Demand-Controlled Ventilation Using CO Sensorsinfohouse.p2ric.org/ref/43/42844.pdf ·...

Demand-Controlled VentilationUsing CO2 SensorsPreventing energy losses from over-ventilation while maintaining indoor air quality

Executive Summary

Demand-controlled ventilation (DCV) using carbon dioxide (CO2) sensing is a combination of twotechnologies: CO2 sensors that monitor CO2 levels in the air inside a building, and an air-handlingsystem that uses data from the sensors to regulate the amount of ventilation air admitted. CO2 sen-sors continually monitor the air in a conditioned space. Given a predictable activity level, such asmight occur in an office, people will exhale CO2 at a predictable level. Thus CO2 production in thespace will very closely track occupancy. Outside CO2 levels are typically at low concentrations ofaround 400 to 450 ppm. Given these two characteristics of CO2, an indoor CO2 measurement can beused to measure and control the amount of outside air at a low CO2 concentration that is being intro-duced to dilute the CO2 generated by building occupants. The result is that ventilation rates can bemeasured and controlled to a specific cfm/person based on actual occupancy. This is in contrast to thetraditional method of ventilating at a fixed rate regardless of occupancy.

Building codes require that a minimum amount of fresh air be provided to ensure adequateair quality. To comply, ventilation systems often operate at a fixed rate based on an assumedoccupancy (e.g., 15 cfm per person multiplied by the maximum design occupancy). The resultis there often is much more fresh air coming into buildings than is necessary. That air must beconditioned, resulting in higher energy consumption and costs than is necessary with appro-priate ventilation. In humid climates, excess ventilation also can result in uncomfortable humid-ity and mold and mildew growth, making the indoor air quality (IAQ) worse rather than better.

A lack of adequate fresh air, on the other hand, can make building occupants drowsy anduncomfortable. To avoid the problems of too much or too little fresh air, the heating, venti-lation, and air-conditioning (HVAC) system can use DCV to tailor the amount of ventilationair to the occupancy level. CO2 sensors have emerged as the primary technology for monitor-ing occupancy and implementing DCV. Energy savings come from controlling ventilationbased on actual occupancy versus whatever the original design assumed.

Application Domain

CO2 sensors have been available for about 12 years. An estimated 60,000 CO2 sensors are sold annu-ally for ventilation control in buildings, and the market is growing. There is a potential for millionsof sensors to be used, since any building that has fresh air ventilation requirements might potentiallybenefit from the technology.

CO2-based DCV has the most energy savings potential in buildings where occupancy fluctuates duringa 24-hour period, is unpredictable, and peaks at a high level—for example, office buildings, govern-ment facilities, retail stores and shopping malls, movie theaters, auditoriums, schools, entertainmentclubs and nightclubs.

CO2 sensors are considered a mature technology and are offered by all major HVAC equipment and con-trols companies. The technology is recognized in ASHRAE Standard 62, the International Mechanical

Leading by example,saving energy andtaxpayer dollarsin federal facilities

DOE/EE-0293

Internet: www.eere.energy.gov/femp/

No portion of this publication may be altered in any form withoutprior written consent from the U.S. Department of Energy, EnergyEfficiency and Renewable Energy, and the authoring national laboratory.

Handheld CO2 sensor with

data logging.

Code (which establishes minimum regu-lations for mechanical systems), andsome state and local building codes.Although the first CO2 sensors sold wereexpensive, unreliable, and difficult tokeep calibrated, manufacturers say theyhave largely resolved those problems.The unit cost of sensors has droppedfrom $400 to $500 a few years ago to$200 to $250 (not including installa-tion). As market penetration increases,prices are expected to fall further.

Several manufacturers produce CO2 sen-sors for DCV. Most manufacturers ofthermostats and economizers are inte-grating CO2 sensors into their products,and major manufacturers of packagedrooftop HVAC systems offer factory-installed CO2 sensors as an option.

Benefits

DCV saves energy by avoiding the heat-ing, cooling, and dehumidification ofmore ventilation air than is needed. CO2

sensors are the most widely acceptedtechnology currently available for imple-menting DCV. Additional benefits ofCO2-based DCV include

• Improved IAQ—By increasing ven-tilation if CO2 levels rise to an un-acceptable level,

• Improved humidity control—Inhumid climates, DCV can preventunnecessary influxes of humidoutdoor air that makes occupantsuncomfortable and encouragesmold and mildew growth.

Estimated Savings

The potential of CO2-based DCVfor operational energy savings has

been estimated in the literature atfrom $0.05 to more than $1 per squarefoot annually. The highest paybackcan be expected in high-density spacesin which occupancy is variable andunpredictable (e.g., auditoriums, someschool buildings, meeting areas, andretail establishments), in locations withhigh heating and/or cooling demand,and in areas with high utility rates.

Design Considerations

CO2 sensing is a fairly simple technol-ogy, and installation of the sensors them-selves is not complicated. IncludingCO2-based DCV in a new HVACinstallation should not add significantlyto the difficulty of commissioning thesystem. However, retrofitting an existingsystem for DCV may be more problem-atic, particularly for an older system withpneumatic controls. Applying a CO2-based DCV strategy using ASHRAE 62is more complicated than simply install-ing CO2 sensors and using them to con-trol dampers. In variable-air-volumesystems, particularly, fairly complexcalculations and control algorithmsmay be necessary to program the con-trol system properly for DCV. The useof a more complex control algorithmoften provides increased savings andimproved IAQ; while it increases thelevel of commissioning, the results out-weigh the extra initial time and expense.

Maintenance Impact

Maintenance of the sensors themselvesis not generally reported to be a prob-lem. Manufacturers offer sensors thatrecalibrate themselves automaticallyand that are guaranteed not to need

calibration for up to 5 years. However,it is recommended that calibrationbe checked periodically by comparingsensor readings during a several-hourperiod when the building is unoccu-pied with readings from the out-door air. Many sensor models areable to sense calibration problemsand alert maintenance personnel ifthey are malfunctioning.

Costs

Costs for sensors have dropped byabout 50% over the last several years.Sensors typically cost about $250 to$260 each, uninstalled. For a newsystem, the installed cost will gener-ally be about $600 to $700 per zone.For a retrofit system, the cost willdepend on what type of control sys-tem the building has. A controls con-tractor estimates installed costs forretrofit applications at from $700 to$900 per zone for systems with anexisting DDC programmable control-ler and from $900 to $1200 per zonefor systems with pneumatic, elec-tronic, or application-specific DDCs.Installation costs for wireless systemsare minimal beyond the cost of theactual sensor and gateway that canserve multiple sensor units.

In addition to the installation of thesensors, other components such asvariable frequency drives and controlinput and output hardware often areneeded to control the whole building,incrementally increasing the overallinstalled project cost beyond just thesensor installation cost.

Disclaimer

This report was sponsored by the United States Department of Energy, Energy Efficiency and Renewable Energy,

Federal Energy Management Program. Neither the United States Government nor any agency or contractor thereof,

nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for

the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents

that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or

service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorse-

ment, recommendation, or favoring by the United States Government or any agency or contractor thereof. The views

and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or

any agency or contractor thereof.

FEDERAL ENERGY MANAGEMENT PROGRAM — 1

Contents

Abstract ............................................................................................................................................. 3About the Technology ....................................................................................................................... 3

Application DomainEnergy-Saving MechanismOther BenefitsVariationsInstallation

Federal-Sector Potential ..................................................................................................................... 7Estimated Savingsand Market PotentialLaboratory Perspective

Application ....................................................................................................................................... 8Application ScreeningWhere to ApplyWhat to AvoidDesign ConsiderationsMaintenance ImpactEquipment WarrantiesCodes and StandardsCostsUtility Incentives and Support

Technology Performance ................................................................................................................. 12Field ExperienceEnergy SavingsMaintenance Impact

Case Study ...................................................................................................................................... 13Facility DescriptionExisting Technology DescriptionNew Technology Equipment SelectionBuilding Upgrade AssessmentLife-Cycle CostPost-Implementation Experience

The Technology in Perspective ........................................................................................................ 19The Technology’s DevelopmentTechnology Outlook

Manufacturers ................................................................................................................................. 20Who is Using the Technology .......................................................................................................... 20For Further Information .................................................................................................................. 21Appendix A Federal Life-Cycle Costing Procedures

and the BLCC Software .............................................................................................. 23

2 — FEDERAL ENERGY MANAGEMENT PROGRAM

This page left blank intentionally


Abstract

Demand-controlled ventilation (DCV)using carbon dioxide (CO2) sensorscombines two technologies: advancedgas sensing and an air-handling systemthat uses data from the sensors to regu-late ventilation. CO2 sensors continu-ally monitor the air in a conditionedspace. Since people exhale CO2, thedifference between the indoor CO2 con-centration and the level outside thebuilding indicates the occupancy and/or activity level in a space and thus itsventilation requirements. The sensorssend CO2 readings to the ventilationcontrols, which automatically increaseventilation when CO2 concentrationsin a zone rise above a specified level.

Either too little or too much fresh airin a building can be a problem. Over-ventilation results in higher energyusage and costs than are necessary withappropriate ventilation while potentiallyincreasing IAQ problems in warm,humid climates. Inadequate ventila-tion leads to poor air quality that cancause occupant discomfort and healthproblems. To ensure adequate air qual-ity in buildings, the American Societyof Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE)recommended a ventilation rate of15–20 cfm per person in ASHRAEStandard 62-1999. To meet the stan-dard, many ventilation systems aredesigned to admit air at the maximumlevel whenever a building is occupied,as if every area were always at full occu-pancy. The result, in many cases, hasbeen buildings that are highly over-ventilated. The development of CO2-based DVC was driven in part bythe need to satisfy ASHRAE 62 with-out overventilating.

Non-dispersive infrared CO2 sensors arethe type most widely used. All majormakers of heating, ventilation, and air-conditioning (HVAC) equipment andHVAC controls offer the sensors, eitheras separate units or as a part of packagedHVAC systems. Although earlier sensorswere plagued by reliability and calibra-tion problems, those issues seem to havebeen largely resolved in newer models.

Typically, a CO2 sensor is installed onthe wall like a thermostat. Newer prod-ucts entering the market combine ther-mostats and CO2 and relative humiditysensors, the three primary indicatorsof human comfort, in one unit. Sensorsmay be installed inside ductwork ratherthan wall-mounted, but in-duct instal-lation is not recommended for all appli-cations. In a conventional “wired” CO2

sensor system, wires are run from thesensors to the HVAC controls or tothe damper actuators. With wirelessCO2 sensors, the data are transmittedto the building automation system viaa wireless gateway for use in the con-trol algorithm. Properly functioningmodulating dampers are necessary.Pneumatic controls may need to bereplaced with electronic or direct digi-tal controls. In a retrofit installation,dampers may need to be repaired orupgraded to work with the sensors.

In all applications of CO2-based DCV,a minimum base ventilation rate mustbe provided at all times the building isoccupied. A higher rate may be neededfor buildings in which building mate-rials, contents, or processes releasechemicals into the air. DCV should beused only in areas where human activ-ity is the main reason for ventilating thespace. Industrial or laboratory spacesare unsuitable for CO2-based ventila-tion control.

The potential of CO2-based DCV forenergy savings is estimated at from$0.05 to more than $1 per square footannually. The highest payback can beexpected in high-density spaces in whichoccupancy is variable and unpredictable(e.g., auditoriums, some school build-ings, meeting areas, and retail estab-lishments), in locations with highheating and/or cooling demand, andin areas with high utility rates. Casestudies show DCV offers greater savingsfor heating than for cooling. In areaswhere peak power demand and peakprices are an issue, DCV can be usedto control loads in response to real-time prices. In those locations, DCVmay enable significant cost savingseven with little or no energy savings.

CO2-based DCV does not interfere witheconomizers or other systems that intro-duce outdoor air into a building forcooling. Economizer operation over-rides DCV when conditions warranteconomizer use. Buildings that useevaporative cooling may not benefitfrom DCV during the cooling season.

