.ugust 2ln^ | www.epa.gov/or
United States
Environmental Protection
Agency
Technology Evaluation Report
Testing and Evaluation
of Handheld Toxic Industrial
Chemical Detectors
Office of Research and Development
National Homeland Security Research Center
-------
EPA 600/R-12/560
August 2012
Technology Evaluation Report
Testing and Evaluation
of Handheld Toxic Industrial
Chemical Detectors
-------
Notice
The U.S. Environmental Protection Agency (EPA), through its Office of Research and
Development's National Homeland Security Research Center, funded and directed this
technology evaluation under Work Assignments 1-10 and 2-10 of EPA Contract Number EP-C-
10-001 with Battelle. This report has been peer and administratively reviewed and has been
approved for publication as an EPA document. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use of a specific product.
Questions or comments should be addressed to:
Shannon Serre
National Homeland Security Research Center
Office of Research and Development
U.S. Environmental Protection Agency
Mail Code E343-06
Research Triangle Park, NC 27711
in
-------
Executive Summary
The U.S. Environmental Protection
Agency's (EPA's) National Homeland
Security Research Center (NHSRC) helps
protect human health and the environment
from adverse impacts of terrorist acts by
carrying out a variety of research activities,
including performance tests on homeland
security technologies. As part of its
mission, NHSRC supports EPA's Regional
On-Scene Coordinators and response teams,
as well as state and local emergency
response agencies, by evaluating
technologies to meet the monitoring needs
of their organizations. In particular, first
responders and emergency management
professionals need reliable, sensitive, and
portable monitoring devices that can rapidly
indicate the presence of hazardous
conditions, including air containing reduced
levels of oxygen, explosive levels of
flammable chemicals in air, or harmful
levels of toxic or corrosive chemicals.
This report describes testing to assess the
performance of commercially available
handheld detectors capable of quantifying
oxygen (02), flammable mixtures (in terms
of the lower explosive limit [LEL] for CH/i),
and six toxic industrial compounds (TICs)
(i.e., H2S, SO2, NH3, C12, PH3, and HCN) at
concentrations that would present a threat to
emergency response personnel. The
evaluation reported here used realistically
hazardous concentrations of the target
species, matched to the detection ranges of
each of the detectors. Testing evaluated the
following quantitative performance
parameters:
• Response and Recovery Time
• Accuracy
• Repeatability
• Response Threshold (i.e.,
detection limit)
• Effect of Operating Conditions
(i.e., temperature and relative
humidity [RH])
• Effect of O2 Deficiency on TIC
Response
• Cold/Hot Start Behavior
• Interference Effects
• Battery Life
Operational factors such as size and weight;
ease of use; clarity of displays, alarms, and
instructions; startup and shutdown
procedures; sensor replacement;
maintenance issues; and design features
affecting handheld operation were also
evaluated. The ease of using each detector
with personal protective equipment
including heavy gloves was also assessed.
Testing was conducted over a temperature
range of approximately 8 to 35 °C and an
RH range from less than 20% to
approximately 80%. Interferent testing was
conducted using vapors of the following six
materials, both in otherwise clean air (to
assess false positive responses) and
comingled with O2, CH4, and each of the six
TICs (to assess false negative responses):
• Latex paint
• Gasoline exhaust hydrocarbons
• Diesel exhaust hydrocarbons
• Ammonia cleaner
• Air freshener
• N,N-diethylaminoethanol
(DEAE) (a boiler and
humidification water additive)
The seven handheld detectors subjected to
testing were:
• BW Technologies GasAlert
Micro 5
• Drager X-am 7000
• Environics ChemPro lOOi
IV
-------
• Industrial Scientific iBRID MX6
• RAE Systems MultiRAE Pro
• RKI Instruments Eagle 2
• Sperian PHD6
All of the tested detectors except the
Environics ChemPro lOOi employed a
galvanic cell for percent 62 measurement, a
catalytic bead sensor for LEL, and
electrochemical (EC) cells for TIC
detection. Those six detectors could not
incorporate sensors to detect all of the target
gases at once, so each detector was
purchased with a set of sensors installed and
additional sensors were substituted into the
detectors as needed to conduct the testing.
The ChemPro lOOi employed a multi-sensor
measurement approach that includes open-
loop ion mobility spectrometry along with
semiconductor, metal oxide semiconductor,
and field effect sensors and temperature,
RH, pressure, and flow sensors. The
ChemPro lOOi was not designed to
determine atmospheric C>2 or LEL and,
unlike the six other detectors, provided a
qualitative reading of signal intensity rather
than a measured concentration (e.g., in
ppm).
In total, the testing reported here involved
seven handheld detectors, eight target gases,
six interferents, and six different
temperature/RH conditions, as well as
specific tests involving three cold start
conditions and two levels of reduced 62.
The test results on each performance
parameter are summarized below.
ES.l Response and Recovery Time
Response and recovery time were
determined as the elapsed time to achieve a
stable detector reading after the start or end,
respectively, of a target gas challenge. The
response and recovery times of the seven
handheld detectors in determination of TICs
are summarized in Figures ES-1 and ES-2,
respectively. Each figure shows the mean,
median, and ±1 standard deviation (SD)
range of all the response times recorded for
each detector in all testing with the six TICs.
Figure ES-1 shows that the ChemPro lOOi
exhibited the fastest response overall in
testing with the six TICs, and the iBRID
MX6 exhibited the slowest response overall
with those TICs. Median response times in
the TIC testing ranged from approximately
20 seconds with the ChemPro lOOi to
approximately 100 seconds with the iBRID
MX6. The other five detectors exhibited
response times in TIC testing that were
closely similar and intermediate between
those of the ChemPro lOOi and the iBRID
MX6, e.g., median TIC response times of
approximately 40 to 50 seconds. In testing
of six detectors with O2 and CFL (not shown
in Figure ES-1), relatively faster response
was observed as compared to the TIC
responses. With C>2, response times for all
six detectors were typically less than 30
seconds, and the Eagle 2 often responded in
less than 10 seconds. With CtL;, response
times for most of the six detectors were less
than 30 seconds, with the GasAlert Micro 5
-------
180
140
-S 120
c
o
« 100
0)
E
80
0)
g 60
O
Q.
a) 40
Mean Response Time
^Median Response Time
+ Mean + Std. Dev.
— Mean - Std. Dev.
GasAlert X-am 7000 ChemPro iBRID MultiRAE Eagle 2 PHD6
Figure ES-1. Summary of response time results in TIC testing.
800
700
600
1/1
•a
c
8 500
0)
i/i
aT
E 400
Mean Recovery Time
Median Recovery Time
+ Mean + Std. Dev.
— Mean - Std. Dev.
o
u
QJ
300
200
100
GasAlert X-am 7000 ChemPro iBRID MultiRAE Eagle 2 PHD6
Figure ES-2. Summary of recovery time results in TIC testing.
VI
-------
always responding within 20 seconds and
the Eagle 2 often responding in 10 seconds
or less. The X-am 7000 response times for
CH4 ranged from about 30 to nearly 50
seconds.
Figure ES-2 shows that the GasAlert Micro
5, ChemPro lOOi, Eagle 2, and PHD6
exhibited the fastest recovery overall in
testing with the six TICs, and the iBRID
MX6 exhibited the slowest recovery overall
with those TICs. Median recovery times in
the TIC testing ranged from approximately
50 seconds with the GasAlert Micro 5 to
approximately 360 seconds with the iBRID
MX6. In testing of six detectors with O2 and
CFLj (not shown in Figure ES-2), relatively
faster recovery was observed as compared to
the TIC recoveries. With C>2, recovery times
for most of the six detectors were typically
less than 30 seconds, and the MultiRAE Pro
and Eagle 2 often recovered in
approximately 10 seconds or less. However,
the recovery times for the Sperian PHD6
with O2 were usually more than 40 seconds
and ranged up to more than 250 seconds.
With CH4, recovery times for the six
detectors were usually less than 25 seconds,
but the GasAlert Micro 5, X-am 7000,
iBRID MX6, MultiRAE Pro, and PHD6 all
showed recovery times for CH4 that
exceeded 280 seconds in testing conducted
at35°C.
ES.2 Accuracy
Quantitative accuracy (QUA) was
determined for all detectors except the
Environics ChemPro lOOi, which provided a
qualitative indication of response intensity
rather than a quantitative concentration
reading. Figure ES-3 summarizes the QUA
results determined for the other six detectors
in all testing with the six TICs, C>2, and CH/i.
That figure shows the mean, median, and ±1
SD range of all the QUA values recorded for
each detector in all testing, excluding any
readings that resulted from a pegged
overrange response on a detector. Thus,
Figure ES-3 does not include values such as
the 111% QUA recorded for the MultiRAE
Pro with FbS, which resulted from the
monitor pegging at a reading of 99.9 ppm
when challenged with 90 ppm of H^S.
Figure ES-3 shows that the mean QUA
values for the six detectors over all target
gases ranged from 91% for the MultiRAE
Pro to 125% for the iBRID MX6, and the
median QUA values ranged from 95% for
the MultiRAE Pro to 113% for the iBRID
MX6. However, Figure ES-3 is based on
only about two-thirds of the possible QUA
results for the X-am 7000 due to non-
quantitative overrange indications by that
detector in some tests. The same is true for
the MultiRAE Pro and Eagle 2 due to
exclusion of fixed quantitative readings
exhibited during overrange conditions on
those detectors. The exclusion of these
readings means that QUA values for those
three detectors might be significantly higher
if quantitative readings above the nominal
full-scale value could be obtained from the
detectors. In contrast, the iBRID MX6 and
Sperian PHD6 never reported an overrange
condition in any test. The PHD6 in
particular achieved mean and median QUA
values near 100% and a relatively narrow
range of QUA results around 100%, as
indicated by the ±1 SD range in Figure ES-
3.
vn
-------
180
160
140
120
£100 -
u
re
I 80
v
£ 60
re
§ 40
a
20
• Mean QUA
4 Median QUA
+ Mean QUA+ SD
-Mean QUA-SD
GasAlert X-am 7000
iBRID
MultiRAE
Eaele 2
PHD6
Figure ES-3. Summary of QUA results in TIC, Oi, and CELt testing (QUA not determined
for ChemPro lOOi). Data shown exclude any readings indicating a constant overrange
condition of a detector.
Identification accuracy (IA) was 100% (i.e.,
the detectors correctly identified the gas
challenge in all trials) in almost all tests.
Other than in tests at the lowest challenge
concentrations, the only cases of IA less
than 100% were with the ChemPro lOOi,
which failed to respond in some tests with
SO2, NH3, C12, and HCN that involved
interferent vapors or temperature and RH
conditions other than 22°C and 50% RH.
ES.3 Repeatability
For the six detectors other than the ChemPro
lOOi, repeatability was consistently within
5% relative standard deviation (RSD) in
detection of H2S, SO2, PH3, HCN, O2, and
CH4. A few exceptions of repeatability up
to approximately 10% RSD occurred with
the Eagle 2 with HCN and with the PHD6
with CH4. Repeatability results were
substantially higher (usually within 10%
RSD, with occasional values of 20% or
more) for all six detectors with NHs and C12.
Repeatability for these six detectors was not
affected by interferent vapors or by test
conditions of temperature and RH.
Repeatability values for the ChemPro lOOi
were constrained by the detector's l-to-3-
bar intensity indication and, in most cases,
the ChemPro lOOi gave the same intensity
response with all five challenges in a test
Vlll
-------
(i.e., repeatability = 0% RSD). However,
the presence of interferent vapors and test
conditions other than room temperature and
50% RH sometimes degraded the
repeatability of ChemPro lOOi response.
ES.4 Response Threshold
With few exceptions, all detectors tested
exhibited response thresholds of less than 3
ppm for H2S and NHs, less than 5 ppm for
SC>2 and HCN, less than 1 ppm for Cb and
PHa, and less than 0.2% by volume (i.e., less
than 4% of the LEL) for CH4. The
exceptions were that the BW GasAlert
Micro 5 showed a response threshold in the
range of 1 to 3 ppm for Cb, the RAE
MultiRAE Pro showed a response threshold
in the range of 0.2 to 0.5% for CH4, and the
Environics ChemPro lOOi showed response
thresholds in the range of 20 to 50 ppm for
SO2, 10 to 50 ppm for NH3, and 3 to 10 ppm
for C\2. The observed response thresholds
are generally far below the immediately
dangerous to life and health (IDLH) levels
for the target TICs; even the ChemPro lOOi
response thresholds for 862, NHs, and Cb
are at least a factor of two less than the
respective IDLH levels. Except in the case
of NHs, the response threshold testing
reported above did not extend to low enough
concentrations to prove detection at the
acute (i.e., 1 hour) Reference Exposure
Level values for these TICs.
ES.5 Effect of Operating Conditions
With all seven detectors the performance
factors most affected by variations in
temperature and RH conditions were
response and recovery times, which were
usually lengthened by conditions other than
normal room temperature and 50% RH.
Effects of temperature and RH on response
and recovery times were seen less frequently
with the ChemPro lOOi than with the other
six detectors. The performance factors least
affected by variations in temperature and
RH were QUA, IA, and repeatability.
Effects on QUA occurred with several
detectors (this performance parameter was
not determined for the ChemPro lOOi),
whereas the majority of effects on IA and
repeatability occurred with the ChemPro
lOOi.
ES.6 Effect of O2 Deficiency on TIC
Response
The RKI Eagle 2 showed no significant
differences in any performance parameter
for H2S with reduced O2 levels, and none of
the detectors showed any significant
differences in IA for H2S at reduced 62
levels. Significant effects of O2 level on
response time, recovery time, and QUA for
H2S were seen with some detectors. The
response time for H2S was shortened at the
16% O2 level with both the BW GasAlert
Micro 5 and Industrial Scientific iBRID
MX6, but was increased (i.e., nearly
doubled) with the Drager X-am 7000 at both
19% and 16% O2. The recovery time for
H2S was greatly increased at 16% 02 for the
Environics ChemPro lOOi and at both 19%
and 16% O2 for the Industrial Scientific
iBRID MX6. The QUA for H2S declined
consistently with reduced O2 levels for the
BW GasAlert Micro 5, Drager X-am 7000,
and Industrial Scientific iBRID MX6.
ES.7 Cold/Hot Start Behavior
In most cases, response times, QUA, IA, and
repeatability for detection of H2S were
affected only minimally by rapid startup
after storage overnight at room, cold, or hot
temperature. The delay times between
powering up each detector and being ready
to begin monitoring similarly showed little
impact from the storage condition before
IX
-------
startup. However, recovery times were
lengthened with several detectors, especially
after rapid startup from room temperature or
cold conditions. Repeatability was degraded
with the ChemPro lOOi after cold starts from
all three storage conditions.
ES.8 Interference Effects
All of the seven detectors showed false
positive (FP) responses in some tests when
sampling an interferent vapor in otherwise
clean air. Gasoline and diesel exhaust
hydrocarbons and paint vapors were the
interferents that most frequently caused FP
responses. The MultiRAE Pro was the
detector most subject to interference effects,
showing FP responses with all six
interferents in testing with H2S, O2, and
CFLj, and FP responses with at least one
interferent with every target gas. The
ChemPro lOOi and iBRID MX6 also showed
FP responses with at least one interferent
with every target gas with which they were
tested. The X-am 7000 and GasAlert Micro
5 were the detectors least subject to FP
responses. The X-am 7000 showed no FP
responses at all in testing with H2S, PHs,
HCN, and O2. The GasAlert Micro 5
showed no FP responses at all in testing with
H2S, C12, PH3, HCN, and CH4.
The false negative (FN) rates that resulted
from the interferents were almost always
zero. In fact, for six of the seven detectors
(i.e., the GasAlert Micro 5, X-am 7000,
iBRID MX6, MultiRAE Pro, Eagle 2, and
PHD6) the FN rate was zero with every
interferent in every test. FNs were observed
with the ChemPro lOOi in tests with SO2,
NHa, C12, and HCN. Gasoline engine
exhaust hydrocarbons caused FN with the
ChemPro lOOi with all four of these TICs,
and ammonia cleaner, air freshener, and
diesel exhaust also caused FN responses in a
few tests with the ChemPro lOOi.
ES.9 Battery Life
The battery life of the seven detectors is
illustrated in Figure ES-4, and ranged from
less than 10 hours for the ChemPro lOOi and
Drager X-am 7000 to nearly 46 hours for the
RKI Eagle 2 unit E2A505. The two Eagle 2
units exhibited the longest and third-longest
periods of battery life, but the battery life of
Unit E2A505 was more than twice as long
as that of Unit E2 A410. Thi s difference i s
attributed largely to the greater power
demand of the LEL sensor in Unit E2A410.
ES.10 Operational Factors
The following are brief summaries of key
positive and negative operational factors
reported by the test operators for each
handheld detector.
BW Technologies GasAlert Micro 5. This
detector was small, lightweight, and easy to
use, and large font on the display made it
easy to read. Operating menus were easy to
understand, calibration menus less so.
-------
RKI Eagle 2 (E2A505)
Indus. Sci. iBRID MX6
RKI Eagle 2 (E2A410)
BWIGasAlertMicroS
RAEMultiRAEPro
Sperian PHD6
DragerX-am 7000
ChemPro lOOi
10 20 30
Battery Life (hours]
40
50
Figure ES-4. Summary of battery life test results.
The operating manual was troublesome
because required key sequences were
sometimes not located together on the same
page.
Drager X-am 7000. This detector was
relatively heavy and boxy in shape, making
it uncomfortable to hold in the hand for
more than a few minutes. The display area
was large and easily readable. Operating
menus were easy to understand and the
detector was easy to use and had numerous
user-defined options. However, the
operating manual did not appear to cover all
of the features or operations of the unit.
Environics ChemPro lOOi. This detector
was easy to operate, with intuitive menus,
and had large control buttons that could be
manipulated correctly even when wearing
heavy gloves. The ChemPro lOOi required
confidence checks with a chemical vapor
source provided with the detector. Those
checks were simple to perform and the
detector responded quickly to the confidence
check. The ChemPro lOOi was relatively
sensitive to the test conditions (temperature
and RH) and occasionally had difficulty
maintaining its baseline operating condition
when moved during testing, causing false
alarms and requiring that the operator reset
the baseline. The MOS sensor in the first
ChemPro lOOi unit failed during testing, and
a replacement ChemPro lOOi unit was
provided by the manufacturer.
Industrial Scientific iBRID MX6. This
detector had logical and self-explanatory
menus, but the menus were difficult to
navigate because the buttons on this detector
were small and clustered tightly together.
This was especially a problem when wearing
heavy gloves. The display of the iBRID
MX6 was weakly backlit and the display
XI
-------
font was small, making readings difficult to
discern. This detector also responded
relatively slowly to daily bump checks.
RAE Systems MultiRAE Pro. This
detector was easy to operate by following
the instruction manual, the menus were
clearly understandable, and the display was
easy to read. However, it was difficult to
determine the full-scale ranges of the
sensors installed in the MultiRAE Pro
without seeking technical support or online
information from the manufacturer. The use
of heavy gloves made it difficult to feel
when the control buttons had been
successfully pressed. Multiple EC sensors
could fit into the C>2 sensor location of this
detector, but would not work in that
location. The operator would not know that
the sensor was not working until the detector
had been reassembled and powered up.
RKI Instruments Eagle 2. Three separate
units of this detector had to be purchased to
conduct testing because the necessary
sensors could not be interchanged within a
single unit. The Eagle 2 was relatively large
and heavy, but its design and built-in handle
made it comfortable to use. The display was
clear and legible but did not indicate the
status of the batteries. Operation of this
detector while wearing heavy gloves was
difficult, as it was hard to feel when the
control buttons had been successfully
pressed.
Sperian PHD6. This detector's display was
easy to read, but the detector's alarms would
change the display, interfering with
concentration readings. Testing staff
adjusted the alarm values to avoid this issue
during testing. Selection of a particular
sensor on the calibration menu required
toggling through multiple menu steps.
Operation of the detector's control buttons
and performance of the pump test were
difficult when wearing heavy gloves. The
sample inlet tubing of the PHD6 connects at
the bottom of the detector, and thus the
connection point is directed toward the user
when the detector is held in the hand,
potentially leading to pinching or snagging
of the inlet tubing. The battery charger of
the PHD6 makes electrical contact by
gravity and sometimes did not make proper
contact.
xn
-------
Acknowledgments
Contributions of the following individuals and organizations to the development of
this document are gratefully acknowledged.
