Product Testing Report

Duct - Mounted Relative Humidity Transmit ters

April 2004

Table of Contents Introduction

1 Introduction Humidity transmitters are common components
in building control systems and their performance
2 Classification of Humidity Sensors can significantly impact energy use in heating,
How is Humidity Measured?
ventilating, and air-conditioning (HVAC) systems.
Types of Sensors
In particular, relative humidity transmitters are
Capacitive Humidity Sensors
used to monitor supply and return air conditions
Resistive Humidity Sensors from air handling units, monitor conditions in
occupied spaces, and control humidification
4 Choosing the Right and dehumidification processes as well as
economizer cycles. In the case of economizers,
Humidity Transmitter relative humidity and temperature measurements
Accuracy of outdoor and return air conditions are used to
Response Time calculate the enthalpies of the two air streams.
Resistance to Contaminants The air stream with the least energy content is
Transducer Performance then selected to provide building cooling. If one
Maintenance and Calibration or both of the computed enthalpies is wrong, as
Long-term Stability can happen when humidity transmitters are not
Humidity Transmitter Cost accurate, significant energy penalties can result
from cooling of the incorrect air stream.
6 NBCIP Testing
Manufacturer Information Growing concern about indoor environmental
Testing Methods quality issues, including mold, mildew and
moisture problems also emphasize the need for
10 Test Results accurate humidity sensing devices. In many
Accuracy cases, humidity control is the primary means of
Repeatability system control in laboratories, humidity sensitive
Linearity spaces, museums, and computer rooms. The
Hysteresis consequences of poor humidity transmitter
performance in these applications can outweigh
15 Conclusions energy concerns and result in costly damage.
Next Steps
Because of their impact on energy use and the
16 Terms and Definitions quality of the indoor environment, the National
Building Controls Information Program (NBCIP)
tested and evaluated the performance of resistive
Program Sponsors and capacitive duct-mounted relative humidity
transmitters used in typical building HVAC
Iowa Energy Center applications. NBCIP developed an experimental
NSTAR Electric and Gas Corporation procedure for testing the humidity transmitters,
U.S. Environmental Protection Agency and used this procedure to test products from six
leading manufacturers.

NBCIP 1
This NBCIP Product Testing Report provides
an overview of factors to consider when
Measuring Humidity
purchasing a humidity transmitter, presents Relative Humidity
manufacturer data for transmitters selected for Relative humidity is the ratio of the amount of water vapor
testing, describes the test procedure and test in air to the maximum amount of water vapor that the
apparatus used by NBCIP to evaluate transmitter air can hold at a given temperature and pressure.
performance, and presents test results for It is expressed in percentage and can be determined
accuracy, repeatability, linearity, and hysteresis as follows:
for each humidity transmitter model tested. P v x 100
%RH =
Ps
Throughout this report, the humidity sensing
element is referred to as the sensor. A where: Pv: Actual partial pressure of water vapor
humidity transmitter is a device that consists of in moist air (bar; kPa)
a sensor as well as a transducer that converts Ps: Saturation pressure of water vapor at
the sensor reading into an output signal. the same given temperature (bar; kPa)

Air is saturated at 100% relative humidity, which means

it cannot hold any more water. Raising the temperature
Classification of without changing the amount of water in the air reduces
Humidity Sensors the relative humidity. The relative humidity decreases
because warmer air can hold more water than cold air. At
The selection of the most appropriate 100% relative humidity, the dry bulb, wet bulb, and dew
humidity sensor for a particular application point temperatures are equal.
requires knowledge of the parameter to be
measured, how the sensor works, and its Specific Humidity
operating characteristics. Specific humidity is the mass of water vapor in air per unit
mass of dry air. It is usually expressed as lb (kg) of water
How Humidity is Measured per lb (kg) of dry air or grains of water per lb of dry air
Water vapor in air or any other gas is termed (7000 grains = 1 lb):
humidity, while water content in liquids and M
W= w
solids is usually designated as moisture. Mg
Different approaches to measuring humidity where: Mw: mass of water vapor (lb; kg; grains)
are based on quantifying different physical Mg: mass of the dry air (lb; kg)
properties of the mixture of water vapor and
air, which is referred to as moist air. Humidity Specific humidity is sometimes referred to as the
can be expressed in a variety of forms with the humidity ratio.
most common forms being percent relative
humidity (RH), specific humidity, absolute Absolute Humidity
humidity, dew point and vapor pressure. Refer The absolute humidity (i.e., vapor concentration or vapor
to the side bar Measuring Humidity for a density) is the mass of water vapor per unit volume of
description of each of these parameters. moist air. It is represented as follows:

Types of Sensors AB = M w
The most widely used humidity transmitters V
in HVAC control applications measure relative where: AB: Absolute humidity
humidity and utilize capacitive and resistive (grains/ft 3 ; grams/m3)
sensors. These low-cost transmitters consist of Mw: Mass of water vapor (grains; grams)
an integrated sensor and transducer assembly. V: Volume of moist air (ft3; m3)
The sensor provides a measure of the relative
humidity while the transducer generates an Dew Point
electronic output signal representative of the Dew point is the temperature at which water begins
sensed humidity. Both capacitive and resistive to condense out of the air when cooled at constant
humidity transmitters range in accuracy from ±5% pressure. When air is cooled to its dew point, it reaches
to ±1% for HVAC applications. The corresponding 100% relative humidity and the water vapor is in a state
cost range is $50 to $1000 for the capacitive type of saturation. Additional cooling of air causes some of
and $100 to $1500 for the resistive type. the water vapor in the air to change to the liquid phase.

NBCIP 2
NBCIP 2
This NBCIP Product Testing Report addresses (Continued from page 2)
the performance of resistive and capacitive Liquid water molecules accumulate, droplets form, and
transmitters only. Information on other humidity water condenses out of the air. At this new condition,
sensing technologies can be found in the ASHRAE the air contains less water vapor, has a lower specific
Handbook of Fundamentals (ASHRAE 2001). humidity and a lower dew point temperature; however, it
is still at 100% relative humidity.
Capacitive Humidity Sensors
The main components of a capacitive humidity Vapor Pressure
sensor are shown in Figure 1. A capacitor is The partial pressure of water vapor is that part of the
formed by depositing a polymer or metal oxide total pressure contributed by water vapor, and is also the
film between a conductive material (lower pressure exerted by free water molecules at the surface
electrode) and a porous conductive material of a solid or liquid. For most HVAC processes, the vapor
(upper electrode) onto a glass, ceramic, or silicon pressure of interest is that of water in contact with itself,
substrate. The polymer layer adsorbs water which is the saturation pressure normally found in the
molecules as they permeate through the porous thermodynamic tables. However, water in contact with
upper electrode. The dielectric constant of the other substances (e.g., wood, paper, salt) also has a vapor
polymer layer changes as it adsorbs moisture, pressure, which may be different from the vapor pressure
causing the capacitance of the two electrodes to of water in contact with itself.
increase. The change in capacitance is directly
proportional to the relative humidity.

a conductive polymer or a salt. As the polymer

Upper electrode coating adsorbs moisture, ions are released
causing the electrical resistance to change. The
Polymer film resistance decreases as humidity increases.
This change in electrical resistance of the
Lower electrode polymer is measured by the sensor.
Substrate
Resistive humidity sensors are generally
considered economical to manufacture, have
Figure 1. long-term stability, are operational over a
Schematic of a capacitive-type humidity sensor. broad humidity range (15 to 90% RH), perform
(Adapted from Yamatake) well at high humidity, and are resistant to
surface contaminants. However, resistive
Capacitive humidity sensors are generally sensors are considered to be less accurate
considered to be accurate at low humidity (less at low humidity (less than 15% RH), have
than 15% RH) and high ambient temperatures. slow response times (on the order of tens of
High quality capacitive sensors have fast response seconds to minutes) because the moisture
times at low humidity, on the order of seconds. must fully permeate the conductive polymer
The main limitation of capacitive sensors is the layer before the resistance reading is affected,
small change in capacitance compared to the and are sensitive to chemicals that are similar
capacitance of the cable connecting the sensor. to the polymer material (Brownawell 1989).
This requires signal conditioning to occur close Furthermore, polymer based sensors have a
to the sensor, usually at distances within 10 feet. strong temperature dependence, requiring units
Also, capacitive humidity sensors are sensitive
to contaminants and chemicals, are generally
considered to be inaccurate at high humidity Conductive
(over 85% RH) due to saturation of the sensing polymer film
material, and require periodic recalibration.