Costs for sensors have dropped by about50% over the last several years as thetechnology has matured and becomemore widely used. Sensors typically costabout $250 to $260 each, uninstalled.For a new system, the installed costwill generally be about $600 to $700per zone. For a retrofit system, the costwill depend on what type of controlsystem the building has and the degreeof difficulty of installing signal andpower wiring for a wired system. Acomplete wireless sensor system canbe deployed quickly and without addi-tional cost as such systems are batterypowered. Given the advances in bat-tery technology and microprocessor-controlled power management, sensorscan be expected to operate for 2–3 yearsbefore they require a battery change.

About the Technology

Demand-controlled ventilation (DCV)using carbon dioxide (CO2) sensing isa combination of two technologies:CO2 sensors that monitor the levels ofCO2 in the air inside a building, andan air-handling system that uses datafrom the sensors to regulate the amountof outside air admitted for ventilation.DCV operates on the premise thatbasing the amount of ventilation airon the fluctuating needs of buildingoccupants, rather than on a pre-set,fixed formula, will save energy and atthe same time help maintain indoor airquality (IAQ) at healthy levels.

CO2 sensors continually monitor theair in a conditioned space. Becausepeople constantly exhale CO2, the dif-ference between the indoor CO2 con-centration and the level outside thebuilding indicates the occupancy and/or activity level in a space and thus itsventilation requirements. (An indoor/


outdoor CO2 differential of 700 ppmis usually assumed to indicate a venti-lation rate of 15 cfm/person; a differ-ential of 500 ppm, a 20 cfm/personventilation rate, etc.) The sensors sendCO2 readings to the air handling sys-tem, which automatically increasesventilation when CO2 concentrationsin a zone rise above a specified level.

Building codes specify that a minimumamount of fresh air be brought into abuilding to provide for adequate airquality. The American Society of Heat-ing, Refrigerating and Air-ConditioningEngineers (ASHRAE) recommends aventilation rate of 15-20 cfm per personin ASHRAE Standard 62.1 To complywith codes, building ventilation systemsoften operate at constant or predeter-mined rates regardless of the occupancylevel of the building. Many systems ven-tilate buildings at the maximum levelall the time, as if every area were alwaysfully occupied. Others are programmedto accommodate expected occupancy,varying the ventilation rate by time ofday but without regard to the actualoccupancy level. The result often ismore ventilation air coming into build-ings than is necessary. That air mustbe heated, cooled, or dehumidified/humidified, resulting in higher energy

consumption and costs thanwould be necessary withappropriate ventilation. Inaddition, in humid climates,an overload of fresh air canresult in uncomfortablehumidity and mold andmildew growth, makingthe indoor air qualityworse rather than better.

A lack of adequate freshair, on the other hand, canmake building occupantsdrowsy and uncomfortableas occupancy-related con-taminants accumulate.A CO2 level of around1100 ppm (a differentialof 700 ppm, assuming theoutdoor air CO2 level is

around 400 ppm) indicates that theventilation rate has dropped belowacceptable levels and that contami-nants in the air are increasing.

To avoid the problems of too much ortoo little fresh air, a heating, ventila-tion, and air-conditioning (HVAC)system can employ DCV to adjust theamount of ventilation air supplied toan indoor space according to the occu-pancy level. CO2 sensors have emergedas the primary technology for monitor-ing occupancy and implementing DCV.

Application Domain

CO2 sensors have been available forabout 12 years. An estimated 60,000CO2 sensors are sold annually for venti-lation control in buildings, and a manu-facturer estimates the market is growingat about 40 to 60% per year. There isa potential for millions of sensors to beused, since any building that has freshair ventilation requirements couldpotentially benefit from this refinedcontrol technology.

DCV using CO2 sensors has the mostenergy-saving potential in buildingswhere occupancy fluctuates during a24-hour period, is unpredictable, andpeaks at a high level, for example, office

buildings, government facilities, retailstores and shopping malls, movie the-atres, auditoriums, schools, entertain-ment clubs and nightclubs. In buildingswith more stable occupancy levels, DCVcan ensure that the target ventilationrate per person is being provided at alltimes. DCV is more likely to reduceenergy costs in areas with high utilityrates or climate extremes.

CO2-based DCV can operate in con-junction with economizers or othersystems that introduce outdoor airinto a building for heating or cooling.However, energy savings may be lesswhere economizers are in use, depend-ing on climate, occupancy schedule,and building type.

Buildings that use evaporative coolingmay not benefit from DCV during thecooling season, and those that use heatexchangers to transfer heat betweenincoming and outgoing air may notrealize significant energy savings.

In the next revision of its building code,California will begin requiring CO2-based DCV in all buildings housing25 people or more per 1000 ft2. A pro-posed change in the Oregon buildingcode would require DCV for HVACsystems with ventilation air require-ments of at least 1500 cfm, servingareas with an occupant factor of 20 orless. ASHRAE 90 requires CO2 sensorsfor DCV in high-density applications.The U.S. Green Building Code givespoints in its Leadership in Energy andEnvironmental Design (LEED) ratingsystem for use of CO2-based ventilationcontrol in buildings.

CO2 sensors are considered a maturetechnology and are offered by all majorHVAC equipment and controls compa-nies. The technology is recognized inASHRAE Standard 62, the InternationalMechanical Code (which establishesminimum regulations for mechanicalsystems), and some state and localbuilding codes. The first CO2 sensorssold were expensive, unreliable, and

1 An ASHRAE standard is usually designated by a standard number and the year in which it was last revised. Standard 62 is under continuous mainte-nance and is constantly being revised.

Figure 1. The relationship between CO2 and ventilationrates, assuming office-type activity.


difficult to keep calibrated accurately.However, manufacturers say they havelargely resolved those problems-theunits currently available generally areself-calibrating and similar to thermo-stats in price and reliability. A few yearsago, sensors cost around $400 to $500each; the cost now is usually about$200 to $250 per sensor (not includinginstallation). As market penetrationincreases, sensor prices are expectedto fall further.

Several manufacturers produce CO2 sen-sors for use in DCV. Most manufacturersof thermostats and economizers are inte-grating CO2 sensors into their products,and major manufacturers of packagedrooftop HVAC systems offer factory-installed CO2 sensors as an option.

CO2-based DCV is suitable only whenthere is a means of automatically adjust-ing the ventilation air supply (e.g.,variable-speed fans or some variabledamper arrangement). If this control isnot presently available, the savings fromDCV may justify the modifications toaccommodate this degree of control.

CO2 sensors designed for DCV aresuitable only for controlling occupant-related ventilation. DCV does noteliminate the need for a base rate ofventilation to prevent degradation ofair quality from contaminant sourcesunrelated to building occupancy, suchas emissions from building materials.Thus CO2-based DCV may not beappropriate—or may require higher tar-get ventilation settings-in new buildingsor others where there are contaminantsnot related to human occupancy, as itmay not provide sufficient fresh air todilute those contaminants. The CO2

sensors used for DCV are not appro-priate to monitor CO2 for medical orindustrial purposes that demand pre-cise air quality control.

DCV should be used only in areas wherehuman activity is the main reason forventilating the space. Industrial orlaboratory spaces that are subject toindoor air quality (IAQ) degradationfrom a wide variety of sources are unsuit-able for CO2-based ventilation control.

that has a light source at one end and alight detector at the other. Selectiveoptical fibers mounted over the lightdetector permit only light at the 4.26-micron wavelength absorbed by CO2

to pass through to the detector. AsCO2 levels rise, more infrared light isabsorbed and less light is detectable.

Photo-acoustic CO2 sensors also areavailable. In these sensors, also, airdiffuses into a chamber in the sensorsand is exposed to light at the wave-length absorbed by CO2. As the CO2

molecules absorb light energy, theyheat the air chamber and causes pres-sure pulses. A piezo-resistor senses thepulses and transmits data to a proces-sor that calculates the CO2 level.

Electrochemical sensors measure thecurrent transmitted across a gap filledwith an electrochemical solution. CO2

decreases the pH of the solution, free-ing conductive metal ions. A weakelectrical current can then flow acrossthe gap; the current signals an increasein the CO2 level.

Mixed-gas sensors also can detect CO2

along with other gases in the air, butthey have not proved to be effectivefor DCV because they do not measureCO2 specifically and cannot be tied to

Energy-Saving Mechanism

The energy savings from CO2 sensorsfor DCV result from the avoidance ofheating, cooling, and dehumidifyingfresh air in excess of what is neededto provide recommended ventilationrates. Many HVAC systems ventilateat a constant fixed level, usually thelevel prescribed for full occupancy, andthus provide more fresh air per occu-pant than the designed ventilation ratemuch of the time. Moreover, systemsusing fixed ventilation rates cannotaccurately account for unanticipatedair infiltration into a building (e.g., fromleakage or opened windows) and adjustthe fresh air intake accordingly.

DCV based on CO2 sensing allows real-time control of the ventilation levelsaccording to building occupancy. If abuilding is only 50% full, then only50% of the design-rate ventilation air,not 100%, is pulled in. CO2 sensors arethe most widely accepted technologycurrently available for implementingDCV. They do this by increasing theventilation rate whenever the CO2 levelin a space reaches a predetermined levelthat represents a differential betweenthe indoor and outdoor CO2 levels. Theoutdoor CO2 level is slightly dependenton local conditions and elevation, butcan generally be assumed to be around400 ppm. An indoor level of 1100 ppmof CO2 thus represents a 700-ppm dif-ferential and indicates a ventilation rateof 15 cfm per person in the occupiedspace. A differential of 500 ppm in thesame space would indicate a ventilationrate of 20 cfm per person.

The technology most often used in CO2

sensors is non-dispersive infrared spec-troscopy. It is based on the principlethat every gas absorbs light at specificwavelengths. Carbon dioxide sensorscalculate CO2 concentrations bymeasuring the absorption of infraredlight (at a wavelength of 4.26 microns)by CO2 molecules. The sensor appara-tus incorporates a source of infraredradiation, a detector, and electronics todetect the absorption. Air from the areabeing monitored diffuses into a chamber

Figure 2. Typical non-dispersive infraredspectroscopic CO2 sensor.


• Improved humidity control: In humidclimates, DCV can prevent unnec-essary influxes of humid outdoor airthat causes occupants to be uncom-fortable and encourages the growthof mold and mildew.

• Records of air quality data: Sensorreadings can be logged to provide areliable record of proper ventilationin a building. Such records can beuseful in protecting building ownersagainst ventilation-related illness ordamage claims.

• Reduced operational running timesfor the major HVAC equipment:Improving the ability to conditionthe building could delay start-timesof the HVAC equipment duringmorning pre-conditioning periodsby as much as several hours on aMonday morning in humid climates,resulting in incremental energy andcost savings.

Variations

DCV can be implemented using meth-ods other than CO2 sensing to indicateoccupancy levels or air quality condi-tions. For example, humidity sensors,motion detectors, particle counters,volatile organic contaminant sensors,and mixed-gas sensors can be used toregulate DCV. Time-controlled venti-lation (e.g., using programmed timeclocks) is also an option for buildingsthat are occupied only during certaintimes, for example, office buildings

and schools. Based on a review of theliterature, it appears that CO2 sensorsare becoming the industry standard fortypical DCV applications.

CO2 sensors are of three main types:infrared, electrochemical, and photo-acoustic. Based on the literature, infra-red sensors appear to be the type mostcommonly used for DCV applications.

Combination sensing units are avail-able that package wall-mounted CO2

sensors with thermostats or withhumidity sensors.

There are numerous variations in thetypes of HVAC systems and controlsystems with which CO2-based DCVis implemented. CO2-based DCVcapability can be added to an existingsystem by installing sensors and con-necting them with the air handlingsystems. Increasingly, new HVAC sys-tems are being factory-equipped withinput and controls strategies to acceptCO2-based DCV.

Installation

Typically, a CO2 sensor is installedon the wall like a thermostat. Newerproducts entering the market com-bine thermostats and CO2 and relativehumidity sensors, the three primaryindicators of human comfort, in oneunit. Sensors may be installed insideductwork instead of on a wall, butin-duct installation is not recommendedfor applications where the sensor would

ventilation rates as explicitly as a CO2

sensor can.