U.S. Environmental Protection Agency (EPA)
Michael Boykin
Steven Merritt
Lukas Oudejans
Len Zintak
Battelle Memorial Institute
Xlll
-------
Contents
NOTICE iii
EXECUTIVE SUMMARY iv
ACKNOWLEDGMENTS xiii
APPENDICES xv
FIGURES xv
TABLES xvi
ABBREVIATIONS/ACRONYMS xviii
1.0 INTRODUCTION 1
2.0 EXPERIMENTAL METHODS 2
2.1 Performance Parameters 2
2.1.1 Response and Recovery Time 2
2.1.2 Accuracy 2
2.1.3 Repeatability 2
2.1.4 Response Threshold 3
2.1.5 Operating Conditions 3
2.1.6 Oxygen Deficiency and TIC Response 3
2.1.7 Cold/Hot Start Behavior 3
2.1.8 Interference Effects 3
2.1.9 Battery Life 3
2.1.10 Operational Characteristics 3
2.2 Target Compounds 4
2.3 Detectors Tested 5
2.4 Testing Parameters 9
2.4.1 Test Conditions 9
2.4.2 Chemical Interferences 9
2.4.3 Test Matrix 11
2.4.4 Test System and Procedures 15
2.5 Data Acquisition 19
3.0 STATISTICAL CALCULATIONS 21
3.1 Response and Recovery Time 21
3.2 Repeatability 21
3.3 Accuracy 21
3.4 Response Threshold 22
3.5 Effect of Operating Conditions 22
3.6 Cold/Hot Start Behavior 22
3.7 Interference Effects 23
3.8 Battery Life 23
4.0 QUALITY ASSURANCE/QUALITY CONTROL 25
4.1 Data Quality Indicators 25
4.2 Audits 26
xiv
-------
4.2.1 Performance Evaluation Audit 26
4.2.2 Technical Systems Audit 27
4.2.3 Data Quality Audit 28
4.2.4 QA/QC Reporting 28
4.3 Data Review 28
5.0 RESULTS 29
5.1 Response and Recovery Time 29
5.2 Accuracy 39
5.3 Repeatability 49
5.4 Response Threshold 58
5.5 Effect of Operating Conditions 58
5.6 Effect of Oxygen Deficiency on TIC Response 68
5.7 Cold/Hot Start Behavior 69
5.8 Interference Effects 71
5.9 Battery Life 76
5.10 Operational Factors 76
6.0 SUMMARY 82
6.1 Response and Recovery Time 82
6.2 Accuracy 84
6.3 Repeatability 85
6.4 Response Threshold 86
6.5 Effect of Operating Conditions 86
6.6 Effect of O2 Deficiency on TIC Response 86
6.7 Cold/Hot Start Behavior 87
6.8 Interference Effects 87
6.9 Battery Life 87
6.10 Operational Factors 87
Appendices
APPENDIX A: NOMINAL UPPER RANGE LIMITS OF THE TESTED DETECTORS
APPENDIX B: EXAMPLE OF LABORATORY DATA RECORDING SHEETS
Figures
Figure ES-1. Summary of Response Time Results in TIC Testing vi
Figure ES-2. Summary of Recovery Time Results in TIC Testing vi
Figure ES-3. Summary of QUA Results in TIC, O2, and CH4 Testing (QUA not
determined for ChemPro lOOi) viii
Figure ES-4. Summary of Battery Life Test Results xi
xv
-------
Figure 2.3-1. Handheld Detectors Tested; a: BW Technologies GasAlert Micro 5, b:
Draeger XAM 7000, c: Environics ChemPro lOOi, d: Industrial Scientific
iBRTD MX6, e: RAE Instruments MultiRAE Pro, f: RKI Instruments Eagle
2, g: SperianPHD6 7
Figure 2.4-1. Schematic of Test System 16
Figure 2.4-2. View of Two Test Systems in Adjacent Laboratory Hoods for Testing of
Handheld Detectors 17
Figure 2.4-3. Example of Glass Bell Tube Placed over Inlet Tube of a Detector 18
Figure 5.9-1. Summary of Battery Life Test Results 76
Figure 6.1-1. Summary of Response Time Results in TIC Testing 83
Figure 6.1-2. Summary of Recovery Time Results in TIC Testing 84
Figure 6.2-1. Summary of QUA Results in TIC, O2, and CH4 Testing 85
Figure 6.9-1. Summary of Battery Life Test Results 88
Tables
Table 2.2-1. Target Gases Used to Evaluate Handheld Multigas Detectors 4
Table 2.4-1. Summary of Temperature andRH Conditions for Testing 9
Table 2.4-2. Interferences Used in Testing of Handheld Multigas Detectors 11
Table 2.4-3. Summary of Quantitative Evaluations Conducted 12
Table 2.4-4. Summary of Tests Conducted with Each Detector and Target Gas 13
Table 2.4-5. Summary of TIC Challenge Concentrations (ppm) Used with Each Detector 14
Table 2.4-6. Summary of O2 and CH4 Concentrations Used in %O2 and LEL Testing 15
Table 4.1-1. Summary Statistics of Reference Method Results 25
Table 4.2-1. Summary of PE Audit Results 27
Table 5.1-1. Summary of Mean Response and Recovery Times (seconds) with H2S 31
Table 5.1-2. Summary of Mean Response and Recovery Times (seconds) with SO2 32
Table 5.1-3. Summary of Mean Response and Recovery Times (seconds) withNH? 33
Table 5.1-4. Summary of Mean Response and Recovery Times (seconds) with C12 34
Table 5.1-5. Summary of Mean Response and Recovery Times (seconds) with PHs 35
Table 5.1-6. Summary of Mean Response and Recovery Times (seconds) with HCN 36
Table 5.1-7. Summary of Mean Response and Recovery Times (seconds) with O2 37
Table 5.1-8. Summary of Mean Response and Recovery Times (seconds) with CH4 38
Table 5.2-1. Summary of Quantitative Accuracy and Identification Accuracy (percent) with
H2S 41
Table 5.2-2. Summary of Quantitative Accuracy and Identification Accuracy (percent) with
S02 42
Table 5.2-3. Summary of Quantitative Accuracy and Identification Accuracy (percent) with
NH3 43
Table 5.2-4. Summary of Quantitative Accuracy and Identification Accuracy (percent) with
C12 44
Table 5.2-5. Summary of Quantitative Accuracy and Identification Accuracy (percent) with
PH3 45
xvi
-------
Table 5.2-6. Summary of Quantitative Accuracy and Identification Accuracy (percent) with
HCN 46
Table 5.2-7. Summary of Quantitative Accuracy and Identification Accuracy (percent) with
O2 47
Table 5.2-8. Summary of Quantitative Accuracy and Identification Accuracy (percent) with
CH4 48
Table 5.3-1. Summary of Repeatability (percent RSD) with H2S 50
Table 5.3-2. Summary of Repeatability (percent RSD) with SO2 51
Table 5.3-3. Summary of Repeatability (percent RSD) with NH3 52
Table 5.3-4. Summary of Repeatability (percent RSD) with C12 53
Table 5.3-5. Summary of Repeatability (percent RSD) with PH3 54
Table 5.3-6. Summary of Repeatability (percent RSD) with HCN 55
Table 5.3-7. Summary of Repeatability (percent RSD) with O2 56
Table 5.3-8. Summary of Repeatability (percent RSD) with CH4 57
Table 5.4-1. Summary of Response Threshold Results 58
Table 5.5-1. Performance Parameters under Different Temperature and Relative Humidity
Conditions with H2S 60
Table 5.5-2. Performance Parameters under Different Temperature and Relative Humidity
Conditions with SO2 61
Table 5.5-3. Performance Parameters under Different Temperature and Relative Humidity
Conditions with NH3 62
Table 5.5-4. Performance Parameters under Different Temperature and Relative Humidity
Conditions with C12 63
Table 5.5-5. Performance Parameters under Different Temperature and Relative Humidity
Conditions with PH3 64
Table 5.5-6. Performance Parameters under Different Temperature and Relative Humidity
Conditions with HCN 65
Table 5.5-7. Performance Parameters under Different Temperature and Relative Humidity
Conditions with O2 66
Table 5.5-8. Performance Parameters under Different Temperature and Relative Humidity
Conditions with CH4 67
Table 5.6-1. Performance Differences Observed in H2S Detection at Reduced O2 68
Table 5.7-1. Summary of Performance Parameters under Fully Warmed Up and Cold Start
Conditions 70
Table 5.8-1. Summary of False Positive (FP) and False Negative (FN) Rates with H2S 72
Table 5.8-2. Summary of False Positive (FP) and False Negative (FN) Rates with SO2 72
Table 5.8-3. Summary of False Positive (FP) and False Negative (FN) Rates with NH3 73
Table 5.8-4. Summary of False Positive (FP) and False Negative (FN) Rates with C12 73
Table 5.8-5. Summary of False Positive (FP) and False Negative (FN) Rates with PH3 74
Table 5.8-6. Summary of False Positive (FP) and False Negative (FN) Rates with HCN 74
Table 5.8-7. Summary of False Positive (FP) and False Negative (FN) Rates with O2 75
Table 5.8-8. Summary of False Positive (FP) and False Negative (FN) Rates with CH4 75
xvn
-------
Abbreviations/Acronyms
CH4 methane
CI number of correct identifications
C\2 chlorine
CO carbon monoxide
COR Contracting Officer's Representative
DEAE N,N-diethylaminoethanol
DHS Department of Homeland Security
EC electrochemical
EPA U.S. Environmental Protection Agency
FN false negative
FP false positive
GC/FID gas chromatography with flame ionization
detection
HCN hydrogen cyanide
H2S hydrogen sulfide
I number of samples with interferent in air
IA identification accuracy
IDLH immediately dangerous to life and health
IMS ion mobility spectrometry
LEL lower explosive limit
MOS metal oxide semiconductor
NH? ammonia
NHSRC National Homeland Security Research Center
NR number of negative responses
O2 oxygen
PE performance evaluation
PH3 phosphine
PPE personal protective equipment
PR number of positive responses
QA quality assurance
QC quality control
QMP quality management plan
QUA quantitative accuracy
RH relative humidity
RSD relative standard deviation
SD standard deviation
SO2 sulfur dioxide
STEL short term exposure limit
T temperature
TI number of samples with interferent and target gas
TIC toxic industrial compound
TSA technical systems audit
TWA time-weighted average
VOC volatile organic compound
xvin
-------
1.0 Introduction
The U.S. Environmental Protection
Agency's (EPA's) National Homeland
Security Research Center (NHSRC) is
helping protect human health and the
environment from adverse impacts resulting
from acts of terror. NHSRC works in
partnership with recognized testing
organizations, with stakeholder groups
(buyers, vendor organizations, scientists,
and permitters), and with individual
technology developers in carrying out
performance tests on homeland security
technologies. In response to the needs of
stakeholders, NHSRC conducts research and
evaluates the performance of innovative
homeland security technologies by
developing test plans, conducting
evaluations, collecting and analyzing data,
and preparing peer-reviewed reports. All
evaluations are conducted in accordance
with rigorous quality assurance (QA)
protocols to ensure the generation of high
quality data and defensible results.
NHSRC-supported research provides
unbiased, third-party information
supplementary to vendor-provided
information that is useful to decision makers
in purchasing or applying the evaluated
technologies. Stakeholder involvement
ensures that user needs and perspectives are
incorporated into the evaluation design to
produce useful performance information for
each evaluated technology.
Responding to an accident, fire, or
deliberately caused chemical release can
expose first responders to hazardous
conditions, including air containing reduced
levels of oxygen, explosive levels of
flammable chemicals in air, or harmful
levels of toxic or corrosive chemicals. To
minimize such exposures, first responders
and emergency management professionals
need reliable, sensitive, and portable
monitoring devices that can rapidly indicate
the presence of multiple chemical and
environmental hazards at the same time.
EPA's NHSRC supports EPA's Regional
On-Scene Coordinators and response teams,
as well as state and local emergency
response agencies, by evaluating
technologies to meet this monitoring need.
The test results presented in this report are
part of NHSRC's efforts to identify and
verify the performance of portable hazard
detectors for use by such organizations.
The objective of the testing described in this
report was to assess the performance of
commercially available handheld detectors
capable of quantifying oxygen (02),
flammable mixtures (in terms of the lower
explosive limit [LEL]), and multiple toxic
industrial compounds (TICs) at
concentrations that would present a threat to
emergency response personnel. The
evaluations used realistically hazardous
concentrations of the target species, and
assessed response time, accuracy,
repeatability, effects of potential
interferents, and effects of normal
temperature and relative humidity (RH)
variations. Operational factors such as
battery lifetime, startup time under normal,
cold, and hot conditions, and clarity of
displays and alarms were evaluated. The
ease of using each detector with personal
protective equipment (PPE) including heavy
gloves was also assessed. In performing
this technology evaluation, the procedures
specified in the peer-reviewed test/QA plan
developed for this test, and complied with
quality requirements in the NHSRC Quality
Management Plan (QMP) were followed.
-------
2.0 Experimental Methods
Seven commercially available handheld
multigas detectors were tested with O2, a
flammable gas (methane, CFLj), and selected
TICs under a realistic range of conditions
and procedures of use. This section presents
the experimental design, test procedures,
and test and reference methods.
2.1 Performance Parameters
The following performance parameters were
evaluated:
• Response and Recovery Time
• Accuracy
• Repeatability
• Response Threshold (i.e.,
detection limit)
• Effect of Operating Conditions
(temperature and RH)
• Effect of O2 Deficiency on TIC
Response
• Cold/Hot Start Behavior
• Interference Effects
• Battery Life
• Operational Factors.
2.1.1 Response and Recovery Time
Response time (also known as rise time) is
the length of time required for a handheld
detector to provide a stable quantitative
reading after the onset of a challenge with a
target gas. The response time was evaluated
because response personnel need a rapid
indication of chemical hazard and
concentration.
Recovery time (also known as fall time) is
the length of time required for the detector
to return to a stable baseline quantitative
reading after a challenge ends. Recovery
time was evaluated because it limits how
rapidly the detector can provide an accurate
reading of a safe (no-hazard) condition or a
new response to a hazard condition. This
parameter is relevant when, for example,
different levels of contamination are present
in different places at a response scene, and
the detector must clear before it could be
used reliably in another place.
Both response and recovery time were
recorded in repetitive challenges with each
target gas. For both response and recovery
time, a stable detector reading was defined
as a reading that did not change over
approximately 20 seconds, as observed by
the test operator.
2.1.2 Accuracy
Accuracy is the degree of quantitative
agreement between the target gas
concentration indicated by a handheld
detector and the known challenge
concentration. Quantitative accuracy
(QUA) was evaluated by direct comparison
of known challenge concentrations and
quantitative detector responses.
Identification accuracy (IA) was determined
by reviewing detector responses to evaluate
whether the detector accurately identified
the target gas being sampled.
2.1.3 Repeatability
Repeatability is the degree of consistency of
the response of a handheld detector to
repeated challenges with the same target gas
concentration under uniform test conditions.
This parameter is important as an indication
of the reliability of an individual response
from the detector. Repeatability was
determined with each target gas by means of
the same repetitive challenges used to
determine response and recovery time.
-------
2.1.4 Response Threshold
The response threshold is the approximate
concentration below which a handheld
detector does not detect a target gas (i.e.,
does not provide a reading different from its
baseline reading). It is important to
determine whether the response threshold of
a detector is low enough that an absence of
detector response can be taken to indicate
the absence of a hazard. Challenge gas
concentrations were stepped downward to
estimate the response threshold of each
handheld detector. Response threshold was
determined at normal conditions of
approximately 22°C and 50% RH.
2.1.5 Operating Conditions
Emergency response situations can occur in
any weather, so handheld multigas detectors
used by responding personnel must be
capable of providing correct readings under
a wide range of ambient conditions.
Consequently, challenge gas mixtures were
sampled at selected temperature and RH
conditions to investigate the effect of such
conditions on detector performance.
2.1.6 Oxygen Deficiency and TIC
Response
Some TICs, such as H^S, are detected by
oxidation within the electrochemical (EC)
sensors used in many handheld detectors.
For such TICs, sensor response may depend
on the concentration of C>2 in the air, and
detector performance may be degraded in air
of lower than normal C>2 content.
Consequently, each handheld detector was
challenged with H^S at C>2 levels below the
normal 20.9% to test for this behavior.
2.1.7 Cold/Hot Start Behavior
Monitoring instruments may need to provide
full operational capabilities on short notice
in emergency response situations.
Consequently, the handheld multigas
detectors were tested for the delay time that
is required between turning the instrument
on and readiness for hazard detection, and
for the accuracy and speed of response
under such use. This rapid startup behavior
was determined for three separate startup
conditions: after overnight startup from
room temperature, from cold storage, and
from hot storage of the detector.
2.1.8 Interference Effects
In emergency response situations, relatively
innocuous chemical compounds or mixtures
present in the air may interfere with (i.e.,
mask or alter) the response of a handheld
detector. Examples of such potential
interferences may be cleaning supplies, paint
fumes, or vehicle exhaust. The effect of
potential interferences was assessed because
such compounds can potentially produce
two types of errors with the handheld
detectors: (1) erroneous reporting of the
presence of a target gas when none is
present (false positives [FPs]) or (2)
reduction in sensitivity or masking of
response to target gases of interest (false
negatives [FNs]). To investigate both types
of error, interference effects were evaluated
by sampling potential interferences both in
otherwise clean air, and in air containing the
target gases.
2.1.9 Battery Life
Handheld multigas detectors operate on
battery power when in use in the field, and
the length of battery life is critical to
uninterrupted response operations. Battery
life was determined by operating each
handheld detector continuously, starting
with a fully charged battery, until the battery
was fully depleted and the detector stopped
operating.
2.1.10 Operational Characteristics
-------
Key operational characteristics of the
handheld detectors were evaluated by
observations of test personnel and, if
necessary, by inquiry to the respective
vendors. The operational factors included
the readability of displays; ease of operation
with and without PPE (i.e., heavy gloves);
logic and simplicity of operational functions
and software menus; data recording
capabilities; and cost. The costs for each
handheld detector were assessed based on
the purchase price of the detector, any
additional sensors needed for testing, and
any replaceable or maintenance items.
Testing was not of sufficient duration to test
long-term maintenance or operational costs
of the technologies.
2.2 Target Compounds
Table 2.2-1 lists the target gases used in
testing the handheld multigas detectors. The
determination of LEL was addressed by
using CH4 in air at concentrations at or
below 25% of its LEL of 5% in air (i.e.,
concentrations at or below 1.25% methane
in air). Six TICs were used to represent a
range of gaseous chemical hazards.
Table 2.2-1. Target Gases Used to Evaluate Handheld Multigas Detectors
Hazard Category
Oxygen-Depleted Environment
Lower Explosive Limit
Toxic Industrial Chemicals (TICs)
Target Gas
Oxygen (O2)
Methane (CH4)
Hydrogen Sulfide (H2S)
Sulfur Dioxide (SO2)
Ammonia (NHs)
Chlorine (C12)
Phosphine (PH3)
Hydrogen Cyanide (HCN)
-------
2.4 Detectors Tested
The seven handheld detectors tested are
described below and illustrated in Figure
2.3-1. All seven detectors were purchased
from the manufacturers with internal air
sampling capability to actively draw the
challenge gas mixtures to their sensors.
Each detector was operated according to the
manufacturer's instructions, as indicated in
the operating manuals provided in electronic
form for each detector. Operations included
daily confidence checks or "bump" tests
specified by the vendor to confirm that a
detector was operating properly. Positive
response to a target or surrogate chemical
was required in such checks before testing
could start with a given detector.
Six of the seven detectors tested employed a
galvanic cell for percent 62 measurement, a
catalytic bead sensor for LEL, and EC cells
for TIC detection. Those six detectors could
not incorporate sensors to detect all the
target gases at once, so each detector was
purchased with a set of sensors installed and
additional sensors were then substituted into
the detectors as needed to conduct the
testing. The configuration of each detector
(i.e., the set of sensors installed in the
detector) was recorded throughout the
testing process. The seventh detector
employed a completely different
measurement principle based on a
proprietary open-loop ion mobility
spectrometry (IMS) approach. The
following descriptions note specific features
or requirements of each detector that
affected how the detector was used in
testing.
BW Technologies GasAlert Micro 5. The
GasAlert Micro 5 was 14.5 x 7.4 x 3.8 cm
(5.7 x 2.9 x 1.5 in) in size and weighed
approximately 370 g (13 oz). This detector
(Serial No. M5-XWHS-R-P-D-Y-N-00) was
operated on internal rechargeable battery
power, which was recharged overnight. The
detector was used with the optional pump
module, and drew sample in at
approximately 0.45 L/min through the pump
connector and a 15 cm length of the
Teflon®-lined Tygon tubing supplied with
the pump module. However, the sample
probe and Tygon tubing supplied with the
pump module were not used. This detector
was capable of holding a maximum of four
sensors. The C>2 and LEL sensors were
permanently installed. The other sensors
installed in the GasAlert Micro 5 were as
follows: Cb and SC>2 sensors, during testing
with those two TICs; NHs and HCN sensors,
during testing with those two TICs; and PH3
and H2S sensors, during testing with those
two TICs, O2, and CH4 (LEL). The
purchase price of the GasAlert Micro 5 and
sensors was approximately $2,600.
Driiger X-am 7000. The X-am 7000 was
15 x 14 x 7.5 cm (5.9 x 5.6 x 3 in) in size
and weighed 600 g (21 oz). This detector
(Serial No. ARBM-0503) was operated on
internal rechargeable battery power, which
was recharged overnight. The detector was
used with the internal pump and pump
adapter, and drew in sample at
approximately 0.66 L/min through a 5 cm
length of the inlet tubing provided. The
sample probe obtained with the detector was
not used. This detector was capable of
holding a maximum of four sensors. The C>2
and LEL sensors were permanently
installed. The other sensors installed in the
X-am 7000 were as follows: Cb and SO2
sensors, during testing with those two TICs;
NHa and HCN sensors, during testing with
those two TICs; a PH3 sensor during testing
with that TIC; and PH? and H2S sensors,
during testing with H2S, C>2, and CH4 (LEL).
The purchase price of the X-am 7000 and
sensors was approximately $5,000.
-------
e
g
Figure 2.3-1. Handheld detectors tested; a: BW Technologies GasAlert Micro 5,
b: Drager X-am 7000, c: Environics ChemPro lOOi, d: Industrial Scientific iBRID MX6,
e: RAE Systems MultiRAE Pro, f: RKI Instruments Eagle 2, g: Sperian PHD6.
-------
Environics ChemPro 1001. The ChemPro
1001 was 23 x 10 x 5.7 cm (9 x 4 x 2 in) in
size and weighed 880 g (31 oz). This
detector had internal rechargeable batteries,
but at the manufacturer's request was kept in
operating mode and connected to line power
during all tests except the battery lifetime
test. Brief periods of operation on battery
power showed no differences in response
compared to operation on line power,
however this comparison was not a focus of
testing. The ChemPro lOOi's internal pump
drew sample in at approximately 1.3 L/min.
The Field Monitoring Cap provided with the
instrument was used as the instrument's inlet
in all testing. This approach was chosen
because the intent of testing was to assess
hazard identification in the field, and
because of the absence of any physical
connection of the detector to the test
apparatus (see Section 2.4.4) that would
have required use of the detector's Fixed
System Monitoring Cap. The ChemPro lOOi
was designed to detect all six of the target
TICs, but did not have capability for O2 or
LEL measurement. The ChemPro lOOi uses
a multi-sensor measurement technology that
includes open-loop IMS; semiconductor,
metal oxide semiconductor (MOS), and field
effect sensors; and temperature, RH,
pressure, and flow sensors. The First
Responder library of the ChemPro lOOi was
used in testing, as this library was most
applicable to the intent of the testing and
provided identification of the target TICs.
Unlike the other six detectors, the ChemPro
lOOi does not provide quantitative
indications of TIC concentration (e.g., ppm
values). Instead the ChemPro lOOi provided
a qualitative indication of response intensity
(i.e., one, two, or three bars) when
responding to a TIC. The purchase price of
the ChemPro lOOi was approximately
$15,800.
Environics representatives required that
Contractor personnel take a brief training
session in operation and testing of the
ChemPro lOOi. That training session was
conducted by teleconference before any
testing took place. The ChemPro lOOi was
subjected to a confidence check consisting
of a sensor test before every test procedure,
using the "test tube" source of chemical
vapors (1-propanol and
diisopropylmethylphosphonate) provided
with the detector. No testing of the
ChemPro lOOi took place unless the detector
display indicated "Test Passed" upon
completion of the sensor test.
