Content uploaded by Michael CW Kintner-Meyer
Author content
All content in this area was uploaded by Michael CW Kintner-Meyer on May 02, 2015
Content may be subject to copyright.
Wireless Sensors: Technology and Cost-Savings for Commercial Buildings
M. Kintner-Meyer, M. Brambley, T. Carlon, and N. Bauman
Pacific Northwest National Laboratory
1
ABSTRACT
Two projects underway for the U.S. Department of Energy Office of Building
Technology, State and Community Programs, aim to adapt, test and demonstrate wireless
sensors and data acquisition for heating, ventilating, and air-conditioning (HVAC) in
commercial buildings. One project focuses on built-up systems in medium to large
buildings; the second on applications for rooftop units in small- to medium-size facilities.
Beyond mobility, which is the driver for many wireless applications, the key promise
of wireless technology in building operation is to reduce the cost of installing data acquisition
and control systems by eliminating the wires. Installation of wiring can represent 20% to
80% of the cost of a sensor point in an HVAC system. The availability of low-cost wireless
sensor systems could not only reduce sensor costs overall, but also lead to increased use of
sensors necessary to establish and maintain highly efficient and effective building operations.
In this paper, the authors present the technical characteristics and costs of off-the-
shelf wireless sensor and data-acquisition systems and describe how they can be adapted to
commercial buildings. The paper provides a brief overview of wireless communication
standards and discusses their appropriateness to HVAC control applications. The authors
describe two wireless technology demonstration projects and discuss their cost
competitiveness to conventional wired system. The paper provides a general discussion on
cost competitiveness of wireless versus wired control networks for retrofit and new
construction and concludes with some future prospects for wireless technologies for
buildings applications.
Introduction
While long promised as an emerging technology for the building automation industry,
wireless applications in HVAC controls are still in their infant stage at best and are not
common practice. A 1999 expert roundtable of HVAC industry professionals unanimously
agreed that the wireless sensing of indoor conditions will be inevitable, promoting more
localized and personalized control of indoor climates (Ivanovich and Gustavson 1999).
Experts agree that the driving argument for the deployment of wireless sensors will be cost
advantages and the flexibility to relocate thermostats and sensors as the interior building
layout adapts to the organizational changes of the tenants and occupants that require ever
changing space requirements. While the mobility of wireless sensors is irrefutable, the cost
of the wireless technology at the current time may still be too high to penetrate the market
more widely.
1
Operated for the U.S. Department of Energy by Battelle Memorial Institute und Contract DE-AC06-
76RL01830.
Information and Electronic Technologies: Promises and Pitfalls - 7.121
For any new technology to penetrate the market place it either must be significantly
less expensive than the existing technology or it must have additional features that provide a
competitive advantage and justify the same cost as the technology to be replaced. While
mobility is the compelling driver for the impressive inroads of wireless technologies in the
communication and computer LAN markets, the need for mobility in building control
applications remains limited. This means that wireless technologies must compete
predominantly on the basis of cost.
This paper discusses the cost aspects for the installation of the wireless sensors in two
very different retrofit applications that are part of two demonstration projects currently
underway at Pacific Northwest National Laboratory (PNNL). These applications were
selected to explore the competitiveness of wireless sensors in a range of typical applications
in which wireless technology may successfully compete. The paper presents a cost
comparison between wireless and wired sensor networks and discusses the key drivers for the
cost competitiveness of wireless technologies in buildings. The paper concludes with a
discussion on future trends of wireless sensing and control applications in buildings.
Before describing the DOE demonstration projects, it is important to understand key
technical features and characteristics of relevant wireless technologies that are currently
available. Therefore, the paper provides an overview of wireless sensor and control products,
followed by a discussion of two demonstration projects.
Existing Commercial Wireless Sensing And Data Acquisition Technology
Wireless communication can be accomplished using any of a number of different
communication schemes and protocols. Selection of these for data acquisition for HVAC
monitoring, diagnostics, and control, today and for the foreseeable future, is likely to be
driven primarily by cost. The following sections provide brief descriptions and assessments
of applicability for three wireless communication technology classes that are differentiated
by their communication modulation and protocols.
Bluetooth
Bluetooth (Official Bluetooth Website 2002; Bhagwat 2001; Bluetooth SIG Inc.