Sensors generally are either wall-mounted in the space to be monitoredor mounted inside the duct system.(Wall-mounted sensors usually are rec-ommended because duct-mounted unitsprovide data on the average CO2 con-centration in multiple spaces rather thanon the CO2 levels in individual areas.)

The sensors monitor CO2 concentra-tions continually and send data to thesystem for controlling the ventilationequipment. Various types of control sys-tems can be used to incorporate CO2

sensing: simple setpoint control thatactivates a fan or damper when CO2

levels exceed a setpoint; proportionalcontrol, in which sensor data adjustventilation air volume through a rangeof levels; proportional-integral control,in which fresh air intake is controllednot only by the CO2 level but also bythe rate at which the level is changing;and two-stage controls for zone-basedsystems in which both temperature sen-sors and CO2 sensors control ventilation.

Other Benefits

Potential secondary benefits of CO2-based DCV include:

• Improved IAQ: By increasing thesupply of fresh air to the buildingif CO2 levels rise to an unacceptablelevel, the technology could preventunder-ventilation that results inpoor air quality and stuffy rooms.

All of these types of systems can be modified to accomplish a DCV strategy

Control Type Minimum Modification Recommended Modification

All Apply ASHRAE Standard 62-1999 with CO2 Provide CO2 sensors in all zones, using thesensors provided in zones with high critical highest zone to increase the fresh air providedoutside air (OA) calculations by the associated air-handling unit (AHU)

Pneumatic Provide electronic-to-pneumatic transducer Upgrade the AHU to DDCto limit the OA damper position programmable control

Electronic Provide electronic device to limit the OA Upgrade the AHU to DDCdamper position programmable control

DDC–application specific Provide electronic device to limit the OA Upgrade the AHU to DDCdamper position programmable control

DDC-programmable Modify the control program to provide Same as minimumnew DCV strategy


average readings from several differentareas. With hardwired sensors (thoserequiring both power and signal wiring),wires are run from the sensors to thebuilding’s HVAC control system or tothe actuator that controls the fresh airsupply. Wireless sensors require neitherpower nor signal wiring, and the sensordata are fed into the HVAC controlsvia a gateway using one of several com-munication protocols.

In a building retrofit, dampers may needto be repaired or upgraded to work withthe sensors. Properly functioning modu-lating dampers are necessary. Pneumaticcontrols will need to be replaced withelectronic or direct digital controls(DDCs); it may also be worthwhileto replace existing electronic controlswith DDCs. Actuator modules thatdo not have input points for the sen-sors will need to be upgraded.

Part of the preparation for implement-ing DCV is to monitor the air in theoutdoor areas surrounding the build-ing for at least a week to establish theCO2 concentration in the ventilationair. The differential between the CO2

levels of the outdoor and indoor airwill determine the amount of freshair needed to meet ventilation require-ments. CO2 concentrations in all zonesinside the conditioned space also should

be monitored for several days to indi-cate existing ventilation levels in thezones. This monitoring may identifyproblems with air-handling systemsthat need to be corrected before DCVis implemented.

To ensure that the sensors are correctlycalibrated and working properly, ahand-held monitor should be used tocheck CO2 concentrations inside thebuildings regularly for a few days afterthe system begins operating. The read-ings from the monitor should be com-pared with the readings the sensors aresending to the ventilation system con-trols. Sensors that prove to be improp-erly calibrated should be recalibrated.

In all applications of CO2-based DCV,a minimum ventilation rate should beprovided at all times the building isoccupied. A rate of 20–30% of the origi-nal design ventilation rate for the spaceat maximum occupancy is often recom-mended as a new baseline ventilationrate. A higher rate may be needed forbuildings in which building materials,contents, or processes release chemicalsinto the air.

Federal-Sector Potential

Federal Technology Alerts target tech-nologies that appear to have significant

untapped federal-sector potential andfor which some installation experienceexists. CO2 sensors are recognized ashaving potential to reduce energyconsumption and costs resulting fromover-ventilation of buildings whileguarding against under-ventilation.

Estimated Savingsand Market Potential

CO2-sensors for demand-controlledventilation are recognized as havingpotential to reduce energy consump-tion, costs, and emissions.

Demand-controlled ventilation has notbeen assessed by the New TechnologyDemonstration activities. There are noknown estimates of the savings poten-tial of CO2-based DCV for the federalbuilding sector. Therefore, this reportcannot adequately quantify the energy-savings potential of the application ofthis technology for the federal sector.

The literature about DCV includesnumerous estimates and models of thesavings potential of DCV, as well asaccounts of monitored savings in spe-cific applications. The predicted andactual savings vary widely dependingon climate, type of HVAC system withwhich DCV is implemented, occupancypatterns in the space in which is it imple-mented, and other operating conditions.The capability of the building staff tokeep equipment adequately maintainedand operating properly also may affectsavings significantly.

The potential of CO2-based DCV foroperational energy savings has beenestimated in some of the literature atfrom $0.05 to more than $1 per squarefoot annually. The highest payback canbe expected in high-density spaces inwhich occupancy is variable and unpre-dictable (e.g., auditoriums, some schoolbuildings, meeting areas, and retailestablishments), in locations with highheating and/or cooling demand, and inareas with high utility rates.

A report by Lawrence Berkeley NationalLaboratory cited five case studies in largeoffice buildings with CO2-based DCV,all of which reported energy savings

Figure 3. CO2 equilibrium levels at various ventilation rates.


that resulted in payback times of from0.4 to 2.2 years. Two of the studies werecomputer simulations. One of those,conducted in 1994, simulated a 10-flooroffice building located in Miami,Atlanta, Washington, D.C., New York,and Chicago. The simulation predictedlarge gas savings for heating and smallerelectricity savings, resulting in predictedpayback times for the different loca-tions of from 1.4 to 2.2 years.

A 1999 study modeled the impactof DCV and economizer operationon energy use in four building types(office, retail, restaurant, school) inthree locations representing differentclimates: Atlanta; Madison, Wisconsin;and Albuquerque. For cooling, predictedsavings attributed to DCV dependedgreatly on location-savings were largerin Atlanta and Madison because humid-ity made economizer operation lessbeneficial. In low-humidity Albuquer-que, economizer operation was muchmore significant than DCV in reduc-ing cooling energy demand. In all threelocations, DCV resulted in large savingsin heating energy—27%, 38%, and42% for the office building in Madison,Albuquerque, and Atlanta, respectively;from 70% in Madison to over 80% inAtlanta and Albuquerque for the school;and over 90% in all three locations forthe retail and restaurant spaces. Similarresults were obtained for 17 other U.S.locations modeled. In all locations, theoffice building showed the most mod-est savings.

A recent proposal for a change to theOregon building code to require DCVin some applications cites a DOE2analysis of implementing a DCV strat-egy in a middle school gymnasium.It predicts energy cost savings of$3700 per year from DCV and asimple payback of about 6 months.

No estimate of market potential forCO2-based DCV in the federal sectorwas available.

Laboratory Perspective

Research on CO2-based DCV at thenational laboratories has been limitedso far. The technology is generally

regarded as a valid operational strategythat offers potential for energy andcost savings and protection of IAQ inmany facilities. A study conducted byLawrence Berkeley National Laboratoryconcludes that sensor-regulated DCVis generally cost-effective in buildingsin which the measured parameter (e.g.,CO2) is the dominant emission, theoccupancy schedule and levels are var-ied and unpredictable, and the heating/cooling requirements are large. Moni-toring of CO2-based DCV systems isunder way at some national laboratoriesto quantify savings and analyze effectson IAQ.

Some caveats have emerged from labo-ratory experience. Calibration drift hasbeen observed to be a problem in somesensors. The management of energymanagement systems so that theyadmit ventilation air at an appropri-ate CO2 level is another. A researcherat Oak Ridge National Laboratorywho works with building projects atfederal facilities noted that a skilled,well-trained building maintenancestaff is essential to proper functioningof the sensors and associated controlsystems. If other HVAC control sys-tems in a building frequently functionimproperly, there probably will beproblems with CO2 sensors and DCVcontrols, also. He advises that a testinstallation be tried first to ensure thatthe building staff understand how thedevices work, that the devices functionproperly, and that the staff can handlea larger DCV implementation.

Application

This section addresses technical aspectsof applying the technology. The rangeof applications and climates in whichthe technology can be best applied areaddressed. The advantages, limitations,and benefits in each application areenumerated. Design and integrationconcerns for the technology are dis-cussed, including equipment andinstallation costs, installation details,maintenance impacts, and relevantcodes and standards. Utility incentivesand support are also discussed.

Application Screening

DCV based on CO2 sensing offers themost potential energy savings in build-ings where occupancy fluctuates dur-ing a 24-hour period, is somewhatunpredictable, and peaks at a highlevel. It is also more likely to reduceenergy costs in locales that requireheating and cooling for most of theyear and where utility rates are high.

Savings opportunities are the greatest inbuildings with low average occupancylevels compared with the design occu-pancy levels. Larger savings are likelyin buildings that supply 100% outsideair to conditioned spaces than in build-ings that supply a mixture of outsideand recirculated air.

Types of buildings in which CO2-basedDCV is likely to be cost-effective includelarge office buildings; assembly rooms,auditoriums, and lecture halls; largeretail buildings and shopping malls;movie theaters; restaurants, bars andnightclubs; banks; outpatient areas inhospital; and hotel atriums or lobbies.

In primary and secondary schools, whereoccupancy is variable but predictable,time-controlled DCV may be more

Figure 4. Percent of total ventilation capac-ity percent building occupancy.


cost-effective. College classroom build-ings and large lecture rooms with vari-able occupancy through the day and theweek are more appropriate candidates.

CO2-based DCV does not interfere witheconomizers or other systems that intro-duce outdoor air into a building forcooling. Economizer operation over-rides DCV when conditions warranteconomizer use. The energy and costsavings for such arrangements dependon climate, occupancy schedule, andbuilding type. Warm, humid climates(e.g., the Southeast) offer the mostpotential to save cooling energy withDCV because high humidity reducesopportunities for economizer cooling.In dry climates such as the desert South-west, economizer operation may savemore cooling energy than DCV. Addi-tional cooling savings from DCV maybe insignificant in such climates, andDCV without economizer coolingmay even result in an energy penalty.

Buildings that use evaporative coolingmay not benefit from DCV during thecooling season.

Case studies generally show greatersavings for heating than for coolingin all climates. Heating savings aregreatest where heating the ventila-tion air accounts for a large portionof the energy demand. A buildingwith exchangers to transfer heatbetween incoming and exhaust air hasless potential for savings from DCVbecause the heat exchangers reducethe energy penalty of ventilation air.

In areas where peak power demand andpeak prices are an issue, DCV can beused to control loads in response to real-time prices. In those locations, DCVmay enable significant cost savingseven with little or no energy savings.

DCV can be implemented only in build-ings in which the outside air supply canbe automatically adjusted. If the air-handling system lacks that capability,it must be upgraded before CO2-basedDCV can be implemented. It is rec-ommended that pneumatic controls bereplaced with electronic or digital con-trols to implement CO2-based DCV.

The cost of installing CO2 sensors forDCV depends on how easily the exist-ing control system can incorporatethem. If the existing system has digitalcontrols and available input points onthe control modules for wired sensors,costs will be lower and implementa-tion easier. Wireless sensors may bemore easily integrated into an existingcontrol system, as no new input con-trol modules are needed.

Modeling is recommended, if feasible,to estimate the energy savings andcost-effectiveness of DCV in a spe-cific situation.

CO2 from human respiration must bethe dominant pollutant in a space forCO2-based DCV to be appropriate.Otherwise, it could lead to insufficientventilation. Occupancy-based DCV isinappropriate for spaces with high lev-els of contaminants unrelated to humanoccupancy, including industrial orlaboratory spaces. If CO2-based DCVis used in a new building or other spacecontaining materials that release irri-tating emissions (e.g., a clothing orcarpet store), the target ventilationrate must be high enough to ensurethat emissions are adequately diluted.