Resistive Humidity Sensors Metal electrodes

The main components of a resistive humidity
Substrate
sensor are shown in Figure 2. Resistive
humidity sensors are composed of interlocked
metal electrodes that are deposited on a Figure 2.
substrate. The substrate is then coated Schematic of a resistive-type relative humidity sensor.
with a moisture-sensitive material, such as (Adapted from Yamatake)

NBCIP 3
to be temperature compensated. The use of a Selecting a humidity transmitter based solely
water-soluble coating causes resistive sensors on manufacturer stated accuracy provides
to be less accurate if condensation occurs. only a partial view of transmitter performance.
Recent advances in sensor technology address Transmitters that are calibrated at laboratory
condensation issues by using a ceramic substrate controlled temperature and humidity conditions
coated with a polymer/ceramic mixture. may not reflect actual field conditions to which
the transmitter will be subjected. It is therefore
important to know the calibration standard
that was used, the temperature and humidity
Choosing the Right condition(s) at which the transmitter was
Humidity Transmitter calibrated, how transmitter performance is
affected by varying temperature and humidity,
Specifying or selecting a humidity transmitter and how ageing of the sensor will affect
is dictated by the requirements of its intended its performance.
application. Transmitter selection requires
careful consideration of several factors and how Most manufacturers calibrate and report
important they are to the application. These transmitter accuracy at or around 77°F (25°C).
factors include: All humidity transmitters are temperature
dependent, and transmitter performance can
• Accuracy, including repeatability, linearity, be affected when subjected to extremes in
and hysteresis temperature. Temperature compensation can
• Environmental conditions to which the be specified when the application requires
transmitter will be exposed a higher degree of accuracy or operation at
• Response time high or low temperatures. Two temperature
• Resistance to contaminants compensation options are possible: electronic
• Transducer performance compensation or control of sensor temperature.
• Maintenance and calibration requirements
• Long-term stability Extreme humidity conditions can also affect
• Transmitter cost transmitter performance. At high humidity
levels (e.g., greater than 90% RH) even a small
Accuracy temperature change can result in condensation
Accuracy is an important parameter in the on the sensor’s surface. Exposure to high
selection of humidity transmitters; however, humidity conditions could degrade sensor
there is usually a direct relationship between accuracy. Manufacturers suggest several
accuracy and cost. It is important to note that methods to prevent condensation, including
a particular application may not necessarily minimizing temperature variations by ensuring
require the most accurate and thus the most good air mixing, warming the sensor so that the
expensive humidity transmitter. In applications probe temperature remains above the dew point
that require strict humidity control, such as temperature, or utilizing a protective coating
humidity-sensitive spaces, museums, and that prevents condensation from occurring on
laboratories, a humidity transmitter with ±1 the sensor. Furthermore, advances in sensor
or ±2% RH accuracy may be appropriate. materials make it possible to select humidity
Otherwise, a ±5% RH accuracy level may be sensors that are specifically designed to
sufficient for most HVAC applications that are operate in extreme humidity environments.
non-energy critical, depending on the mounting
configuration of the transmitter. For example, The term accuracy also encompasses
space or wall-mounted humidity transmitters other measures of performance such as
that are used simply to monitor relative hysteresis, linearity and repeatability. In some
humidity conditions in enclosed spaces may be cases, the accuracy stated in manufacturer
specified at ±5% RH accuracy. However, if the product literature includes some or all of
RH measurements are used for HVAC control, these performance measures, while some
such as in a chilled water reset strategy, then an manufacturers report these parameters
accuracy of ±3% RH is recommended. separately. Consideration of all these
Duct-mounted humidity transmitters are parameters will provide a clearer understanding
exposed to high airflow conditions, associated of humidity transmitter behavior and will aid in
with high energy content, justifying sensor selecting the most appropriate transmitter for
accuracy of ±3% RH. a given application.

NBCIP 4
Response Time manufacturers to purchase the sensing element,
Many industrial processes require humidity design and fabricate the transducer, and package
sensors with fast response times, on the these components as a relative humidity
order of seconds. In contrast, most building transmitter. It is important to recognize that the
air-conditioning processes have response transducer plays a critical role in determining the
times on the order of minutes, and therefore, accuracy, noise levels and other performance
slower humidity sensor response times can be characteristics of relative humidity transmitters.
acceptable. In general, resistive RH sensors
are considered to have slower response times Maintenance and Calibration
than capacitive sensors, particularly at low Maintenance requirements depend on the type
humidity levels. The response time of resistive of technology used in the sensor, environmental
sensors can range from less than two minutes conditions and exposure to contaminants.
to as much as 15 minutes under conditions of Capacitive and resistive humidity transmitters
low humidity and low air flow. This is because require calibration checks and occasional
water vapor must fully permeate the conductive cleaning. In general, manufacturers recommend
polymer layer before the resistance reading that transmitters be checked and calibrated
is affected. In contrast, capacitive humidity once a year. High accuracy transmitters, as
sensors are considered to have fast response well as transmitters that are subjected to high
times, on the order of seconds, because the temperature and/or humidity conditions or
porous electrode layer quickly adsorbs water harsh environments should be checked and
vapor. However, lower quality capacitive recalibrated at a frequency of every six months
sensors, in which the porous electrode is less or as recommended by the manufacturer.
uniform or contains fewer pores, may have slow
response times, on the order of minutes. Humidity transmitters that operate over a
narrow temperature and humidity range can
Resistance to Contaminants usually be checked and recalibrated at a single
The accuracy, response time, and life condition (single point calibration). In contrast,
expectancy of a humidity sensor can transmitters that operate over a broad range of
be adversely affected by contaminants. conditions need to be checked and recalibrated
Contaminants include dust, particulates, at several temperature and humidity conditions
chemicals, and chemical vapors. Sensor (multipoint calibration).
accuracy can degrade as contaminants
permeate the sensor. A high level of Calibration methods vary, as do their cost
contaminants may permanently damage the and the resulting accuracy of calibration. The
sensor while at lower levels the sensors can most accurate, and most expensive, method
usually be cleaned or reactivated. Bulk polymer involves calibrating the transmitter against
resistive humidity sensors are considered to be a known national standard in an approved
inherently resistant to surface contaminants. laboratory. ASHRAE Standard 41.6, Method
However, while sensor accuracy may remain for Measurement of Moist Air Properties,
unaffected, the response time may increase. identifies three calibration standards: primary,
Filters can be used to keep humidity sensors secondary, and working standards. Primary
free from dust; however, sensor accuracy standards are based on fundamental principles
may be affected. Manufacturer’s contaminant and produce known humidity conditions
compatibility charts should be consulted if the using base units. One application of primary
sensor will be applied in a harsh environment. standards is to calibrate other instruments,
including secondary devices. In the U.S., the
Transducer Performance national primary standard is the gravimetric
The relative humidity transmitters of the hygrometer developed by the National Institute
type tested by NBCIP have two primary of Standards and Technology (NIST). This
components: 1) a sensing element; and 2) a device was selected as the national primary
transducer. The transducer converts the sensed standard because it is extremely precise in
resistance or capacitance to an output voltage its measurements; however, it is difficult,
and may also amplify and filter the signal. In expensive, and labor-intensive to use. NIST
the case of humidity transmitters with a current therefore developed the two-pressure humidity
output, the sensed resistance or capacitance is generator, which is used as a surrogate for the
converted to a current. It is not uncommon for primary standard. Secondary standards are