Two units of the ChemPro lOOi were used in
testing. The first unit (S/N
06CPil03701538) was used throughout
testing with SO2, C12, NH3, and HCN, but
displayed an unrecoverable "functional
exception D08:2057" on July 25, 2011, near
the end of testing with PHa. That unit was
returned to Environics, and a replacement
unit (S/N 06CPi 102201497) was promptly
received. The replacement unit was then
used to complete the final two tests with
PH3, and for all testing with H2S. The
original unit sometimes responded relatively
slowly, and occasionally failed a sensor test,
giving the error message "No MOS signal
detected." This message apparently referred
to the metal oxide sensor, and a problem
with that sensor may have been the ultimate
cause of the first ChemPro lOOi unit's
failure. The replacement unit never failed
the confidence check.
Industrial Scientific iBRID MX6. The
iBRID MX6 was 13.5 x 7.7 x 4.3 cm (5.3 x
3 x 1.7 in) in size and weighed
approximately 409 g (14.4 oz). This
detector (Serial No. 1101397-002) was
operated on internal rechargeable battery
power, which was recharged overnight. The
detector had an internal pump which drew in
-------
sample at approximately 0.34 L/min through
a 5 cm length of Teflon® tubing. The
sample probe obtained with the detector was
not used. This detector was capable of
holding a maximum of five sensors. The C>2
and LEL sensors, and a carbon monoxide
(CO) sensor, were permanently installed.
The other sensors installed in the iBRTD
MX6 were as follows: H^S and SO2 sensors,
during testing with SO2; C\2 and 862
sensors, during testing with Cb; NH? and
HCN sensors, during testing with those two
TICs; and PH? and H^S sensors, during
testing with those two TICs, 62, and CH4
(LEL). The purchase price of the iBRTD
MX6 and sensors was approximately
$4,000.
RAE Systems MultiRAE Pro. The
MultiRAE Pro was 19.3 x 9.7 x 6.6 cm (7.6
x 3.8 x 2.6 in) in size and weighed 880 g (31
oz). This detector (Serial No. PGM-6240)
was operated on internal rechargeable
battery power, which was recharged
overnight. The detector had an internal
pump which drew in sample at
approximately 0.40 L/min through a filter
and a 6 cm length of Teflon tubing. This
detector was capable of holding a maximum
of five sensors. In almost all tests with the
six TICs, sensors for CO, LEL, and volatile
organic compounds (VOCs) were installed
in the MultiRAE Pro. The other sensors
installed in the MultiRAE Pro were as
follows: tbS and 862 sensors, during testing
with SC>2; tbS and NHa sensors, during
testing with NH?; Cb and HCN sensors,
during testing with C12; PH3 and HCN
sensors, during testing with those two TICs;
and PH3 and H2S sensors, during almost all
testing with H2S. However, during the final
tests with H2S (consisting of the cold start
tests, see Table 2.4-4) the MultiRAE Pro
held sensors for LEL, VOCs, C>2, PHs and
H2S. That same set of sensors was in the
MultiRAE Pro in all testing with O2 and
CH4 (LEL). The MultiRAE Pro gave CH4
readings in % LEL, rather than in %CH4 by
volume. The %LEL readings were
converted to %CH4 for QUA determination
based on the fact that the LEL for CH4 is 5%
by volume in air. The purchase price of the
MultiRAE Pro and sensors was
approximately $7,300.
RKI Instruments Eagle 2. The Eagle 2
was the largest and heaviest of the detectors
tested, measuring 24.1 x 13.5 x 15 cm (9.5 x
5.3 x 5.9 in) in size and weighing 1.73 kg
(61 oz). The vendor of the Eagle 2 indicated
that the sensors for the various target gases
were not all compatible with one another.
Consequently, it was necessary to buy three
separate units of the detector to achieve
detection of all of the target gases for
testing. One unit of the Eagle 2 (E2A505
Type 3112) was equipped with sensors for
SO2, PH3, and HCN. A second unit
(E2A504 Type 2011) was equipped with
sensors for Cb and NHa, and the third unit
(E2A410 Type 3001) was equipped with
sensors for H2S, 62, and CH4. Each Eagle 2
unit had an internal pump which drew in
sample at approximately 0.78 L/min through
an approximately 30 cm length of the
sample hose provided with the unit. That
hose was connected by stainless steel quick-
disconnect fittings between the sample inlet
of the Eagle 2 unit and the hydrophobic
probe filter provided with the unit. The
Eagle 2 units operated on replaceable
batteries (C cells) rather than on
rechargeable batteries. The total purchase
price of the three units of the Eagle 2 with
installed sensors was approximately $6,700.
Sperian PHD6. The PHD6 detector
measured 21.6 x 7.9 x 6.1 cm (8.5 x 3.1 x
2.4 in) and weighed 499 g (17.6 ounces).
This detector (Serial No. 531104032) was
operated on internal rechargeable battery
power, which was recharged overnight. The
-------
detector had an internal pump which drew in
sample at approximately 1.0 to 1.3 L/min
through an approximately 30 cm length of
Teflon® tubing connected to the detector's
inlet port. The sample probe provided with
the detector was not used. This detector was
capable of holding a maximum of five
sensors. The O2 and LEL sensors were
permanently installed. The other sensors
installed in the PHD6 were as follows: SO2,
NH3, and H2S sensors, during testing with
SO2 and NH3; C12, PH3, and HCN sensors,
during testing with those three TICs; and
C12, PH3, and H2S sensors, during testing
with H2S, O2, and CH4 (LEL). The PHD6
CH4 readings were displayed in %LEL,
rather than in %CH4 by volume (the PHD6
manual indicates that either unit can be
used). The %LEL readings were converted
to %CH4 for QUA determination based on
the fact that the LEL for CH4 is 5% by
volume in air. The purchase price of the
PHD6 and sensors was approximately
$2,500.
2.4 Testing Parameters
2.4.1 Test Conditions
Table 2.4-1 summarizes the temperature and
RH conditions used in testing. The same
test procedures were followed with each
target gas at each of the test conditions
denoted by an "X" in Table 2.4-1. The test
gas mixture supplied to the handheld
detectors undergoing testing had the
indicated RH, and both the challenge gas
delivery system and the handheld detectors
were maintained at the indicated test
temperature. As Table 2.4-1 shows, the test
conditions included low, medium, and high
RH at room temperature, medium RH at low
temperature, and medium and high RH at
high temperature.
Table 2.4-1. Summary of Temperature and RH Conditions for Testing
RH (%)
< 20
50 (±5)
80 (±5)
Temperature (°C)
8 (±3)
~
X
~
22 (±3)
X
X
X
35 (±3)
~
X
X
2.4.2 Chemical Interferences
Table 2.4-2 lists the six chemical mixtures
or compounds used to test the interference
response of the handheld chemical detectors:
latex paint fumes, ammonia cleaner, air
freshener, N,N-diethylaminoethanol (DEAE;
a boiler water additive found in indoor air
via humidification systems), simulated
gasoline exhaust, and simulated diesel
exhaust. Each of these interferents was
delivered to each detector along with each
target gas, and also alone in otherwise clean
air. Interferent testing used one interferent
at a time.
For the latex paint, ammonia cleaner, and air
freshener, delivery of the interference
involved sweeping saturated vapors from the
whole commercial product (obtained at a
retail outlet) into an air stream. For the
DEAE, delivery of the interference involved
sweeping saturated vapors from the neat
chemical (i.e., > 95% purity, obtained from
a commercial supplier) into an air stream.
-------
For these four interferences, the interferent
vapor generation consisted of a flow of
approximately 100 cm3/min of clean air
passing over a stirred aliquot (< 0.5 L) of the
interferent product or chemical in a glass
flask (approximately 2 L volume). The 100
cm /min flow became saturated with the
interferent vapor, and was then diluted in the
approximately 10 L/min clean air flow to the
test plenum in the test apparatus described in
Section 2.5. The simulated diesel and
Table 2.4-2. Interferences Used in Testing of Handheld Multigas Detectors
Interferent Category
Indoor contaminant
Indoor contaminant
Indoor contaminant
Indoor contaminant
Vehicle exhaust
Vehicle exhaust
Interferent
Latex paint fumes
Ammonia cleaner
Air freshener
N,N-diethylaminoethanol (DEAE)
Simulated gasoline exhaust
Simulated diesel exhaust
Source
Vapor from whole
commercial product
Vapor from whole
commercial product
Vapor from whole
commercial product
Vapor from neat chemical
Compressed gas standard
Compressed gas standard
gasoline exhaust interferences were
delivered by dilution of commercially
prepared compressed gas standards (Scott
Specialty Gases, Plumsteadville, PA) that
contain numerous individual hydrocarbon
compounds known to be present in the
respective exhaust composition. The
standards used were Department of
Homeland Security (DHS) approved Diesel
Exhaust Interferent Standard (part no.
MDHS0002-T-30AL) and DHS approved
Gasoline Exhaust Interferent Standard (part
no. MDHS0003-T-30AL).
2.4.3 Test Matrix
Table 2.4-3 summarizes the quantitative
evaluations conducted, in terms of the
performance parameters, the objective of
each parameter, and the basis of evaluating
each parameter. The test procedures
provided information on several
performance parameters simultaneously.
Operational factors were evaluated based on
qualitative observations that occurred in the
test procedures, so no testing specifically to
address those factors is included in Table
2.4-3. As the footnote to Table 2.4-3
indicates, the response threshold for the
target gas O2 was not evaluated because the
handheld detectors are intended to detect
departures of atmospheric O2 below its
normal level of approximately 20.9% by
volume; the minimum amount of O2 that can
be detected is unimportant.
The evaluations summarized in Table 2.4-3
were implemented by a series of tests
carried out with each detector, and with each
of the six TICs, O2, or CH4 as the target gas.
Table 2.4-4 shows the matrix of tests, briefly
describing each of the 20 different tests and
indicating the nature of each test in terms of
the test conditions and interferent (if any).
Tests 1 to 4 involved successively stepping
down in target gas concentration to assess
response threshold. Tests 8, 9, 11, 12, 14,
and 15 involved the interferent vapors
described in Section 2.4.2. Tests 16 to 20
involved testing at temperature and RH
conditions other than room temperature and
10
-------
involved testing at temperature and RH
conditions other than room temperature and
50% RH, as described in Section 2.4.1.
Tests 5 and 6 in Table 2.4-4 tested detection
of H2S in a reduced Oz atmosphere, and
Tests 7, 10, and 13 investigated cold start
performance with F^S as the target gas.
The seven detectors had widely differing
response ranges for the six TICs, as shown
by the range values summarized in
Appendix A. Consequently, testing with a
TIC as the target gas used TIC challenge
concentrations adapted to the ranges of each
detector. Table 2.4-5 lists the concentration
of each TIC that was used in each test with
each detector. This table illustrates the
downward steps in TIC concentrations in
Tests 1 through 4, and reiterates the fact that
Tests 5 to 7, 10, and 13 were conducted only
with H2S. In some cases, the upper range
limit of a detector for a TIC was lower than
the range limits of other detectors, so that
detector was not challenged at the highest
TIC concentrations. For example, Table
2.4-5 shows that the RKI Instruments Eagle
2 could not be tested with SO2 at 100, 50, or
20 ppm in Tests 1 to 3, respectively; all
testing of that detector with SO2 used the 5
ppm concentration introduced in Test 4.
11
-------
Table 2.4-3. Summary of Quantitative Evaluations Conducted"
Performance
Parameter
Response Time
Recovery Time
Accuracy
Repeatability
Response
Threshold
O2 Deficiency
Effects
Temperature
and RH Effects
Cold/Hot Start
Behavior
Interferent
Effects
Battery Life
Objective
Determine rise time of detector
response
Determine fall time of detector
response
Characterize agreement of
detector readings with reference
results
Characterize ability of detector to
correctly identify the target gas
Characterize consistency of
detector readings with constant
target gas concentration
Estimate minimum concentration
that produces detector response
Evaluate impact of reduced 62
environment on TIC detection
Evaluate effect of temperature and
RH on detector performance
Evaluate effect of storage
temperature on detector
performance at startup
Evaluate effect of contaminants
that may interfere with detector
performance
Determine useful operating life of
detectors on battery power
Basis for Comparison
Elapsed time to stabilization of detector
readings after onset of target gas
challenge
Elapsed time to stabilization of detector
readings after removal of target gas
challenge
Compare detector readings to known
challenge concentration
Compare detector indication to known
identity of target gas
Relative standard deviation of multiple
detector readings with constant
challenge
Stepping down in target gas
concentration until no response occurs0
Challenges with constant H2S
concentration at different C>2 levels
Conducting target gas challenges at
different temperature and RH
conditions
Same as above for response/recovery
times, repeatability, and accuracy, after
startup from storage
Sample interferents in clean air and
along with target gases
Continuous operation of detector to
depletion of batteries
(a) Testing consisted of five challenges with each target gas concentration at each test condition, alternating with
five clean air challenges.
(b) Stable reading defined as no change in detector reading for approximately 20 seconds.
(c) This parameter was not determined for O2.
12
-------
(d)
Table 2.4-4. Summary of Tests Conducted with Each Detector and Target Gas
Test
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Test Conditions
T (°C)/RH (%)a
22/50
22/50
22/50
22/50
22/50
22/50
22/50
22/50
22/50
22/50
22/50
22/50
22/50
22/50
22/50
22/20
22/80
8/50
35/50
35/80
Interferent
—
—
—
—
—
—
—
Paint Vapors
Gasoline Exhaust
Ammonia Cleaner
Diesel Exhaust
Air Freshener
DEAE
—
—
—
—
—
Additional Description
Base test
Step down in concentration
Step down in concentration
Step down in concentration
Conducted in 19% O2 atmosphere
(with H2S only)
Conducted in 16% O2 atmosphere
(with H2S only)
Cold start test after room T storage
(with H2S only)
Interferent testing
Interferent testing
Cold start test after low T
(approximately 8°C) storage
(with H2S only)
Interferent testing0
Interferent testing
Cold start test after high T
(approximately 40°C) storage
(with H2S only)
Interferent testing
Interferent testing
Testing of T/RH effects
Testing of T/RH effects
Testing of T/RH effects
Testing of T/RH effects
Testing of T/RH effects
(a) Test temperature controlled ± 3°C, test RH controlled ± 5 %RH.
(b) False positive and false negative responses assessed with interferent in clean air and in challenge with each
target gas, respectively.
(c) Interferent testing with ammonia cleaner was not conducted with C\2 as the target gas, to avoid formation of
paniculate matter.
13
-------
Table 2.4-5. Summary of TIC Challenge Concentrations (ppm) Used with Each Detector
Test Number
1
2
3
4
5 to 7, 10, 13
8,9, 11,12, 14
to20a
TIC
H2S
S02
NH3
C12
PH3
HCN
H2S
SO2
NH3
C12
PH3
HCN
H2S
SO2
NH3
C12
PH3
HCN
H2S
S02
NH3
C12
PH3
HCN
H2S
H2S
SO2
NH3
C12
PH3
HCN
BW GasAlert
Micro 5
90
100
100
10
—
-
30
50
50
3
-
15
10
20
10
1
5
5
3
5
3
—
1
-
90
90
50
100
10
5
15
Driiger
X-am 7000
90
100
100
10
50
50
30
50
50
3
20
15
10
20
10
1
5
5
o
J
5
3
—
1
~
90
90
50
100
10
20
50
Environics
ChemPro
lOOi
90
100
100
10
50
-
30
50
50
3
20
15
10
20
10
1
5
5
o
J
5
3
—
1
-
90
90
50
100
10
20
15
Industrial
Scientific
iBRID MX6
90
100
100
10
~
~
30
50
50
3
~
15
10
20
10
1
5
5
3
5
3
—
1
~
90
90
50
100
10
5
15
RAE Systems
MultiRAE
Pro
90
—
100
10
~
50
30
-
50
3
20
15
10
20
10
1
5
5
o
J
5
3
—
1
-
90
90
20
50
10
20
50
RKI
Instruments
Eagle 2
90
~
~
~
~
~
30
~
50
3
~
15
10
—
10
1
—
5
3
5
3
—
1
~
90
90
5
50
3
1
15
Sperian
PHD6
90
—
100
10
~
100
30
-
50
3
20
15
10
20
10
1
5
5
o
J
5
3
—
1
-
90
90
20
50
10
20
50
(a) With the exception that Test 11 (ammonia cleaner as interferent) was not conducted with C12 as the TIC.
-------
Table 2.4-6 shows the concentrations of C>2
and CH4 that were used in testing of
detectors (except the ChemPro lOOi) for
%C>2 and %LEL determination. The C>2
concentration was 19% in nearly all of the
C>2 tests, and the testing evaluated whether
that reduced 62 content could be accurately
determined over the range of T/RH
conditions and interferents. An C>2 level of
16% was used in Test 2 to simulate more
severe 62 depletion. Those same 62 levels
were used in Tests 5 and 6 to assess the
impact of reduced C>2 on H^S detection.
Methane levels of 1.25%, 0.5%, and 0.2%
by volume were used in the LEL testing,
corresponding to 25%, 10%, and 4%,
respectively, of the LEL for CFL;.
Table 2.4-6. Summary of Oi and CH4 Concentrations Used in %Oi and LEL Testing
Test Number3
1
2
3
5
6
8,9, 11,12, 14 to 20
Target
Gas
02
CH4
O2
CH4
CH4
O2
02
02
CH4
Concentration
(%)
19
1.25
16
0.5
0.2
19
16
19
1.25
(a) Tests 4, 7, 10, and 13 not conducted with reduced O2 level or with CH4 as target gas.
2.4.4 Test System and Procedures
The handheld detectors were tested using
test systems represented schematically in
Figure 2.4-1. The test system consists of a
challenge gas delivery system, a Nafion®
humidifier, two challenge plenums, a clean
air plenum, RH sensors, thermocouples, and
mass flow meters. The appropriate target
gas generation system, typically a
compressed gas cylinder, was selected for
the gas of interest. The target gas was then
mixed with a humidified dilution air flow
entering the challenge plenums. The test
system allows the temperature and RH of
the clean air and the challenge gas mixtures
to be controlled, multiple challenge
concentrations to be delivered, and
interferent vapors to be introduced along
with the target gases.
Two such test systems were installed in
adjacent laboratory hoods and used to
conduct testing of all seven handheld
detectors simultaneously. Figure 2.4-2 is a
photograph of the laboratory showing the
two test systems in the adjacent hoods, and
the two mass flow control modules (the
black boxes at the right center of the figure)
that controlled the clean and challenge gas
flow rates, the interferent delivery flow rate,
the humidifier flow rate, and the plenum
temperatures. The laptop computer atop
each mass flow control module continually
displayed and recorded the temperatures
15
-------
MFC Mass Flow Controller
MF Mass Flow Meter
MV Multiport Valve
Temperature
Controlled Chamber
{Pj Pressure Sensor
(l) Temperature Sensor
-non I Temperature and Relative
lm"l Humidity Sensor
© One Way Check Valve
Figure 2.4-1. Schematic of test system.
-------
Figure 2.4-2. View of two test systems in adjacent laboratory hoods for testing of handheld
detectors.
and RH readings at multiple points in the
test system. The MultiRAE Pro, Eagle 2,
and PHD6 were tested in the system in the
hood at the left in Figure 2.4-2, and the
GasAlert Micro 5, X-am 7000, iBRID MX6,
and ChemPro lOOi were tested in the system
in the hood at the right in Figure 2.4-2.
The seven detectors were not connected
directly to the test systems. Instead, clean
air or challenge gas mixtures were supplied
to each detector through an individual glass
bell-shaped tube that surrounded the intake
tube of the detector but had an internal
diameter much larger than the outer
diameter of the intake tube of the detector.
These bell tubes thus provided challenge gas
flow to each detector in excess of the
detector's intake requirement, but without
pressurization or flow disturbances. This
arrangement is illustrated in Figure 2.4-3,
which shows the detail of placement of the
Drager X-am 7000 inlet tube within the
glass bell tube connected to the flow system.
Each glass bell was connected to a four-way
valve, with which the clean air or challenge
gas could be selected for delivery to the
detector. Flow measurements conducted
before any testing took place confirmed that
excess sample flow was provided to each
detector, and that no detector responses
occurred due to the valve switching with
only clean air in the system.
17
-------
Figure 2.4-3. Example of glass bell tube placed over inlet tube of a detector
(Drager X-am 7000).
Each test with a target gas began with all
detectors in a test system sampling clean air.
The target gas mixture was then delivered to
one detector. The length of time needed by
that detector to achieve a stable quantitative
response was recorded as the response time,
and the final quantitative response of the
detector was recorded. A maximum of 3
minutes (180 sec) was allowed for the
detector to achieve its reading; if an alarm or
stable reading was not achieved within 3
minutes, the detector was switched back to
sampling clean air, and the response time
was recorded as >180 sec. The length of
time needed by the detector to return to its
baseline reading after switching back to
clean air was recorded as the recovery time.
No strict limitation was placed on the length
of the recovery time, because the challenge
gas was delivered to a second detector as the
first detector was switched back to clean air.
Thus, the second detector was responding to
the gas challenge as the first was recovering
from it. In some cases, a detector did not
completely return to its baseline reading
despite a lengthy recovery time after a
challenge. In those cases, the recovery time
is reported as a "greater than" (>) value in
seconds. The sequence of successively
challenging one detector at a time was
repeated in each test until all detectors in the
test system had been subjected to five
alternating challenges with clean air and the
challenge gas mixture.
In testing with methane and the six TICs, the
background readings of the detectors were
determined with clean air of the same RH as
18
-------
the challenge mixture. In those cases, the
target gas was not present in the sample gas
when the background reading was obtained.
In testing with O2 as the target gas, clean
humidified air was also used as the
background gas, but the target gas was
present at its normal atmospheric level
(approximately 20.9%).
Reference methods were used to quantify
the target gas concentrations in the challenge
plenum both before and after delivery of the
target gas mixture to the detectors to
confirm the challenge concentrations used.
For H2S, SO2, NH3, C12, and PH3, the
reference methods were EC detectors made
by different manufacturers than the detectors
being tested and calibrated independently of
the test gas standards. For HCN, the
reference method was gas chromatography
with flame ionization detection (GC/FID),
implemented using an HP 5890 GC in the
test laboratory. Challenge gas samples were
transferred from the test system to the GC
sample loop in Tedlar gas sampling bags.
The GC/FID was calibrated with a dedicated
HCN gas standard. The reference method
for O2 was a commercial galvanic cell,
calibrated with air. The reference method
for CH4 was a commercial LEL sensor made
by a different manufacturer than the tested
detectors, and calibrated with a dedicated
CH4 standard.