An HVAC system with CO2-basedDCV needs to be capable of a dailypre-occupancy (e.g., early morning)purge of inside air to avoid exposingoccupants to emissions (e.g., frombuilding materials) that accumulatewhile ventilation is at a low level.

Although DCV was developed mainlyas an energy-efficiency technology, italso is useful to ensure acceptable IAQ.In a properly operating system, CO2-based DCV ensures that the targetventilation rate per person is beingprovided at all times. Thus it may beappropriate for spaces where there areIAQ concerns. It may also help controlthe growth of mold and mildew byreducing unnecessarily large influxesof humid outdoor air.

Where to Apply

• Buildings where occupancy fluctu-ates during a 24-hour period, is

somewhat unpredictable, and peaksat a high level.

• Locales that require heating andcooling for most of the year.

• Areas where utility rates are high.

• Areas where peak power demandand prices are high.

• Buildings with low average occu-pancy compared with the designoccupancy.

• Large office buildings; assemblyrooms, auditoriums, and lecturehalls; large retail buildings and shop-ping malls; movie theaters; restau-rants, bars and nightclubs; banks;outpatient areas in hospital; andhotel atriums or lobbies.

• Warm, humid climates.

• Buildings in which heating/cooling/dehumidifying the ventilation airaccounts for a large portion of theenergy demand.

• Buildings in which the outside airsupply can be automatically adjusted.

• Buildings with HVAC systems thathave, or can be upgraded to, elec-tronic or digital controls.

• Spaces in which CO2 from humanrespiration is the only or dominantpollutant.

• Spaces that do not have high levelsof contaminants unrelated to humanoccupancy (e.g., industrial or labo-ratory spaces).

• Buildings in which poor IAQ resultingfrom under-ventilation or excessivehumidity resulting from under- orover-ventilation is a concern.

Precautions

The following precautions should bekept in mind in considering the useof CO2-based DCV.

• Sensors should be placed wherethey will provide readings that areclose to actual conditions in thespaced to be controlled.

• Wall-mounted sensors are generallypreferable to duct-mounted sensors.


• Sensors should not be mountedin locations where people willregularly breathe directly on them(e.g., at standing level near the cof-fee machine).

• Don’t buy carelessly—choose sensorswith a reputation for performing wellin terms of self-calibration, drift,accuracy, and reliability.

• A competent, well-trained mainte-nance staff is important to a suc-cessful implementation.

• In a retrofit, it is important to makea thorough study of the HVAC sys-tem and troubleshoot for existingventilation problems before install-ing a DCV system.

• A new base ventilation rate should beprovided at any time the building isoccupied. An often-cited guideline is20 to 30% of the original design rate.

• Applications in moderate climates,especially where economizer operationcontributes substantially to cooling,may show little energy/cost savings.

• Spaces where there are high levels ofcontaminants not related to occu-pancy may demand a higher venti-lation rate than would be providedby DCV based solely on CO2 levels.

DCV in a new HVAC installationshould not add significantly to the dif-ficulty of getting the system into opera-tion. However, retrofitting an existingsystem to work with DCV may be moreproblematic, particularly for an oldersystem with pneumatic controls.

The sensors typically are mounted onwalls like thermostats. Some manufac-turers offer thermostat/sensor combina-tions. Units that monitor temperature,CO2, and humidity are also available;they are useful with systems that includedesiccant dehumidification to controlthe humidity load in ventilation air.

Data from the sensors are fed to thebuilding’s HVAC control system or toan actuator that controls the amount ofventilation air that is admitted. In aretrofit installation, it may be necessaryto repair or upgrade dampers so theywill work in a more dynamic, modu-lating fashion in response to the sensors.Properly functioning dampers that canbe automatically controlled are essen-tial. Pneumatic controls will need tobe replaced with electronic controls orDDCs; it may also be worthwhile toreplace existing electronic controls withDDCs. Actuator modules that do nothave input points for the sensors willneed to be upgraded to add input points.

• During the first year or so in a newbuilding, it is advisable to maintaina higher ventilation rate than CO2

sensing would indicate to ensureproper dilution of emissions fromnew materials.

Design Considerations

CO2 sensing is a fairly simple technol-ogy, and installation of the sensorsthemselves is not complicated. Sensorvoltage, power, and control outputrequirements are similar to those usedcommonly by thermostats. For wiredsensor installations, the type of wire usedfor the signal wiring is often critical;some controls manufacturers specifythe wire gauge, type, and shielding/grounding requirements to preventsignal irregularities and errors. Withwireless sensors, these considerationsare not a concern, as data are transmit-ted over a Federal CommunicationsCommission-approved frequency. Thewireless sensors are self-powered anduse on-board power management toalert the building operator when thebattery needs to be changed.

Most HVAC equipment suppliers arenow offering systems designed to accom-modate DCV and accept readings fromCO2 sensors, so including CO2-based

Air handler unit and variable-air-volume controls are used for communication between the sensors and the air-handling system.

Control Type Control Medium Sequence

Pneumatic Air pressure from 3 to 15 psi Hard-piped: changes require additionalhardware and physical modifications tothe control air tubing

Electronic Electronic signals from 2 to 10 V DC, 0 to Hard-wired: changes require additional10 V DC, 4 to 20 mA are the most common. hardware and physical modifications toOther signals are 3 to 9 V DC, 6 to 9 V DC, the control wiring0 to 20 V phase cut

Direct digital control (DDC)— Electronic signals from 2 to 10 V DC, 0 to Programmed into an electronic controllerapplication-specific 10 V DC, 4 to 20 mA are the most common. that can be configured: some systems may

Other signals are 3 to 9 V DC, 6 to 9 V DC, need additional hardware to accomplish0 to 20 V phase cut, and via wireless transmis- desired sequencession to multi-protocol enabled gateways

DDC—programmable Electronic signals from 2 to 10 V DC, 0 to Programmed into an electronic controller10 V DC, 4 to 20 mA are the most common. that can be reprogrammed to accomplishOther signals are 3 to 9 V DC, 6 to 9 V DC, desired sequences0 to 20 V phase cut and via wireless transmis-sion to multi-protocol enabled gateways


The application of a CO2-based DCVstrategy using ASHRAE 62 is morecomplicated than simply installing CO2

sensors and using them to controldampers. In variable-air-volume sys-tems, particularly, fairly complex cal-culations and control algorithms maybe necessary to program the controlsystem properly for DCV.

DCV is compatible with economizersor other systems that can bring out-door air into a building for cooling.Economizer operation overrides DCVwhen conditions favor economizeruse. DCV used in conjunction withan enthalpy economizer (one regu-lated by relative humidity) probablywill save more energy than DCV witha temperature-regulated economizer.

Heat exchangers that transfer heatbetween supply air and exhaust airreduce the energy demand caused bylarge inflows of ventilation air; there-fore, DCV may not reduce energyuse in buildings with heat exchangers.Evaporative cooling systems may notbenefit from DCV during the cool-ing season.

Maintenance Impact

Maintenance of the sensors them-selves is not reported to be a problem.Although earlier sensor models hadreliability problems, the literaturegenerally reports that newer modelstypically are reliable and accurate.Manufacturers offer sensors thatrecalibrate themselves automaticallyand that are guaranteed not to needcalibration for up to 5 years. However,it is recommended that calibration bechecked periodically by comparing sen-sor readings during a several-hour periodwhen the building is unoccupied withreadings from the outdoor air. Manysensor models are able to sense calibra-tion problems and alert maintenancepersonnel if they are malfunctioning.

Some users report problems in get-ting the sensors calibrated initially. Ahand-held monitor should be used toensure that the installed sensors aremeasuring CO2 concentrations in theirzones accurately.

CO2-based DCV is a more sophisticatedtechnology than many maintenance per-sonnel are accustomed to, a buildingresearcher noted. A facility with a well-trained staff who know how to maintainand troubleshoot controls is the bestcandidate for a CO2-based DCV system.

Equipment Warranties

The prospective user should ask poten-tial suppliers, contractors, and installersabout warranties for specific equipmentmodels. Warranties for CO2 sensorsvary widely among different manufac-turers. Those reviewed offer warrantyperiods for parts and labor rangingfrom 90 days from the date of shipmentto 5 years from the date of purchase.A warranty period of 12 to 18 monthsfor parts and labor appears to be fairlycommon. Some sensors are guaranteedto remain calibrated for at least 5 years,and at least one manufacturer offers alifetime calibration guarantee.

Codes and Standards

The use of CO2-based DCV is acceptedin both ASHRAE 62 and the Interna-tional Mechanical Code (IMC). TheIMC is referenced by the BuildingOfficials and Code AdministratorsInternational, the Southern BuildingCode Congress International, and theInternational Code Conference ofBuilding Officials, which togetherestablish the model code languageused in local and state building codesthrough the United States. The com-mentary on IMC Section 403.1 states,“The intent of this section is to allowthe rate of ventilation to modulate inproportion to the number of occupants.CO2 detectors can be used to sense thelevel of CO2 concentrations which areindicative of the number of occupants...and this knowledge can be used to esti-mate the occupant load in a space.”

ASHRAE 62 recommends DCV for allventilation systems with design outsideair capacities of greater than 3000 ft3

per inch serving areas with an averagedesign occupancy density of more than100 people per 1000 ft2.

The California Building Standards Codewas amended in June 2001 to requireCO2-based DCV in some high-densityapplications during periods of partialoccupancy. A proposed change to theOregon Building Code (NR-HVAC-7)would require HVAC systems to includeprovision for DCV during periods whenspaces are only partially occupied.

Costs

Costs for sensors have dropped by about50% over the last several years as thetechnology has matured and becomemore widely used. Sensors typically costabout $250 to $260 each, uninstalled.For a new system, the installed cost willgenerally be about $600 to $700 perzone. For a retrofit system, the costwill depend on what type of controlsystem the building has. A controlscontractor estimates installed costs forretrofit applications at from $700 to$900 per zone for systems with anexisting DDC programmable control-ler and from $900 to $1200 per zonefor systems with pneumatic, electronic,or application-specific DDCs. Installa-tion costs for wireless systems are mini-mal beyond the cost of the actual sensorand gateway that can serve multiplesensor units.

In addition to the installation of thesensors, other components such asvariable frequency drives, controlinput and output hardware are oftenneeded to control the whole building,incrementally increasing the overallinstalled project cost beyond just thesensor cost.

Utility Incentives and Support

No specific information was foundregarding utility incentives that are inplace for CO2-based DCV. Some states(e.g., California) are considering suchan incentive program. Most electricand natural gas utility companies haverebate programs to promote energy-efficiency technologies that resultin overall improvement in buildingperformance, and DCV systems wouldquality for incentives under some ofthose programs.


DCV is one of the technologies thatutilities might consider for financialincentives in working with federal cus-tomers through a utility energy servicecontract on energy-efficiency projectsat federal sites.

Technology Performance

Field Experience

Three building operators whose facili-ties have installed CO2 sensors forDCV and a controls engineer whosecompany oversees HVAC installationswere contacted for this report regard-ing their experience with the use ofCO2-based DCV. All were generallypleased with the performance theyhave observed, although one reportedproblems with getting the sensors prop-erly calibrated and wired.

Purdue University has installed CO2

sensors in 12 large auditoriums andlecture halls (100 to 500 seats) to addressboth air quality issues and energy costs.The sensors were added to existing airhandling systems, and modulating out-side damper actuators were added wherethey were not already in place. The exist-ing air-handling units were 15 to 50 yearsold. Some of the rooms had pneumaticdampers with electronic controls; thosecontrols were not replaced unless theywere in bad condition. The large audi-toriums had DDC systems. All theexisting control modules had enoughavailable inputs to add the sensors.