NBCIP 5
also based on fundamental principles; however,
the accuracy of their results depends heavily
on proper use. Working standards refer to NBCIP Testing
devices that are widely used in commercial
and industrial applications. These include, Manufacturer Information
among others, resistive and capacitive humidity NBCIP consulted trade magazines and HVAC
transmitters. Since the operation of working specifiers’ guides to identify manufacturers of
devices is not based on fundamental principles, duct-mounted relative humidity transmitters.
they must be periodically calibrated against a NBCIP reviewed manufacturers’ product
primary or secondary standard. literature and selected only duct-mounted
relative humidity transmitters having an accuracy
In the U.S., devices whose calibration can be of ±3% RH and 0-10V output signal for testing.
traced back to a primary standard are said to be
NIST-traceable. Factory or laboratory calibrated NBCIP purchased three humidity transmitters of
transmitters are supplied with a calibration the same model from six leading manufacturers
certificate that reports the unit’s accuracy at for a total of 18 transmitters, of which half were
the time of manufacture or shipment. A resistive type sensors and half were capacitive
transmitter that has been calibrated long before type sensors. Technical information for the
it is put in use may be less accurate due to transmitters, obtained from manufacturer
possible ageing. product literature, is reported in Table 1. A
sample of the humidity transmitter models
Calibration can also be carried out on site using that were purchased are shown in Figure 3.
portable devices such as a portable meter, Transmitters were purchased anonymously from
saturated bath solutions, or a portable humidity the manufacturer or authorized distributor in
generator. Portable devices also require three separate batches over a period of several
periodic calibration in an approved laboratory. weeks to increase the possibility that they
Depending on the standard or technology originated from different manufacturing lots.
used, portable calibration devices may prove The transmitters were purchased over a period
to be less accurate than laboratory devices. ranging from March to June 2002. At the
However, they can be useful in performing spot- time of purchase, manufacturers had not been
checks of transmitter performance on site. informed about NBCIP’s intentions to perform
independent tests on the humidity transmitters.
Long-term Stability Manufacturers were contacted at the time of
Long-term stability refers to how transmitter writing this report and were asked to verify the
accuracy drifts over time and determines the correctness of product information presented
frequency of calibration. A stable humidity in Table 1. Manufacturers were not given an
transmitter would require fewer calibrations. opportunity to view the test results prior to
Most of the secondary humidity transmitters, public release of this report.
such as resistive and capacitance types,
require frequent calibration (e.g., every six
months to one year), depending on the use of
the humidity transmitter.

Humidity Transmitter Cost

First costs will vary depending on transmitter
accuracy and the type of application. In addition
to initial cost, the long-term costs, including
sensor replacement and maintenance, must
be considered.

Figure 3.
Sample of duct-mounted humidity transmitters
selected from six manufacturers for NBCIP testing;
from left to right: MAMAC Systems Inc., Building
Automation Products Inc., Johnson Controls
Inc., Vaisala, General Eastern Inc., Automation
Components Inc.

NBCIP 6
 Table 1. Technical information for duct-mounted relative humidity transmitters tested by NBCIP, as reported in manufacturer product literature.
Automation Building Automation General MAMAC
Components Inc. Products Inc. Eastern Inc. Johnson Controls Inc. Systems Inc. Vaisala
Manufacturer (ACI) (BAPI) (GEA) (JCI) (MAMAC) (Vaisala)

Model Number A/RH3-D BA/H310-D MRH-3-D HT-6703-0N00P HU-224-3-VDC HMD50U

Sensing Element Resistive Impedance type Bulk polymer All-polymer Bulk polymer Thin film polymer
Resistive Capactive Capacitive Capacitive
Price Paid by NBCIP ($US)1 $125 $150 $194 $248 $275 $180

Accuracy ±3% RH ±3% RH ±3% RH at 77˚F (25˚C) ±3% RH at 77˚F (25˚C) ±3% RH Better than±3% RH at
20% to 95% RH at 15% to 95% RH 20% to 95% RH 20% to 80% RH 10% to 90% RH 68˚F (20˚C)
± 5% RH at 77˚F (25˚C) 10% to 90% RH
77˚F (25˚C) 4
10% to 20% and 80% to 90% RH

Operating Range -10˚F to 160˚F -20˚F to 160˚F -20˚F to 140˚F 32˚F to 122˚F -30˚F to 130˚F 14˚F to 140˚F
(-23.3˚C to 71˚C) (-28˚ to 70˚C) 6 (-29˚C to 60˚C) (0˚C to 50˚C) (-35˚C to 55˚C) (-10˚C to 60˚C)
0% to 100% RH 0% to 100%RH 0% to 95% RH 0% to 100% RH 10% to 90% RH 0% to 100% RH
Linearity NA ±0.1% of span Included in accuracy NA Included in accuracy Included in accuracy 4

Hysteresis < 0.4% RH < 0.4% RH <1% RH NA ±1% RH Included in accuracy 4

Included in accuracy

Repeatablity 0.5% RH NA 0.5% RH NA Included in accuracy Included in accuracy 4

Included in accuracy
Long Term Stability <2% RH drift over <2% RH drift over <1% RH drift per year NA NA ± 2% RH over 2 years
5 years 5 years 6

Response Time 30 sec. for 63% step 30 sec. for NA Within 5% RH in 15 min. for NA < 20 sec. at 68˚ F (20˚ C) in
63% step 6 10-30%,30-90%, 40-90% RH still air 4

Sensitivity 0.1% RH NA 0.1% RH NA NA NA

Supply Voltage 2 15- 36 VDC for 250 Ohm 24 VDC 12-36 VDC 12-40 VDC 12-35 VDC/12-24 VAC for
14-30 VDC for 0-10 VDC output,
load, 18-36 VDC for 500 24 VAC 20-30 VAC for 0-10 VDC output, 12-35 VAC 0-1 VDC output,
Ohm load5, or 24 VAC 16-30 VDC for 4-20mA output 15-35 VDC/15-24 VAC
for 0-10 VDC output

Output3 0-5 VDC, 0-10 VDC, 0- 10 VDC 0-5 VDC, 0-10 VDC, 0-10 VDC, 0-5 VDC, 0-1 VDC, 0-10 VDC
4-20mA 4-20mA 4-20mA 0-10 VDC Selectable voltage
Jumper select Jumper select Jumper select Jumper select 3-wire output

Storage Temperature -10˚ F to 160˚F (-23.3˚C to -20˚ F to 160˚ F -85˚ F to 158˚ F -40˚F to 176˚ F -30˚F to 130˚F -40˚ F to 140˚F
71˚C) 0% to 100% RH 4 (-28˚C to 70˚C) 6 (-65˚C to 70˚C) (-40˚C to 80˚C) (-35˚C to 55˚C)4 (-40˚C to 60˚C)

Warranty 2 years NA 2 years 3 years 4 2 years 1 year 4

Notes:
NA indicates that the information was not available in the manufacturer’s product literature.
1. Price shown in the table represents actual invoice cost to NBCIP. Your actual cost may differ depending on volume purchased, reseller, trade discounts and other pricing structures.
2. NBCIP tested the duct-mounted humidity transmitters using a supply voltage of 24 VDC.
3. NBCIP tested the duct-mounted humidity transmitters using an output voltage of 0-10 VDC. Other voltage outputs are available as indicated in the table.
4. Information in italics font was not available in manufacturer’s product literature. Information in italics font was reported directly by the manufacturer.