Interferent testing involved only one
interferent at a time. The target gas source
was independently controlled such that the
interferent could be introduced to the
flowing gas streams either in the absence or
the presence of the target gas. This allowed
interference effects to be evaluated with the
interferent alone, and with an interferent and
target gas together. Testing with the
interferent alone allowed evaluation of false
positive responses; testing with the
interferent and target gas together allowed
evaluation of false negatives. False positive
testing began with alternating sampling of
clean air and the interferent alone in
otherwise clean air, for a total of up to five
times each, in a procedure analogous to that
described above. However, if no false
positive response was observed after three
such test cycles, the false positive testing
was truncated at that point.
2.5 Data Acquisition
Recorded data during testing included the
times and conditions of steps in testing; the
identities of the test personnel; calibration
data and challenge gas results for the
reference methods; the responses (or lack
thereof) and response and recovery times of
the handheld detectors in each portion of the
test; and observations about ease of use,
cost, etc. These data were recorded by the
test personnel in laboratory record books
and data forms.
The acquisition of data from the handheld
detectors was tailored to the expected use of
those instruments as portable rapid-response
indicators of hazardous conditions. In such
use, the visual display of readings, coupled
with an audible or visual alarm, is the
primary data output. Consequently, test data
including both quantitative readings and the
occurrence of alarms were recorded
manually by the test personnel, on color-
coded data forms for each detector that were
prepared before testing began. An example
of such a completed data form is shown in
Appendix B. The first page of the form
records information on the date, time,
conditions, and nature of the test, the sensor
configuration of the detector in question,
and the identities of the testing personnel.
The second page records the calibration data
for the reference method used, the reference
measurement results on the target gas
mixture, and other information such as the
air flow rate of the interferent vapor source.
19
-------
The third page of the form documents the
challenge mixture, and records the detector
responses, response times, and recovery
times in successive sampling of clean air
and the challenge mixture. The contractor
Work Assignment Leader reviewed all such
data forms upon completion, and required
that any corrections be made promptly by
the testing staff.
Test personnel also filled out a test summary
form for each test that included the target
gas and test number; the identity of the test
system and mass flow control module used;
the handheld detectors being tested; the
results of daily bump tests with individual
monitors and of the confidence check of the
ChemPro lOOi; the start and end times of the
test; the test system mass flow rates at the
start and end of the test; the name of the
datalogger file that recorded all temperature,
RH, and flow readings during the test; and
the battery life indication of each detector at
the end of testing. Those summary forms
were filled out by hand, and were pasted
into the laboratory record book at the
completion of each test by means of their
peel-off adhesive backing.
All test data were transferred from the hand-
written data forms into a Microsoft® Access
database, which organized the test
information, detector responses, and
reference method results for each test
procedure. Organization of the data in this
way allowed evaluation of the performance
parameters clearly and consistently. The
accuracy of entering manually-recorded data
into the database was checked at the time the
data were entered, and a portion of the data
were also checked by the contractor QA
Manager as part of the Data Quality Audit
(Section 4.2.3).
20
-------
3.0 Statistical Calculations
The quantitative performance parameters
defined in Section 2.1 were evaluated by
statistical calculations using the test data.
These calculations were built into the
Access database compiled from the test data,
so that calculations were completed
automatically as data were entered into the
spreadsheets. The following sections define
the calculations that were conducted for
each performance parameter.
3.1 Response and Recovery Time
The data collected to evaluate response time
were the measured time periods required for
each detector to reach a stable reading after
initiation of a gas challenge. Response time
(in seconds) was measured in each of five
replicate test runs at each test condition with
each target gas, and the mean, range, and
standard deviation (SD) of the five response
times in each test were tabulated.
The corresponding data collected to evaluate
recovery time were the measured time
3.2
periods required for each detector to return
to a stable baseline reading after removal of
a gas challenge. Recovery time (in seconds)
was measured in each of five replicate test
runs at each test condition with each target
gas, and the mean, range, and SD of the five
recovery times in each test were tabulated.
When a detector failed to reach a stable
reading within 180 seconds after the start of
a gas challenge, the response time was
recorded as ">180 seconds" and the test
procedure was continued with the next clean
air sampling period. Detectors were allowed
periods of up to 15 minutes to return to
baseline after removal of a gas challenge,
while challenges were delivered to other
detectors. Failure to reach a stable baseline
reading was recorded as a ">" recovery
time. For statistical analysis, all ">"
recovery and response times were assigned
their numerical values, i.e., the ">" notation
was dropped.
Repeatability
Repeatability was calculated in terms of the percent relative standard deviation (RSD) of the
quantitative readings from five successive test runs with a detector at each test condition and
target gas concentration. That is:
Repeatability = (SD/Mean) x 100%
(1)
where SD is the standard deviation of the five quantitative readings and Mean is the arithmetic
average of those five readings.
3.3 Accuracy
The QUA of each handheld detector was calculated as a percentage in terms of the ratio of the
detector's quantitative reading to the known concentration of the target gas challenge. That is:
QUA = (Detector Reading/Known Concentration) x 100%
(2)
21
-------
QUA was calculated as the mean of the quantitative detector responses in the five replicate runs
at each test condition and target gas concentration. When a detector gave a quantitative reading
for a target gas, even though in a constant overrange condition, the QUA was calculated but
results were flagged as being underestimates of the true QUA value.
Accuracy was also assessed in terms of the percentage of tests in which each handheld detector
properly identified the target gas being delivered. IA was calculated as follows:
IA = (CI/#Tests) x 100%
(3)
where CI is the number of target gas challenges in which the detector correctly indicated the
target gas, and #Tests is the number of target gas challenges. IA was calculated in this way for
each handheld detector for each test condition and target gas concentration. That is, #Tests was
typically 5, because of the five replicates in each such test scenario. The overall IA was also
calculated by applying Equation 3 to all tests conducted with each detector.
3.4 Response Threshold
No statistical calculations were needed to
estimate the response threshold of each
handheld detector for the target gases. After
five replicate tests at approximately 22 °C
and 50% RH at an initial gas concentration
(see Table 2.4-5), the concentration was
reduced and five more replicate gas
challenges (interspersed with clean air
challenges) were conducted. This process
was repeated until a concentration was
reached at which the detector failed to
respond to the target gas in at least three of
the five challenges, or the challenge gas
concentration was as low as could
reasonably be delivered and confirmed by
the test procedures. The response threshold
is reported as an upper limit, i.e., less than or
equal to the lowest concentration tested.
3.5 Effect of Operating Conditions
The effects of temperature and RH on the
performance parameters of response time,
recovery time, repeatability, QUA, and IA
were determined by comparing the
quantitative measures of these parameters in
tests conducted at different temperature/RH
conditions. For response time, a significant
effect of test conditions was inferred when
there was no overlap between the mean (± 1
SD) ranges of the response times determined
at two different temperature/RH conditions.
The same criterion was used to judge
temperature/RH effects on recovery time.
For repeatability, QUA, and IA, a significant
effect of test conditions was inferred when
the metric calculated by means of Equations
1, 2, or 3 above, respectively, differed by
more than 20% between two sets of test
conditions.
3.6 Cold/Hot Start Behavior
The effects of storage temperature on the
performance parameters of response time,
recovery time, repeatability, QUA, and IA
were determined by comparing the
quantitative measures of these parameters in
tests conducted with H2S after overnight
storage at different conditions. One
22
-------
test run (i.e., five challenge/clean air
replicates) was conducted at the start of a
test day immediately after the detector had
been removed from cold, hot, or room
temperature storage overnight. Storage
under these three conditions took place in
three successive overnight periods, and the
detectors were tested on the corresponding
three successive mornings. The time from
initial power-up of each detector until the
detector was ready to begin monitoring was
recorded as the detector delay time. Each
detector then received a challenge gas
consisting of 90 ppm H2S in air at 22°C and
50% RH, and the response time and reading
of the detector were recorded. The
challenge gas was then replaced with clean
air and the recovery time of the detector was
recorded. Five successive alternating
readings of challenge gas and clean air were
obtained and used to determine the response
time, recovery time, repeatability, QUA, and
IA of the detector after startup from the
storage condition in question.
The results for response time, recovery time,
repeatability, QUA, and IA of the detector
after cold, hot, and room temperature
overnight storage were compared. For
response time, a significant effect of storage
conditions was inferred when there was no
overlap between the mean (± 1 SD) ranges
of the response times determined with two
different storage conditions. The same
criterion was used to judge storage condition
effects on recovery time. For repeatability,
QUA, and IA, a significant effect of storage
conditions was inferred when the metric
calculated by means of Equations 1, 2, or 3
above, respectively, differed by more than
20% between two sets of results obtained
after different storage conditions.
3.7
Interference Effects
Interference effects were calculated in terms
of the rates of false positive and false
negative responses from each handheld
detector.
False positive rates were determined based
on the response of the handheld detectors to
air containing only the interferent vapors.
For each detector, the false positive rate (FP)
was calculated as:
FP = (PR/I)x 100%
where PR is the number of positive
responses observed when sampling air
containing the interferent, and I is the total
number of such samples (I usually = 5).
This calculation was done for each detector
with each target gas.
False negative rates were determined based
on the absence of handheld detector
response to a known concentration of each
target gas, when the interferent was present
along with the target gas. For each detector,
the false negative rate (FN) was calculated
as:
FN = (NR/TI)x 100%
where NR is the number of negative
responses (i.e., failures to detect a hazardous
concentration), and TI is the total number of
samples tested containing both the target gas
and the interferent (TI = 5). This calculation
was done for each detector with each target
gas.
3.8 Battery Life
The battery life of each handheld detector
was quantified in terms of the hours and
minutes of continuous operation achieved
before the battery was depleted. Battery life
was determined by starting with a fully
charged battery or set of batteries and
operating the detector until the battery
supply was exhausted and the detector shut
down.
23
-------
requirement helped ensure that testing was
In addition, in all testing, the battery status always conducted with properly operating
indication of each handheld detector was detectors. Testing of a detector did not take
noted on the data recording form at the place unless its batteries showed a sufficient
beginning and end of each set of test runs. charge indication.
In addition to providing information about
battery depletion during testing, this
24
-------
4.0 Quality Assurance/Quality Control
QA/quality control (QC) procedures were
performed in accordance with the applicable
QMP and the test/QA plan for this
evaluation. QA/QC procedures and
associated results are summarized below.
4.1 Data Quality Indicators
The testing reported here consisted of more
than 800 separate tests, each consisting of
five alternate blank and challenge runs with
one detector, one challenge gas, and one
combination of T/RH conditions, interferent,
and startup condition. All of those tests met
the requirements for data quality stated in
the test/QA plan. Specifically, all
temperature and RH conditions in testing
were within 3°C and 5% RH, respectively,
of the relevant target conditions stated in
Table 2.4-1.
The reference methods similarly confirmed
that the delivered challenge concentrations
closely matched the target concentrations
listed in Table 2.4-5. This fact is illustrated
in Table 4.1-1, which shows the number,
mean, SD, RSD, and median of the
reference method results for each target gas
at each concentration at which enough
testing was done to develop these statistics.
Table 4.1-1. Summary Statistics of Reference Method Results
Gas
02
CH4
H2S
SO2
NH3
C12
PH3
HCN
Target
Concentration
19%
1.25 %
90ppm
50 ppm
20ppm
5 ppm
100 ppm
50 ppm
10 ppm
3 ppm
20 ppm
50 ppm
15 ppm
Reference Method Results
Number
88
64
136
44
48
48
48
48
96
46
95
96
96
Mean"
19.0
1.22
89.1
49.9
18.7
4.63
99.5
50.3
9.9
2.83
19.3
48.5
15.1
SDa
0.07
0.04
2.29
2.52
0.94
0.57
7.32
3.02
0.51
0.11
0.74
1.54
0.64
RSD (%)
0.4
3.7
2.6
5.1
5.0
12.3
7.4
6.0
5.1
4.0
3.9
3.2
4.2
Median"
19.0
1.22
89.0
50.0
18.5
5.0
98
51
9.9
2.80
19.0
48.7
15.3
(a) Units as indicated in Target Concentration column.
Table 4.1-1 shows that mean and median
reference method results for all target gases
closely matched the target concentrations.
Also the uniformity of concentrations was
maintained, as indicated by the RSD values
in Table 4.1-1, most of which are less than
about 5 percent. The delivered target gas
concentrations were well within the target
delivery tolerance of 20 percent specified in
the test/QA plan. In fact 97.7 percent of the
reference measurements for all target gases
at all concentrations fell within 10% of the
target concentration, and 76.9% fell within 5
percent of the target concentration.
25
-------
4.2 Audits
4.2.1 Performance Evaluation Audit
A performance evaluation (PE) audit was
conducted to assess the quality of the
measurements and gas challenges made in
this project. The audit addressed only those
reference measurements that factored into
the data used for evaluation, i.e., the
handheld detectors were not the subject of
the PE audit. The PE audit was performed
by analyzing a standard or comparing to a
measurement device that was independent of
standards used during the testing. Table 4.2-
1 summarizes the PE audits that were done
and indicates the PE audit standard or
measurement device, the standard or device
used in testing, the value or concentration
level at which the PE audit comparison was
done, the target degree of agreement for the
PE audit, and the observed degree of
agreement with the PE audit standard. This
audit was conducted by the contractor's
testing staff.
The PE audit standards for methane and the
six TICs were gaseous standards of those
compounds, obtained from commercial
suppliers and distinct from the standards
used for reference method calibrations. The
PE audits for those gases were conducted by
diluting the PE audit standard and the test
standard to the same concentration and
analyzing both by the reference method.
The PE audit standard for O2 was ambient
air, with a known O2 content of 20.9%. The
PE standard for temperature and RH was a
calibrated monitoring device for those
parameters.
Table 4.2-1 shows that the PE audit results
for O2, CH4, SO2, NH3, PH3, HCN,
temperature, and RH were all well within
the respective target range of agreement, and
the PE audit result for C12 was only slightly
outside the target range. However, the PE
audit result for H2S was substantially greater
than the target. The PE audit comparison
for H2S (and for some of the other TICs)
was challenging because the PE audit gas
standard differed widely in concentration
from the test gas standard with which it was
compared. This difference required quite
different dilution steps to prepare the desired
TIC concentration for the PE audit. A
comparably large PE audit result (i.e.,
agreement within about 18%) was originally
found for HCN when comparing a 500 ppm
PE audit standard to the 1% HCN test
standard. However, the agreement shown in
Table 4.2-1 (i.e., within 2.7%) was found for
HCN when a PE audit gas of 1% HCN
concentration was obtained. Similar
agreement would be expected in the PE
audit results for H2S had a PE audit standard
closer to 1% concentration been available in
the course of this project. That is, the
relatively high PE audit result for H2S is
believed to be due to the difference in PE
and test gas standards, and not to the
accuracy of the test gas standard itself.
Overall, Table 4.2-1 confirms the validity of
the test gas standards and measurement
devices used in the test.
26
-------
Table 4.2-1. Summary of PE Audit Results
Parameter
O2
CH4
H2S
SO2
NH3
C12
PH3
HCN
Temperature3
RHa
PE Audit
Standard or
Device
Ambient Air
99% CH4
Cylinder
923103L
l,OOOppmH2S
Cylinder
ALM065847
l,OOOppmSO2
Cylinder
ALM058997
28% NH3
Cylinder
ALM033321
490 ppm C12
Cylinder XA6266
500 ppm PH3
Cylinder
CC88366
0.998% HCN
Cylinder D735
Vaisala C20972
Vaisala C20972
Test Standard
or Device
Drager
PAC-III
10% CH4
Cylinder
ALM035787
0.999% H2S
Cylinder
ALM016184
1% SO2
Cylinder
A2316
27.8% NH3
Cylinder
ALM055009
1.00%C12
Cylinder
B6237
1%PH3
Cylinder
A1809
1%HCN
Cylinder
1A9405
Vaisala
C21552
Vaisala
C20749
Vaisala
C21552
Vaisala
C20749
Test Value
or Condition
20.9% O2
1% CH4
50 ppm H2S
100 ppm SO2
278 ppm NH3
10 ppm C12
20 ppm PH3
50 ppm HCN
21 °C
50 % RH
Target
Agreement
± 1% O2
± 10 %
± 10 %
± 10 %
± 10 %
± 10 %
± 10 %
± 10 %
±2°C
± 5% RH
Actual
Agreement
0.0% O2
2.3 %
17.5 %
4.0 %
0.9%
10.5 %
5.8 %
2.7%
0.1 °C
0.0 °C
0.2 % RH
0.2 % RH
(a) Dual entries indicate audit of the temperature/RH monitoring units in the two test systems.
4.2.2 Technical Systems Audit
The contractor QA Manager conducted a
technical systems audit (TSA) of the test
procedures in the test laboratory on July 26,
2011, to ensure that the evaluation was
being conducted in accordance with the
test/QA plan and the QMP. As part of the
TSA, test procedures were compared to
those specified in the test/QA plan, and data
acquisition and handling procedures were
reviewed. Observations from this TSA were
documented in a report which was submitted
to the Work Assignment Leader for
response. No adverse findings resulted from
this TSA. However, two deviations were
prepared and approved documenting slight
differences between actual test procedures
and those stated in the test/QA plan. One
deviation addressed the use of Tedlar bags
rather than a gas-tight syringe for collection
of gas samples for reference analysis. The
other deviation addressed the procedure for
conducting cold-start tests on the handheld
27
-------
detectors, which was incorrectly described
in the test/QA plan. TSA records were
permanently stored with the contractor QA
Manager.
4.2.3 Data Quality Audit
At least 10% of the data acquired during the
evaluation were audited. A contractor QA
auditor traced the data from the initial
acquisition, through reduction and statistical
analysis, to final reporting to ensure the
integrity of the reported results. All
calculations performed on the data
undergoing the audit were checked.
4.2.4 QA/QC Reporting
Each audit was documented in accordance
with the QMP. The results of the audits
were submitted to EPA (i.e., to the NHSRC
Quality Assurance Manager and the EPA
Contracting Officer's Representative
[COR]).
4.3 Data Review
As described in Section 2.5, all detector test
conditions, reference method results, and
detector responses were recorded by the
testing personnel in laboratory record books,
on test summary forms, and on pre-printed
data forms that were color-coded for the
seven detectors (see example in Appendix
B). The testing personnel initialed and dated
every page of every data form during the
data recording process. All record books,
data forms, and summary forms were then
subjected to a QC/technical review by the
Work Assignment Leader, who clarified and
corrected any incomplete or unclear entries
through discussions with test personnel and
review of other records (e.g., datalogger
files). The hard copy data forms were then
scanned and converted to electronic (i.e.,
pdf) format. The data were then entered into
a Microsoft® Access database and used in
assessing detector performance. All data
recording and review were performed by
contractor staff. Entry of data from the data
sheets into the Access database was
performed by subcontractor staff, under the
supervision of, and subject to review by, the
contractor staff.
28
-------
5.0 Results
This section summarizes the performance
results for each detector with each challenge
gas. The following sections address the
several performance parameters stated in
Section 2.1.
5.1 Response and Recovery Time
Tables 5.1-1 through 5.1-8 summarize the
mean response and recovery times observed
with each detector in each test with H2S,
S02, NH3, C12, PH3, HCN, O2, and CH4,
respectively. For each detector and test
condition, the mean response and recovery
time are shown in seconds. Note that each
detector was given up to 180 seconds to
respond to the challenge gas mixture before
switching back to the clean air challenge.
Consequently, 180 seconds is the maximum
value recorded for response time. The
testing procedures allowed considerable
time for recovery after a challenge, so the
maximum recorded recovery times of the
detectors were several minutes long.
Table 5.1-1 through 5.1-8 show that the
response and recovery times of the detectors
varied widely depending on the challenge
gas and test conditions, but that response
times were generally much shorter than
recovery times. Relatively rapid response
and recovery were observed with all
detectors with O2 and CH4 (Tables 5.1-7 and
5.1-8, respectively); relatively slow response
and recovery were seen with all detectors
with NH3 (Table 5.1-3).
Table 5.1-1 shows that the ChemPro lOOi
and RAE MultiRAE Pro responded most
rapidly to H2S (i.e., usually within 20
seconds), but the MultiRAE Pro recovery
times were much longer than those of the
ChemPro lOOi. The iBRID MX6 exhibited
relatively long response and recovery times
with H2S.
Table 5.1-2 shows that the Eagle 2 and
PHD6 responded most rapidly to SO2, and
the ChemPro lOOi responded rapidly in
some tests but showed no response in others.
Response times with SO2 were longest with
the iBRID MX6, and recovery times were
longest with the iBRID MX6, ChemPro
lOOi, MultiRAE Pro, and Eagle 2.
Table 5.1-4 shows widely varying response
and recovery times for C12, with the notable
finding that recovery times were shorter than
response times in many tests with the
GasAlert Micro 5 and iBRID MX6. The
ChemPro lOOi responded relatively rapidly
in some tests but showed no response in
others. The longest recovery times after C12
challenges were observed with the X-am
7000, MultiRAE Pro, and Eagle 2.
Table 5.1-5 indicates the shortest response
and recovery times (i.e., often less than 15
seconds) with PH3 were with the GasAlert
Micro 5 and Eagle 2 detectors. The
detectors generally responded to PH3
challenges relatively rapidly (compared to
response times with other TICs), but some
recovery times exceeding 300 seconds were
observed, especially with the PHD6
detector.
Table 5.1-6 shows that the X-am 7000 and
ChemPro lOOi responded most rapidly to
HCN, with response times often
approximately 20 seconds or less. The
iBRID MX6 exhibited the slowest
response, often not reaching stable response
within 300 seconds. Recovery times with
HCN often ranged from about 300 to 500
29
-------
seconds with the iBRID MX6, MultiRAE
Pro, PHD 6, and X-am 7000. Both response
and recovery times with HCN were usually
less than 100 seconds for the GasAlert
Micro 5 and Eagle 2.