Purdue’s controls systems engineerreported that there were minor prob-lems with retrofitting the older air-handling systems, but they were thesame kinds of problems that wouldhave surfaced with a conventionalHVAC retrofit-a lack of original orupdated building plans and plans thatdid not match the existing systems.“We had to do some legwork to verifythe existing configurations. It wouldbe easier with new systems,” she said.

Purdue has found the equipmentinstalled to be reliable, and it has per-formed as expected. There has been

a negligible impact on the mainte-nance staff because the sensors installedrecalibrate themselves automatically.Purdue had previously installed sensorsfrom another manufacturer that provedunsatisfactory because they could notbe calibrated properly or could notmaintain their calibration, the controlsengineer said. “The calibration issueis very important in sensor selection,”she said. The maintenance for theother equipment installed has beencomparable to maintenance on thesystems replaced.

Once the sensors began operating,they revealed that some of the lecturerooms had been underventilated bythe old systems. The sensors workedwell in resolving IAQ issues. Purduehas not monitored energy use sincethe sensors and new dampers wereinstalled, but trended data have shownthat the ventilation dampers modulateto lower or minimal positions when thelecture rooms are unoccupied or partlyoccupied and on weekends. Before thesensors were installed, the dampers wereopen to ventilate for full occupancy dur-ing the occupied cycle time (6 A.M.to 10 P.M. weekdays and 8 A.M. to4 P.M. Saturdays).

Purdue continues to install CO2-basedDCV in new applications in large roomswith variable occupancy. It will be addedin a renovation of the air handling sys-tem in a 900-person-capacity ballroomof the Purdue Memorial Union and isbeing reviewed for application in din-ing halls on campus.

A CO2 sensor was installed in an officebuilding on the Beaufort Marine CorpsAir Station in South Carolina to regu-late the makeup air system. It is thefirst of several DCV retrofits plannedon the base as part of an energy servicesperformance contract and a MarineCorps-funded controls system upgradefor the station. The air station is locatedin the low country of South Carolina,a particularly humid climate with aheavy cooling load.

The office building, a one-story41,354 ft2 facility with no windows,contains offices and a flight simulator.It was designed for 500 people, andthe ventilation system supplied enoughair to meet ASHRAE standards for 500,but the occupancy is rarely over 100,said Neil Tisdale, utilities director forthe base. “That’s a lot of air to be con-ditioning for no reason.” The sensorwas added as part of a makeup-air-system retrofit. The system alreadyhad modulating dampers and DDCs.Adding CO2-based DCV was fairlysimple—installing a sensor and modi-fying the control program to regulatethe damper with input from the sen-sor. The CO2 sensor was placed in thereturn air path.

The system has been operating for ayear, Tisdale said, and he is not awareof any maintenance problems or mal-functions. He ventured a rough esti-mate that the DCV system reduces thecooling load for the building by 20 tons,saving roughly of 12 megawatt hoursof power annually. The load for thebuilding’s chiller, sized to provide coolair for 500 people, was reduced somuch that part of the chilled waterwas diverted and used to cool an adja-cent 11,000 ft2 building.

A large hospital in Houston, Texas, hasinstalled CO2 sensors in its auditoriumsto ensure good IAQ and control energycosts, according to the hospital’s man-ager of energy services. The sensorswere added to existing systems, all ofwhich had digital controls; integratingthe sensors was not difficult, he said.The sensor input is not yet being usedto control the dampers automatically;building staff are monitoring the sen-sors to see whether they will control airchanges properly before switching toautomatic control. “We’re going to takeit a step at a time and expand gradu-ally,” the energy services manager said.The switch to direct control of damp-ers by the sensors will take “minutes”to implement and will probably occurduring the coming year, he said.


When the hospital begin installing thesensors, it was discovered that about halfof them were not wired properly andneeded to be rewired, he said. Manyof the units were calibrated incorrectlyand had to be recalibrated using a hand-held meter. “Some of them were show-ing 4000 to 5000 ppm of CO2. Youcan’t just set them up and assume they’llwork properly,” he cautioned.

However, he expects the sensors andDCV to perform well now that theinitial problems have been addressed.They may be added in other areas suchas busy foyers and other gathering areasonce the conference room installationis in full operation.

The owner of a digital controls companywho oversees HVAC installations sayshis company’s clients who have installedCO2-based DCV systems have beengenerally pleased with their performanceand found DCV to be a valuable addi-tion. A new HVAC installation thatincorporates CO2 sensors and DCV is ofminimal cost and difficulty and requiresonly a few additional calculations to setup the air handling program properly,he said. He cautions that sensors shouldbe installed to cover every CO2 zone inan area because high occupancy, and acorresponding high CO2 level, in onezone may not be reflected in the sensorreadings from the adjacent zone.

Although his firm has not done formalstudies of savings from DCV, it has seena reduction in outside air requirementsin all DCV installations because noneof the facilities are at the design occu-pancy all the time. The largest energysavings are likely in buildings designedto meet ASHRAE 62 requirements,he said, because they are likely to beadmitting more unneeded outside air.

He notes that if a building has previ-ously been chronically under-ventilated,installing DCV might increase ratherthan decrease energy usage because itwould bring in more outside air. How-ever, DCV will correct IAQ problemsand reduce liability for IAQ-relatedillnesses in such situations and may cor-rect problems with mold growth.

The maintenance impact of installingCO2 sensors is usually minimal if thesensors are self-calibrating, he said.Sensor calibration can be checked peri-odically by comparing indoor sensorreadings with metered outside air read-ings after a building has been unoccu-pied for about 5 hours.

Energy Savings

None of the building operators inter-viewed had measured energy consump-tion before and after the installation ofCO2 sensors. However, both Purdue staffand the controls company owner haveobserved lower outside air requirementsin facilities using DCV, which generallyresults in reduced energy demand.

Maintenance Impact

The maintenance impact of install-ing CO2 sensors has been minimalat Purdue. The sensors used are self-calibrating and have performed reliably.The hospital in Texas reported problemswith getting several of the sensors cali-brated initially and had to use hand-heldmonitors to recalibrate them. In addi-tion, some of the sensors were wiredimproperly and had to be corrected.

Case Study

This case study describes the method-ology used to assess the cost and energysavings implications of retrofitting abuilding for CO2-based ventilation con-trol. This methodology can be appliedto any building that has a track recordof energy usage. In this case, the pre-liminary assessment of the buildingshowed significant energy savings wereavailable. The performance of the build-ing after 6 months of operating withCO2 control is also presented and showsthat the predicted performance wasslightly conservative compared withthe actual savings realized.

Facility Description

This case study addresses a privatelyowned (non-government) 30-storyClass A office building in Birminghamthat was retrofitted with a CO2-based

ventilation control system in 2001. Ithad an existing state-of-the-art digitalbuilding control system installed severalyears earlier that was fully functional.The building was also upgraded toqualify for the EnergyStar label awardedby EPA to buildings having met quali-fying energy efficiency standards.

Utility costs for the building were verylow, $0.48/kWh during the base yearof 1999; overall energy costs for the baseyear were $1.61 ft2/year, representing117,992 Btu/ft2/year. This energy per-formance was compared with that ofother similar buildings in that regionusing the DOE Energy InformationAdministration (EIA) energy intensityindices, which provide a general guideto average energy usage for differenttypes of buildings in a region. The casestudy building was found to consumeabout 30% more than those found inthe EIA indices. The fact that the energycost was higher than average providedsome preliminary indication that despitethe low utility costs and EnergyStarrating, there were some opportunitiesfor further energy conservation.

Of the total energy consumed by thebuilding in the year 2000, 84.55%was spent on electricity for a total of13,966,500 kWh or $670,742 annually;the maximum peak load was 3,318 kW.Of the remaining total energy consumedby the building, 13.81% was spent onsteam for a total of 7,810,000 lb at acost of $122,581 annually.

Existing TechnologyDescription

The building HVAC system wasdesigned with two air-handling units(AHUs) per floor, located on oppositesides of the building. Variable air vol-ume (VAV) boxes served interior zones,and linear diffusers were located onperimeter zones. The linear diffusershave re-heat capability; the interior VAVboxes do not. Each AHU fan motorwas controlled by a variable-frequencydrive (VFD) to supply VAV boxes onthe floor. Each AHU was on a time-of-day schedule provided by the automa-tion system, wherein the units were


scheduled on and off to correspondwith the staff occupancy for that par-ticular floor on a Monday throughFriday basis, with limited Saturdayhours of operation. Sunday was mostoften scheduled off all day for mostfloors. Because of the nature of the cli-ent operations, a few floors operated24 hours per day, 7 days per week.

Outside air is drawn into each AHU ateach floor through a ducted grille. Thecombined general building and bath-room exhaust are provided by twin fans(of differing sizes) exhausted at the rooflevel by connecting all floors throughtwin vertical shafts. Both exhaustfans were of a constant-speed design,exhausting the full design load airregardless of the intake of outside airinto the building. Although a limitednumber of floors operated 24 hoursper day, both exhaust fans ran con-stantly at the full design load. Thisfeature probably also contributed tothe higher than normal operating costfor the building.

The building had a DDC building airsystem that was deemed capable ofexecuting a DCV strategy with someadditional programming, and had suf-ficient input/output capacity to addthe required components (e.g., CO2

sensors, peripheral I/O modules).

New Technology EquipmentSelection

The building automation system (BAS),an existing Siemens System 600, hadbeen installed several years previously.It was deemed capable of accepting thedata from the new CO2 sensors, able toexecute the new control strategy, andto have sufficient spare point capacityto provide the required input/output(I/O) from/to the new devices. Fournew Telaire series 8002, dual-beam CO2

sensors were installed per floor, excepton the second floor, which had a num-ber of enclosed meeting/conferencerooms. One CO2 sensor was installedin each of these rooms. One outsideCO2 sensor was installed to provideaccurate indoor/outdoor CO2 differen-tial readings.To facilitate the additional

I/O point requirements, new moduleswere installed within each S600 cabinet.For the CO2 sensors, analog input mod-ules accepting a 4-20 mA signal wereneeded, for the new fully proportionalelectric actuators, replacing the old relay(open/closed) actuators required ana-log modules driving a 0-10 V signal.

The existing building exhaust fans wereretrofitted with Magnatek VFDs to bal-ance the aggregate total of outside-airintake. The Magnatek VFDs were speci-fied with factory-installed Siemens S600FLN (field level network) cards to facili-tate the connection to the S600 BASwithout having to ‘hardwire’ each I/Opoint into the S600 control cabinet; thisarrangement saved labor and the expenseof individual I/O modules while simul-taneously providing much more motorperformance data to the building owner.

Completing the BAS componentchanges were the addition of oneSetra building static pressure sensorthat was essential to ensure a positivepressure in the whole building relativeto the outside.

No other major control componentswere required for the case study build-ing; however, the application of a newand fairly complex control strategy wasessential to execute the DCV applica-tion. The DCV control algorithm isthe most critical aspect of the project,as failure to implement an effectivecontrol sequence will yield lower energysavings performance.

Building Upgrade Assessment

Building owners must be able to evalu-ate the energy-saving potential andcost of specific initiatives to weighthe value of various building upgradeoptions. The methodology used toassess the potential of CO2-based ven-tilation control involved five basic steps.

1. Spot measurement of CO2 levels

2. Trend logging of CO2 and otherenvironmental factors

3. Estimation of the savings potential

4. Implementation assessment

5. Payback analysis

1. Spot measurement of CO2

As indicated previously, CO2 concentra-tions can be correlated to cfm/personventilation rates inside a space. For a pre-liminary assessment of ventilation ratesfor this building, a number of spot mea-surements of CO2 concentrations weremade on each floor. A hand-held CO2

monitor was used that is capable of cal-culating the cfm/person ventilation ratebased on inside/outside differential CO2

concentrations. The monitor assumesoutside concentrations are 400 ppm,but it will also allow the baseline con-centration to be set based on an actualoutside measurement.

Spot measurements for CO2 were madein the mid-morning to late morning andlate afternoon hours on weekdays, afteroccupancy had stabilized in the build-ing. They were taken while the buildingwas not operating in economizer mode.