NBCIP 7
5. The manufacturer’s literature indicates a supply voltage of 17-36 VDC for 500 Ohm load. A supply voltage of 18-36 VDC was reported directly by the manufacturer.
6. Manufacturer reported that product literature is in the process of being updated. Information in italics font was reported directly by the manufacturer.
Prior to performance testing, NBCIP subjected
all transmitters to the manufacturer stated
Two-Pressure
voltage input to ensure they were in full Humidity Generator
working order. The voltage output signal The Thunder Scientific 2500 Humidity Generator,
from each transmitter was also verified. One shown in Figure 4, can supply accurate and known
transmitter from Building Automation Products humidity values based on the fundamental principle of
Inc. (BAPI) failed this initial test and was the “Two-Pressure” generator developed by NIST. The
replaced by a newly purchased transmitter. device has an accuracy of ±0.5% RH over a humidity
range of 10%RH to 98% RH and a temperature range
Testing Methods of 32°F to 158°F (0°C to 70°C). This device can
Performance testing of the relative humidity be used for instrument calibration, evaluation and
transmitters was performed at the Center verification on a continuous basis. A copy of the
for Building Energy Research, Iowa State statement of accuracy and traceability to NIST for the
University, under NBCIP’s direction. Testing Thunder Scientific 2500 Humidity Generator is included
was performed from July to August 2002. in Appendix A.
NBCIP tested the humidity transmitters for
accuracy, repeatability, linearity, and hysteresis
using a Thunder Scientific Model 2500 Humidity
Generator (Thunder Scientific, 2000) that
employs an operating principle traceable to the
National Institute of Standards and Technology
(NIST). Refer to the sidebar, Two-Pressure
Humidity Generator, for more information.

NBCIP tested three relative humidity Figure 4.

transmitters at a time inside the humidity Thunder Scientific 2500 Humidity Generator
generator. The transmitters were placed inside The two-pressure method is shown in Figure 5. Air,
a custom made manifold that directs the or some other gas such as nitrogen, is saturated (in a
conditioned air over the sensing element of the saturator) with water vapor at a given pressure and
humidity transmitters. The manifold is made temperature. The saturated gas then flows through
out of copper to achieve a uniform temperature an expansion valve where it is reduced at constant
rapidly. Transmitters were placed in Slots 1, temperature to chamber pressure. If the temperature
2 and 3 of the manifold as shown in Figure 6. of the gas is held constant during pressure reduction,
The manifold was placed inside the chamber the humidity at the chamber pressure may be
of the humidity generator. The conditioned air approximated as the ratio of the two absolute
from the saturator of the humidity generator pressures, as shown below:
enters the manifold through the inlet port and
passes over the sensing element of the humidity PC
% RH ~~ x 100
transmitters before exiting the manifold. A PS
temperature probe is located at the center of the Relative humidity produced in the test chamber does
manifold to measure the manifold temperature. not depend on measuring the amount of water vapor
in the test chamber; rather it is dependent upon the
The humidity transmitters were powered by measurement of fundamental base units of pressure
a 24 VDC input from a Hewlett Packard (HP) and temperature only. The operation of the two-
Model 6253A DC power supply. The stability pressure humidity generator is based on fundamental
of the power supply is better than ±0.1% over principles and produces known humidity conditions
its full range. An HP Model 3455-A digital DC using base units.
voltmeter was used to measure the output of
the power supply to ensure that the appropriate Ps Pc
voltage was supplied to the transmitters.
Test
Air Supply Saturator Exhaust
The relative humidity transmitters tested Chamber
Expansion
provide a 0-10 VDC output. The voltage output Valve
leads from a transmitter were terminated at a Figure 5.
National Instruments SCB-68 shielded input/ Principle of the two-pressure humidity generator
output connector block, which was interfaced to (Thunder Scientific, 2000).
NBCIP 8
NBCIP 8
a data acquisition card via a shielded cable. The Inlet
shielded connector block and cable provide low- Temperature Probe
noise signal termination. Twisted-pair shielded
wire (American Wire Gauge 22) was used
for making connections between the power Outlet
supply, humidity transmitters, and connector
block to help reduce interference from other
instruments. The National Instruments
PCI-6035E data acquisition card used to
measure the output of the relative humidity Slots 1, 2, 3
transmitters has 16-bit resolution on analog
inputs and is calibrated traceable to NIST. Figure 6.
Schematic of the humidity transmitter manifold
The data acquisition card was interfaced to developed by Thunder Scientific.
LabVIEW software, which provided a convenient
graphical interface for controlling and
monitoring data collection. LabVIEW was used Figure 7 illustrates an example data set and
to scale the digital representation of the 0-10 V the numbering scheme used to identify the
output from the relative humidity transmitters data points.
into the measured relative humidity ranging
from 0-100%. The humidity transmitters were tested at
steady-state conditions inside the humidity
NBCIP tested the humidity transmitters at five generator. The transmitter output was sampled
levels of relative humidity, 10%, 30%, 50%, at a frequency of 1 kHz. The 1000 samples
70% and 90% RH, for each of three different collected each second were averaged to
temperatures, 59°F, 77°F, and 95°F (15ºC, 25ºC produce a single recorded value for each
and 35ºC, respectively), to reflect the conditions second. NBCIP recorded the humidity readings
encountered in a typical building HVAC system. each second for 45 minutes after the generator
The following procedure was used: had satisfied the steady-state criteria. Data
recorded at 45 minutes were then used for
• The test temperature and relative humidity further analysis.
were initially set to 59°F (15ºC) and 10% RH. At
59°F (15ºC), the relative humidity was increased
up to 90% RH in 20% RH increments. These
measurements are referred to as the first 100
100
forward measurements. 9090
5
Measured Relative Humidity (%)

• After reaching 90% RH, the tests were 8080

Measured Relative Humidity (%)

reversed, with the relative humidity being Reverse

6
7070 Reverse
decreased from 90% RH to 10% RH in measurements
measurements
20% RH increments while maintaining the 6060
Y-Axis

4
test temperature at 59°F (15ºC). These 7
measurements are referred to as the reverse 5050 10
First forward
First forward
measurements. 4040 measurements
measurements
3
8
• Once the 10% RH level was attained, the 3030 Second forward
Second forward
relative humidity was increased again to measurement
measurement

50% RH at 59°F (15ºC). This measurement 2020 9 2

is referred to as the second forward 1010 1

measurement.
00
• The above procedure was repeated once for 0 10 20 30 40 50 60 70 80 90 100
77°F (25ºC) and 95°F (35ºC) test temperatures.
0 40 50Humidity
30 Relative
10 20 Actual 60 70 (%) 80 90 100
Figure 7. Actual Relative Humidity (%)
This procedure resulted in 10 data points at Example data set to illustrate numbering scheme
each temperature for each transmitter tested. used to identify data points for NBCIP tests.

NBCIP 9
 Components Inc. and General Eastern also
showed a wide range in accuracy at the lowest
Test Results level of relative humidity (10% RH).