30
-------
Table 5.1-1. Summary of Mean Response and Recovery Times (seconds) with HiSa
Test
Number
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Test Description
Base test, 90 ppm
Step down, 30 ppm
Step down, 10 ppm
Step down, 3 ppm
H2S, 19%O2
H2S, 16% O2
H2S, room T start
Paint vapors
Gasoline exhaust
H2S, low T start
Ammonia cleaner
Diesel exhaust
H2S, high T start
Air freshener
DEAE
RoomT, <20%RH
Room T, 80% RH
Low T, 50% RH
High T, 50% RH
High T, 80% RH
BW GasAlert
Micro 5
50
35
26
19
49
38
43
45
49
66
37
44
33
37
43
53
43
41
36
35
46
25
14
7
45
47
51
50
62
55
55
63
62
51
53
50
50
52
45
48
Driiger
X-am 7000
29
101
75
50
56
55
24
31
29
127
38
24
25
25
22
19
91
20
76
180
95
59
29
19
119
106
139
99
109
403
118
110
93
104
94
94
107
100
190
340
Environics
ChemPro lOOi
19
18
19
19
18
18
18
18
18
19
18
18
19
19
18
18
19
18
18
18
30
50
34
101
46
102
92
47
56
71
68
55
157
24
216
29
34
28
58
66
Industrial
Scientific
iBRID MX6
121
68
49
44
82
70
88
68
73
86
71
63
94
73
70
88
103
105
84
81
342
122
69
38
701
754
>849b
581
565
>884b
635
655
892b
695b
589
353
582
672
460
590
RAE Systems
MultiRAE
Pro
18
19
14
71
14
14
18
15
41
22
15
15
19
13
14
16
14
15
14
16
319
162
67
29
279
281
324
328
367
420
319
316
300
348
323
323
309
317
351
359
RKI
Instruments
Eagle 2
20
17
15
180
29
27
38
12
62
36
13
14
51
11
14
16
15
14
14
28
32
20
13
9
34
34
33
32
32
60
30
31
36
33
32
39
33
42
34
35
Sperian
PHD6
30
28
25
42
34
32
41
26
99
37
27
27
39
23
28
27
28
25
28
29
107
74
50
23
113
116
112
110
117
119
110
108
115
113
110
117
114
130
114
117
(a) Entries are mean response time and mean recovery time, in seconds, from five replicate challenges in each test.
(b) Response not fully cleared after 15 minutes or more on clean air following one or more TIC challenges.
-------
Table 5.1-2. Summary of Mean Response and Recovery Times (seconds) with SOia
Test
Number
1
2
3
4
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 100 ppm
Step down, 50 ppm
Step down, 20 ppm
Step down, 5 ppm
Paint vapors
Gasoline Exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
RoomT, <20%RH
RoomT, 80% RH
LowT, 50% RH
HighT, 50% RH
High T, 80% RH
BW GasAlert
Micro 5
76
33
36
11
180
48
59
40
135
180
29
99
95
85
172
126
71
37
12
88
44
63
44
86
84
52
126
128
119
176
Drager X-am
7000
64
40
44
32
80
134
52
93
82
98
32
134
45
97
150
711
503b
287
130
582
495
>496
431
>585
645
>597
>600
>616
>675
>510
Environics
ChemPro lOOi
21
28
NR
NR
24
NR
17
17
16
16
NR
33
20
31
19
165
383
NR
NR
156
NR
390b
365b
494b
356b
NR
30
560b
79
106
Industrial
Scientific
iBRID MX6
91
32
30
15
165
180
62
178
106
183
37
63
60
55
41
403
254
128
53
430
455
289
466
352
404
240
359
>548
273
248
RAE Systems
MultiRAE
Pro
NT
NT
37
35
44
31
28
30
31
27
39
55
45
55
75
NT
NT
260
124
315
215
219
254
230
201
401
261
345
273
405
RKI
Instruments
Eagle 2
NT
NT
NT
13
12
12
12
11
13
13
12
23
12
15
33
NT
NT
NT
342
232C
178
221
244
261
206
460
354
369
265
308
Sperian
PHD6
NT
NT
22
15
27
17
15
14
16
15
24
23
28
38
29
NT
NT
117
52
88
75
80
74
73
74
117
134
92
141
95
to
(a) Entries are mean response time and mean recovery time, in seconds, from five replicate challenges in each test. NR = no response, NT = not tested.
(b) Response not fully cleared after 15 minutes or more on clean air following one or more TIC challenges.
(c) Based on less than five recovery time values.
-------
Table 5.1-3. Summary of Mean Response and Recovery Times (seconds) with
Test
Number
1
2
3
4
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 100 ppm
Step down, 50 ppm
Step down, 10 ppm
Step down, 3 ppm
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
Room T, <20% RH
Room T, 80% RH
LowT, 50% RH
HighT, 50% RH
High T, 80% RH
BW GasAlert
Micro 5
154
>180
159
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
800b'c
893
198
118
>750C
>910b'c
>810C
>672C
> 828b'c
>818b'c
>823b'c
> 1320b'c
692
1049b
>330C
Drager X-am
7000
>180
>180
>180
>180C
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
321
252
194
66C
390
283
334
339
321
243
198
>1132b
291
>1026b
>900b'c
Environics
ChemPro lOOi
54
120
137
NR
28
30
49
22
41
45
33
66
36
59
83
118
143
52
NR
330
171
158
168
169
174
98
381
119
250
332
Industrial
Scientific
iBRID MX6
101
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
>780C
>1152b
>810b'c
>917b
>765C
>915b'c
>825C
>630C
>915b'c
>840b'c
>825b'c
>1230b,c
>840b'c
>1112b'c
>900b'c
RAE Systems
MultiRAE
Pro
>180
82
169
66
>180
>180
>180
>180
>180
>180
139
>180
>180
>180
>180
>468
>355
274
91
>360
>504
>420
>408
>420
>420
>558b
>450
>426
>408
>426
RKI
Instruments
Eagle 2
NT
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
>180
NT
>567
200
99
>360
>468
>408
>384
>396
>348
>764b
>420
>408
>408
>396
Sperian
PHD6
>180
161
>180
161
>180
133
>180
131
>180
>180
155
>180
>180
>180
>180
>492b
>517
182
47
>360
>471
>384
>377
>432
>360
>590b
>366
>370
>408
>372
(a) Entries are mean response time and mean recovery time, in seconds, from five replicate challenges in each test. NR = no response, NT = not tested.
(b) Response not fully cleared after 15 minutes or more on clean air following one or more TIC challenges.
(c) Based on less than five response or recovery time values.
-------
Table 5.1-4. Summary of Mean Response and Recovery Times (seconds) with Cba
Test
Number
1
2
3
8
9
12
14
15
16
17
18
19
20
Test Description
Base test, 10 ppm
Step down, 3 ppm
Step down, Ippm
Paint vapors
Gasoline exhaust
Diesel exhaust
Air freshener
DEAE
RoomT, <20%RH
Room T, 80% RH
Low T, 50% RH
High T, 50% RH
High T, 80% RH
BW GasAlert
Micro 5
54
29
NR
46
36
87
42
38
180
180
51
152
180
22
5
NR
21
20
33
19
24
35
27
38
25
14
Drager X-am
7000
>180
169
44
>180
>180
119
>180
>180
104
>180
41
>180
>180
>726b
320
16
>600C
>686C
122
>600C
>735b'c
>675C
>702b,c
124
>888b
211
Environics
ChemPro lOOi
24C
17c,d
NRd
21
NR
NRd
28
40
24
69C
47
NR
NR
116C
55c'd
NRd
39
NR
NRd
128
86
169
43C
75
NR
NR
Industrial
Scientific
iBRID MX6
>180
158
59
111
180
23
107
133
12
32
>180
161
>180
48
12
8
71
86
49
90
88
29
43
168
37
31
RAE Systems
MultiRAE
Pro
63
105
>180
86
78
33
40
37
19
71
25
43
>180
>320
208
111
>401
>436
>328
>335
410
>325
>417
>540
188
191
RKI
Instruments
Eagle 2
NT
38
34
68
71
80
>180
>180
61
>180
102
>180
>180
NT
>424
>396
>574b
>426
>391
>408
>390
>370
>324
>468
>392
>348
Sperian
PHD6
69
118
>180
86
95
28
22
19
18
19
>180
13
15
57
18
10
43
43
27
32
24
27
30
41
26
33
(a) Entries are mean response time and mean recovery time, in seconds, from five replicate challenges in each test. NR = no response, NT = not tested.
(b) Response not fully cleared after 15 minutes or more on clean air following one or more TIC challenges.
(c) Based on less than five response or recovery time values.
(d) Alarmed during clean air sampling in at least one challenge.
-------
Table 5.1-5. Summary of Mean Response and Recovery Times (seconds) with
Test
Number
1
2
3
4
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 50 ppm
Step down, 20 ppm
Step down, 5 ppm
Step down, 1 ppm
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
Room T, <20% RH
Room T, 80% RH
LowT, 50% RH
HighT, 50% RH
High T, 80% RH
BW GasAlert
Micro 5
NT
NT
12
8
6
6
5
5
5
5
8
6
8
6
5
NT
NT
14
3
7
8
9
7
8
8
10
9
13
7
5
Driiger
X-am 7000
33
37
32
50
29
25
26
23
28
27
29
29
26
33
32
31
24
23
7
23
25
25
25
24
24
24
24
25
22
23
Environics
ChemPro lOOi
17
78
18
18
17
18
18
18
18
18
17
20
18
16
18
273
212
55
43
231
142
138
138
151
147
360
186
366
117
78
Industrial
Scientific
iBRID MX6
NT
NT
>180
69
50
50
57
51
64
62
127
>180
59
>180
48
NT
NT
56
16
42
38
45
37
47
41
45
134
45
51
78
RAE Systems
MultiRAE Pro
NT
43
56
19
24
22
19
25
16
23
24
22
25
19
18
NT
220
59
26
141
177
256
124
83
226
107
82
104
105
128
RKI
Instruments
Eagle 2
NT
NT
NT
10
10
10
11
11
10
9
9
10
10
10
10
NT
NT
NT
11
12
36
13
54
90
12
14
12
14
14
14
Sperian
PHD6
NT
125
66
50
60
60
65
67
80
57
101
86
106
>180
83
NT
>425
137
41
>330
>348
>320
>322
>367
>309
>420
>331
>300
>372
>396
(a) Entries are mean response time and mean recovery time, in seconds, from five replicate challenges in each test. NR = no response, NT = not tested.
-------
Table 5.1-6. Summary of Mean Response and Recovery Times (seconds) with HCNa
Test
Number
1
2
o
J
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 50 ppm
Step down, 15 ppm
Step down, 5 ppm
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
Room T, <20% RH
Room T, 80% RH
Low T, 50% RH
High T, 50% RH
High T, 80% RH
BW GasAlert
Micro 5
NT
65
31
26
44
41
44
24
29
52
54
58
31
27
NT
79
24
47
63
59
67
53
52
74
66
136
34
36
Drager X-am
7000
20
>180
69
16
17
17
17
17
17
16
21
15
20
41
>345C
>765C
>600C
>328
291
339
291
305
301
>340C
266
27
291
364
Environics
ChemPro lOOi
NT
18
21
30
NR
17
16
17
17
27
26
37
13C
18C
NT
133
120
96
NR
123
73
92
91
63C
119
78
109C
163C
Industrial
Scientific
iBRID MX6
NT
>180
154
84
>180
>180
>180
89
90
>180
>180
>180
>136
148
NT
>679C
287
342
>533
435
>545
>392
348
>420C
>680b
>570C
225
142
RAE Systems
MultiRAE Pro
117
53
77
111
115
93
119
91
86
54
45
69
86
119
240
82
83
>410
>502
>474
>394
>398
>396
274
296
>413
>361
>352C
RKI
Instruments
Eagle 2
NT
149
51
102
80
83
63
82
52
48
45
55
48
71
NT
102
17
52
46
43
41
97
39
62
45
113
36
29
Sperian
PHD6
52
53
135
39
41
28
42
35
37
102
93
43
29
97
>437
181
141
>438
>485
>447
>474
>418
>405
>335
>342
>398
179
186
(a) Entries are mean response time and mean recovery time, in seconds, from five replicate challenges in each test. NR = no response, NT = not tested.
(b) Response not fully cleared after 15 minutes or more on clean air following one or more TIC challenges.
(c) Based on less than five response or recovery time values.
-------
Table 5.1-7. Summary of Mean Response and Recovery Times (seconds) with Oia
Test
Number
1
2
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 19%O2
Step down, 16% O2
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
RoomT, <20%RH
Room T, 80% RH
Low T, 50% RH
High T, 50% RH
HighT, 80% RH
BW GasAlert
Micro 5
NT
NT
21
21
20
NT
18
21
16
21
23
13
20
NT
NT
11
12
10
NT
13
12
10
18
9
12
28
Drager X-am
7000
NT
NT
33
30
29
NT
28
26
24
39
42
28
33
NT
NT
19
20
19
NT
20
20
24
22
19
19
26
Industrial Scientific
iBRID MX6
NT
NT
25
29
25
NT
32
32
22
28
30
18
29
NT
NT
50
26
43
NT
30
29
27
39
47
26
46
RAE Systems
MultiRAE Pro
10
9
16
8
27
17
10
6
13
11
12
9
8
10
15
9
12
11
11
10
12
11
10
10
10
17
RKI Instruments
Eagle 2
9
9
9
7
9
8
8
6
24
7
10
7
14
9
10
8
9
10
8
7
8
7
9
8
7
7
Sperian PHD6
22
29
46
9
43
13
13
9
30
20
23
16
27
63
147
55
>443
55
53
24
235
34
62
256
19
40
(a) Entries are mean response time and mean recovery time, in seconds, from five replicate challenges in each test. NT = not tested.
-------
Table 5.1-8. Summary of Mean Response and Recovery Times (seconds) with
Test
Number
1
2
3
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 1.25%
Step down, 0.5%
Step down, 0.2%
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
Room T, <20% RH
Room T, 80% RH
Low T, 50% RH
HighT, 50% RH
HighT, 80% RH
BW GasAlert
Micro 5
16
14
15
16
15
15
NT
13
15
19
15
15
19
18
12
10
8
15
13
16
NT
13
13
20
13
13
>375b
>375b
Drager X-am
7000
35
35
31
NT
NT
NT
NT
NT
NT
41
46
47
44
NT
38
28
21
NT
NT
NT
NT
NT
NT
33
49
37
>285b
NT
Industrial Scientific
iBRID MX6
27
25
24
27
26
25
NT
26
27
26
39
27
28
30
19
21
15
20
21
22
NT
22
21
20
26
22
24
>345b
RAE Systems
MultiRAE Pro
20
18
>180b
22
15
21
NT
20
23
22
11
24
24
22
13
12
8
17
12
22
NT
17
11
8
>387
10
>353
>355
RKI Instruments
Eagle 2
10
9
8
21
14
18
NT
15
11
10
10
9
10
9
17
14
14
16
19
15
NT
16
19
18
382
19
16
23
Sperian PHD6
10
12
11
14
19
20
NT
14
21
11
16
10
30
28
18
14
12
20
19
19
NT
20
19
19
20
20
16
>326
oo
(a) Entries are mean response time and mean recovery time, in seconds, from five replicate challenges in each test. NT = not tested.
(b) Based on less than five response or recovery time values.
-------
5.2 Accuracy
Tables 5.2-1 through 5.2-8 summarize the
QUA and IA observed with each detector in
each test with H2S, SO2, NH3, C12, PH3,
HCN, O2, and CH4, respectively. Both
measures of accuracy are shown in percent,
as calculated using Equations 1 and 2, with
100% representing ideal performance. QUA
was not determined in testing of the
ChemPro lOOi with the six TICs, as that
detector provides a qualitative indicator of
the intensity of response (i.e., one to three
bars) rather than a concentration
measurement. In addition, IA was
determined for the ChemPro lOOi based on
its indication of "Toxic Hazard" because
that detector does not identify the specific
TIC being detected when operated in the
First Responder library as in this test.
Table 5.2-1 shows that the QUA values for
the GasAlert Micro 5 and iBRTD MX6 for
H2S were high, usually exceeding 150%,
whereas QUA for the PHD6 were almost
entirely in the range of 100 to 130%. The
X-am 7000, MultiRAE Pro, and Eagle 2
gave overrange or pegged full-scale readings
in many tests, even though the 90 ppm H2S
challenge concentration was within their
nominal detection range.
Table 5.2-2 shows that QUA was near 100%
in all tests with SO2 with the GasAlert Micro
5, X-am 7000, iBird MX6, and PHD6.
QUA values were also near 100% with the
MultiRAE Pro and Eagle 2, but many of
those values resulted from pegged full-scale
reading from those detectors.
Table 5.2-3 shows that QUA values for NH3
were consistently near 100% with the X-am
7000, MultiRAE Pro, Eagle 2, and PHD6.
The QUA results for the GasAlert Micro 5
and iBRID MX6 were consistently less than
100%, with some of the iBRID MX6 results
falling below 50%.
Table 5.2-4 shows that the QUA values for
C12 with the GasAlert Micro 5, MultiRAE
Pro, and PHD6 were most consistently near
100%. Some QUA values with the Eagle 2
were near 100%, but values also ranged as
low as 25% with air freshener vapor as
interferent. QUA values for C12 were
variable with the X-am 7000 (i.e., 29 to
148%) and the iBRID MX6 (i.e., 74 to
210%).
Table 5.2-5 shows that QUA values for PH3
were consistently near 100% with the X-am
7000 and iBRID MX6, and were also near
100% with the MultiRAE Pro and Eagle 2,
but many of those values resulted from
pegged full-scale readings from those
detectors. The QUA values from the PHD6
were often relatively low, and the GasAlert
Micro 5 displayed an overrange condition in
most tests.
Table 5.2-6 shows that QUA values for
HCN were consistently near 100% with the
GasAlert Micro 5, iBRID MX6, and PHD6,
but were consistently below 100% with the
MultiRAE Pro, and more so for the Eagle 2.
The X-am 7000 showed an overrange
indication in almost all tests with HCN.
Table 5.2-7 shows QUA values near 100%
for O2 with all detectors. Table 5.2-8 shows
the Eagle 2 and PHD6 produced QUA
results closest to 100% for CH4, with the
GasAlert Micro 5, X-am 7000, and iBIRD
MX6 QUA values often over 150%. Most
MultiRAE Pro QUA values were well below
100%. The CH4 tests were done after all
other testing was completed, so the LEL
sensors in most of the detectors had been
previously exposed to all other test
challenges. Failure of the LEL sensor
39
-------
caused several tests to be cancelled with the
X-am 7000, and LEL sensor failure is
suspected as the explanation for the low
QUA values with the MultiRAE Pro, as
QUA results for CH4 with that detector
declined in chronological order of the tests.
IA was 100% (i.e., the detectors correctly
identified the gas challenge in all trials) in
almost all tests. Excluding the lowest
concentration step-down tests (i.e., Tests 3
and 4), IA was 100% for all challenges
under all test conditions with the GasAlert
Micro 5, X-am 7000, iBRID MX6,
MultiRAE Pro, Eagle 2, and PHD6. Other
than in the step-down tests, the only cases of
IA less than 100% were with the ChemPro
lOOi, which failed to indicate a hazard, or to
respond at all, in some tests with 862, NHs,
C12, and HCN (Tables 5.2-2, 5.2-3, 5.2-4,
and 5.2-6, respectively). Those cases with
the ChemPro lOOi occurred in tests that
involved interferent vapors, or temperature
and RH conditions other than 22°C and 50%
RH.
40
-------
Table 5.2-1. Summary of Quantitative Accuracy and Identification Accuracy (percent) with HiSa
Test
Number
1
2
o
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Test Description
Base test, 90 ppm
Step down, 30 ppm
Step down, 10 ppm
Step down, 3 ppm
H2S, 19%O2
H2S, 16% O2
H2S, room T start
Paint vapors
Gasoline exhaust
H2S, low T start
Ammonia cleaner
Diesel exhaust
H2S, high T start
Air freshener
DEAE
RoomT, <20%RH
Room T, 80% RH
LowT, 50% RH
HighT, 50% RH
High T, 80% RH
BW GasAlert
Micro 5
QUA
179
183
168
133
152
126
178
177
199
172
169
193
170
178
200
202
203
204
172
173
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Drager
X-am 7000
QUA
OR
127
110
73
108
92
OR
OR
OR
89
OR
OR
OR
OR
OR
OR
94
OR
OR
59
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Environics
ChemPro
lOOi
QUA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Industrial
Scientific
iBRID MX6
QUA
163
168
165
140
129
109
148
152
171
141
144
166
142
153
171
181
181
178
158
158
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
RAE Systems
MultiRAE Pro
QUA
lllb
135
139
97
lllb
lllb
lllb
lllb
102
lllb
lllb
lllb
lllb
lllb
lllb
lllb
lllb
lllb
lllb
lllb
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
RKI
Instruments
Eagle 2
QUA
108
107
112
17
102
102
101
111°
79
101
111°
111°
95
111°
111°
111°
111°
111°
111°
104
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Sperian
PHD6
QUA
126
126
130
100
117
118
113
133
94
114
129
131
114
141
128
130
128
137
129
128
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
(a) Entries are mean quantitative accuracy (QUA) and mean Identification Accuracy (IA) from five replicate challenges in each test.
(b) MultiRAE read 99.9 ppm (indicating overrange condition) in all challenges with 90 ppm H2S.
(c) Eagle 2 read 100.0 ppm (indicating overrange condition) in all challenges with 90 ppm H2S.
NA Not Applicable. OR overrange (i.e., offscale) indication, no numerical reading.
-------
Table 5.2-2. Summary of Quantitative Accuracy and Identification Accuracy (percent) with SOi"
Test
Number
1
2
3
4
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 100 ppm
Step down, 50 ppm
Step down, 20 ppm
Step down, 5 ppm
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
Room T, <20% RH
RoomT, 80% RH
LowT, 50% RH
High T, 50% RH
High T, 80% RH
BW
GasAlert
Micro 5
QUA
99
104
104
120
97
88
99
92
98
96
103
100
101
100
98
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Drager X-am
7000
QUA
100
107
106
117
102
97
104
101
103
101
109
100
109
105
99
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Environics
ChemPro lOOi
QUA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
IA
100
100
0
0
100
0
80
100
60
100
0
80
80
80
100
Industrial
Scientific
iBRID MX6
QUA
100
107
108
140
100
92
103
96
101
99
104
103
105
105
105
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
RAE Systems
MultiRAE
Pro
QUA
NT
NT
100b
116
100b
100b
100b
100b
100b
100b
100b
100b
100b
100b
100b
IA
NT
NT
100
100
100
100
100
100
100
100
100
100
100
100
100
RKI
Instruments
Eagle 2
QUA
NT
NT
NT
120C
120C
120C
120C
120C
116
114
120C
118
120C
113
117
IA
NT
NT
NT
100
100
100
100
100
100
100
100
100
100
100
100
Sperian
PHD6
QUA
NT
NT
109
118
105
108
115
113
113
113
105
108
105
103
108
IA
NT
NT
100
100
100
100
100
100
100
100
100
100
100
100
100
to
(a) Entries are mean quantitative accuracy (QUA) and mean Identification Accuracy (IA) from five replicate challenges in each test.