2001) is a royalty-free technology specification for short-range wireless communication
among devices. It uses the 2.4 GHz industrial, scientific, and medical (ISM) radio band,
which is available for license-free use worldwide (FCC, Part 15, 1998). In the U.S. and most
other countries, this band extends from 2400 MHz to 2483.5 MHz. The protocol uses a
frequency-hopping spread spectrum technique, where the radio hops through the 79 channels
in a pseudo-random sequence at a rate of 1600 hops per second. This provides excellent
immunity to interference and contributes to security of the transmissions. The maximum
data rate is 781 kbps. The maximum transmission range for a home environment is 10
meters and for an outside environment can reach 30 meters.
Bluetooth devices can form small ad hoc nets known as piconets. A piconet consists
of up to eight Bluetooth devices. Communication can be extended to more devices by
interconnecting piconets.
The intended purpose of Bluetooth is to provide a universal standard for connecting a
broad set of wireless devices. Bluetooth includes definitions for a set of application-level
7.122
profiles for 13 applications, which are necessary to implement user functions. These include
among others cordless telephone, LAN access, FAX, and serial-cable emulation. The latter
is likely to serve building sensor data acquisition in the near-term. The Bluetooth radio is
intended to be a low-cost device, which will become even lower cost when deployed in
billions of units (which is projected over the next 5 years).
IEEE 802.11b
The IEEE 802.11 (IEEE 1997) is a family of standards for wireless local area
networks (LANs) operating in the 2.4 GHz frequency band. Standard IEEE 802.11b (IEEE
1999) is an extension to 802.11 covering wireless LANs transmitting at up to 11 Mbps in the
2.4 GHz band.
IEEE 802.11b devices may connect in ad hoc networks (i.e., networks requiring no
base station) or in infrastructure mode with a fixed access point, which connects to a
stationary LAN. Roaming is provided between multiple access points. LAN connections are
available in some hotels, airports, restaurants, and other locations. Devices using 802.11b
have a maximum range of about 500 meters outdoors at a data rate of 1 Mbps. Maximum
ranges at higher data rates are more typically 100 meters outdoors and about 50 meters
indoors.
The data rates provided by 802.11b far exceed the requirements of most building data
collection needs. As a result, 802.11b-based devices have much larger bandwidth and greater
electrical power consumption than required for wireless data acquisition. Unless the cost of
wireless LAN systems becomes competitive with alternative wireless communication, they
are unlikely to see use for this purpose.
Wireless Serial Communication
Wireless data acquisition for industrial and agricultural applications is currently
provided primarily with serial communication. Communication is at much lower bandwidth
than wireless LAN systems but is generally sufficient for data collection from most sensors.
Data rates range up to 115.3 kbps, although most wireless serial units operate at 19.2 kbps
and lower. A number of different license-free bands are used (some having greater
limitations than others), including 300 MHz, 433 MHz, 900 MHz, and 2.4 GHz. Maximum
transmission distances vary from about 100 feet to many miles (Fern and Tietsworth 1999).
Generally, lower bandwidth and less sophisticated modulation schemes are used to lower
costs when compatible with the installation environment and data transfer needs. In many
cases, a sensor may need to be polled only once every several minutes, with each
transmission requiring only a few bits; therefore significant cost reductions can be achieved
by matching the wireless technology used with the specific application.
Selected Commercially Available Technologies
Table 1 provides representative characteristics and costs for selected wireless data
acquisition systems. The selection is based on a preliminary survey of appropriate wireless
technologies. It is not an exhaustive survey. For the serial communication products of Table
1, maximum communication distance is a significant cost variable. Products with lower
Information and Electronic Technologies: Promises and Pitfalls - 7.123
communication ranges are significantly less expensive than those that transmit over several
miles (No. 3 and 4). The Bluetooth product listed in Table 1, represents currently available
products. The cost target for a Bluetooth radio chip is significantly lower when mass-
produced for consumer products.
Table 1. Characteristics of Selected Commercially Available Wireless Technology
No
Freq.
Band
[MHz]
Com-
munication
Standard
Maximum
Com-
munication
Distance
Power Source
Point-to-
Pont or
Point-to-
Multi-point
Cost
1 433 Not known Approximately
200 ft.
