($250 / (360 ft_/ton x 4 tons)).

But improving installation methods andquality control is the preferred way toreduce duct leakage. Costs for this areuncertain, but are similar to aerosol-basedsealing.

The life of a properly installed airdistribution system should be as long as therest of the building, and shouldnt requiremaintenance. Some duct assembly methods

are prone to early failure, however, asdescribed above.

If these methods can be adequatelyidentified and prohibited, the life of the ductleakage reductions should be quite long.Life of the aerosol-based sealing technologyis uncertain, but six years of testing resultedin no failures (Modera 2000b).

Commissioning of the air distributionsystem by performing leakage tests isessential to verify that the system has been

properly installed. Leakage tests should bebased on methods prescribed in ASHRAEStandard 152P. Additional algorithms arecurrently being developed to appropriatelyaddress commercial applications.

In summary, the savings available fromcommercial air distribution systemefficiency improvements are much largerthan the uncertainties associated with theanalytical tools. Particularly for smallerbuildings, where a large fraction of the airdistribution system is located in

unconditioned space, it is clear thataddressing these systems in the commercialportions of the Standards is warranted.

Key Stakeholders

There are several key stakeholders with aninterest in air distribution system efficiency.These include architects, engineers,contractors, building owners, occupants,policy makers and code officials. Changes inair distribution system efficiencyrequirements will most directly affect thecontractors who install the systems and thecode officials who verify and approve them.Building owners, occupants, contractors,architects and engineers will also be affectedby the slight increases in buildingconstruction costs.

Finally, owners and occupants will benefit

from the dramatic reductions in energyconsumption and, depending on demandbilling methods (averaging period), demandreductions.

Implementation Strategiesand Recommendations

Under the Standards, residential buildingsreceive an energy credit for tight ductsystems. Verification of duct performance

by testing is required. Duct performance innonresidential buildings is not addressed.

Energy savings calculations forimprovements in residential duct efficiencyare adequately described, particularly inASHRAE Standard 152P. Thesecalculations, however, assume that the airhandler cycles with the heating or coolingdevice. In addition, the Standard assumesthat the thermal and air barriers of thebuilding are collocated, i.e., that a duct isinside or outside both barriers. This is not

the case in nonresidential buildings.

There is a need for additional research toprovide better tools to evaluate ductefficiency improvements in nonresidentialbuildings. To validate the overall energysavings available through air distributionsystem improvements, additional research isalso needed to characterize the air


14/16



distribution systems in commercialbuildings.

Even without that research, however, thesavings available from air distributionsystem efficiency improvements are so large

that early inclusion in the Standards iswarranted. As such, the implementationstrategies and recommendations are asfollows:

Update the Alternative Calculation

Method (ACM) to Credit Duct Sealing

The Non-Residential ACM should bemodified to include the impact of ductleakage and insulation levels on heatingequipment efficiency and cooling equipmentefficiency for individual single-zone unitary(packaged) equipment serving 5000-ft_ orless via ductwork located in the spacebetween an insulated ceiling and the roof.

The impacts of duct leakage and ductinsulation on HVAC system efficiencyshould be calculated using the algorithmssimilar to those used for low-rise residentialbuildings. (see Appendix F of the 1998Residential ACM). The impact of ductefficiency should be accounted for in thestandard building as well as the proposed

building.In nonresidential calculations, assumptionsfor the standard buildings should assume aleakage of 22% of fan flow, split equallybetween supply and return ducts (same asresidences). If the proposed nonresidentialbuilding is diagnostically tested by a HomeEnergy Rating System (HERS) rater to havetight ducts (less than 6% of fan flow forsupply and return ducts combined), then theduct efficiency of the proposed building canbe calculated using 8% total duct leakage

split equally between supply and returnducts. Similarly, if the proposed building isverified by a HERS rater to have ductinsulation levels above the standard R-4.2level, then the duct efficiency of theproposed building shall be calculated usingthe documented insulation level.

Support Air Distribution Sealing

through Utility Programs

The benefits of tighter ducts innonresidential buildings are potentiallysubstantial. California utilitieswith their

extensive experience in designing andimplementing residential duct sealingprogramscould transfer this experience tononresidential programs. This would lead tothe development of additional informationon the costs and benefits to improvingnonresidential air distribution systems. Itwould also provide immediate benefits toratepayers and business owners by reducingthe energy and demand costs associated withpoorly sealed air distribution systems.

Research Air Distribution SystemPerformance and Improve Tools

In the longer term, prescriptive requirementsshould be developed for all commercialbuildings. Additional research will beneeded, however, before such requirementscan be developed. This research includes:

Better characterization of buildings toestablish appropriate baseline values;

Work to develop duct efficiencymeasurement protocols for commercialbuildings; and

Development of modeling tools that canbe used to predict the impact ofcommercial building duct efficiencyimprovements, and account for:

- fan operation (continuous orcycling),

- system type,- relative location of the ducts,

thermal barrier and air barrier,- number of floors,

- duct construction method,- climate, and- occupancy.

In particular, DOE2.1E needs to besignificantly enhanced to allow widespreaduse for duct efficiency modeling. Currently,there are considerable problems with thealgorithms, and the keywords are complex,


15/16


CASE STUDY PAGE 11

difficult to use, and highly sensitive to error.Algorithm problems include:

All leakage occurs on the supply side ofthe system.

All losses go to a single zone.

Only one system can have losses to agiven zone (multiple systems located ina single plenum cannot be accuratelymodeled).