(b) MultiRAE read 19.9 ppm (indicating overrange condition) in all challenges with 20 ppm SO2.
(c) Eagle 2 read 6.0 ppm (indicating overrange condition) in all challenges with 5 ppm SO2.
NA Not Applicable. NT Not Tested.
-------
Table 5.2-3. Summary of Quantitative Accuracy and Identification Accuracy (percent) with NHsa
Test
Number
1
2
3
4
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 100
ppm
Step down, 50
ppm
Step down, 10
ppm
Step down, 3 ppm
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
Room T, <20%
RH
Room T, 80% RH
Low T, 50% RH
High T, 50% RH
HighT, 80% RH
BW GasAlert
Micro 5
QUA
80
79
100
147
66
62
68
61
54
60
70
67
69
73
77
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Drager X-am
7000
QUA
109
80
80
78
88
89
97
85
94
90
96
66
89
92
87
IA
100
100
100
60
100
100
100
100
100
100
100
100
100
100
100
Environics
ChemPro
lOOi
QUA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
IA
100
100
20
0
100
60
100
100
100
100
100
100
100
100
100
Industrial
Scientific
iBRID MX6
QUA
OR
104
136
167
54
38
47
39
48
36
76
81
53
88
88
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
RAE Systems
MultiRAE
Pro
QUA
97
105
92
100
83
86
90
82
80
83
97
93
86
95
94
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
RKI
Instruments
Eagle 2
QUA
NT
96
69
83
88
95
91
91
76
89
87
85
84
82
76
IA
NT
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Sperian
PHD6
QUA
92
98
74
87
88
102
90
104
76
92
92
88
88
94
86
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
(a) Entries are mean quantitative accuracy (QUA) and mean Identification Accuracy (IA) from five replicate challenges in each test.
NA Not Applicable. NT Not Tested. OR overrange (i.e., offscale) indication, no numerical reading.
-------
Table 5.2-4. Summary of Quantitative Accuracy and Identification Accuracy (percent) with Cha
Test
Number
1
2
3
8
9
12
14
15
16
17
18
19
20
Test Description
Base test, 10 ppm
Step down, 3 ppm
Step down, Ippm
Paint vapors
Gasoline exhaust
Diesel exhaust
Air freshener
DEAE
Room T, <20% RH
Room T, 80% RH
Low T, 50% RH
HighT, 50% RH
HighT, 80% RH
BW GasAlert
Micro 5
QUA
74
33
NR
102
98
96
96
114
90
82
106
70
64
IA
100
100
NR
100
100
100
100
100
100
100
100
100
100
Drager X-am
7000
QUA
76
43
15
123
121
77
118
148
100
58
OR
64
29
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
Environics
ChemPro lOOi
QUA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
IA
80
20
NR
100
NR
NR
100
100
100
20
100
NR
NR
Industrial
Scientific
iBRID MX6
QUA
117
107
84
121
119
147
120
139
140
119
210
74
61
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
RAE Systems
MultiRAE
Pro
QUA
109
100
84
102
104
114
113
115
123
107
115
111
90
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
RKI
Instruments
Eagle 2
QUA
NT
100b
130
100b
100b
100b
25
91
100b
57
100b
79
47
IA
NT
100
100
100
100
100
100
100
100
100
100
100
100
Sperian
PHD6
QUA
104
99
92
101
101
106
109
112
112
113
85
140
128
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
(a) Entries are mean quantitative accuracy (QUA) and mean Identification Accuracy (IA) from five replicate challenges in each test.
(b) Eagle 2 read 3.0 ppm (indicating owerrange condition) on all challenges at 3 ppm C12.
NA Not Applicable. NT Not Tested. NR No Response. OR overrange (i.e., offscale) indication, no numerical reading.
-------
Table 5.2-5. Summary of Quantitative Accuracy and Identification Accuracy (percent) with
Test
Number
1
2
3
4
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 50 ppm
Step down, 20 ppm
Step down, 5 ppm
Step down, 1 ppm
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
Room T, <20% RH
RoomT, 80% RH
LowT, 50% RH
High T, 50% RH
High T, 80% RH
BW GasAlert
Micro 5
QUA
NT
NT
OR
100
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
IA
NT
NT
100
100
100
100
100
100
100
100
100
100
100
100
100
Drager
X-am 7000
QUA
92
108
100
100
105
104
115
106
115
104
100
100
95
100
100
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Environics
ChemPro
lOOi
QUA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Industrial
Scientific
iBRID MX6
QUA
NT
NT
94
133
105
101
114
98
107
99
98
103
98
98
114
IA
NT
NT
100
100
100
100
100
100
100
100
100
100
100
100
100
RAE Systems
MultiRAE Pro
QUA
NT
100b
101
130
100b
100b
100b
100b
100b
100b
100b
100b
100b
100b
100b
IA
NT
100
100
100
100
100
100
100
100
100
100
100
100
100
100
RKI
Instruments
Eagle 2
QUA
NT
NT
NT
100C
100C
100C
96
100C
100C
98
100C
100C
100C
100C
100C
IA
NT
NT
NT
100
100
100
100
100
100
100
100
100
100
100
100
Sperian
PHD6
QUA
NT
66
60
94
78
80
89
65
90
80
62
62
65
64
61
IA
NT
100
100
100
100
100
100
100
100
100
100
100
100
100
100
(a) Entries are mean quantitative accuracy (QUA) and mean Identification Accuracy (IA) from five replicate challenges in each test.
(b) MultiRAE Pro read 19.9 ppm (indicating overrange condition) in all challenges with 20 ppm PH3.
(c) Eagle 2 read 1.0 ppm (indicating overrange condition) in all challenges with 1 ppm PH3.
NA Not Applicable. NT Not Tested. OR overrange (i.e.,offscale) indication, no numerical reading.
-------
Table 5.2-6. Summary of Quantitative Accuracy and Identification Accuracy (percent) with HCNa
Test
Number
1
2
3
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 50 ppm
Step down, 15 ppm
Step down, 5 ppm
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
RoomT, <20%RH
RoomT, 80% RH
LowT, 50% RH
HighT, 50% RH
HighT, 80% RH
BW GasAlert
Micro 5
QUA
NT
109
108
107
107
112
107
109
104
109
100
107
99
117
IA
NT
100
100
100
100
100
100
100
100
100
100
100
100
100
Driiger
X-am 7000
QUA
OR
141
162
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Environics
ChemPro
lOOi
QUA
NT
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
IA
NT
100
100
100
0
100
100
100
100
100
100
100
60
80
Industrial
Scientific
iBRID MX6
QUA
NT
117
129
116
114
115
125
115
107
112
109
115
105
113
IA
NT
100
100
100
100
100
100
100
100
100
100
100
100
100
RAE Systems
MultiRAE Pro
QUA
90
93
118
86
85
92
87
85
84
82
86
63
97
98
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
RKI
Instruments
Eagle 2
QUA
NT
87
92
68
65
63
66
73
66
79
73
50
91
73
IA
NT
100
100
100
100
100
100
100
100
100
100
100
100
100
Sperian
PHD6
QUA
100b
112
140
109
108
116
110
111
109
107
116
111
110
98
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
100
(a) Entries are mean quantitative accuracy (QUA) and mean Identification Accuracy (IA) from five replicate challenges in each test.
(b) PHD6 read 100 ppm (indicating overrange condition) on all challenges at 100 ppm HCN in Test 1. Tests 8 to 20 conducted with 50 ppm HCN with this
detector.
NA Not Applicable. NT Not Tested. OR Overrange (i.e.,offscale) indication, no numerical reading.
-------
Table 5.2-7. Summary of Quantitative Accuracy and Identification Accuracy (percent) with
Test
Number
1
2
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 19%O2
Step down, 16% O2
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
Room T, <20% RH
Room T, 80% RH
Low T, 50% RH
HighT, 50% RH
HighT, 80% RH
BW GasAlert
Micro 5
QUA
NT
NT
98
97
98
NT
98
97
99
98
99
98
97
IA
NT
NT
100
100
100
NT
100
100
100
100
100
100
100
Drager X-am
7000
QUA
NT
NT
100
99
100
NT
99
99
101
100
100
100
100
IA
NT
NT
100
100
100
NT
100
100
100
100
100
100
100
Industrial
Scientific
iBRID MX6
QUA
NT
NT
98
98
98
NT
98
98
99
98
97
98
98
IA
NT
NT
100
100
100
NT
100
100
100
100
100
100
100
RAE Systems
MultiRAE Pro
QUA
99
95
100
92
100
97b
98
92
99
99
99
99
98
IA
100
100
100
100
100
100b
100
100
100
100
100
100
100
RKI Instruments
Eagle 2
QUA
99
98
100
92
99
97
98
91
101
97
99
98
100
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
Sperian PHD6
QUA
100
99
100
94
101
98
99
93
100
100
100
100
100
IA
100
100
100
100
100
100
100
100
100
100
100
100
100
(a) Entries are mean quantitative accuracy (QUA) and mean Identification Accuracy (IA) from five replicate challenges in each test.
(b) Based on four challenges due to depletion of interferent supply before last challenge.
NT Not Tested.
-------
Table 5.2-8. Summary of Quantitative Accuracy and Identification Accuracy (percent) with
Test
Number
1
2
3
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 1.25%
Step down, 0.5%
Step down, 0.2%
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
Room T, <20% RH
Room T, 80% RH
LowT, 50% RH
HighT, 50% RH
High T, 80% RH
BW GasAlert
Micro 5
QUA
136
180
250
171
174
171
NT
176
173
152
160
168
176
195
IA
100
100
100
100
100
100
NT
100
100
100
100
100
100
100
Drager X-am
7000
QUA
164
167
190
NT
NT
NT
NT
NT
NT
165
177
124
176
NT
IA
100
100
100
NT
NT
NT
NT
NT
NT
100
100
100
100
NT
Industrial Scientific
iBRID MX6
QUA
163
218
320
203
204
203
NT
212
203
188
181
220
184
191
IA
100
100
100
100
100
100
NT
100
100
100
100
100
100
100
RAE Systems
MultiRAE Pro
QUAb
34
60
20
52
48
56
NT
56
40
23
110
77
95
88
IA
100
100
20
100
100
100
NT
100
100
100
100
100
100
100
RKI Instruments
Eagle 2
QUA
112
141
184
92
106
92
NT
100
96
118
128
128
126
120
IA
100
100
100
100
100
100
NT
100
100
100
100
100
100
100
Sperian PHD6
QUA
106
132
140
117
127
116
NT
122
110
147
110
154
98
78
IA
100
100
100
100
100
100
NT
100
100
100
100
100
100
100
oo
(a) Entries are mean quantitative accuracy (QUA) and mean Identification Accuracy (IA) from five replicate challenges in each test.
(b) Inspection of results shows MultiRAE Pro QUA values decline in chronological order of tests with CH4; possible sensor failure.
NT Not Tested.
-------
5.3 Repeatability
Tables 5.3-1 through 5.3-8 summarize the
repeatability observed with each detector in
each test with H2S, SO2, NH3, C12, PH3,
HCN, O2, and CH4, respectively.
Repeatability was calculated as percent RSD
of the five replicate responses in each test,
according to Equation 3. It should be noted
that for most of the handheld detectors
repeatability was determined from
concentration readings provided by the
detectors, but for the ChemPro lOOi
repeatability was determined from the 1-to-
3-bar intensity indications provided by that
detector. Thus, the repeatability results for
the ChemPro lOOi are not directly
comparable to those of the other detectors.
The BW Gas Alert Micro 5 and the Drager
X-am 7000 gave no quantitative values for
offscale (i.e., overrange) readings, however
other detectors continued to indicate a
quantitative numerical reading even when in
an overrange condition. Such occurrences
are flagged by means of footnotes in Tables
5.3-1 through 5.3-8 to distinguish them from
instances in which a detector gave five
identical on-scale readings (both
occurrences result in a calculated
repeatability value of 0.0% RSD).
For the detectors other than the ChemPro
lOOi, Tables 5.3-1 through 5.3-8 show
repeatability values that were consistently
less than 5% RSD with most detectors in
detection of H2S, SO2, PH3, HCN, O2, and
CH4. A few exceptions of relatively higher
repeatability results (i.e., up to
approximately 10% RSD) occurred with the
Eagle 2 with HCN (Table 5.3-6), and with
the PHD6 with CH4 (Table 5.3-8). On the
other hand, repeatability results were
substantially higher (usually below 10%
RSD, with occasional values of 20% or
more) for all detectors with NH3 and C12
(Tables 5.3-3 and 5.3-4). Repeatability was
not affected by interferent vapors or by test
conditions other than room temperature and
50% RH.
Repeatability values for the ChemPro lOOi
were constrained by the detector's l-to-3-
bar intensity indication and, in most cases,
the ChemPro lOOi gave the same intensity
response with all five challenges in a test
(i.e., repeatability = 0% RSD). However,
the presence of interferent vapors, and test
conditions other than room temperature and
50% RH, sometimes reduced the
repeatability of ChemPro lOOi response.
This observation was most evident with H2S
(Table 5.3-1), NH3 (Table 5.3-3), and PH3
(Table 5.3-5).
49
-------
Table 5.3-1. Summary of Repeatability (percent RSD) with H2Sa
Test
Number
1
2
o
5
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Test Description
Base test, 90 ppm
Step down, 30 ppm
Step down, 10 ppm
Step down, 3 ppm
H2S, 19%O2
H2S, 16% O2
H2S, room T start
Paint vapors
Gasoline exhaust
H2S, low T start
Ammonia cleaner
Diesel exhaust
H2S, high T start
Air freshener
DEAE
RoomT, <20%RH
Room T, 80% RH
LowT, 50% RH
High T, 50% RH
HighT, 80% RH
BW GasAlert
Micro 5
0.0
0.8
2.7
0.0
0.7
0.8
0.5
0.3
0.7
0.6
0.6
0.7
1.3
1.3
0.6
0.2
0.8
0.0
0.4
0.0
Drager X-am
7000
OR
2.8
3.3
1.8
1.1
0.7
OR
OR
OR
22.9
OR
OR
OR
OR
OR
OR
OR
OR
OR
19
Environics
ChemPro
100ib
0.0
0.0
0.0
0.0
0.0
0.0
50
34.2
0.0
34.2
0.0
0.0
63.6
0.0
63.6
0.0
63.6
0.0
34.2
0.0
Industrial
Scientific
iBRID MX6
0.6
1.0
6.1
1.7
0.7
1.0
0.8
0.4
0.7
0.8
0.9
0.7
0.2
1.5
0.5
1.0
0.7
0.1
0.3
0.2
RAE Systems
MultiRAE Pro
0.0C
2.6
2.3
0.0C
o.oc
o.oc
o.oc
o.oc
0.4
o.oc
o.oc
o.oc
o.oc
o.oc
o.oc
o.oc
o.oc
o.oc
o.oc
o.oc
RKI
Instruments
Eagle 2
0.8
1.4
2.4
0.0
0.5
0.2
1.1
o.od
0.4
2.2
0.0d
o.od
0.5
0.0d
0.2
0.0d
0.7
0.0(d)
0.0(d)
0.0
Sperian PHD6
1.1
1.2
0.0
0.0
0.8
0.9
0.5
0.7
0.7
0.0
0.8
0.4
0.9
1.9
0.4
0.7
0.5
1.1
0.8
0.5
(a) Entries are percent relative standard deviation (RSD) of quantitative responses from five replicate challenges in each test. NT Not Tested.
(b) ChemPro lOOi reports intensity (1, 2, or 3 bars) rather than TIC concentrations. Therefore repeatability not comparable to that indicated for other detectors.
(c) MultiRAE Pro read 99.9 ppm (indicating overrange condition) on all challenges at 90 ppm H2S.
(d) Eagle 2 read 100 ppm (indicating overrange condition) on all challenges at 90 ppm H2S.
OR overrange (i.e.,offscale) indication, no numerical reading.
-------
Table 5.3-2. Summary of Repeatability (percent RSD) with SO2a
Test
Number
1
2
o
5
4
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 100 ppm
Step down, 50 ppm
Step down, 20 ppm
Step down, 5 ppm
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
Room T, <20% RH
RoomT, 80% RH
Low T, 50% RH
HighT, 50% RH
High T, 80% RH
BW GasAlert
Micro 5
1.1
0.9
2.2
0.0
1.1
1.9
1.8
1.0
2.2
1.5
1.1
0.0
1.8
0.0
2.5
Drager X-am
7000
0.5
0.8
0.9
0.9
0.5
0.8
0.9
0.8
1.6
0.4
0.8
1.9
0.5
1.5
3.8
Environics
ChemPro
100ib
0.0
0.0
NR
NR
39.3
NR
0.0
0.0
0.0
0.0
NR
0.0
0.0
0.0
0.0
Industrial
Scientific
iBRID MX6
0.2
0.5
0.7
0.6
0.5
1.0
0.5
0.6
1.4
0.8
0.5
0.4
0.5
1.2
0.5
RAE Systems
MultiRAE Pro
NT
NT
0.0C
2.9
0.0C
0.0C
0.0C
0.0C
0.0C
0.0C
0.0C
0.0C
o.oc
o.oc
o.oc
RKI
Instruments
Eagle 2
NT
NT
NT
0.0d
o.od
o.od
o.od
o.od
1.7
1.9
o.od
2.9
0.0
3.7
3.8
Sperian PHD6
NT
NT
0.4
1.4
0.4
0.4
2.6
1.1
1.9
1.0
0.9
2.3
0.8
0.6
1.2
(a) Entries are percent relative standard deviation (RSD) of quantitative responses from five replicate challenges in each test.
(b) ChemPro lOOi reports intensity (1, 2, or 3 bars) rather than TIC concentrations. Therefore repeatability not comparable to that indicated for other detectors.
(c) MultiRAE Pro read 19.9 ppm (indicating overrange condition) on all challenges at 20 ppm SO2.
(d) Eagle 2 read 6.0 ppm (indicating overrange condition) on all challenges at 5 ppm SO2.
NR No Response. NT Not Tested.
-------
Table 5.3-3. Summary of Repeatability (percent RSD) with NH3a
Test
Number
1
2
o
5
4
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 100 ppm
Step down, 50 ppm
Step down, 10 ppm
Step down, 3 ppm
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
Room T, <20% RH
RoomT, 80% RH
Low T, 50% RH
HighT, 50% RH
High T, 80% RH
BW GasAlert
Micro 5
20.2
3.9
10.0
12.5
4.8
4.9
5.8
6.6
45.1
5.3
3.1
7.8
3.3
3.2
4.8
Drager X-am
7000
20.6
8.7
25
24.9
10.6
4.7
6.4
6.5
9.8
5.6
4.9
20.4
7.4
6.1
15
Environics
ChemPro
100ib
0.0
0.0
NR
NR
0.0
34.8
34.4
0.0
39.3
34.4
0.0
0.0
34.4
0.0
0.0
Industrial
Scientific
iBRID MX6
OR
3.7
8.4
0.0
0.8
1.2
1.0
3.3
3.8
2.5
8.6
5.6
1.6
1.7
1.0
RAE Systems
MultiRAE Pro
0.9
1.7
4.9
0.0
2.1
2.0
3.8
0.0
1.8
1.3
1.8
1.9
4.0
1.2
2.8
RKI
Instruments
Eagle 2
NT
4.5
9.4
0.0
2.2
1.4
5.1
1.7
2.5
1.8
2.8
2.5
2.5
0.7
5.8
Sperian PHD6
3.3
6.2
12
21.2
5.3
2.6
6.9
1.7
6.1
5.1
4.2
5.0
7.7
3.3
4.8
(a) Entries are percent relative standard deviation (RSD) of quantitative responses from five replicate challenges in each test.
(b) ChemPro lOOi reports intensity (1, 2, or 3 bars) rather than TIC concentrations. Therefore repeatability not comparable to that indicated for other detectors.
NR No Response. NT Not Tested. OR overrange (i.e., offscale) indication, no numerical reading.
to
-------
Table 5.3-4. Summary of Repeatability (percent RSD) with Cha
Test
Number
1
2
o
5
8
9
12
14
15
16
17
18
19
20
Test Description
Base test, 10 ppm
Step down, 3 ppm
Step down, Ippm
Paint vapors
Gasoline exhaust
Diesel exhaust
Air freshener
DEAE
Room T, <20% RH
RoomT, 80% RH
Low T, 50% RH
HighT, 50% RH
High T, 80% RH
BW GasAlert
Micro 5
7.4
0.0
NR
8.2
8.6
5.7
5.7
7.8
0.0
5.5
10.8
0.0
8.6
Drager X-am
7000
15.1
18.5
0.0
14.3
10.1
OR
12.3
6.4
4.2
25.0
OR
15.2
23.9
Environics
ChemPro
100ib
0.0
NC
NR
0.0
NR
NR
0.0
0.0
0.0
NC
0.0
NR
NR
Industrial
Scientific
iBRID MX6
13.5
5.3
6.0
8.7
7.0
6.5
5.9
6.0
2.9
6.6
10.8
4.7
7.7
RAE Systems
MultiRAE Pro
3.7
6.3
10.7
3.7
5.0
3.9
4.1
4.6
4.0
8.0
3.0
2.7
11.4
RKI
Instruments
Eagle 2
NT
0.0C
2.3
0.0C
o.oc
o.oc
8.1
3.7
0.0C
2.9
0.0C
2.5
5.0
Sperian PHD6
2.3
5.1
4.3
1.1
2.8
1.5
2.6
2.2
2.0
2.0
0.6
2.7
2.6
(a) Entries are percent relative standard deviation (RSD) of quantitative responses from five replicate challenges in each test.
(b) ChemPro lOOi reports intensity (1, 2, or 3 bars) rather than TIC concentrations. Therefore repeatability not comparable to that indicated for other detectors.
(c) Eagle 2 read 3.0 ppm (indicating overrange condition) on all challenges at 3 ppm C12.
NC Not Calculated (response in only one challenge). NR No Response. NT Not Tested. OR overrange (i.e., off scale) indication, no numerical reading.