Transmitter: 24 VAC
Receiver: DC power
supply connected to
120 VAC
Point-to-
multi-point
Transmitters:
$300
Receiver: $600
2 900 Serial
(FHSS)
2500 ft open
field
Transmitter: 2/3 A size
LiMnO
2
(for example
Duracell DL123A)
Receiver: 24 VAC
Point-to-
multi-point
Transmitter with
air temperature
sensor: $100
Repeater: $375
Receiver: $450
3 900 Serial
(DSSS)
15 miles line
of sight
11-25 VDC Point-to-
point and
point-to-
multi-point
Transmitter:
$1428
Point-to-point
bridge: $995
Point-to-multi-
point: $1995
4 900 Serial 35 miles line
of sight
10.5 to 18.0 VDC Point-to-
multi-point
Transmitter:
$1775
Receiver: $1775
5 2,400 Serial 150 ft line of
sight
10 to 30 VDC Point-to-
point
Transmitter:
$800
Receiver: $800
6 2,400 Bluetooth 30 ft to 320 ft 5 VDC
Transmitter: 5 VDC:
4 AA alkaline batteries
Point-to-
multipoint
Bluetooth enable
wireless
monitoring unit:
$1,795
PCMCIA
Bluetooth radio
card: $395
FHSS: frequency hopping spread spectrum
DSSS: direct sequence spread spectrum techniques
U.S. DOE Demonstration Projects of Wireless Sensors in Buildings
To investigate the performance and cost of wireless sensor and control technologies
in buildings, PNNL is conducting two demonstration projects. The first project focuses on a
wireless temperature sensor network in a 70,000 ft
2
office building with a heavy steel-
concrete structure and a central plant HVAC system. A total of 30 zone temperature sensors
are networked and integrated into the existing Johnson Controls HVAC and lighting control
network. The temperature data provide input for a chilled-water reset algorithm designed for
the reduction of peak demand and overall electric energy. This demonstration is typical for
an in-building retrofit application to enhance zone temperature control for improved thermal
comfort and overall HVAC system efficiency. The heavy steel-concrete structure is a
7.124
difficult environment for radio frequency transmission. We chose it to explore the effect of
high attenuation on the performance and cost of wireless technology.
The second project focuses on small commercial buildings with rooftop units. The
wireless technology communicates system conditions from individual packaged units to a
central station located on the roof for overall HVAC system diagnostics. The results of the
diagnostics will then be communicated wirelessly to an Internet service provider for viewing
of the results or alarm notification via email or other notification.
In-Building Central Plant Retrofit Application
The demonstration building is a heavy steel-concrete office building with a total floor
area of about 70,000 ft
2
distributed over three floors. It is located on the campus of PNNL.
The HVAC system consists of a central cooling, boiler, and ventilation system with 100
variable-air-volume (VAV) boxes distributed in the ceiling throughout the building. The
central energy management and control system (EMCS) controls the central plant and the
lighting system. Zone temperature control is performed by means of stand-alone and non-
programmable thermostats controlling individual VAV boxes. The centralized control
system lacks zone temperature information and control of the VAV boxes. The long-term
goal of PNNL facility management is to network the 100 VAV boxes into the central control
infrastructure to improve controllability of the indoor environment. As an intermediate step
toward this end, a wireless temperature sensor network with 30 temperature transmitters was
installed to provide zone air temperature information to the EMCS. The wireless temperature
sensor network consists of a series of Inovonics wireless products including a beta version of
an integration module that interfaces to a Johnson Controls N2 network bus
2
. The zone air
temperatures are then used as input for a chilled-water reset algorithm designed to improve
the energy efficiency of the centrifugal chiller under part-load conditions and reduce the
building’s peak demand.
Description of the Wireless Temperature Sensor Network
The wireless network consists of a commercially available wireless temperature
sensor solution from Inovonics Wireless Corporation. It encompasses 30 temperature
transmitters, 3 repeaters, 1 receiver, and a beta version of the ”Translator,” Inovonics’ new
product for the integration of their wireless temperature sensors into Johnson Controls N2
networks. The layout of the wireless temperature network is shown in Figure 1.
The operating frequency of the wireless network is 902 to 928 MHz, which requires
no license per FCC Part 15 Certification (FCC Part 15, 1998). The technology employs
spread spectrum frequency hopping techniques to enhance the robustness and reliability of
the transmission. The transmitter has an open field range of 2500 feet and is battery-powered
with a standard 123 size 3-volt LiMnO
2
battery with a nominal capacity of 1400 mAh. The
battery life depends on the rate of transmission that can be specified in the transmitter. The
manufacturer estimates the battery life up to 5 years with a 10-min update rate. The
transmitter has a battery test procedure with a ‘low-battery’ notification via the wireless
network. This feature will alert the facility operator through the EMCS that the useful life of
the battery in a specific transmitter is approaching its end. The repeater is powered by the
120 VAC from the wall outlet with a battery backup. There are three repeaters, one installed
2
N2 bus is Johnson Controls network protocol.