Accuracy Repeatability
Manufacturer stated accuracy for the humidity Repeatability is the degree to which a humidity
transmitter models tested is ±3% RH. General transmitter produces the same measurement
Eastern Inc., Johnson Controls Inc., and when subjected repeatedly to the same
Automation Components Inc. report accuracy conditions as they are approached from the
at 77°F (25ºC), Vaisala at 68°F (20ºC), while same direction. The test conditions defined by
Building Automation Products Inc. and MAMAC NBCIP included three measurements at
Systems Inc. do not report the temperature 50% RH, two of which were taken when the
at which the accuracy is stated (Table 1). relative humidity was increasing (e.g., the
The manufacturer stated humidity range at forward measurements). NBCIP assessed
which the accuracy is ±3% RH varies from the error in repeatability of the humidity
manufacturer to manufacturer and is indicated in transmitters at 50% RH and at each of the three
Table 1. test temperature conditions, namely, 59°F, 77°F
and 95°F (15°C, 25°C and 35°C), using pooled
Figures 8 to 13 show the results of NBCIP’s measurements of all three transmitters for a
evaluations of humidity transmitter accuracy given manufacturer. NBCIP defined the error in
for each manufacturer. NBCIP calculated the repeatability as the difference between the two
deviation of the measured relative humidity pooled forward measurements at 50% RH, as
from the actual relative humidity for each shown in Figure 14. Referring to Figure 14, the
transmitter tested and at each test condition error in repeatability is the first pooled forward
(refer to the Terms and Definitions section measurement (data point 3) minus the second
at the end of this report for an explanation pooled forward measurement (data point 10).
of terms related to test data). NBCIP
then calculated and plotted the average The error in repeatability of the transmitters
deviation, referred to as the pooled deviation, tested by NBCIP is shown in Table 2. Note that
which combines the forward and reverse humidity transmitter manufacturers may define
measurements of all three transmitters from and report repeatability differently than NBCIP;
a single manufacturer at each test condition, therefore, the results in Table 2 should be used
namely 10, 30, 50, 70 and 90% RH and at 59°F, to compare transmitter performance among
77°F and 95°F (15°C, 25°C and 35°C). Similar manufacturers, rather than comparing individual
figures were produced for each manufacturer. results with manufacturer reported data.
The accuracy of the humidity generator The error in repeatability of transmitters from
introduces a maximum uncertainty of Automation Components Inc. was consistently
±0.5% RH in the calculated deviations. within ±0.1% RH at all three test temperature
conditions. The error in repeatability of humidity
The humidity transmitter model from transmitters from Building Automation Products
Automation Components Inc. consistently Inc., General Eastern Inc., and Vaisala is less
performed within the ±3% RH accuracy than ±0.6% RH. Transmitters from Johnson
rating at all three test temperature conditions Controls Inc. and MAMAC Systems Inc. are
and at all humidity conditions within the repeatable within ±1.5% RH, with
manufacturer stated humidity range (shown by the highest error in repeatability occurring at
the shaded area in Figure 8). The transmitter 77°F (25°C).
model from Vaisala performed within the
specified accuracy over a 10% to 80% humidity
range at each test temperature condition
(Figure 13). The transmitter model from
Building Automation Products Inc. consistently
performed outside the specified accuracy of
±3% RH (Figure 9). Transmitters from General
Eastern Inc. (Figure 10), Johnson Controls Inc.
(Figure 11), and MAMAC Systems Inc. (Figure
12) show a strong temperature dependence.
Transmitter models from Automation

NBCIP 10
The shaded area in Figures 8 to 13 shows the range in RH for which the stated accuracy is ±3%, according
to manufacturer product literature. NBCIP tested the performance of the humidity transmitters over a range
of 10% to 90% RH and at three different temperatures, 59°F, 77°F and 95°F (15°C, 25°C and 35°C). Refer to
Table 1 for product information reported in manufacturer literature.
12 12
59ºF (15ºC) 59ºF (15ºC)
Actual Relative Humidity (%)

Actual Relative Humidity (%)

9 77ºF (25ºC) 9 77ºF (25ºC)
95ºF (35ºC) 95ºF (35ºC)

) %
) %

6 6

( yti di mu H evi t al e Rl a ut c A
( yti di mu H evi t al e Rl a ut c A
Deviation from

Deviation from
3 3

morf noi t ai ve D
morf noi t ai ve D

0 0

-3 -3

-6 -6

-9 -9

-12 -12
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

Actual Relative Humidity (%) Actual Relative Humidity (%)

Actual Relative Humidity (%) Actual Relative Humidity (%)

Figure 8. Figure 9.
NBCIP measured accuracy of Automation NBCIP measured accuracy of Building
Components Inc. humidity transmitters, Automation Products Inc. humidity transmitters,
model A/RH3-D. model BA/H310-D.

12 12
59ºF (15ºC)
Actual Relative Humidity (%)

59ºF (15ºC)
Actual Relative Humidity (%)
9 77ºF (25ºC) 9 77ºF (25ºC)
95ºF (35ºC) 95ºF (35ºC)
) %

) %

6 6
( yti di mu H evi t al e Rl a ut c A

( yti di mu H evi t al e Rl a ut c A
Deviation from

Deviation from

3 3
morf noi t ai ve D

morf noi t ai ve D

0 0

-3 -3

-6 -6

-9 -9

-12 -12
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

Actual Relative Humidity (%) Actual Relative Humidity (%)

Actual Relative Humidity (%) Actual Relative Humidity (%)

Figure 10. Figure 11.

NBCIP measured accuracy of General Eastern Inc. NBCIP measured accuracy of Johnson Controls Inc.
humidity transmitters, model MRH-3-D. humidity transmitters, model HT-6703-0N00P.

12 12
59ºF (15ºC) 59ºF (15ºC)
Actual Relative Humidity (%)

Actual Relative Humidity (%)

9 77ºF (25ºC) 9 77ºF (25ºC)

95ºF (35ºC) 95ºF (35ºC)
) %
) %

6 6
( yti di mu H evi t al e Rl a ut c A
( yti di mu H evi t al e Rl a ut c A
Deviation from

Deviation from

3 3
morf noi t ai ve D
morf noi t ai ve D

0 0

-3 -3

-6 -6

-9 -9

-12 -12
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

Actual Relative Humidity (%) Actual Relative Humidity (%)

Actual Relative Humidity (%) Actual Relative Humidity (%)

Figure 12. Figure 13.

NBCIP measured accuracy of MAMAC Systems Inc. NBCIP measured accuracy of Vaisala humidity
humidity transmitters, model HU-224-3-VDC. transmitters, model HMD50U.

NBCIP 11
 Table 2. NBCIP test results for repeatability of relative humidity transmitters at 50% RH.
Error in Repeatability
at 50% RH (% RH)
Manufacturer Model Number 59˚ F 77˚ F 95˚ F
(15˚ C) (25˚ C) (35˚ C)
Automation Components Inc. A/RH3-D 0.0 0.1 0.1
Building Automation Products Inc. BA/H310-D 0.2 0.0 0.6
General Eastern Inc. MRH-3-D 0.4 0.2 0.0
Johnson Controls Inc. HT-6703-0N00P 0.3 1.3 0.1
MAMAC Systems Inc. HU-224-3-VDC 0.3 1.4 1.1
Vaisala HMD50U 0.2 0.3 0.5
Note: Only the magnitude of the error in repeatability is shown in the table; consequently, negative signs have been omitted.

100
100 100
100
Error in Linearity
9090 90
90
5
Measured Relative Humidity (%)

8080 80
80
Measured Relative Humidity (%)

Humidity (%)
6
Reverse
7070
Relative Humidity
Reverse
measurements 70
70
measurements
First forward
First forward Actual Best fit
6060 measurements
measurements 60
60 transmitter straight line
Y-Axis

4
7 response
MeasuredRelative

5050 10 50
50
Error
Errorinin
4040 repeatability
repeatability 40
40
3
Measured

8
3030 Second forward
Second forward 30
30
measurement
measurement
2020 9 2 20
20
1010 1 10
10
00 00
0 10 20 30 40 50 60 70 80 90 100 00 10 20
20 30 40 50 50 60 70 70 80 90
90 100
0 10 20 30 40 50 60 70 80 90 100
Actual Relative Humidity (%) Actual Relative
Actual Relative Humidity
Humidity (%)
(%)
Figure 14. Actual Relative Humidity (%) Figure 15.
Example of repeatability of a relative humidity Example of linearity error of a relative humidity
transmitter. NBCIP evaluated repeatability at 50% RH, transmitter. In this example the maximum difference
corresponding to the difference in value between between the measured realative humidity and the
data point 3 and data point 10. best fit straight line occurs at 90% RH.

Linearity
Humidity transmitters generally do not show a the best-fit straight line of the measured data
1:1 linear relationship between the input (actual through the origin, as illustrated in Figure 15. In
conditions) and output (measured value) over this example, the maximum difference between
the working range. The term linearity denotes the measured RH and the best fit straight line
the extent to which the transmitter input and occurs at 90% RH.
output can be approximated by a linear function.
A highly non-linear humidity transmitter can The linearity of the transmitters tested by
result in poor control of HVAC processes. NBCIP at 77°F (25°C) is shown in Figures 16
Knowledge of a transmitter’s linearity to 21. Similar figures showing linearity results
characteristics can be embedded in a controller for tests performed at 59°F and 95°F (15°C and
to best represent and compensate for the 35°C) are included in Appendix B. Each figure
transmitter’s true behavior. NBCIP assessed shows the equation of the best-fit straight line of
the error in linearity of the humidity transmitters the measured data as well as the R2 parameter.
as the maximum difference between the Refer to the side bar, Explanation of Linearity
pooled measured relative humidity of three Figures, for more information.
transmitters from a particular manufacturer and

NBCIP 12
NBCIP tested the linearity of the humidity transmitters over a range of 10% to 90% RH and at three
different temperatures, 59˚F, 77˚F and 95˚F (15°C, 25°C and 35°C). Linearity plots for transmitters tested at
77˚F (25°C) are shown in Figures 16 to 21. Refer to Appendix B for results at 59˚F and 95˚F (15°C and 35°C).
Refer to Table 1 for product information reported in manufacturer literature.