-------
Table 5.3-5. Summary of Repeatability (percent RSD) with PH3a
Test
Number
1
2
o
5
4
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 50 ppm
Step down, 20 ppm
Step down, 5 ppm
Step down, 1 ppm
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
Room T, <20% RH
RoomT, 80% RH
Low T, 50% RH
HighT, 50% RH
High T, 80% RH
BW GasAlert
Micro 5
NT
NT
OR
0.0
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
Drager X-am
7000
0.0
2.5
0.0
0.0
0.0
2.2
0.0
2.1
3.1
2.2
0.0
0.0
0.0
0.0
0.0
Environics
ChemPro
100ib
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
21.2
0.0
34.2
34.2
0.0
34.2
0.0
Industrial
Scientific
iBRID MX6
NT
NT
5.7
0.8
2.7
2.0
3.0
1.8
2.1
2.6
14.0
5.0
3.1
6.1
1.2
RAE Systems
MultiRAE Pro
NT
0.0C
1.0
0.0
o.oc
o.oc
o.oc
o.oc
o.oc
o.oc
o.oc
o.oc
o.oc
o.oc
o.oc
RKI
Instruments
Eagle 2
NT
NT
NT
0.0d
o.od
o.od
5.2
o.od
o.od
3.1
0.0d
o.od
o.od
o.od
o.od
Sperian PHD6
NT
1.1
0.0
5.3
0.4
0.3
0.9
0.3
0.7
1.2
2.9
0.7
5.3
2.0
1.3
(a) Entries are percent relative standard deviation (RSD) of quantitative responses from five replicate challenges in each test.
(b) ChemPro lOOi reports intensity (1, 2, or 3 bars) rather than TIC concentrations. Therefore repeatability not comparable to that indicated for other detectors.
(c) MultiRAE Pro read 19.9 ppm (indicating overrange condition) on all challenges with 20 ppm PH3.
(d) Eagle 2 read 1.0 ppm (indicating overrange condition) on all challenges at 1 ppm PH3.
NT Not Tested. OR overrange (i.e., offscale) indication, no numerical reading.
-------
Table 5.3-6. Summary of Repeatability (percent RSD) with HCNa
Test
Number
1
2
o
5
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 50 ppm
Step down, 15 ppm
Step down, 5 ppm
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
RoomT, <20%RH
Room T, 80% RH
LowT, 50% RH
High T, 50% RH
HighT, 80% RH
BW GasAlert
Micro 5
NT
3.4
10.2
0.0
0.0
2.7
0.0
3.4
3.5
3.4
0.0
0.0
3.0
3.1
Drager X-am
7000
OR
2.4
2.4
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
Environics
ChemPro
100ib
NT
0.0
0.0
0.0
NR
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Industrial
Scientific
iBRID MX6
NT
1.8
1.4
1.4
0.6
1.5
0.8
1.3
1.6
1.0
0.5
1.3
1.3
0.5
RAE Systems
MultiRAE Pro
2.6
0.0
3.7
2.1
1.8
2.3
3.0
2.1
1.4
0.7
1.3
1.4
0.9
3.0
RKI
Instruments
Eagle 2
NT
2.7
5.4
6.1
5.8
7.2
7.1
9.2
6.9
6.6
7.5
16.6
6.7
7.2
Sperian PHD6
0.0C
2.0
2.3
1.0
0.6
0.8
0.8
1.9
0.3
0.4
0.3
1.0
0.8
2.4
(a) Entries are percent relative standard deviation (RSD) of quantitative responses from five replicate challenges in each test.
(b) ChemPro lOOi reports intensity (1, 2, or 3 bars) rather than TIC concentrations. Therefore repeatability not comparable to that indicated for other detectors.
(c) PHD6 read 100 ppm (indicating overrange condition) on all challenges at 100 ppm HCN in Test 1. Tests 8 to 20 conducted with 50 ppm HCN with this
detector.
NR No Response. NT Not Tested. OR overrange (i.e.,offscale) indication, no numerical reading.
-------
Table 5.3-7. Summary of Repeatability (percent RSD) with O2a
Test
Number
1
2
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 19%O2
Step down, 16% O2
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
Room T, <20% RH
RoomT, 80% RH
Low T, 50% RH
HighT, 50% RH
High T, 80% RH
BW GasAlert
Micro 5
NT
NT
0.0
0.2
0.0
NT
0.0
0.0
0.0
0.0
0.2
0.0
0.2
Drager X-am
7000
NT
NT
0.0
0.0
0.3
NT
0.0
0.0
0.0
0.0
0.0
0.0
0.2
Industrial
Scientific
iBRID MX6
NT
NT
0.2
0.0
0.2
NT
0.0
0.0
0.0
0.3
0.0
0.3
0.0
RAE Systems
MultiRAE Pro
0.0
0.3
0.2
0.0
0.2
0.3
0.2
0.0
0.0
0.3
0.0
0.0
0.0
RKI
Instruments
Eagle 2
0.2
0.0
0.0
0.0
0.0
0.0
0.3
0.0
0.0
0.0
0.0
0.2
1.6
Sperian PHD6
0.0
0.7
0.2
0.0
0.0
0.0
0.3
0.0
0.2
0.2
0.0
0.0
0.2
(a) Entries are percent relative standard deviation (RSD) of quantitative responses from five replicate challenges in each test.
NT Not Tested.
-------
Table 5.3-8. Summary of Repeatability (percent RSD) with CH4a
Test
Number
1
2
3
8
9
11
12
14
15
16
17
18
19
20
Test Description
Base test, 1.25%
Step down, 0.5%
Step down, 0.2%
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
RoomT, <20%RH
Room T, 80% RH
LowT, 50% RH
High T, 50% RH
HighT, 80% RH
BW GasAlert
Micro 5
0.0
0.0
0.0
2.3
1.8
2.3
NT
0.0
2.3
0.0
0.0
0.0
0.0
2.0
Drager X-am
7000
0.5
1.2
0.0
NT
NT
NT
NT
NT
NT
0.0
0.5
0.6
0.5
NT
Industrial
Scientific
iBRID MX6
0.5
0.0
0.0
0.4
0.4
0.4
NT
0.4
0.4
0.4
0.4
0.7
0.0
0.0
RAE Systems
MultiRAE Pro
6.5
0.0
NC
0.0
0.0
0.0
NT
0.0
0.0
7.8
2.0
6.8
1.9
0.0
RKI
Instruments
Eagle 2
0.0
4.3
1.2
0.0
2.3
0.0
NT
0.0
0.0
2.0
0.0
3.1
1.9
0.0
Sperian PHD6
5.8
6.4
9.8
2.9
2.6
2.4
NT
1.8
2.0
2.3
4.2
3.0
9.9
10.6
(a) Entries are percent relative standard deviation (RSD) of quantitative responses from five replicate challenges in each test.
NC Not calculated; MultiRAE Pro responded in only one of five challenges. NT Not Tested.
-------
5.4 Response Threshold
Table 5.4-1 summarizes the results of the
response threshold tests for all seven
detectors with the six TICs and with CH/i.
Note that response threshold was not
determined for C>2, as the purpose of C>2
measurement is to determine departures
below normal atmospheric 62 content.
Also, the CH4 response threshold was not
determined for the ChemPro lOOi, as that
detector does not provide an indication of
LEL.
Table 5.4-1. Summary of Response Threshold Results
Challenge
Gas
H2S
SO2
NH3
C12
PH3
HCN
CH4
BW
GasAlert
Micro 5
<3 ppm
<5 ppm
<3 ppm
1-3 ppm
< 1 ppm
<5 ppm
< 0.2%
Drager
X-am 7000
< 3 ppm
< 5 ppm
< 3 ppm
< 1 ppm
< 1 ppm
<5 pm
< 0.2%
Environics
ChemPro
lOOi
<3 ppm
20-50 ppm
10-50 ppm
3 -10 ppm
< 1 ppm
< 5 ppm
NA
Industrial
Scientific
iBRID
MX6
< 3 ppm
< 5 ppm
< 3 ppm
< 1 ppm
< 1 ppm
< 5 ppm
< 0.2%
RAE
Systems
MultiRAE
Pro
<3 ppm
<5 ppm
<3 ppm
< 1 ppm
< 1 ppm
<5 ppm
0.2-0.5%
RKI
Instruments
Eagle 2
< 3 ppm
< 5 ppm
< 3 ppm
< 1 ppm
< 1 ppm
< 5 ppm
< 0.2%
Sperian
PHD6
<3 ppm
<5 ppm
<3 ppm
< 1 ppm
< 1 ppm
<5 ppm
< 0.2%
NA Not Applicable.
Table 5.4-1 shows (by means of entries
indicated as < values) that most of the
detectors had response thresholds below the
lowest challenge concentration for most of
the challenge gases. The Drager X-am
7000, Industrial Scientific iBRID MX6, RKI
Eagle 2, and Sperian PHD6 exhibited
response thresholds that were below the
lowest challenge concentration for all gases
listed. Response thresholds that differ from
those of the other detectors are highlighted
by shaded cells in Table 5.4-1. TheBW
GasAlert Micro 5 and RAE MultiRAE Plus
responded in only one of five challenges at
the lowest challenge concentration for Cb
and for CH/i, respectively. The MultiRAE's
response threshold for CH4 may have been
affected by the sensor issue with that
detector that is noted in Section 5.2. The
Environics ChemPro lOOi exhibited
response thresholds for SC>2, NHa, and Cb
that were substantially higher than those of
the other detectors for those TICs. The great
majority of the observed response thresholds
are far below the immediately dangerous to
life and health (IDLH) levels for the target
TICs, and even the ChemPro lOOi response
thresholds for 862, NHs, and CL2 are at least
a factor of 2 less than the respective IDLH
levels. Except in the case of NHa, the
response threshold testing reported above
did not extend to low enough concentrations
to prove detection at the acute (i.e., 1 hour)
Reference Exposure Level values for these
TICs.
5.5 Effect of Operating Conditions
Tables 5.5-1 through 5.5-8 summarize the
effects of temperature and RH on the
performance parameters of each detector in
each test with H2S, SO2, NH3, C12, PH3,
HCN, O2, and CH4, respectively. The
performance parameters included in this
comparison are the response and recovery
time, QUA, IA, and repeatability. Shaded
cells in Tables 5.5-1 through 5.5-8 highlight
results that were significantly different in
Tests 16 through 20 (conducted over a range
58
-------
of temperature and RH conditions) from the
corresponding results in Test 1 (conducted at
room temperature and 50% RH). In this
comparison, response and recovery times
were judged to be significantly different if
their mean (±1 SD) ranges did not overlap.
Accuracy and repeatability results were
judged significantly different if they differed
by 20% or more.
Tables 5.5-1 through 5.5-8 show that with
all the detectors response and recovery times
were the performance factors most
frequently affected by variations in
temperature and RH conditions. Most often
response and recovery times were
lengthened by conditions other than normal
room temperature and 50% RH, but
reductions in response and recovery times
were also observed, e.g., in a few cases with
PH3 and HCN (Tables 5.5-5 and 5.5-6).
Significant effects of temperature and RH on
response and recovery times occurred less
frequently with the ChemPro lOOi than with
the other detectors. QUA, I A, and
repeatability were less frequently affected
by variations in temperature and RH. The
effects on QUA occurred with several
detectors (QUA was not calculated for the
ChemPro lOOi), whereas most effects on IA
and repeatability occurred with the ChemPro
lOOi, consistent with the observations noted
in Section 5.3 regarding the repeatability of
that detector.
The overall indication from Tables 5.5-1
through 5.5-8 is that varying conditions of
temperature and RH are unlikely to
adversely affect the detectors' identification
of a hazard, or the accuracy and repeatability
of quantifying hazard concentrations.
However, the detectors are likely to respond
more slowly and take longer to clear after a
positive response when the temperature and
RH differ widely from normal room
conditions (approximately 22 °C and 50%
RH).
59
-------
Table 5.5-1. Performance Parameters under Different Temperature and Relative Humidity Conditions with
Performance
Parameter
Response Time
(seconds)
Recovery Time
(seconds)
QUA
(%)
IA
(%)
Repeatability13
(%RSD)
Condition
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
BW
GasAlert
Micro 5
50
53
43
41
36
35
46
50
50
52
45
48
179
202
203
204
172
173
100
100
100
100
100
100
0.0
0.2
0.8
0.0
0.4
0.0
Drager X-am
7000
29
19
91
20
76
180
95
94
107
100
190
340
OR
OR
94
OR
OR
59
100
100
100
100
100
100
OR
OR
OR
OR
OR
19
Environics
ChemPro lOOi
19
18
19
18
18
18
30
29
34
28
58
66
NA
NA
NA
NA
NA
NA
100
100
100
100
100
100
0.0
0.0
63.6
0.0
34.2
0.0
Industrial
Scientific
iBRID MX6
121
88
103
105
84
81
342
353
582
672
460
590
163
181
181
178
158
158
100
100
100
100
100
100
0.6
1.0
0.7
0.1
0.3
0.2
RAE Systems
MultiRAE
Pro
18
16
14
15
14
16
319
323
309
317
351
359
1113
1113
1113
1113
1113
1113
100
100
100
100
100
100
0.0
0.0
0.0
0.0
0.0
0.0
RKI
Instruments
Eagle 2
20
16
15
14
14
28
32
39
33
42
34
35
108
111°
111
111°
111°
104
100
100
100
100
100
100
0.8
0.0
0.7
0.0
0.0
0.0
Sperian
PHD6
30
27
28
25
28
29
107
117
114
130
114
117
126
130
128
137
129
128
100
100
100
100
100
100
1.1
0.7
0.5
1.1
0.8
0.5
(a) MultiRAE Pro read 99.9 ppm (indicating overrange condition) on all challenges.
(b) ChemPro lOOi reported intensity readings (i.e., 1, 2, or 3 bars) rather than TIC concentrations.
(c) Eagle 2 read 100.0 ppm (indicating overrange condition) in all challenges with 90 ppm H2S.
OR overrange (i.e., offscale) indication, no numerical reading. NA Not Applicable.
Therefore repeatability not comparable to that indicated for other detectors.
-------
Table 5.5-2. Performance Parameters under Different Temperature and Relative Humidity Conditions with SOi
Performance
Parameter
Response Time
(seconds)
Recovery Time
(seconds)
QUA
(%)
IA
(%)
Repeatability13
(%RSD)
Condition
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
BW
GasAlert
Micro 5
33
29
99
95
85
172
71
52
126
128
119
176
99
103
100
101
100
98
100
100
100
100
100
100
1.1
1.1
0.0
1.8
0.0
2.5
Drager
X-am 7000
40
32
134
45
97
150
503
>597
>600
>616
>675
>510
100
109
100
109
105
99
100
100
100
100
100
100
0.5
0.8
1.9
0.5
1.5
3.8
Environics
ChemPro
lOOi
228
NR
33
20
31
19
383
NR
30
560
79
106
NA
NA
NA
NA
NA
NA
100
0
80
80
80
100
0.0
NR
0.0
0.0
0.0
0.0
Industrial
Scientific
iBRID MX6
32
37
63
60
55
41
254
240
359
>548
273
248
100
104
103
105
105
105
100
100
100
100
100
100
0.2
0.5
0.4
0.5
1.2
0.5
RAE Systems
MultiRAE
Pro
37
39
55
45
55
75
260
401
261
345
273
405
100a
100a
100a
100a
100a
100a
100
100
100
100
100
100
0.0
0.0
0.0
0.0
0.0
0.0
RKI
Instruments
Eagle 2
13
12
23
12
15
33
342
460
354
369
265
308
120C
120C
118
120C
113
117
100
100
100
100
100
100
0.0
0.0
2.9
0.0
3.7
3.8
Sperian
PHD6
22
24
23
28
38
29
117
117
134
92
141
95
109
105
108
105
103
108
100
100
100
100
100
100
0.4
0.9
2.3
0.8
0.6
1.2
(a) MultiRAE Pro read 19.9 ppm (indicating overrange condition) on all five challenges.
(b) ChemPro lOOi reported intensity readings (i.e., 1, 2, or 3 bars) rather than TIC concentrations.
(c) Eagle 2 read 6.0 ppm (indicating overrange condition) in all challenges with 5 ppm SO2.
NA Not Applicable. NR No Response.
Therefore repeatability not comparable to that indicated for other detectors.
-------
Table 5.5-3. Performance Parameters under Different Temperature and Relative Humidity Conditions with
Performance
Parameter
Response Time
(seconds)
Recovery Time
(seconds)
QUA
(%)
IA
(%)
Repeatability3
(%RSD)
Condition
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
BW
GasAlert
Micro 5
154
>180
>180
>180
>180
>180
800
>823
>1320
692
1049
>330
80
70
67
69
73
77
100
100
100
100
100
100
20.2
3.1
7.8
3.3
3.2
4.8
Drager
X-am 7000
>180
>180
>180
>180
>180
>180
321
198
>1132
291
>1026
>900
109
96
66
89
92
87
100
100
100
100
100
100
20.6
4.9
20.4
7.4
6.1
15.0
Environics
ChemPro lOOi
54
33
66
36
59
83
118
98
381
119
250
332
NA
NA
NA
NA
NA
NA
100
100
100
100
100
100
0.0
0.0
0.0
34.4
0.0
0.0
Industrial
Scientific
iBRID MX6
101
>180
>180
>180
>180
>180
>780
>825
>1230
>840
>1112
>900
OR
76
81
53
88
88
100
100
100
100
100
100
OR
8.6
5.6
1.6
1.7
1.0
RAE Systems
MultiRAE
Pro
82
139
>180
>180
>180
>180
>355
>558
>450
>426
>408
>426
105
97
93
86
95
94
100
100
100
100
100
100
1.7
1.8
1.9
4.0
1.2
2.8
RKI
Instruments
Eagle 2
>180
>180
>180
>180
>180
>180
>567
>764
>420
>408
>408
>396
96
87
85
84
82
76
100
100
100
100
100
100
4.5
2.8
2.5
2.5
0.7
5.8
Sperian
PHD6
>180
155
>180
>180
>180
>180
>492
>590
>366
>370
>408
>372
92
92
88
88
94
86
100
100
100
100
100
100
3.3
4.2
5.0
7.7
3.3
4.8
(a) ChemPro lOOi reported intensity readings (i.e., 1, 2, or 3
NA Not Applicable. OR overrange (i.e., offscale) indication,
bars) rather than TIC concentrations. Therefore repeatability not comparable to that indicated for other detectors.
no numerical reading.
-------
Table 5.5-4. Performance Parameters under Different Temperature and Relative Humidity Conditions with
Performance
Parameter
Response Time
(seconds)
Recovery Time
(seconds)
QUA
(%)
IA
(%)
Repeatability3
(%RSD)
Condition
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8 C/50 % RH
35C/50%RH
35C/80%RH
BW
GasAlert
Micro 5
54
180
180
51
152
180
22
35
27
38
25
14
74
90
82
106
70
64
100
100
100
100
100
100
7.4
0
5.5
10.8
0.0
8.6
Drager X-am
7000
>180
104
>180
41
>180
>180
>726
>675
>702
124
>888
211
76
100
58
OR
64
29
100
100
100
100
100
100
15.1
4.2
25
OR
15.2
23.9
Environics
ChemPro lOOi
24
24
69
47
NR
NR
116
169
43
75
NR
NR
NA
NA
NA
NA
NA
NA
80
100
20
100
0
0
0.0
0.0
NC
0
NR
NR
Industrial
Scientific
iBRID MX6
>180
12
32
>180
161
>180
48
29
43
168
37
31
117
140
119
210
74
61
100
100
100
100
100
100
13.5
2.9
6.6
10.8
4.7
7.7
RAE Systems
MultiRAE
Pro
63
19
71
25
43
>180
>320
>325
>417
>540
188
191
109
123
107
115
111
90
100
100
100
100
100
100
3.7
4.0
8.0
3.0
2.7
11.4
RKI
Instruments
Eagle 2
38
61
>180
102
>180
>180
>424
>370
>324
>468
>392
>348
100
100
57
100
79
47
100
100
100
100
100
100
0.0
0.0
2.9
0.0
2.5
5.0
Sperian
PHD6
69
18
19
>180
13
15
57
27
30
41
26
33
104
112
113
85
140
128
100
100
100
100
100
100
2.3
2.0
2.0
0.6
2.7
2.6
(a) ChemPro lOOi reported intensity readings (i.e., 1, 2, or 3
OR overrange (i.e., offscale) indication, no numerical reading
bars) rather than TIC
. NA Not Applicable,
concentrations. Therefore repeatability not comparable to that indicated for other detectors.
NR No Response. NC Not Calculated (response in only one challenge).
-------
Table 5.5-5. Performance Parameters under Different Temperature and Relative Humidity Conditions with
Performance
Parameter
Response Time
(seconds)
Recovery Time
(seconds)
QUA
(%)
IA
(%)
Repeatability13
(%RSD)
Condition
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8 C/50 % RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
BW
GasAlert
Micro 5
12
8
6
8
6
5
14
10
9
13
7
5
OR
OR
OR
OR
OR
OR
100
100
100
100
100
100
OR
OR
OR
OR
OR
OR
Drager X-am
7000
33
29
29
26
33
32
31
24
24
25
22
23
92
100
100
95
100
100
100
100
100
100
100
100
0.0
0.0
0.0
0.0
0.0
0.0
Environics
ChemPro lOOi
17
17
20
18
16
18
273
360
186
366
117
78
NA
NA
NA
NA
NA
NA
100
100
100
100
100
100
0.0
34.2
34.2
0.0
34.2
0.0
Industrial
Scientific
iBRID MX6
>180
127
>180
59
>180
48
56
45
134
45
51
78
94
98
103
98
98
114
100
100
100
100
100
100
5.7
14
5.0
3.1
6.1
1.2
RAE Systems
MultiRAE
Pro
43
24
22
25
19
18
220
107
82
104
105
128
100a
100a
100a
100a
100a
100a
100
100
100
100
100
100
0.0
0.0
0.0
0.0
0.0
0.0
RKI
Instruments
Eagle 2
10
9
10
10
10
10
11
14
12
14
14
14
100C
100C
100C
100C
100C
100C
100
100
100
100
100
100
0.0
0.0
0.0
0.0
0.0
0.0
Sperian
PHD6
125
101
86
106
>180
83
>425
>420
>331
>300
>372
>396
66
62
62
65
64
61
100
100
100
100
100
100
1.1
2.9
0.7
5.3
2.0
1.3
(a) MultiRAE Pro read 19.9 ppm (indicating overrange condition) on all challenges.
(b) ChemPro lOOi reported intensity readings (i.e., 1, 2, or 3 bars) rather than TIC concentrations.
(c) Eagle 2 read 1.0 ppm (indicating overrange condition) in all challenges with 1 ppm PH3.
OR overrange (i.e., offscale) indication, no numerical reading. NA Not Applicable.
Therefore repeatability not comparable to that indicated for other detectors.