Information and Electronic Technologies: Promises and Pitfalls - 7.125
on each floor. Because the repeater is line powered, the repeater operates at high power and
provides up to 4 miles of open field range. The receiver and the translator are installed in the
mechanical room. The translator connects the receiver with the N2 bus.
Figure 1. Layout of Wireless Sensor Network: The Building Has Three Identical
Floors (Shown Is Only One Floor)
T
T
T
Mechanical Room
North Wing
South Wing
Wireless network
y 30 temp. transmitters (10 per floor)
y 3 repeaters (1 per floor)
y 1 receiver
y 1 translator
Repeater
T T
T
T TT
T
JCI N2 Bus
Receiver
Translator
Design and Installation Considerations of the Wireless Network
Installation of the wireless network requires a radio frequency (RF) survey for the
placement of the repeaters to ensure that the received signal strength is sufficient for robust
operation of the wireless network. The RF surveying is an essential engineering task in the
design of the wireless network topology. Care must be given to the RF survey or the wireless
system may lack robustness in transmission. The signal attenuation in metal-rich indoor
environments caused by metal bookshelves, filing cabinets, or structural elements such as
metal studs or bundles of electric or communication wiring placed in the walls can pose a
significant challenge to achieving robust wireless communication. Background RF noise
emitted from microwave ovens and other sources can also impair the transmission such that
the receiver cannot distinguish noise from the real signal. There is no practical substitute for
RF surveying a building because each building is unique with respect to its RF attenuation
characteristics.
For the 70,000 ft
2
PNNL building, an engineer performed the RF survey in about 4
hours. This provided sufficient time for investigating several scenarios, whereby the metal
bookshelves were placed in the direct pathway between transmitter and receiver. The result
7.126
of the RF survey was a recommendation of three repeaters, one for each floor of the building
(see Figure 1).
Rooftop Unit Application—Small Commercial Building Demonstration
The second part of the DOE wireless project underway at Pacific Northwest National
Laboratory focuses on configuring, testing, evaluating and demonstrating wireless
technology for use with packaged rooftop HVAC units, commonly used on small- and
medium-size commercial buildings. The target application for wireless could system
monitoring, diagnostics, and remote control of packaged units. In the initial phase of this
project, commercially available wireless technology has been characterized, and selected
systems, showing the greatest potential for cost-effective and technically-successful
application, configured for testing.
Application of wireless RF technology to collect data from packaged rooftop HVAC
units relaxes some of the demands imposed by in-building applications of wireless
communication. Equipment can be physically located so direct lines of sight are preserved
and obstructions minimized. By simply positioning antennas sufficiently above the roof, all
transmitting antennas can “see” their corresponding receiving antenna. If this is not possible
and reliable communication cannot be established, repeaters can be used to extend
communication distances and improve the reliability of communication. As a consequence,
lower transmission power can be used, greater sources of interference can be tolerated, and
communication protocols with less sophisticated means for ensuring reliable data
transmission can be used. As a result, system and component costs are likely to be lower for
rooftop wireless data acquisition than for in-building systems.
A representative wireless data acquisition system (WDAS) for rooftop units is shown
schematically in Figure 2. A wireless data acquisition system may serve many individual
packaged HVAC units. Equipment on the HVAC unit includes: 1) sensors, 2) signal
conditioners for the sensor signals, 3) at least one transmitter, and 4) an antenna for each
transmitter. Sensors are selected based on the data needs for planned monitoring,
diagnostics, or control. For example, diagnostic monitoring of outdoor-air control and
economizing might be performed using measurements of outdoor-air, return-air, and mixed-
air temperatures plus a signal indicating the on/off status of the unit’s supply fan (see, for
example, Brambley et al. 1998; Katipamula et al. 1999). Several sensors might be matched
with one signal conditioner or several signal conditioners with one transmitter to minimize
hardware costs. Connections between the sensors and signal conditioning hardware within
the HVAC unit are likely to be wired because equipment costs today are too high to permit
cost-effective RF transmission from each individual sensor. In the future, transmitters may
be packaged as part of individual sensors, but such equipment is not available commercially
today. Electrical power for the data collection equipment can generally be provided at the
packaged unit by tapping into the electrical power supplied for operation of the HVAC unit.
Receiving and data processing equipment can be located at a central location (e.g., on
the rooftop). The antenna for the receiver is located with as direct a line of sight to the
transmitting antennas as possible. Data is transferred from the receiver to a computer for
processing via a suitable communication protocol. This might be RS-232 serial, RS-485,
USB, Ethernet, or other protocol. The selection depends on the capabilities of the receiver
unit and the computer and can be selected to minimize the cost of the components. This
Information and Electronic Technologies: Promises and Pitfalls - 7.127
equipment must be located near a source of power, which is usually available on
commercial-building rooftops.