100 100
y = 0.9975x y = 0.8784x
90 90

Measured and Estimated

2
Measured and Estimated

2
R = 0.9995 R = 0.9963

Relative Humidity (%)

Relative Humidity (%)

80 80

det a m
det a m

) %
) %

70 70

( yti di mu H evi t al e R
( yti di mu H evi t al e R

i t s E dna der us ae M
i t s E dna der us ae M

60 60
50 50
40 40
30 30
20 20
10 10
0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

Actual RelativeHumidity
Actual Relative Humidity(%)(%) Actual Relative
Actual RelativeHumidity
Humidity(%)(%)
Figure 16. Figure 17.
NBCIP measured linearity of Automation Components NBCIP measured linearity of Building Automation
Inc. humidity transmitters, model A/RH3-D tested at Products Inc. humidity transmitters, model BA/H310-D
77˚F (25°C). tested at 77˚F (25°C).

100 100
y = 0.9977x y = 1.0079x
90 90
Measured and Estimated

Measured and Estimated

2 2
R = 0.9901 R = 0.9895
Relative Humidity (%)

Relative Humidity (%)

80 80
det a m
det a m

) %
) %

70 70
( yti di mu H evi t al e R
( yti di mu H evi t al e R

i t s E dna der us ae M
i t s E dna der us ae M

60 60
50 50
40 40
30 30
20 20
10 10
0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

Actual Relative
Actual RelativeHumidity
Humidity(%)(%) Actual
ActualRelative
RelativeHumidity
Humidity(%) (%)
Figure 18. Figure 19.
NBCIP measured linearity of General Eastern Inc. NBCIP measured linearity of Johnson Controls Inc.
humidity transmitters, model MRH-3-D tested at humidity transmitters, model HT-6703-0N00P tested
77˚F (25°C). at 77˚F (25°C).

100 100
y = 0.9558x
90 y = 0.9871x 90 2
R = 0.9997
Measured and Estimated
Measured and Estimated

2
R = 0.9868
80
Relative Humidity (%)
Relative Humidity (%)

80
det a m
det a m

) %
) %

70 70
( yti di mu H evi t al e R
( yti di mu H evi t al e R

i t s E dna der us ae M

60
i t s E dna der us ae M

60
50 50
40 40
30 30
20 20
10 10
0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

Actual RelativeHumidity
Actual Relative Humidity(%)(%) Actual
ActualRelative
RelativeHumidity
Humidity(%)(%)
Figure 20. Figure 21.
NBCIP measured linearity of MAMAC Systems Inc. NBCIP measured linearity of Vaisala humidity
humidity transmitters, model HU-224-3-VDC tested transmitters, model HMD50U tested at 77˚F (25˚C).
at 77˚F (25°C).

NBCIP 13
Table 3. NBCIP test results for error in linearity of relative humidity transmitters.

Error in Linearity (%RH)

Manufacturer Model Number 59˚ F 77˚ F 95˚ F
(15˚ C) (25˚ C) (35˚ C)
Automation Components Inc. A/RH3-D 4.8 1.2 1.2
Building Automation Products Inc. BA/H310-D 2.7 2.2 2.3
General Eastern Inc. MRH-3-D 5.6 5.8 7.8
Johnson Controls Inc. HT-6703-0N00P 5.4 3.9 2.8
MAMAC Systems Inc. HU-224-3-VDC 4.9 3.8 5.1
Vaisala HMD50U 0.6 0.7 1.0
Note: Only the magnitude of the error in linearity is shown in the table; consequently, negative signs have been omitted.

The maximum difference between the

measured value and the best-fit straight line Explanation of
are summarized in Table 3 for each humidity
transmitter model tested and at all three Linearity Figures
temperatures. Note that humidity transmitter A best-fit straight line through the origin has the
manufacturers may define and report linearity relationship y = ax, where a is the slope of the line, x is
differently than NBCIP; therefore, the results the actual relative humidity, and y is the predicted relative
in Figures 16 to 21, figures in Appendix B, and humidity based on the best-fit straight line. R2 is a
values in Table 3 should be used to compare measure of the agreement between the measured relative
transmitter performance among manufacturers, humidity data from transmitters and predicted values of
rather than comparing individual results with the relative humidity obtained from the best-fit straight
manufacturer reported data. line. R2 can take a value between 0 and 1. An R2 value
close to 1 indicates a good fit of the data to the line.
Figures 16 to 21 show that all relative humidity
transmitters tested at 77°F (25°C) exhibit a
linear relationship between input and output.
With the exception of the transmitter model 100
100
from Building Automation Products Inc., the
9090
best-fit straight line for the transmitters tested 5
Measured Relative Humidity (%)

is very close to an ideal 1:1 linear relationship 8080

Measured Relative Humidity (%)

(i.e., the slopes of the best-fit straight lines are 6

7070 Reverse
Reverse
nearly equal to one). Table 3 shows that the measurements
measurements
transmitters from Vaisala have the lowest error 6060 First
Firstforward
forward
measurements
measurements
in linearity, while transmitters from General
Y-Axis

4
7
Eastern Inc. have the highest error in linearity 5050
10
across all test temperatures. Hysteresis
Hysteresis
4040
3
8
Hysteresis 3030 Second forward
Second forward
Hysteresis is the error in measurement when measurement
measurement
the same humidity condition is approached from 2020 9 2
a lower and then higher humidity condition.
NBCIP assessed the hysteresis of each 1010 1
manufacturer’s humidity transmitters using the 00
pooled forward measurements and the pooled 0 10 20 30 40 50 60 70 80 90 100
reverse measurements, which were available at 0 30 40
10 20 Actual 60 70(%)80 90 100
50 Humidity
Relative
30%, 50% and 70% at each test temperature. Figure 22.
The hysteresis of a manufacturer’s humidity Actual Relative
Example of hysteresis Humidity
of a relative (%)
humidity
transmitter at each temperature was calculated transmitter. NBCIP evaluated the hysteresis at 30%,
as the maximum difference of the pooled 50%, and 70% RH and reported the maximum value.
values obtained for the forward and reverse In this example, the maximum value of hysteresis
measurements, as illustrated in Figure 22. In occurs at 50% RH, corresponding to the difference in
the example shown in Figure 22, the maximum value between data point 7 and data point 3.

NBCIP 14
Table 4. NBCIP test results for maximum hysteresis of relative humidity transmitters.