The results of the spot measurementsshowed that most areas of the buildingunder study had CO2 concentrationsbelow 700-800 ppm, correspondingto ventilation rates in the range of28 to 35 cfm/person. These rates werewell over the original design target of20 cfm/person. The chart shows thecorrelation between peak CO2 levels

Figure 5. Handheld CO2 sensor with cfm/person calculation and data logging.


Figure 6. CO2 to ventilation rate conversion,assuming 400 ppm outside and office-typeactivity (1.2 MET).

database every 15 minutes for a periodof a week (at least 672 data points perparameter measured). Typically, data-logging sensors come with software toadjust, download and graph the results.

Several locations in the building weremonitored over 7 days. Results of trendlogging verified that the building wasover-ventilated compared with theoriginal design. The ventilation ratecan be determined by looking at por-tions of the graph that show extendedperiods of operation where the CO2

levels have stabilized; these indicatethat the amount of CO2 produced bypeople has reached equilibrium withthe ventilation rate of the space. Theseperiods are represented by the flat areasat the peak of each of the daily trendlogs. In almost all locations, CO2 levelsand calculated ventilation rates weresimilar to those recorded during spotmeasurements. In general, peak levelsaround 930 ppm indicate a ventilationrate of 20 cfm/person; peak levels near1100 ppm would indicate ventilationrates of 15 cfm/person. At the peakvalues recorded in the chart (674 ppmon average), the effective delivered

amount of fresh air per person was morethan 34 cfm/person, or 170% of thedesign need of 20 cfm/person. Duringperiods of lower occupancy throughoutthe day and during Saturday morningoperation, the effective delivered cfm/person considerably exceeded design needsand resulted in the use of excess energyto condition the surplus outside air.

Over-ventilation in the space may bedue to either space densities being beloworiginal design conditions or the upwardadjustment of outside air delivery to thebuilding over design conditions. Thebuilding operator indicated that duringsummer months, occupants in thebuilding often complained of uncom-fortable humidity levels, indicating thatperhaps the existing system was intro-ducing more outside air than the sys-tem was originally designed for. Thisoften happens when building operators“tweak” the building control system orair intakes to respond to complaints orto better tune the “feel” of the building.Based on hundreds of measurementsin buildings throughout the country,over-ventilation appears to be a com-mon problem in office buildings.

Figure 7. CO2 trend-logged results from 7 days of monitoring of one location in the building.

Nov 2000

and cfm/person ventilation rates, assum-ing office-type activity and an outsidelevel of 400 ppm. Based on the spotmeasurements, it appeared that thisbuilding could be a good candidate forenergy savings through better controlof ventilation. These results warrantedfurther investigation.

2. Trend logging of environmentalfactors

Trend logging was conducted over7 typical days in the building to ensurethat representative conditions werebeing measured. Ideally, measurementsshould be made in the major occupancyzones on each floor. If many locationsare involved and monitoring devices arelimited, multiple measurement sessionsmay be necessary. Devices used formeasurement should measure CO2 andshould be able to log concentrations to a

Cou

rtes

y of

Tel

aire

CO2 & Ventilationcfm/person CO2

UnderVentilated

Ideal

OverVentilated

Outside air

5

10

15

20

25

30

35

40

50

100

2500

1500

1200

1100

1000

900

850

800

750

700

650

600

500

400

Cou

rtes

y of

Tel

aire


3. Savings potential/pre-projectenergy analysis

A number of CO2 sensor manufacturersoffer software to analyze the cost savingsthat will result from applying a CO2

ventilation control based on occupancyversus a strategy of fixed ventilation dur-ing all occupied hours. These programstypically use local hourly weather dataand energy cost data. They focus on theenergy required to heat and cool vari-ous quantities of outside air deliveredto the building without undertaking afull-blown energy performance analysisof the building. A similar analysis can beperformed using more elaborate build-ing analysis programs such as DOE2.

Using one manufacturer’s CO2 software,the case study building was modeledfloor-by-floor to assess the potentialannual energy savings. Ventilation ratesdetermined from the trend logging wereused to calculate the current fixed ven-tilation rate by multiplying the cfm/person of ventilation air by the typicalpeak occupancy on the floor observedduring the trend-logged period. ForCO2 control, the modeling programassumed a control algorithm that wouldproportionately modulate air deliverybased on the CO2 concentration, a typi-cal ventilation control strategy used withCO2 sensing. It also allowed for simu-lated occupancy patterns to be variedevery 30 minutes to reflect typical occu-pancy variations through the day. Theprogram correlates the hourly ventilationload with hourly normalized outdoortemperature and dewpoint climactic datato provide an accurate assessment of

the HVAC load during all hours whileaccounting for days when economizeroperation results in zero savings as aresult of the ‘“free cooling” effect.

For the case study building, the CO2

modeling program projected that annualsavings of excess of $81,293 would beachievable based on normalized climaticdata. Based on energy costs in 2000,these savings are equivalent to a 10%reduction in total energy costs, anaverage of $3000 in savings per floorannually, and $0.22 per square footof gross area per year.

4. Implementation assessment

Most floors in the building had openfloor plans, with perimeter offices insome cases. In most cases, office doorsremained open during occupied hours,allowing free flow of ventilation airbetween the open central area and theoffices. For these floors, it was decidedto place one CO2 sensor on the wall ineach quadrant of the building. Therationale for such spacing was that eachsensor would cover an area of no greaterthan 3600 ft2 of open floor, and two sen-sors would provide input to each AHU.To ensure that all spaces were adequatelyventilated, the ventilation is controlledbased on the highest (or worst-case) levelof the paired CO2 sensors. It is impor-tant to note that duct sensors were notused because they tend to reflect an aver-age ventilation rate rather than what isactually occurring in the space.

One of the floors had a large numberof enclosed meeting/training/video-conferencing rooms that were not often

all occupied at the same time. One CO2

sensor was installed in each of thoserooms. In many instances, inter-zonaltransfer of air from spaces with lowerCO2 levels will moderate the amountof outside air that must be deliveredto the floor, particularly if all roomsare not occupied.

The AHU on each floor required that therelay-operated outside air damper actua-tor be replaced with a fully modulatingelectrical motorized actuator to allowmodulation of air delivery to each floor.To alleviate the constant negative pres-sure relationship in the building, theoriginal building static pressure sensorlocated at the highest point withinthe building (in the elevator shaft, forunknown reasons) was disconnected.A new building static pressure sensorwas located off the lobby level, on the leeside of the building and of the prevail-ing winds to minimize false positive ornegative readings. For the twin exhaustfans, VFDs were added to balance theoutside air intake with the requiredexhaust while maintaining a positive-pressure relationship with the outside.The new outside air CO2 sensor wasinstalled on the roof, away from anybuilding exhaust and the effects ofstreet-level fluctuations in CO2.

Critical tasks a building control sys-tem or a supporting control and logicsystem must be able to perform forCO2 control include these:

• Take input from a number of CO2

sensors on a floor and determinethe highest level.

• Modulate a signal to an actuator orvariable-speed fan serving each floorto proportionally modulate the deliv-ery of air based on the CO2 concen-tration measured on the floor.

• Control the central building airintake so demand to the entirebuilding can be modulated basedon the demand of each floor.(Often a pressure sensor in themain supply air trunk can ensureadequate air delivery.)

Summary of projected energy savings from CO2 control

Energy Electricity Steam TotalkWh Costa ($) Therms Costb

Base 13,966,500 670,800 147,756 122,591 793,391

With CO2 12,513,896 601,075 134,758 111,023 712,098

Savings 1,452,604 69,725 12,998 11,568 81,293

Savings (%) 10.4 8.8 10.2a Electricity cost = $0.48 per kWh.b Steam cost = $0.89 per therm (natural gas source @81% efficiency).


In summary, the work required toupgrade the building included

• Installing a minimum of four CO2

sensors and two VFD drives oneach of the air intakes for each floorand interfacing these devices to thebuilding control system.

• Programming the building controlsystem to take the CO2 transmittersignal and regulate air delivery on eachfloor based on in-space CO2 levels.

• Moving and replacing the buildingstatic pressure sensor.

• Installing variable-speed drives onthe building exhaust fans.

It is important to note that for CO2

control to work in a building, all otherHVAC-related systems must be in goodoperating order. CO2-related investiga-tions may identify problems not previ-ously recognized and add cost to anupgrade project. In this building, theassessment identified problems withthe location of the building pressuriza-tion sensor, which had to be relocatedand replaced. Logging of CO2 concen-trations in the spaces also indicatedthat the time-of-day operating scheduleson some floors did not match actualoccupancy patterns, and a change wasrecommended. In some cases, a regularcheck of logged CO2 data from perma-nently installed sensors can help keeptime-of-day schedules relevant to currentoccupancy patterns. In some cases, CO2

data have been used to detect end-of-day occupancy and initiate setbackoperation when inside levels approachoutside concentrations.

5. Payback analysis

The total cost of the building upgrade,including equipment and labor, wasestimated at $178,800, which alsoincluded $15,000 for the for the pre-project trend analysis. Based on theprojected cost savings of $81,293 theCO2 upgrade project was projected toyield a 2.2-year payback. Based on this

methodology and analysis, the projectwas initiated by the building owner inthe late spring of 2001 and completedin July of that year.

Life-Cycle Cost Analysis

A life-cycle cost analysis of this projectusing BLCC 5.1-03 was performedusing energy and cost data for the build-ing collected for the year 2000 but basedon an April 2003 start date, as requiredby the BLCC program. This procedureprobably results in an analysis based onslightly lower costs in the analysis thanprevail currently. The study period of15 years was selected because this is thetypical life of most electronic controldevices. If a longer period is desired, theuser should account for replacementof the sensors at the end of year 15(approximate cost is $250 to $350 perreplacement sensor, including labor).In performing the CO2 portion of thestudy, no annual or periodic costs wereassumed. In this case, we assumed useof self-calibrating sensor that requireno maintenance or calibration over theiroperating life. Some sensors do requireperiodic calibration at 3 to 5 years, andusers are urged to consider this fact inthe selection and cost analysis of their

particular installation. The cost ofcalibration will vary depending on themanufacturer and procedures required.About a third of the sensors sold todayhave a self-calibrating feature that elimi-nates maintenance requirements.

The total cost of the building upgrade,including equipment and labor, wasestimated at $178,800 (including$15,000 for pre-project trend analy-sis). The projected annual cost savingswas $81,293. The upgrade project wasprojected to yield a 2.2-year payback.

Post-Implementation Experience

After 6 months of operation, energydata from the building were collectedto determine how the CO2 retrofit andother improvements to the buildingwere performing. Unfortunately, be-cause of changes resulting from a com-pany reorganization, a full 12 monthsof performance data was not availablefor analysis. Figure 6 provides a sum-mary of the energy usage for each half-year period during 2000 and 2001.The CO2 ventilation control systemwas operated from July to Decemberof 2001. As can be seen from the data,the reductions in energy consumption

Summary of BLCC 5.1-03 life-cycle cost analysis

Base: Current Retrofit of SelfStudy Period 15 years Operational/No Calibrating Saving fromDiscount Rate 3% CO2 Control CO2 Control CO2

Initial Investment: Cash $ $178,800.00 $ (178,800.00) Requirements

Future CostAnnual & non-annual $ $ $ recurring costEnergy related cost $9,414,476 $8,460,875 $953,602Total $9,414,476 $8,460,875 $953,602

Net SavingsPV of non-investment savings $953,602Increased total investment $178,800

Savings-To-Investment Ratio 5.33Adjusted Internal Rate of Return 15.66%Simple Payback 3 yearsDiscounted Payback 3 years


Before-and-after energy performance

Energy Cost 2000 2001Jan–June July–Dec Jan–June July–Dec*

Electricity kwh 6,357,000 7,609,500 6,606,000 6,219,000Electricity ($) $305,136 $365,256 $317,088 $298,512Steam (therm) 71,918 75,838 84,523 22,960Steam ($) $64,007 $67,496 $75,225 $20,434

Total ($) $369,143 $432,752 $392,313 $318,946Annual ($) $801,895 $711,2606 month savings comparing July–Dec 2000 vs 2001Electric ($) $66,744Steam ($) $47,061Total ($) $113,805

*CO2 Control in Operation

Figure 8. Shown are the monthly costs (right scale) for the ‘baseline’ months 2000(light line) and for the months of 2001 (dark line) calculated using 2000 rates. The vari-ance (left scale) is shown by the columns. A major vector change occurred beginningin July (as the first phases of DCV were coming on-line) that is known not to be weatherrelated (the minor variance in Feb/Mar is known to reflect a shift in the weather load).

were in excess of the savings predicted.The difference is probably due to anumber of factors, including these:

• Replacement and relocation ofthe building pressurization sensorprobably impacted energy savings,but savings were not predicted forthis improvement.