Figure 2. Schematic Diagram of a Representative Wireless Data Acquisition System for
Packaged HVAC Units
Display
Connector
Parallel and
Serial Ports
Low-cost, Capability-
match Personal
Computer w/o Display
Diagnostic
Processing (e.g., by the
WBD Software)
Data Acquisition
Receiver
Ethernet and Modem
Connections
Data Storage
Data
Repeaters (as needed)
Antenna
Connection to
Intranet or Internet
or phone line or
wireless modem
Data Transmitter for HVAC Unit 1
Sensor
1
Sensor
2
Sensor
3
Sensor
N1
Signal
Conditioner
Signal
Conditioner
Sensor
1
Sensor
2
Sensor
3
Sensor
N2
Signal
Conditioner
Signal
Conditioner
Data Transmitter for HVAC Unit 2
Sensor
1
Sensor
2
Sensor
3
Sensor
Nm
Signal
Conditioner
Signal
Conditioner
Data Transmitter for HVAC Unit M
Sensors for HVAC
Rooftop Unit 1
Sensors for HVAC
Rooftop Unit 2
Sensors for HVAC
Rooftop Unit M
AntennaAntennaAntenna
The computer can be co-located with the receiving unit or located separately in the
building. When located on a rooftop, no monitor is required. A handheld device, laptop
computer, or a ruggedized LCD monitor can be used temporarily as a user interface during
installation and maintenance. Data storage (disk space), processing (CPU), and
communication capabilities (motherboard and ancillary boards) should be selected to meet
the specific needs of data processing software installed on this computer. Results of
processing (e.g., diagnostic results in text, tables, or graphics) can be made available in the
building or at remote locations using a connection to an intranet or the Internet via direct
wired, wireless LAN, wireless Internet, or dial-up connections. The DOE demonstration
project currently uses a wired LAN connection for communication to staff located in the
building. Plans call for use of a combination of wired LAN, wireless LAN, and wireless
Internet connection for communication locally and remotely later in the project.
Wireless data collection systems of this general architecture have been configured
and are being tested, first in a laboratory, then in field applications. Target buildings include
a small leased office building occupied by PNNL in Washington State and a commercial
office building and two fast-food restaurants in northern California.
7.128
Cost-Effectiveness: In-Building Temperature Sensor Example
For the cost comparison, we considered a wired system design with in-plenum wiring.
The cumulative wiring distances for all temperature sensors are about 3000 feet with the
majority of wiring being loose in-plenum. Assumed are 18 AWG cable for sensor
connections at an approximate cost of $0.07/ft and a labor cost of $1.53 per linear foot of
wiring (RS Means 2001). The cost comparison is shown in Table 2 below.
Table 2. Cost Comparison Between Wired and Wireless Designs for In-Building
Temperatures Sensors
Wired Design Wireless Design
Qty
Cost per Unit Total Cost Cost per Unit Total Cost
Temperature sensor 30 $60 $1,800 $100 $3000
Wiring 3000 ft $1.6 per lin. ft $4,800
Wireless network gear
inc. repeater, receiver,
translator $2475
RF surveying 4 hours $100 $400
Wireless network
configuration 4 hours $100 $400
Total cost $6,600 $6,275
Cost per sensor $220 $209
The cost for the wireless system includes an assumed installer mark-up of 50%. For
the RF surveying and RF installation we estimated the labor rate of an engineer at $100 per
hour. Omitted in the cost comparison are the costs for the sensor configuration in the
Johnson Controls network, which are assumed to be similar, if not equal for both the wired
and the wireless designs. For simplicity, the labor cost for battery change-out, expected to
occur every 5 years, is not included in Table 2. This activity can be estimated at about $200,
assuming a battery cost of $3 per battery and 2 hours of labor for replacing 30 batteries.
The wireless system for this in-building temperature sensor application is about 5%
less expensive than a wired solution. It should be noted that the estimates in Table 2 have
considerable uncertainties introduced by the assumptions of the installer mark-up for the
wireless system and the wiring cost for the layout of the demonstration building. The results
of Table 2 suggest that the wireless system can be a cost-effective solution. Due to
uncertainties in the cost estimates, the wireless system may range from being cost-effective
to marginally cost-effective and potentially slightly more expensive. One of the advantages
of the wireless network is that it can be easily extended with additional temperature sensors
at an incremental cost of a temperature transmitter, up to 100 transmitters. Installations with
more than 100 temperature sensors require additional receivers and translators.