Maximum Hysteresis (% RH)

Manufacturer Model Number 59˚ F 77˚ F 95˚ F
(15˚ C) (25˚ C) (35˚ C)
Automation Components Inc. A/RH3-D 0.6 0.7 0.8
Building Automation Products Inc. BA/H310-D 1.0 0.8 1.3
General Eastern Inc. MRH-3-D 2.6 1.2 0.3
Johnson Controls Inc. HT-6703-0N00P 3.2 2.3 2.1
MAMAC Systems Inc. HU-224-3-VDC 1.5 1.2 1.2
Vaisala HMD50U 0.4 0.5 0.7
Note: Only the magnitude of the maximum hysteresis is shown in the table; consequently, negative signs have been omitted.

difference in pooled measured values occurs manufacturers and tested the transmitters
at 50% RH, corresponding to the difference in for accuracy, repeatability, linearity and
value between data point 7 and data point 3. hysteresis. The results show that some
The maximum hysteresis of the transmitters humidity transmitters are sensitive to varying
tested by NBCIP is shown in Table 4. Note that temperature and humidity conditions. Some
humidity transmitter manufacturers may define of the transmitters tested did not perform
and report hysteresis differently than NBCIP; within the manufacturer stated accuracy. In
therefore, the results in Table 4 should be used certain cases, the transmitter accuracy differed
to compare transmitter performance among significantly from the stated ±3% RH accuracy
manufacturers, rather than comparing individual at temperatures that were higher and lower
results with manufacturer reported data. than the temperature at which accuracy
was reported by manufacturers. Ultimately,
The hysteresis of humidity transmitters from specifying a humidity transmitter requires
Automation Components Inc. and Vaisala does careful consideration of the application in
not exceed 0.8% RH and shows very little which the transmitter is expected to perform
variation with temperature. The hysteresis of and careful analysis of product manufacturer
humidity transmitters from Building Automation information to determine whether a particular
Products Inc. and MAMAC Systems Inc. does product will meet the needs of the application.
not exceed 1.5% RH. The largest value of
hysteresis occurs for humidity transmitters Next Steps
from General Eastern Inc. and Johnson NBCIP is currently testing the long-term
Controls Inc. (hysteresis of 2.6% RH and performance of the relative humidity
3.2% RH, respectively) at a temperature of 59°F transmitters discussed in this report. Two
(15°C) and decreases with increasing humidity transmitters from each of the six
test temperature. manufacturers were installed in an outdoor
air duct of the Iowa Energy Center’s Energy
Resource Station (NBCIP’s test facility). The
transmitters will be exposed to a broad range
Conclusions of environmental conditions for a period of
one year. The transmitters will be tested for
The specification of a relative humidity accuracy every four months to assess sensor
transmitter goes beyond an examination of drift. Additional testing will also measure
manufacturer stated accuracy. Manufacturers’ response time, an important characteristic for
product literature usually states accuracy control applications, as well as the capability
and other performance criteria for a single of the transmitters to withstand extreme
temperature condition, which often does not conditions such as cycling between high and
reflect the conditions the transmitter is likely to low relative humidity conditions and exposure
be subjected to in the field. NBCIP purchased to saturated conditions. Results of this phase
duct-mounted relative humidity transmitters of testing will be available to the public as a
with a stated accuracy of ±3% RH from six supplement to this report.

NBCIP 15
Terms and Definitions
Accuracy: Measured Relative Humidity:
Accuracy is the deviation of the relative The relative humidity reading from a
humidity measured by a humidity transmitter humidity transmitter.
from the actual, or known, humidity.
Pooled Deviation:
Actual Relative Humidity: The pooled deviation at a given test condition
The relative humidity generated by the is determined by averaging the mean deviation
reference standard. of all humidity transmitters of a specific
manufacturer model at the test condition. Note
Deviation: that NBCIP tested three humidity transmitters
Deviation is the difference between the of a given model per manufacturer. As an
measured and actual relative humidity at a given example, consider a test condition of 59°F
test condition, and is defined by the following (15°C) and 10% RH. If the three transmitters of
equation: a given manufacturer I are denoted by A, B, and
Deviation= RHmeasured - RHactual C, the pooled deviation is calculated as:
NBCIP calculated the deviation for each
humidity transmitter tested at each test (MDA + MDB + MDC) I,10,59
PD I,10,59 =
condition. NBCIP also calculated the mean 3
deviation for a given transmitter at a given
test condition by averaging the deviation of the where: PD I,10,59 = pooled deviation for
forward measurement(s) and the deviation of transmitters A, B, C from manufacturer I
the reverse measurement. Referring to at 10% RH and 59°F (15°C)
Figure 7 as an example, at 10% RH, data point 1 MD = mean deviation of each
represents the forward measurement, and data transmitter (A, B, C) from manufacturer I
point 9 represents the reverse measurement.
The deviation of the forward measurement is: Repeatability:
Deviation 1 = RH measured_1 - RH actual_1. Repeatability is the degree to which a humidity
Similarly, the deviation of the reverse transmitter produces the same measurement
measurement is: when subjected repeatedly to the same
Deviation 9 = RH measured _ 9 - RH actual_9. conditions as they are approached from the
The mean deviation is therefore the average of same direction.
Deviation 1 and Deviation 9.

Humidity Sensor:
Refers to the humidity sensing element.

Humidity Transmitter:
A device that consists of a sensor as well as a
transducer that converts the sensor reading into
an output signal.

Hysteresis:
Hysteresis is the error in measurement when
the same humidity condition is approached from
a lower and then higher humidity condition.

Linearity:
Linearity is the extent to which the humidity
transmitter input and output can be
approximated by a linear function through
the origin.

NBCIP 16
Further Reading

ASHRAE, Handbook of Fundamentals, American Society of Heating,

Refrigerating, and Air-Conditioning Engineers, Inc., Atlanta, GA, 2001.

ASHRAE, Humidity Control Design Guide For Commercial and

Institutional Buildings, American Society of Heating, Refrigerating, and
Air-Conditioning Engineers, Inc, Atlanta, GA, 2001.

ASHRAE, Psychrometrics: Theory and Practice, American Society of

Heating, Refrigerating and Air Conditioning Engineers, Inc.,
Atlanta, GA, 1996.

ASHRAE, Standard 41.6-1994: Method for Measurement of Moist

Air Properties, American Society of Heating, Refrigerating and Air
Conditioning Engineers, Inc., Atlanta, GA, 1994.

Brownawell, M., “An RH sensor review, with HVAC considerations”,

Sensors, March 1989, Vol. 6, No. 3, pp. 35-40.

Roveti, D. K., “Choosing a Humidity Sensor: A Review of Three

Technologies”, Sensors Magazine, July 2001.
http://www.sensorsmag.com/articles/0701/54/main.shtml

Thunder Scientific Technical Catalog, Thunder Scientific Corporation,

623 Wyoming S. E. Albuqerque, NM 87123, USA, 2000.

Wiederhold, P. R., “Optimizing RH transmitter performance for HVAC

applications”, Sensors, March 1990, Vol. 7, No. 3, pp. 69-71, 1990.

Weiderhold, P. R., “True Accuracy of Humidity Measurement”, Sensors

Magazine, September 1997.
http://www.sensorsmag.com/articles/0997/humidity /main.shtml

Yamatake, 2004.
http://ibdcom.yamatake.co.jp/products/fp3/background.html

NBCIP 17
Appendix A:
Copy of statement of accuracy and traceability to NIST for
Thunder Scientific 2500 Humidity Generator.

NBCIP 18
Appendix B:
Linearity plots for humidity transmitters tested at 59°F (15°C) and 95°F (35°C).
NBCIP tested the linearity of the humidity transmitters over a range of 10% to 90% RH and at three
different temperatures, 59°F, 77°F and 95°F (15°C, 25°C and 35°C). Linearity plots for transmitters tested
at 59°F (15°C) are shown in Figures 23 to 28. Linearity plots for transmitters tested at 95°F (35°C) are
shown in Figures 29 to 34. Refer to Table 1 for product information reported in manufacturer literature.
100 100
y = 1.0048x y = 0.8831x
90 2 90

Measured and Estimated

Measured and Estimated

R = 0.9933 2
R = 0.9938

Relative Humidity (%)

Relative Humidity (%)

80 80
det a m

det a m
) %

) %
70 70
( yti di mu H evi t al e R

( yti di mu H evi t al e R
i t s E dna der us ae M

60

i t s E dna der us ae M
60
50 50
40 40
30 30
20 20
10 10
0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Actual Relative
Actual RelativeHumidity
Humidity(%)(%) ActualRelative
Actual Relative Humidity
Humidity (%) (%)
Figure 23. Figure 24.
NBCIP measured linearity of Automation Components NBCIP measured linearity of Building Automation
Inc. humidity transmitters, model A/RH3-D tested at Products Inc. humidity transmitters, model
59˚F (15˚C). BA/H310-D tested at 59˚F (15˚C).