• Time-of-day schedules were repro-grammed based on actual occupancy,and building exhaust fans wereadjusted to minimum levels duringunoccupied hours. The change thatprobably contributed additionalenergy savings to savings predicted.

• The year 2001 was milder than2000 and had 20% fewer coolingdegree days and 4% fewer heatingdegree days.

According to the facility manager, thetenants were satisfied with the comfortlevels in the space following the retrofitand no longer complained of highhumidity levels in the building duringsummer months. Because the loggedCO2 data showed the space to be sig-nificantly over-ventilated, reducingventilation with CO2 control reducedthe amount of humid outside air drawninto the building and allowed the cool-ing system to maintain better controlof humidity.

The following charts show energy per-formance, electricity costs, and steamcosts before and after the CO2-basedDCV retrofit.

vari

an

ce

Mo

nth

ly T

ota

ls

Cou

rtes

y of

Tel

aire


The Technologyin Perspective

The Technology’s Development

CO2-based DCV is an emerging tech-nology with currently limited butincreasing market penetration. Theliterature on CO2-based DCV consis-tently predicts that the technology willbecome a common ventilation strategy.

CO2 sensors for DCV regulation area relatively mature product, havingbeen on the market since about 1990.The first CO2 sensors were expensive,unreliable, and difficult to keep cali-brated. However, manufacturers claimto have resolved those problems—unitscurrently available generally are self-calibrating and similar to thermostats

in price and reliability. Although manu-facturers say most sensors are designedto operate for 5 years without mainte-nance, regular inspection to verifycalibration and proper operation isstill recommended.

Interest in DCV was spurred byASHRAE 62, which increased therequirement for ventilation air inbuildings to safeguard IAQ. Buildingmanagers became concerned that theincreased ventilation level was drivingup energy costs by introducing unnec-essarily large amounts of fresh air thathad to be cooled, heated, dehumidified.CO2-controlled DCV offers a methodof regulating ventilation levels so as tosatisfy the requirements of ASHRAE62 without over-ventilating.

DCV based on CO2 sensing is recog-nized as a viable strategy by the HVACand building industries:

• CO2 sensors are sold by all majorHVAC equipment and controlscompanies.

• More than a dozen manufacturersproduce CO2 sensors for DCV.

• Most manufacturers of thermostatsand economizers now integrate CO2

sensors into their products, and majormanufacturers of packaged rooftopHVAC systems offer factory-installedCO2 sensors as an option.

• ASHRAE 90 requires CO2 sensorsfor DCV in high-density applications.

• In the next revision of its buildingcode, California will begin requiringCO2-based DCV in all buildingshousing 25 people or more per1000 ft2.

• A proposed change in the Oregonbuilding code would require DCVfor HVAC systems with ventilationair requirements of at least 1500 cfm,serving areas with an occupant fac-tor of 20 or less.

• ASHRAE Standard 62, the IMC(which establishes minimum regula-tions for mechanical systems), andsome state and local building codesrecognize CO2-based DCV.

• The U.S. Green Building Code givespoints in its LEED rating system foruse of CO2-based DCV.

Technology Outlook

CO2-based DCV appears likely tobecome a commonplace ventilationstrategy for large commercial and insti-tutional buildings as building operatorsseek ways to balance building codeventilation requirements and IAQ con-cerns with the need to control energydemand and operating costs. Equip-ment development is encouraging theadoption of the technology. Prices ofCO2 sensors for DCV prices havedropped by about 50% since 1990,

Figure 9. Shown are the monthly costs (right scale) for the ‘baseline’ months in 2000(lighter line) and the cost for the months of 2001 (darker line) calculated using 2000rates. The variance (left scale) is shown by the columns. Since July, savings haveoccurred each month as a direct result of the CO2 retrofit project, whereas the variancein March 2001 is known to be due to a shift in weather load.

Before and After Steam Usage for Case Study Buildingvari

an

ce

Mo

nth

ly T

ota

ls

Cou

rtes

y of

Tel

aire


AirTest Technologies1520 Cliveden AvenueDelta, BC V3M 6J8Tel: 888-855-8880, 604-517-3888Fax: 604-517-3900www.airtesttechnologies.commike.schell@airtesttechnologies.com

Carrier Corporation6304-T Thompson Rd., P.O. Box 4808Syracuse, NY 13221 4808Tel: 315-432-6000www.carrier.com/

DetectAire, Inc.5973 Encina Road, Suite 109Goleta, California 93107Tel: 805-683 1117www.detectaire.com(DetectAire markets wireless sensors)

Digital Control Systems, Inc.7401-T S.W. Capitol Hwy.Portland, OR 97219 2431Tel: 503-246-8110 Fax: 503-246-6747http://www.dcs-inc.net

Honeywell Control Products11 W. Spring St.Freeport, IL 61032Tel: 815-235-6847 Fax: 815-235-6545Cable: Honeywell-Freeporthttp://www.honeywell.com/sensing

Johnson Controls, Inc.Controls Group507 E. Michigan St., P.O. Box 423Milwaukee, WI 53201 0423Tel: 800-972-8040 Fax: 414-347-0221

Kele3300 Brother Blvd.Memphis, TN 38133Tel: 888-397-5353 Fax: 901-382-2531Email: [email protected]

MSAMSA Bldg., P.O. Box 426Pittsburgh, PA 15230 0426Tel: 412-967-3000Fax: 412-967-3450 or 800-967-0398Cable: MINSAFhttp://www.msanet.com

Telaire Systems, Inc.6489 Calle Real, Dept. TRGoleta, CA 93117Tel: 805-964-1699 Fax: 805-964-2129

Texas Instruments, Inc. Commercial Sensors & Controls Div.34 Forest St., MS 20-22, P.O. Box 2964Attleboro, MA 02703-0964Tel: 800-788-8661, Ext. 400http://www.tisensors.com

Thermo Gas Tech8407 Central Ave.Newark, CA 94560Tel: 888-243-6167 Fax: 510-794-6201Or call: 510-745-8700http://www.thermogastech.com

Vaisala, Inc.100 Commerce WayWoburn, MA 01801 1008Tel: 800-408-5266 Fax: 781-933-8029http://www.vaisala.com

Veris Industries, Inc.10831-T S.W. Cascade Ave.Portland, OR 97223Tel: 503-598-4564 Fax: 503-598-4664http://www.veris.com

Who is Usingthe Technology

Federal SitesThe PentagonRobert BillakDepartment of DefensePentagon Heating and Refrigeration Plant300 Boundary Channel Dr.Arlington, VA 22020Navy AnnexRobert BillakDepartment of DefensePentagon Heating and Refrigeration Plant300 Boundary Channel Dr.Arlington, VA 22020Beaufort Marine Corps Air StationNeil Tisdale, Utilities DirectorBeaufort, South Carolina

Non-Federal SitesPurdue UniversityLuci Keazer, P.E.Facilities Service Department1670 PFSB Ahlers DriveWest Lafayette, IN [email protected] UniversityAdam Joseph Lewis Center for Environmental StudiesLeo Evans122 Elm StreetOberlin, Ohio 44074Reedy Creek Energy Services (The Walt Disney Company)Paul [email protected] Reality ServicesBob [email protected]

and as market penetration increases,are expected to fall further. Most HVACequipment makers are including factoryinstallation of CO2 sensors for DCVas an option in their large packaged sys-tems. At least one major manufacturerhas adapted its equipment line toaccommodate CO2-based DCV inevery piece of ventilation equipmentit makes. This implicit endorsement ofthe technology by equipment makersmakes its adoption simpler and lessexpensive, at least in new installations,and will almost certainly contributeto developing a broader market forCO2-based DCV.

The emergence of wireless, self-poweredmulti-parameter (CO2/temperature/humidity) sensors designed for perma-nent installation is expected to furtherincrease the speed of adoption of DCV.They have the potential to lower thetotal cost of deployment, makinginstallation simpler and faster andimproving paybacks.

As installers, HVAC engineers, andmaintenance personnel accumulatemore hands-on experience with thecontrols and operational issues CO2-based DCV presents, and gain moreexpertise in troubleshooting and resolv-ing problems, the comfort level withthe technology should increase. Theincreased familiarity is likely to increaseits acceptance.

Manufacturers

The following list includes companiesidentified as manufacturers of CO2 sen-sors for demand-controlled ventilationat the time of this report’s publication.The list does not include manufactur-ers of CO2 sensors designed for use insafety equipment or for control of con-ditions in laboratories or industrialprocesses that are not appropriate fordemand-controlled ventilation. Wemade every effort to identify all manu-facturers of the equipment, includingan extensive search of the ThomasRegister; however, this listing is notpurported to be complete or to reflectfuture market conditions.


For Further Information

AssociationsAmerican Society of Heating, Refrigerating and Air-Conditioning EngineersAmerican Society for Testing and MaterialsU.S. Green Building CouncilAmerican Indoor Air Quality Council

Design and Installation GuidesDemand Controlled Ventilation System Design, Carrier Corporation, Syracuse, NY, 2001Application Guide for Carbon Dioxide Measurement and Control, Telaire Corporation, Goleta, California, 1994.

Manufacturer’s Application Notes“Reference Guide for Integration CO2 DCV with VAV Systems,” Telaire Corporation,October 2000. www.telaire.com/telaire.htm.

“Common CO2 Wiring Issues,” Telaire Corporation, October 2000, www.telaire.com/telaire.htm.

PublicationsA.T. DeAlmeida and W.J. Fisk, Sensor-Based Demand Controlled Ventilation, LBNL-40599, UC-1600, Lawrence BerkeleyNational Laboratory, July 1997.*

M.J. Brandemuehl and J.E. Braun, “The Impact of Demand-Controlled and Economiser Ventilation Strategies on EnergyUse in Buildings,” ASHRAE Transactions, vol. 105, pt. 2, pp. 39–50, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Atlanta, 1999.*

M.J. Brandemuehl and J.E. Braun, “The Impact of Demand-Controlled Ventilation on Energy Use in Buildings,” p. 15,paper RAES99.7679 in Renewable and Advanced Energy Systems for the 21st Century, RAES 1999 Proceedings, R. Hogan,Y. Kim, S. Kleis, D. O’Neal, and T. Tanaka, eds., American Society of Mechanical Engineers, New York, 1999.**

S.C. Carpenter, “Energy and IAQ impacts of CO2-Based Demand-Controlled Ventilation, ASHRAE Transactions, vol. 102,pt 2., pp. 80–88, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Atlanta, 1996.**

S.J. Emmerich and A.K. Persily, “Literature Review on CO2-Based Demand-Controlled Ventilation,” ASHRAE Transac-tions, vol. 103, pt. 2, pp. 229–243, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Atlanta,1997.*

Electric Power Research Institute, Office Complexes Guidebook, Innovative Electric Solutions, TR-109450, October 31, 1997.