It should also be mentioned that the transmission rates of wired sensors are much
higher than those for the wireless sensors. Typically in conventional control networks, zone
temperature sensors are polled by the EMCS or a control device every 1 or 2 seconds. The
wireless network installed at the PNNL building updates its temperature every 10 minutes.
The user can define the update rate of the temperature transmitter. The penalty of a higher
update rate is a higher power consumption and, hence, a reduced life-time of the battery.
Thus, battery-operated wireless sensors are not suitable for closed-loop control circuits as
employed in the economizer or air-handler control loops. The typical polling rate in these
Information and Electronic Technologies: Promises and Pitfalls - 7.129
closed-loop applications is less than 1 second. For zone temperature control, however, the
time constant for temperature changes is generally 30 minutes or longer. A 10-minute update
frequency is therefore, sufficient and acceptable.
Cost-Effectiveness: Rooftop-Unit Data Acquisition Example
To compare costs of current technology for wired and wireless data acquisition
systems for rooftop packaged HVAC units, consider the situation shown in Figure 3. Three
rooftop units separated by the indicated distances are shown. For each unit, four sensors are
installed: four temperature sensors (for outside air, return air, mixed air, and supply air) and
one indicator of the on/off status of the supply fan. These measurements are sufficient to
perform diagnostics (or even control) of outside-air control and air-side economizing based
on dry-bulb temperature. Other sensors might be installed for other purposes and increase
the total cost of the system but not make a difference in communication costs between wired
and wireless systems.
Table 3 shows the system costs for a wired base case and two wireless systems
configured from commercially available components—low and high cost. The low-cost
wireless system corresponds to the technology listed as number 1 in Table 1. The high-cost
system corresponds to technology listed as number 5. Key cost differences between the
wired system and the wireless systems are attributable to the communication components.
For the wired case, cable and conduit must be installed to each HVAC unit.
Figure 3. Baseline Wired Data Collection System
For the wireless case, the cable and conduit are replaced with RF transmitters and
receivers. The results show that low-cost wireless data collection has cost advantages over
wired data collection. The high-cost wireless solution is not cost competitive with wired data
collection. These results apply, however, only to the configuration shown in Figure 3.
Shorter cable runs increase the cost advantage for wired data collection. Conversely, longer
Computer
HVAC
#2
HVAC
#3
HVAC
#1
57 ft.
32 ft.
15 ft.
Twisted 1 pair shielded
wire with ½” conduit
Each HVAC unit has 4 thermocouple
sensors ($43.00 each), a current switch
($40.00), 8 channel thermocouple input,
and a digital I/O module.
RS-232 to RS-485 converter
Computer
HVAC
#2
HVAC
#3
HVAC
#1
57 ft.
32 ft.
15 ft.15 ft.
Twisted 1 pair shielded
wire with ½” conduit
Each HVAC unit has 4 thermocouple
sensors ($43.00 each), a current switch
($40.00), 8 channel thermocouple input,
and a digital I/O module.
RS-232 to RS-485 converter
7.130
cable runs (greater distances from the data collection point to the HVAC units) increase the
cost advantage for wireless systems, up to the point where one or more repeaters are
required.
Greater numbers of HVAC units generally will improve the cost-effectiveness of
wireless data acquisition because total linear feet of distances from a central point to the
HVAC units will generally increase. In all cases, the low-cost wireless solutions have a cost
advantage. The Pacific Northwest National Laboratory is testing wireless systems to
determine the limits on technical performance in typical rooftop environments; results of
performance testing will be presented in future publications.