Building and zone infiltration is notsensitive to duct leakage.

The EnergyPlus simulation programcurrently under development by DOE willbe much better at modeling duct efficiencyproblems, although the programs currentdevelopment plans leave some key gaps.EnergyPlus models HVAC systems as anetwork of nodes. Each of the nodesrepresenting the duct system can include airlosses, and these losses will properly affectthe remainder of the duct system as well asthe zones where the node is located and thezones served by the system. Current plans,however, do not include modeling ofthermal losses through the ducts walls, suchas occur when ducts are located in a hotattic. Current plans also do not includespecific duct efficiency input commands.

It may be desirable for California to fundwork to ensure that EnergyPlus will includeall needed capabilities, with straightforwardinput requirements, for use as a compliancetool.

Bibliography

ASHRAE. 2000. ASHRAE Standard 152P Method of test for determining thedesign and seasonal performance ofresidential distribution systems advanced working draft 00/2. AmericanSociety of Heating, Refrigerating andAir Conditioning Engineers.

Provides calculations for determiningthe efficiency of the duct system both atdesign conditions (peak) and forseasonal averages. The calculations for

in this CASE study were done using adraft spreadsheet intended for inclusionwith the ASHRAE Standard 152P.

Delp, W. W., N. E. Matson, E. Tshudy, M.P. Modera, and R. C. Diamond. 1997.

Field investigation of duct systemperformance in California lightcommercial buildings. LawrenceBerkeley National Laboratory, Report#40102.

Eight California light commercialbuildings, containing a total of 15systems, were studied to characterizefactors significant to duct systemperformance. They found that 50% ofthe buildings had the ceiling cavityoutside the thermal and air barrier, and

that the leakage area per square foot offloor area averaged 2.6 times (360%)that of residential buildings. The ductsystems were constructed using methodssimilar to that of residential buildings,with 60% of the buildings using allmetal duct systems, with the remainderflex-octopus. All the buildings usedpackaged rooftop units with theductwork in the ceiling cavity. Most ofthe units operated with cycling fans(apparently all but one).

Felts et al. 1999. Roof top unit technicalpotential assessment, working papers.Pacific Gas and Electric Company.

Characterization of rooftop packagedequipment in northern Californiaresulting from inspections, monitoringand analyses carried out on over 250units in widely varying climates.

Modera, M. P. 2000a. Personalcommunication with Bruce Wilcox.

Described cost of duct sealing usingAeroseal as $250 per system, includingsealing and post-sealing leakage test.

Modera, M. P. 2000b. Email to RogerHedrick.

Described life of Aeroseal duct sealingtechnology. No known seal failures after6 years of installations in residential


16/16



buildings. Accelerated life testing withno failures to >70,000 cycles, whereduct tape failed at 200-1,500 cycles.

Modera, M. P. 1999. Commercial thermaldistribution systems final report for

CIEE/CEC. Lawrence BerkeleyNational Laboratory, Report #44320.

Summarizes the state of research oncommercial duct leakage; compares it toresidential data. Includes equationsdescribing the various loss mechanisms,and discusses the various assumptions.Discusses in detail the ability of DOE-2.1E to model duct losses.

Modera, M., T. Xu, H. Feustel, N. Matson,C. Huizenga, F. Bauman, E. Arens, andT. Borgers. 1999. Efficient thermalenergy distribution in commercialbuildings final report. LawrenceBerkeley National Laboratory, Report#41365.

Describes the results of surveys thatcharacterize the commercial buildingstock in California. Includes distributionsystem type and various components ofsystem energy consumption. Providescase studies on a small subset ofbuildings.

RLW Analytics et al. 1999. California non-residential new construction baselinestudy. California Board for EnergyEfficiency.

Conducted by RLW in association withArchitectural Energy Corporation andthe Heschong Mahone Group.Developed from a database of over 800onsite surveys compiled duringevaluations of NRNC DSM programsoperated by PG&E, SCE and SDG&E.

Provides statewide coverage of the newconstruction marketplace.

Xu, T., R. F. Carrie, D. J. Dickerhoff, W. J.Fisk, J. McWilliams, D. Wang and M. P.Modera. 1999a. Performance of thermal

distribution systems in large commercialbuildings. Lawrence Berkeley NationalLaboratory, Report #44331.

Five systems in three large offices andone supermarket were studied, each ofwhich served at least 2000 m_ (21,000ft_). Air leakage areas (ELA25) weremeasured in the range of 0.1 to 7.7 cm_per square meter of floor area, whichresulted in leakage of up to 1/3 of thefan-supplied airflow in constant volumesystems. Leakage varied by system and

section of the same system. Energyimpacts were due to additional fanpower requirements.

Xu, T., O. Bechu, R. Carrie, D. Dickerhoff,W. Fisk, E. Franconi, O. Kristiansen, R.Levinson, J. McWilliams, D. Wang, M.Modera, T. Webster, E. Ring, Q. Zhang,C. Huizenga, F. Bauman, and E. Arens.1999b. Commercial thermal distributionsystems final report for CIEE/CEC.Lawrence Berkeley NationalLaboratory, Report #44320.

Characterized thermal distributionsystems in large and light commercialbuildings, evaluated adding ductperformance to the commercial portionof Title 24, tested aerosol duct sealingfor large commercial application andevaluated protocols to be applied tolarge commercial systems. Described indetail the commands available in DOE-2.1E for modeling duct leakage,including a discussion of the limitation

of such modeling.

air distribution system case study

Documents