W.J. Fisk and A.T. DeAlmeida, “Sensor-Based Demand Controlled Ventilation: A Review,” Energy and Buildings, 29(1)(March 1), 1998.**

S. Gabel and B. Krafthefer, Automated CO2 and VOC-Based Control of Ventilation Systems Under Real-Time Pricing,EPRI-TR-109117, Electric Power Research Institute, Palo Alto, California, and Honeywell Technology Center, Minneapolis,Minnesota, October 1998.**

D. Houghton, “Demand-Controlled Ventilation: Teaching Buildings to Breathe,” E Source Tech Update TU-95-10, ESource, Boulder, Colorado, 1995.*

D.B. Meyers, H. Jones, H. Singh, P. Rojeski, “An In situ Performance Comparison of Commercially Available CO2 Sensors,pp. 45–53 in Competitive Energy Management and Environmental Technologies: Proceedings, Association of Energy Engi-neers, Atlanta, 1995. **

B.A. Rock and C.T. Wu, “Performance of Fixed, Air-side Economizer, and Neural Network Demand-Controlled Ventilationin CAV Systems, ASHRAE transactions, vol. 104, pt 2., pp. 234–245, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Atlanta, 1998.

* Publication that provided information for this document.** Publication retrieved from NISC/BiobliLine.


K.W. Roth, D. Westpalen, J. Dieckmann, S.D. Hamilton, and W. Goetzler, Energy Consumption Characteristics of Commer-cial Building HVAC Systems Volume III: Energy Savings Potential, TIAX 68370-00, TIAX, LLC, Cambridge, MA, July 2002.*

M.B. Schell and D. Inthout, “Demand Control Ventilation,” ASHRAE Journal, American Society of Heating, Refrigeratingand Air-Conditioning Engineers, Atlanta, February 2001.*

M.B. Schell, S.C. Turner, and R.O. Shim, “Application Of CO2 Demand-Controlled Ventilation Using ASHRAE 62: Opti-mizing Energy Use And Ventilation,” ASHRAE Transactions, vol. 104, pt. 2, pp. 1213-1225, American Society of Heating,Refrigerating and Air-Conditioning Engineers, Atlanta, 1998.*

M. Schell, “Real-Time Ventilation Control,” Heating, Piping, Air-Conditioning Engineering, Interactive Feature, April2002.* www.hpac.com/member/feature/2002/0204/0204schell.htm

Case StudiesAdam Joseph Lewis Center for Environmental Studies, www.oberlin.edu/envs/ajlc/ (Building Systems, Heating and AirQuality, Indoor Air Quality)

Codes and StandardsVentilation for Acceptable Indoor Air Quality, ASHRAE Standard 62-2001, American Society of Heating, Refrigerating andAir-Conditioning Engineers, Atlanta. www.ashrae.org.

Energy Standards for Buildings Except Low-Rise Residential Buildings, ASHRAE Standard 90.1-2001, American Society ofHeating, Refrigerating and Air-Conditioning Engineers, Atlanta. www.ashrae.org.

American Society of Testing and Materials Standard D6245-98, Standard Guide for Using Indoor Carbon Dioxide Concentrationsto Evaluate Indoor Air Quality and Ventilation

2000 International Mechanical Code, International Code Council, Falls Church, Virginia.

2001 California Building Standards Code, Title 24, California Code of Regulations, Part 4: California Mechanical Code,California Building Standards Commission, Sacramento, California.


Appendix A

Federal Life-Cycle Costing Procedures and the BLCC Software

Federal agencies are required to evaluate energy-related investments on the basis of minimum life-cycle costs (10 CFR Part 436).A life-cycle cost evaluation computes the total long-run costs of a number of potential actions, and selects the action that mini-mizes the long-run costs. When considering retrofits, sticking with the existing equipment is one potential action, often calledthe baseline condition. The life-cycle cost (LCC) of a potential investment is the present value of all of the costs associated withthe investment over time.

The first step in calculating the LCC is the identification of the costs. Installed Cost includes cost of materials purchased and thelabor required to install them (for example, the price of an energy-efficient lighting fixture, plus cost of labor to install it). EnergyCost includes annual expenditures on energy to operate equipment. (For example, a lighting fixture that draws 100 watts andoperates 2,000 hours annually requires 200,000 watt-hours (200 kWh) annually. At an electricity price of $0.10 per kWh, this fix-ture has an annual energy cost of $20.) Nonfuel Operations and Maintenance includes annual expenditures on parts and activitiesrequired to operate equipment (for example, replacing burned out light bulbs). Replacement Costs include expenditures to replaceequipment upon failure (for example, replacing an oil furnace when it is no longer usable).

Because LCC includes the cost of money, periodic and aperiodic maintenance (O&M) and equipment replacement costs, energyescalation rates, and salvage value, it is usually expressed as a present value, which is evaluated by

LCC = PV(IC) + PV(EC) + PV(OM) + PV(REP)

where PV(x) denotes “present value of cost stream x,”IC is the installed cost,EC is the annual energy cost,OM is the annual nonenergy O&M cost, andREP is the future replacement cost.

Net present value (NPV) is the difference between the LCCs of two investment alternatives, e.g., the LCC of an energy-savingor energy-cost-reducing alternative and the LCC of the existing, or baseline, equipment. If the alternative’s LCC is less than thebaseline’s LCC, the alternative is said to have a positive NPV, i.e., it is cost-effective. NPV is thus given by

NPV = PV(EC0) – PV(EC

1)) + PV(OM

0) – PV(OM

1)) + PV(REP

0) – PV(REP

1)) – PV(IC)

or

NPV = PV(ECS) + PV(OMS) + PV(REPS) – PV(IC)

where subscript 0 denotes the existing or baseline condition,subscript 1 denotes the energy cost saving measure,IC is the installation cost of the alternative (note that the IC of the baseline is assumed zero),ECS is the annual energy cost savings,OMS is the annual nonenergy O&M savings, andREPS is the future replacement savings.

Levelized energy cost (LEC) is the break-even energy price (blended) at which a conservation, efficiency, renewable, or fuel-switching measure becomes cost-effective (NPV >= 0). Thus, a project’s LEC is given by

PV(LEC*EUS) = PV(OMS) + PV(REPS) – PV(IC)

where EUS is the annual energy use savings (energy units/yr). Savings-to-investment ratio (SIR) is the total (PV) savings of ameasure divided by its installation cost:

SIR = (PV(ECS) + PV(OMS) + PV(REPS))/PV(IC).

Some of the tedious effort of life-cycle cost calculations can be avoided by using the Building Life-Cycle Cost software, BLCC,developed by NIST. For copies of BLCC, call the EERE Information Center (877) 337-3463.


This page left blank intentionally

About FEMP’s New Technology Demonstrations

The Energy Policy Act of 1992 and sub-sequent Executive Orders mandate thatenergy consumption in federal buildingsbe reduced by 35% from 1985 levels bythe year 2010. To achieve this goal, theU.S. Department of Energy’s FederalEnergy Management Program (FEMP)sponsors a series of activities to reduceenergy consumption at federal installa-tions nationwide. FEMP uses new tech-nology demonstrations to acceleratethe introduction of energy-efficient andrenewable technologies into the federalsector and to improve the rate of tech-nology transfer.

As part of this effort, FEMP sponsorsthe following series of publications thatare designed to disseminate informationon new and emerging technologies:

Technology Focuses—brief informationon new, energy-efficient, environmen-tally friendly technologies of potentialinterest to the federal sector.

Federal Technology Alerts—longersummary reports that provide details onenergy-efficient, water-conserving, andrenewable-energy technologies that havebeen selected for further study for pos-sible implementation in the federal sector.

Technology Installation Reviews—con-cise reports describing a new technology

and providing case study results, typi-cally from another demonstration pro-gram or pilot project.

Other Publications—we also issue otherpublications on energy-saving technolo-gies with potential use in the federal sec-tor.

More on FederalTechnology Alerts

Federal Technology Alerts, our signaturereports, provide summary informationon candidate energy-saving technolo-gies developed and manufactured in theUnited States. The technologies featuredin the FTAs have already entered the mar-ket and have some experience but arenot in general use in the federal sector.

The goal of the FTAs is to improve therate of technology transfer of new energy-saving technologies within the federalsector and to provide the right people inthe field with accurate, up-todate infor-mation on the new technologies so thatthey can make educated judgments onwhether the technologies are suitable fortheir federal sites.

The information in the FTAs typicallyincludes a description of the candidatetechnology; the results of its screening

tests; a description of its performance,applications, and field experience to date;a list of manufacturers; and importantcontact information. Attached appendixesprovide supplemental information andexample worksheets on the technology.

FEMP sponsors publication of the FTAsto facilitate information-sharing betweenmanufacturers and government staff.While the technology featured promisessignificant federal-sector savings, the FTAsdo not constitute FEMP’s endorsementof a particular product, as FEMP has notindependently verified performance dataprovided by manufacturers. Nor do theFTAs attempt to chart market activityvis-a-vis the technology featured. Read-ers should note the publication date onthe back cover, and consider the FTAsas an accurate picture of the technologyand its performance at the time ofpublication. Product innovations andthe entrance of new manufacturers orsuppliers should be anticipated since thedate of publication. FEMP encouragesinterested federal energy and facilitymanagers to contact the manufacturersand other federal sites directly, and touse the worksheets in the FTAs to aidin their purchasing decisions.

Federal Energy Management Program

The federal government is the largest energy consumer in the nation. Annually, the total primary energy consumed by the federalgovernment is 1.4 quadrillion British thermal units (quads), costing $9.6 billion. This represents 1.4% of the primary energy consumption inthe United States. The Federal Energy Management Program was established in 1974 to provide direction, guidance, and assistance to federalagencies in planning and implementing energy management programs that will improve the energy efficiency and fuel flexibility of thefederal infrastructure.

Over the years, several federal laws and Executive Orders have shaped FEMP’s mission. These include the Energy Policy and ConservationAct of 1975; the National Energy Conservation and Policy Act of 1978; the Federal Energy Management Improvement Act of 1988; theNational Energy Policy Act of 1992; Executive Order 13123, signed in 1999; and most recently, Executive Order 13221, signed in 2001, andthe Presidential Directive of May 3, 2001.

FEMP is currently involved in a wide range of energy-assessment activities, including conducting new technology demonstrations, to hastenthe penetration of energy-efficient technologies into the federal marketplace.

A Strong Energy Portfolio for a Strong AmericaEnergy efficiency and clean, renewable energy will mean a stronger economy, a cleanerenvironment, and greater energy independence for America. Working with a wide arrayof state, community, industry, and university partners, the U.S. Department of Energy’sOffice of Energy Efficiency and Renewable Energy invests in a diverse portfolio ofenergy technologies.

Produced for the U.S. Departmentof Energy, Energy Efficiency andRenewable Energy, by Oak RidgeNational Laboratory

DOE/EE-0293

March 2004

For More Information

EERE Information Center

1-877-EERE-INF or

1-877-337-3463

www.eere.energy.gov/femp

General Program Contacts

Ted Collins

New Technology Demonstration

Program Manager

Federal Energy Management Program

U.S. Department of Energy

1000 Independence Ave., SW EE-92

Washington, D.C. 20585

Phone: (202) 586-8017

Fax: (202) 586-3000

[email protected]

Steven A. Parker

Pacific Northwest National Laboratory

P.O. Box 999, MSIN: K5-08

Richland, WA 99352

Phone: (509) 375-6366

Fax: (509) 375-3614

[email protected]

Technical Contact and Author

James R. Sand

Oak Ridge National Laboratory

P.O. Box 2008, Bldg. 3147

Oak Ridge, TN 37831-6070

Phone: (865) 574-5819

Fax: (865) 574-9329

[email protected]

Log on to FEMP’s Web site for informationabout New Technology Demonstrations

www.eere.energy.gov/femp/

You will find links to

• A New Technology Demonstration Overview

• Information on technology demonstrations

• Downloadable versions of publications in Adobe PortableDocument Format (pdf )

• A list of new technology projects under way

• Electronic access to a regular mailing list for new productswhen they become available

• How Federal agencies may submit requests to us to assessnew and emerging technologies

Demand-Controlled Ventilation Using CO Sensorsinfohouse.p2ric.org/ref/43/42844.pdf ·...

Documents

Transcript of Demand-Controlled Ventilation Using CO Sensorsinfohouse.p2ric.org/ref/43/42844.pdf ·...