Table 3. Results of Cost Analysis for Rooftop Units
Wired System Low-Cost Wireless
1
High-Cost Wireless
2
Description Quantity Cost Quantity Cost Quantity Cost
Thermocouple Sensors 12 $516 12 $516
Current Switch 3 $120 3 $120
RS-232 Converter 1 $799
Thermocouple Signal Conditioner
($239 each)
3 $717
Digital I/O module ($129 each) 3 $387 $387
Twister pair wiring 104 ft $13
½” Conduit 104 ft $55
Labor for installing sensors (3 hr per
unit)
9 hrs $450 9 hrs $450 9 hrs $450
Labor for installing wire and conduit
(at $7 per ft)
104 ft $729
R.F. transmitting units with sensors and
signal conditioners
3 $900
R.F. receiver unit 1 $600
2.4 GHz wireless radio modem ($800
each)
6 $4800
Thermocouple input transmitter ($239
each)
$717
Total Cost $378
5
$1950 $7000
1
corresponds to technology number 1 in Table 1
2
corresponds to technology number 5 in Table 1
Cost Comparison of Wireless Versus Wired System for Retrofit and New
Construction Applications
The cost-effectiveness of wireless sensor systems in buildings with respect to wired
systems depends on many factors. We define the cost effectiveness as the ratio of capital
cost for a wireless system over the capital cost of a wired system (Cost
wireless
/Cost
wired
). A
ratio of less than unity indicates that wireless technology is more cost effective. Of interest is
only the cost associated with the transport of a signal from point A to B over a given distance
because other components (sensors, controllers, and actuators) are common to both wired and
wireless systems. The cost of the wired system depends primarily on two key factors: 1) on
the degree of difficulty to route the wires and code requirements prescribing shielding and
wire support and 2) on the distance. For simplicity, we neglect the effect of different wire
material. In general, the installation of wiring in new construction is less difficult because of
Information and Electronic Technologies: Promises and Pitfalls - 7.131
the relatively easy accessibility to routing channels. As a consequence, we assume the wiring
cost to be lower for new construction than when performed as a retrofit measure.
The key drivers for the cost of wireless
systems are the signal attenuation and signal
to noise ratio for the transmission. In general, we find that the higher the attenuation in a
building, the more repeaters that are required. The cost model for the wireless system is used
in this analysis corresponds to the second serial wireless technology shown in Table 1. In
addition we estimated cost for the integration into a wireless system of $500.
The cost effectiveness ratio (Cost
wireless
/Cost
wired
) is then a function of distance,
installation type (retrofit versus new construction), and number of repeaters. Figure 4 shows
this relation. Consider the points A, B, C, and D in Figure 4 representing different cost ratios
at a constant distance of 3000 ft for the wiring. For the retrofit example, we establish a
wiring cost of $6,600, assuming a cost per linear foot of $2.2 including wires. For new
construction, we assumed a reduced wiring cost (because of easier access) in the amount of
$2,010 for a cost of $0.67 per linear foot. We assumed that wiring conduits already exist and
thus, the wiring cost excludes the cost associated with installing conduits. Point A (cost
ratio=0.3) represents the cost competitiveness of a wireless system in a retrofit case with no
repeater necessary. Point B (cost ratio=0.9) represents the cost for a building with high
attenuation characteristics, requiring 10 repeaters. Corresponding for new construction are
points C (cost ratio=1.0) and point D (cost ratio=2.9).
Figure 4. Competitiveness of Wireless Sensors and Data Acquisition Systems
Compared to Wired Systems
Competitiveness of Wireless Sensors Network in Retrofit and
New Construction
-
0.50
1.00
1.50
2.00
2.50
3.00
200
600
1,000
1,400
1,800
2,200
2,600
3,000
3,400
3,800
4,200
4,600
5,000
Linear Feet of Wiring
Cost Ratio
[Cost
wireless
/Cost
wired
]
New Construction
Retrofit
C
D
B
A
no repeater
10 repeaters
no repeater
10 repeaters
While the cost-effectiveness analysis is simplified, it illustrates the sensitivity of the
key drivers for wireless technologies in HVAC applications. It indicates that the early
adopters of this technology will implement wireless devises most likely in existing buildings
7.132
that do not pose difficulty in transmission of the RF signal. Likely applications are rooftop
connectivity with line-of-sight transmission or applications in light construction that do not
require repeater devices. Wireless technologies in new construction are not yet commonly
competitive. Today’s wireless technologies are still expensive. Solely battery-operated
wireless sensors currently do not achieve the performance of wired sensors with respect to
update frequencies. To gain significant market shares, wireless technology would need to
compete in both the retrofit as well as in the new construction markets. With lower cost of
wireless technology and available integration products for interconnecting wireless with
wired systems, wireless technologies may become an attractive solution for HVAC control
networks coexisting and augmenting wired systems.
Future Trends
While the mobility feature in conventional commercial HVAC control applications
may remain limited, at least for the short-term, the cost avoidance for wiring will most likely
be the key selling point of the wireless technology. The first adoption of wireless technology
is expected to occur in retrofit applications where the technology extends existing wired
control networks to places where there is are no control networks cables. This includes, for
instances, opportunities for one-way or two-way connectivity among packaged rooftop units
with line-of-sight transmission, permanent or temporary indoor air monitoring, monitoring of
remote equipment (e.g., water pumps, cooling intake valves), and control of outdoor lighting.