100 100
y = 1.0182x y = 1.0396x
90 2 90 2
Measured and Estimated

R = 0.9863
Measured and Estimated

R = 0.9767
80
Relative Humidity (%)

80
Relative Humidity (%)
det a m

det a m
) %

) %

70 70
( yti di mu H evi t al e R

( yti di mu H evi t al e R
i t s E dna der us ae M

i t s E dna der us ae M

60 60
50 50
40 40
30 30
20 20
10 10
0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

ActualRelative
Actual RelativeHumidity
Humidity(%)(%) ActualRelative
Actual Relative Humidity
Humidity (%) (%)

Figure 25. Figure 26.

NBCIP measured linearity of General Eastern Inc. NBCIP measured linearity of Johnson Controls Inc.
humidity transmitters, model MRH-3-D tested at humidity transmitters, model HT-6703-0N00P tested
59˚F (15˚C). at 59˚F (15˚C).

100 100
Measured and Estimated
Measured and Estimated

y = 1.0321x y = 0.9681x
90 90
Relative Humidity (%)
Relative Humidity (%)

2 2
R = 0.9808 R = 0.9996
80 80
det a m
det a m

) %
) %

70 70
( yti di mu H evi t al e R
( yti di mu H evi t al e R

i t s E dna der us ae M
i t s E dna der us ae M

60 60
50 50
40 40
30 30
20 20
10 10
0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

Actual Relative
Actual RelativeHumidity
Humidity(%)(%) Actual
ActualRelative
RelativeHumidity (%) (%)
Humidity
Figure 27. Figure 28.
NBCIP measured linearity of MAMAC Systems Inc. NBCIP measured linearity of Vaisala humidity
humidity transmitters, model HU-224-3-VDC tested transmitters, model HMD50U tested at 59˚F (15˚C).
at 59˚F (15˚C).

NBCIP 19
 100 100
y = 0.9887x y = 0.8565x
90 2 90 2
R = 0.9991
Measured and Estimated

Measured and Estimated

R = 0.9942
80 80
Relative Humidity (%)

Relative Humidity (%)

det a m

det a m
) %

) %
70 70
( yti di mu H evi t al e R

( yti di mu H evi t al e R
i t s E dna der us ae M

i t s E dna der us ae M
60 60
50 50
40 40
30 30
20 20
10 10
0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Actual RelativeHumidity
Actual Relative Humidity (%)
(%) ActualRelative
Actual Relative Humidity
Humidity (%) (%)
Figure 29. Figure 30.
NBCIP measured linearity of Automation Components NBCIP measured linearity of Building Automation
Inc. humidity transmitters, model A/RH3-D tested at Products Inc. humidity transmitters, model
95˚F (35˚C). BA/H310-D tested at 95˚F (35˚C).

100 100
y = 0.9593x y = 0.9806x
90 2 90 2
Measured and Estimated

R = 0.9857

Measured and Estimated

R = 0.9954
80
Relative Humidity (%)

80

Relative Humidity (%)

det a m

det a m
) %

) %
70 70
( yti di mu H evi t al e R

( yti di mu H evi t al e R
i t s E dna der us ae M

i t s E dna der us ae M

60 60
50 50
40 40
30 30
20 20
10 10
0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

ActualRelative
Actual RelativeHumidity
Humidity(%)(%) ActualRelative
Actual Relative Humidity
Humidity (%) (%)

Figure 31. Figure 32.

NBCIP measured linearity of General Eastern Inc. NBCIP measured linearity of Johnson Controls Inc.
humidity transmitters, model MRH-3-D tested at humidity transmitters, model HT-6703-0N00P tested
95˚F (35˚C). at 95˚F (35˚C).

100 100
y = 0.9519x y = 0.9470x
90 90 2
Measured and Estimated

Measured and Estimated

2
R = 0.9894 R = 0.9996
Relative Humidity (%)

Relative Humidity (%)

80 80
det a m
det a m

) %
) %

70 70
( yti di mu H evi t al e R
( yti di mu H evi t al e R

i t s E dna der us ae M
i t s E dna der us ae M

60 60
50 50
40 40
30 30
20 20
10 10
0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

Actual Relative
Actual RelativeHumidity
Humidity(%)(%) ActualRelative
Actual Relative Humidity
Humidity (%) (%)

Figure 33. Figure 34.

NBCIP measured linearity of MAMAC Systems Inc. NBCIP measured linearity of Vaisala humidity
humidity transmitters, model HU-224-3-VDC tested transmitters, model HMD50U tested at 95˚F (35˚C).
at 95˚F (35˚C).

NBCIP 20
Product Testing Report your information cannot be regarded as a
Duct-Mounted Relative representation that the product is or is not
Humidity Transmitters safe to use in any specific situation, or that the
Principal Investigators: Shailesh Joshi, Iowa particular product you purchase will conform to
State University; John House, Iowa Energy the results found in this report.
Center; Michael Pate, Iowa State University;
Ron Nelson, Iowa State University The National Building Controls
NBCIP Project Manager: John M. House Information Program
Report Authors: Maria Corsi and John M. House The National Building Controls Information
Project Technical Advisor: Curtis J. Klaassen Program (NBCIP) was established by the Iowa
Program Technical Director: Maria Corsi Energy Center in 2001 with support from
Communications Manager: Amy Swenson the U.S. Environmental Protection Agency
to facilitate the adoption of energy efficient
The following persons provided technical building control products and strategies
review of this report: William M. Healy, National through testing, demonstration, education
Institute of Standards and Technology; Thomas and dissemination of product information. The
H. Kuehn, University of Minnesota; David Iowa Energy Center has launched this national
Sellers, Portland Energy Conservation, Inc. initiative based on our investment in the Energy
The following persons provided technical review Resource Station, a unique facility that serves as
of the method of test: Kenneth L. Gillespie, NBCIP’s testing laboratory, and the expertise our
Pacific Gas and Electric Company; Thomas H. researchers have acquired in building controls.
Kuehn, University of Minnesota.
Reviewers are listed to acknowledge their NBCIP recognizes that improving the operation
contributions to the final publication. Their of control systems is a major energy saving
approval or endorsement of this report is not opportunity. NBCIP research is aimed at
necessarily implied. understanding the cause of problems with
building controls, evaluating the performance
COPYRIGHT NOTICE. Copyright of control systems and strategies, and
© 2004 Iowa State University, developing best-practice recommendations for
the Iowa Energy Center. All rights implementing energy-efficient
are reserved by Iowa State University, and no control strategies.
portion of this publication or the information
contained herein may be reproduced, NBCIP provides unbiased, accurate, reliable,
disseminated, published, transferred, or manufacturer-specific information on building
excerpted in any way in other publications, control systems to guide and improve
databases, web sites or any other medium the selection, installation, operation and
without express written permission of Iowa maintenance of control products and strategies
State University’s Iowa Energy Center. for building applications.
Permission is granted to an individual to print
copies of all or part of this publication solely for NBCIP establishes building control product
their own personal use and not for any other performance in a consistent manner. The NBCIP
use. Any other use without written permission label signifies that testing was performed
is a violation of United States copyright laws. using an established process that includes
Permission is not granted for use of the peer review of the method of test and the test
data contained in this publication for product results. To ensure objectivity, NBCIP does
publicity purposes. The NBCIP and Iowa Energy not accept funding from control
Center logos may not be used in any advertising product manufacturers.
or publicity, or otherwise to indicate NBCIP’s
or the Iowa Energy Center’s sponsorship of or Contacting NBCIP
affiliation with any product or service. National Building Controls Information Program
Iowa Energy Center, Energy Resource Station
The products described herein have not been DMACC, 2006 S. Ankeny Blvd.
tested for safety. The Iowa Energy Center and Ankeny, IA 50021
Iowa State University make no representations Phone: (515) 965-7055
whatsoever with regard to the safety of Fax: (515) 965-7056
products, in whatever form or combination Email: [email protected]
used, and the results of testing set forth for Visit NBCIP online at: www.buildingcontrols.org

NBCIP 21