The first wireless installations are expected to be monitoring applications that are not time
critical and require only one-way communications. Control applications will initially be
limited to open-loop control function, such as turning equipment on or off. Wireless closed-
loop control applications that require higher update frequencies (less than one second) pose
higher transmission robustness requirements are therefore, expected to mature later. As
wireless technology gains inroads into the HVAC controls markets and sales volumes
increase, cost reductions of the technology will follow. Primary driver of the cost reduction
will be design and manufacturing optimization of RF technology components and further
integration of sensing, signal conditioning and RF communication modules that can be mass
manufactured.
Technological challenges for closed-loop control applications with high update
frequency requirements still remain for battery-operated devices requiring technological
advancements in power management, ultra-low power electronics, and utilization of ambient
power sources and power scavenging.
As with the advent of television, when many feared that it will replace radio
broadcasting, so it is unlikely that wireless technology will replace the entire wired HVAC
controls market. A more likely scenario is that it will complement the conventional wired
controls technology where it makes economic sense. Significant reductions in cost for
wireless sensing will lead to greater use of sensors in building application, which in turn will
lead to better control and maintenance of systems that will improve the overall energy
efficiency of the existing building stock and provide healthier and more productive
workplaces.
Toward this end, the projects described in this paper will identify technology gaps in
order to develop a research, development, and demonstration agenda that will bridge the
technology and cost gaps and demonstrate the value of wireless technology in building
monitoring, operation and maintenance.
Information and Electronic Technologies: Promises and Pitfalls - 7.133
Acknowledgement
The work described in this paper was sponsored by the Office of Building
Technology, State and Community Programs, U. S. Department of Energy as part of the
Building Systems Program at PNNL.
Reference
Bhagwat, P. 2001. “Bluetooth: Technology for Short-Range Wireless Apps.” IEEE
Internet Computing, Vol. 5, No. 3, May-June 2001, pp. 96-103.
Bluetooth SIG Inc. 2001. Specification of the Bluetooth System-Core. Version 1.1,
February 22, 2001. Available online at
http://www.bluetooth.com/pdf/Bluetooth_11_Specifications_Book.pdf.
Brambley, Michael R., Robert G. Pratt, David P. Chassin, and Srinivas Katipamula. 1998.
“Diagnostics for Outdoor Air Ventilation and Economizers.” ASHRAE Journal, Vol.
40, No. 10, pp. 49-55, October 1998.
FCC, Part 15, 1998. Part 15 Radio Frequency Devices. Code of Federal Regulation 47 CFR
Ch. I (10–1–98 Edition), Federal Communications Commission, Washington, D.C.
Fern, David G. and Steven C. Tietsworth. 1999. “Automatic Wireless Communications.”
Sensors, Vol. 16, no. 9 (September 1999).
IEEE 802.11-1997. Standard for Information Technology, Telecommunications and
Information Exchange Between Systems, Local and Metropolitan Area Networks,
Specific Requirements, Part 11: Wireless LAN Medium Access Control (MAC) and
Physical Layer (PHY) Specifications: Higher Speed Physical Layer Extension in the
2.4 GHz Band, Institute of Electrical and Electronic Engineers.
IEEE 802.11-1999. Supplement to Standard for Information Technology--
Telecommunications and Information Exchange Between Systems-- Local and
Metropolitan Area Networks, Specific Requirements--Part 11: Wireless LAN Medium
Access Control (MAC) and Physical Layer (PHY) Specifications: Higher Speed
Physical Layer Extension in the 2.4 GHz Band, Institute of Electrical and Electronic
Engineers.
Ivanovich, M. and D. Gustavson. 1999. "The Future of Intelligent Buildings is Now."
Heating/Piping/Air Conditioning Engineering, May 1999, pp. 73-79.
Katipamula, Srinivas, R. G. Pratt, D. P. Chassin, Z. T. Taylor, K. Gowri, and M. R.
Brambley. 1999. “Automated Fault Detection and Diagnostics for Outdoor-Air
Ventilation Systems and Economizers: Methodology and Results from Field Testing,”
ASHRAE Transactions, Vol. 105, Pt. 1.
The Official Bluetooth Website. 2002. http://www.bluetooth.com/index.asp.
RS Means 2001. Electrical Cost Data 2001. 24
th
Annual Edition, Wiring Methods, 500-
7400. RS Means Company, Inc., Kingston, MA.
7.134