-------
Table 5.5-6. Performance Parameters under Different Temperature and Relative Humidity Conditions with HCN
Performance
Parameter
Response Time
(seconds)
Recovery Time
(seconds)
QUA
(%)
IA
(%)
Repeatability3
(%RSD)
Condition
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
BW
GasAlert
Micro 5
65
52
54
58
31
27
79
74
66
136
34
36
109
109
100
107
99
117
100
100
100
100
100
100
3.4
3.4
0.0
0.0
3.0
3.1
Drager X-am
7000
20
16
21
15
20
41
>765
>340
266
27
291
364
OR
OR
OR
OR
OR
OR
100
100
100
100
100
100
OR
OR
OR
OR
OR
OR
Environics
ChemPro lOOi
18
27
26
37
13
18
133
63
119
78
109
163
NA
NA
NA
NA
NA
NA
100
100
100
100
60
80
0.0
0.0
0.0
0.0
0.0
0.0
Industrial
Scientific
iBRID MX6
>180
>180
>180
>180
>136
148
>679
>420
>680
>570
225
142
117
112
109
115
105
113
100
100
100
100
100
100
1.8
1.0
0.5
1.3
1.3
0.5
RAE Systems
MultiRAE
Pro
117
54
45
69
86
119
240
274
296
>413
>361
>352
90
82
86
63
97
98
100
100
100
100
100
100
2.6
0.7
1.3
1.4
0.9
3.0
RKI
Instruments
Eagle 2
149
48
45
55
48
71
102
62
45
113
36
29
87
79
73
50
91
73
100
100
100
100
100
100
2.7
6.6
7.5
16.6
6.7
7.2
Sperian
PHD6
52
102
93
43
29
97
>437
>335
>342
>398
179
186
100
107
116
111
110
98
100
100
100
100
100
100
0.0
0.4
0.3
1.0
0.8
2.4
(a) ChemPro lOOi reported intensity readings (i.e., 1, 2, or 3
OR overrange (i.e., offscale) indication, no numerical reading
bars) rather than TIC concentrations. Therefore repeatability not comparable to that indicated for other detectors.
. NA Not Applicable.
-------
Table 5.5-7. Performance Parameters under Different Temperature and Relative Humidity Conditions with
Performance
Parameter
Response Time
(seconds)
Recovery Time
(seconds)
QUA
(%)
IA
(%)
Repeatability
(%RSD)
Condition
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
BW
GasAlert
Micro 5
NT
16
21
23
13
20
NT
10
18
9
12
28
NT
99
98
99
98
97
NT
100
100
100
100
100
NT
0.0
0.0
0.2
0.0
0.2
Drager X-am
7000
NT
24
39
42
28
33
NT
24
22
19
19
26
NT
101
100
100
100
100
NT
100
100
100
100
100
NT
0.0
0.0
0.0
0.0
0.2
Industrial Scientific
iBRID MX6
NT
22
28
30
18
29
NT
27
39
47
26
46
NT
99
98
97
98
98
NT
100
100
100
100
100
NT
0.0
0.3
0.0
0.3
0.0
RAE Systems
MultiRAE Pro
10
13
11
12
9
8
10
11
10
10
10
17
99
99
99
99
99
98
100
100
100
100
100
100
0.0
0.0
0.3
0.0
0.0
0.0
RKI Instruments
Eagle 2
9
24
7
10
7
14
9
7
9
8
7
7
99
101
97
99
98
100
100
100
100
100
100
100
0.2
0.0
0.0
0.0
0.2
1.6
Sperian
PHD6
22
30
20
23
16
27
63
34
62
256
19
40
100
100
100
100
100
100
100
100
100
100
100
100
0.0
0.2
0.2
0.0
0.0
0.2
Oi
ON
NT Not Tested.
-------
Table 5.5-8. Performance Parameters under Different Temperature and Relative Humidity Conditions with CELj
Performance
Parameter
Response Time
(seconds)
Recovery Time
(seconds)
QUA
(%)
IA
(%)
Repeatability
(%RSD)
Condition
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
22 C/50 % RH
22 C/< 20 % RH
22 C/80 % RH
8C/50%RH
35C/50%RH
35C/80%RH
BW
GasAlert
Micro 5
16
19
15
15
19
18
12
20
13
13
>375
>375
136
152
160
168
176
195
100
100
100
100
100
100
0.0
0.0
0.0
0.0
0.0
2.0
Drager X-am
7000
35
41
46
47
44
NT
38
33
49
37
>285
NT
164
165
177
124
176
NT
100
100
100
100
100
NT
0.5
0.0
0.5
0.6
0.5
NT
Industrial Scientific
iBRID MX6
27
26
39
27
28
30
19
20
26
22
24
>345
163
188
181
220
184
191
100
100
100
100
100
100
0.5
0.4
0.4
0.7
0.0
0.0
RAE Systems
MultiRAE Pro
20
22
11
24
24
22
13
8
>387
10
>353
>355
34a
23a
110a
77a
95a
88a
100
100
100
100
100
100
6.5
7.8
2.0
6.8
1.9
0.0
RKI Instruments
Eagle 2
10
10
10
9
10
9
17
18
382
19
16
23
112
118
128
128
126
120
100
100
100
100
100
100
0.0
2.0
0.0
3.1
1.9
0.0
Sperian
PHD6
10
11
16
10
30
28
18
19
20
20
16
>326
106
147
110
154
98
78
100
100
100
100
100
100
5.8
2.3
4.2
3.0
9.9
10.6
(a) Inspection of results shows MultiRAE Pro QUA values decline in chronological order of tests with CH4; possible sensor failure.
NT Not Tested.
-------
5.6 Effect of Oxygen Deficiency on
TIC Response
In Tests 5 and 6 with H2S, the detectors
were challenged with 90 ppm of H2S at O2
levels of 19% and 16% in air, respectively
(see Table 2.4-4). The purpose of these tests
was to evaluate whether the response to H2S
was changed by the reduced oxygen level,
relative to the response to the same H2S
concentration delivered in normal air in Test
1. In this comparison, response and
recovery times were judged to be
significantly different if their mean ±1 SD
ranges did not overlap. Accuracy and
repeatability results were judged to be
significantly different if they differed by
20% or more.
The test results show relatively little impact
of the reduced O2 levels on the detector
performance parameters for H2S. The RKI
Eagle 2 showed no significant differences in
any performance parameter for H2S with
reduced O2 levels. Similarly the other six
detectors showed no significant differences
in IA (all detectors identified H2S in all five
replicates in all of tests 1, 5, and 6), or in
repeatability. The few differences found for
different detectors in response time,
recovery time, and QUA are summarized in
Table 5.6-1, where shaded entries indicate
performance in Tests 5 and 6 that differs
significantly from that obtained with H2S in
normal air (Testl).
Table 5.6-1 shows that response time for
H2S was reduced at the 16% O2 level with
both the BW GasAlert Micro 5 and
Industrial Scientific iBRID MX6, but was
increased (i.e., nearly doubled) with the
Drager X-am 7000 at both 19% and 16% O2.
The small differences in response time
shown for the RAE MultiRAE Pro and
Sperian PHD6 are significant by the criteria
noted above but of little practical
significance.
Table 5.6-1 also shows that the recovery
time for H2S was greatly increased at 16%
O2 for the Environics ChemPro lOOi and at
both 19% and 16% O2 for the Industrial
Scientific iBRID MX6. Only small effects
on recovery time were observed for the
Drager X-am 7000 and Sperian PHD6.
Finally, Table 5.6-1 shows that QUA for
H2S declined consistently with reduced O2
levels for the BW GasAlert Micro 5, Drager
X-am 7000, and Industrial Scientific iBRID
MX6.
Table 5.6-1. Performance Differences Observed in H2S Detection at Reduced
Performance Factor
Response Time (sec)
Recovery Time (sec)
Quantitative Accuracy
(%)
Detector
BW GasAlert Micro 5
Drager X-am 7000
Indus. Sci. iBRID MX6
RAE MultiRAE Pro
Sperian PHD6
Drager X-am 7000
Environics ChemPro lOOi
Indus. Sci. iBRID MX6
Sperian PHD6
BW GasAlert Micro 5
Drager X-am 7000
Indus. Sci. iBRID MX6
20.9%a
50 ±4.1
29 ±4.1
121 ±36.9
18±1.1
30 ±2.1
95 ± 8.2
30 ±1.3
342 ± 23
107 ±2.9
179 ±0.0
127±1.2d
163 ±0.9
O2 Level
19%b
49±13.1
56 ±9.0
82 ±15.7
14±1.1
34 ±0.9
119 ±10.8
46 ±3.7
701 ±83
113 ±2.4
152 ±1.0
108 ±1.2
129 ±1.0
16%c
38 ±4.4
55 ±2.9
70 ±5.7
14 ±0.7
32±3.1
106 ±6.0
102 ±6.0
754±111
116±3.4
126 ±1.0
92 ±0.7
109 ±1.2
(a) Testl
(b) TestS
(c) Test 6
(d) Drager X-am 7000 responses to 90 ppm H2S in Test 1 were off scale; quantitative response comparison based on responses
to 30 ppm H2S in Test 2.
68
-------
5.7 Cold/Hot Start Behavior
The performance of the seven detectors was
tested with H2S immediately after starting
up from room temperature, cold (8 °C), and
hot (40 °C) overnight storage in Tests 7, 10,
and 13, respectively (see Table 2.4-4). All
such tests were conducted with 90 ppm of
H2S, delivered in air at 20°C and 50% RH,
and the results were compared to the
corresponding results obtained in the same
test conditions with each detector in a fully
warmed-up state in Test 1. Table 5.7-1
summarizes the results of these tests for
response time, recovery time, QUA, IA, and
repeatability. Shaded cells in Table 5.7-1
indicate results that differ from those
obtained in the corresponding fully warmed-
up test at the same conditions. For response
and recovery time differences were judged
significant if the ±1 SD ranges of the
response or recovery times did not overlap.
For QUA, IA, and repeatability, differences
were judged significant if these metrics
differed by 20% or more.
Table 5.7-1 shows that for most detectors
the delay time between powering up the
detector and being ready to begin
monitoring was not dependent on the storage
condition before startup. For the GasAlert
Micro 5 the delay time increased from 1
minute after room temperature storage to 2
minutes after hot storage, and for the
ChemPro 100, which had the longest delay
times in general, the corresponding increase
was from 4 minutes to 7 minutes delay time.
The delay time of the X-am 7000 was longer
after room temperature storage than after
cold or hot storage.
Table 5.7-1 also shows that response times
for FbS were affected minimally if at all by
cold or hot startup, regardless of storage
conditions, but that recovery times were
lengthened with several detectors, especially
after a cold start from room temperature or
cold conditions. The parameters of QUA,
IA, and repeatability for H^S were largely
unaffected, although the QUA and
repeatability comparisons were limited by
the overrange readings of the X-am 7000
and the MultiRAE Pro. Repeatability
effects were observed with the ChemPro
lOOi after cold starts from all three storage
conditions.
69
-------
Table 5.7-1. Summary of Performance Parameters under Fully Warmed Up and Cold Start Conditions
Performance
Parameter
Startup Delay
(seconds)
Response Time
(seconds)
Recovery Time
(seconds)
QUA
(%)
IA
(%)
Repeatability13
(%RSD)
Start Condition
Room T Cold Start
5°C Cold Start
40°C Cold Start
Warmed Up
Room T Cold Start
5°C Cold Start
40°C Cold Start
Warmed Up
Room T Cold Start
5°C Cold Start
40°C Cold Start
Warmed Up
Room T Cold Start
5°C Cold Start
40°C Cold Start
Warmed Up
Room T Cold Start
8°C Cold Start
40°C Cold Start
Warmed Up
Room T Cold Start
5°C Cold Start
40°C Cold Start
BW
GasAlert
Micro 5
60
60
120
50 ±4
43 ±1
66 ±16
33 ±7
46 ±2
51 ±2
55 ±2
62 ±35
179
178
172
170
100
100
100
100
0.0
0.5
0.6
1.3
Drager X-am
7000
120
60
60
29 ±4
24 ±2
127 ±72
25 ±2
95 ±8
139 ±14
403 ±118
93 ±8
OR
OR
89
OR
100
100
100
100
OR
OR
22.9
OR
Environics
ChemPro lOOi
240
300
420
19 ±1
18 ±1
19 ±2
19 ±2
30 ±1
92 ±76
71 ±49
157 ±71
NA
NA
NA
NA
100
100
100
100
o.ob
50.0b
34.2b
63.6b
Industrial
Scientific
iBRID MX6
<60
<60
60
121 ±37
88 ±7
86 ±14
94 ±25
342 ±23
>849
>884
892 ±683
163
148
141
142
100
100
100
100
0.6
0.8
0.8
0.2
RAE Systems
MultiRAE Pro
120
120
120
18 ±1
18 ±1
22 ±3
19 ±3
3 19 ±43
324 ±48
420 ±189
300 ±23
1113
1113
1113
1113
100
100
100
100
0.0
0.0
0.0
0.0
RKI
Instruments
Eagle 2
60
60
60
20 ±7
38±11
36 ±8
51 ±6
32 ±3
33 ±10
60 ±23
36 ±1
108
101
101
95
100
100
100
100
0.8
1.1
2.2
0.5
Sperian
PHD6
60
60
60
30 ±2
41 ±3
37 ±0.5
39 ±2
107 ±3
112 ±3
119 ±2
115 ±6
126
113
114
114
100
100
100
100
1.1
0.5
0.0
0.9
(a) MultiRAE Pro read 99.9 ppm (indicating overrange condition) on all challenges.
(b) ChemPro lOOi reported intensity readings (i.e., 1, 2, or 3 bars) rather than TIC concentrations. Therefore repeatability not comparable to that indicated for
other detectors.
OR overrange (i.e., offscale) indication, no numerical reading. NA Not Applicable.
-------
5.8 Interference Effects
Each of the six interferents (latex paint
vapors, gasoline exhaust, ammonia cleaner
vapors, diesel exhaust, air freshener vapors,
and DEAE) were supplied to each detector
both in clean air and in air containing one of
the target analytes. When the sensor
configuration of a detector was changed by
replacement of sensors, the sampling of
interferent vapors in otherwise clean air was
repeated, so that FP responses were assessed
in all detector configurations. Tables 5.8- 1
through 5.8-8 summarize the effects of these
interferents on detector response by showing
the FP and FN rates for each interferent with
each detector, in testing with each target
analyte.
Tables 5.8-1 through 5.8-8 show that each of
the seven detectors showed FP responses in
some tests, when sampling one of the
interferent vapors in otherwise clean air.
Gasoline and diesel exhaust hydrocarbons
and paint vapors were the interferents that
most frequently resulted in FP responses,
with ammonia cleaner, air freshener, and
DEAE causing relatively few FP responses.
False positive responses occurred most
frequently when NFL? was the target gas, i.e.,
FP responses for NH? occurred at least twice
as often as for any other target gas.
The MultiRAE Pro was the detector most
subject to interference effects. The
MultiRAE Pro showed FP responses with all
six interferents in testing with F^S, 62, and
CFLj, and FP responses with at least one
interferent with every target gas. The
ChemPro lOOi and iBRTD MX6 also showed
FP responses with at least one interferent
with every target gas with which they were
tested (the ChemPro lOOi was not tested
with O2 or CH4). On the other hand, the X-
am 7000 and GasAlert Micro 5 were the
detectors least subject to FP responses. The
X-am 7000 showed only a few FP responses
in testing with 862, NH3, and Cb, and no FP
responses at all in testing with F^S, PH3,
HCN, and C>2 (that detector could not tested
for interferent effects with CH/i). The
GasAlert Micro 5 showed FP responses with
all six interferents in testing with NH3, only
a few FP responses in testing with SC>2 and
O2, and no FP responses at all in testing with
H2S, C12, PH3, HCN, and CH4.
An important result shown in Tables 5.8-1
through 5.8-8 is that the FN rates that
resulted from the interferents were almost
always zero. In fact, for six of the seven
detectors (i.e., the GasAlert Micro 5, X-am
7000, iBRJD MX6, MultiRAE Pro, Eagle 2,
and PFID6) the FN rate was zero with every
interferent in every test. This result means
that the interferents never prevented those
six detectors from properly identifying the
appropriate hazard. False negatives were
observed with the ChemPro lOOi in tests
with SO2, NH3, C12, and HCN (Tables 5.8-2
through 5.8-4, and 5.8-6, respectively).
Gasoline engine exhaust hydrocarbons were
a cause of FN with the ChemPro lOOi with
all four of these TICs. Ammonia cleaner, air
freshener, and diesel exhaust also caused FN
responses in a few tests with the ChemPro
lOOi.
71
-------
Table 5.8-1. Summary of False Positive (FP) and False Negative (FN) Rates with H2Sa
Test
Number
8
9
11
12
14
15
Test Description
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
BW GasAlert
Micro 5
FP
0
0
0
0
0
0
FN
0
0
0
0
0
0
Driiger
X-am 7000
FP
0
0
0
0
0
0
FN
0
0
0
0
0
0
Environics
ChemPro
lOOi
FP
100
0
100
0
0
0
FN
0
0
0
0
0
0
Industrial
Scientific
iBRID MX6
FP
0
100
0
0
0
0
FN
0
0
0
0
0
0
RAE Systems
MultiRAE Pro
FP
100
100
100
100
100
100
FN
0
0
0
0
0
0
RKI
Instruments
Eagle 2
FP
0
0
0
0
0
0
FN
0
0
0
0
0
0
Sperian
PHD6
FP
0
100
0
80
0
0
FN
0
0
0
0
0
0
(a) Entries are FP and FN rates in percent in each test with H2S and an interferent.
Table 5.8-2. Summary of False Positive (FP) and False Negative (FN) Rates with SO2a
Test
Number
8
9
11
12
14
15
Test Description
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
BW
GasAlert
Micro 5
FP
0
0
0
0
25
0
FN
0
0
0
0
0
0
Drager
X-am 7000
FP
0
100
0
100
0
0
FN
0
0
0
0
0
0
Environics
ChemPro lOOi
FP
100
0
0
0
0
0
FN
0
100
20
0
40
0
Industrial
Scientific
iBRID MX6
FP
0
100
0
0
0
0
FN
0
0
0
0
0
0
RAE Systems
MultiRAE
Pro
FP
100
33
0
33
33
0
FN
0
0
0
0
0
0
RKI
Instruments
Eagle 2
FP
0
100
0
100
100
0
FN
0
0
0
0
0
0
Sperian
PHD6
FP
50
0
0
0
0
0
FN
0
0
0
0
0
0
(a) Entries are FP and FN rates in percent in each test with SO2 and an interferent.
-------
Table 5.8-3. Summary of False Positive (FP) and False Negative (FN) Rates with NH3a
Test
Number
8
9
11
12
14
15
Test Description
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
BW GasAlert
Micro 5
FP
100
100
100
100
100
100
FN
0
0
0
0
0
0
Drager
X-am 7000
FP
20
0
20
0
0
60
FN
0
0
0
0
0
0
Environics
ChemPro lOOi
FP
100
0
20
100
20
20
FN
0
40
0
0
0
0
Industrial
Scientific
iBRID MX6
FP
60
40
100
20
60
40
FN
0
0
0
0
0
0
RAE Systems
MultiRAE
Pro
FP
0
33
0
100
33
0
FN
0
0
0
0
0
0
RKI
Instruments
Eagle 2
FP
0
0
0
33
0
0
FN
0
0
0
0
0
0
Sperian
PHD6
FP
50
0
0
0
0
0
FN
0
0
0
0
0
0
(a) Entries are FP and FN rates in percent in each test with NH3 and an interferent.
Table 5.8-4. Summary of False Positive (FP) and False Negative (FN) Rates with Cl2a
Test
Number
8
9
12
14
15
Test Description
Paint vapors
Gasoline exhaust
Diesel exhaust
Air freshener
DEAE
BW GasAlert
Micro 5
FP
0
0
0
0
0
FN
0
0
0
0
0
Drager X-am
7000
FP
0
100
60
0
0
FN
0
0
0
0
0
Environics
ChemPro lOOi
FP
100
0
0
40
0
FN
0
100
100
0
0
Industrial
Scientific
iBRID MX6
FP
0
100
0
0
0
FN
0
0
0
0
0
RAE Systems
MultiRAE
Pro
FP
100
0
0
0
0
FN
0
0
0
0
0
RKI
Instruments
Eagle 2
FP
0
0
33
0
0
FN
0
0
0
0
0
Sperian
PHD6
FP
0
0
0
0
0
FN
0
0
0
0
0
(a) Entries are FP and FN rates in percent in each test with C12 and an interferent. Ammonia cleaner not used as an interferent with this TIC.
-------
Table 5.8-5. Summary of False Positive (FP) and False Negative (FN) Rates with PH3a
Test
Number
8
9
11
12
14
15
Test Description
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
BW GasAlert
Micro 5
FP
0
0
0
0
0
0
FN
0
0
0
0
0
0
Driiger
X-am 7000
FP
0
0
0
0
0
0
FN
0
0
0
0
0
0
Environics
ChemPro
lOOi
FP
100
0
0
0
0
20
FN
0
0
0
0
0
0
Industrial
Scientific
iBRID MX6
FP
0
100
0
100
0
0
FN
0
0
0
0
0
0
RAE Systems
MultiRAE Pro
FP
100
100
0
20
0
0
FN
0
0
0
0
0
0
RKI
Instruments
Eagle 2
FP
0
100
0
100
100
0
FN
0
0
0
0
0
0
Sperian
PHD6
FP
0
100
0
80
0
0
FN
0
0
0
0
0
0
(a) Entries are FP and FN rates in percent in each test with PH3 and an interferent.
Table 5.8-6. Summary of False Positive (FP) and False Negative (FN) Rates with HCNa
Test
Number
8
9
11
12
14
15
Test Description
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
BW GasAlert
Micro 5
FP
0
0
0
0
0
0
FN
0
0
0
0
0
0
Drager
X-am 7000
FP
0
0
0
0
0
0
FN
0
0
0
0
0
0
Environics
ChemPro
lOOi
FP
100
0
0
0
0
0
FN
0
100
0
0
0
0
Industrial
Scientific
iBRID MX6
FP
0
0
20
0
0
0
FN
0
0
0
0
0
0
RAE Systems
MultiRAE Pro
FP
100
100
0
20
0
0
FN
0
0
0
0
0
0
RKI
Instruments
Eagle 2
FP
0
100
0
100
100
0
FN
0
0
0
0
0
0
Sperian
PHD6
FP
0
100
0
80
0
0
FN
0
0
0
0
0
0
(a) Entries are FP and FN rates in percent in each test with HCN and an interferent.
-------
Table 5.8-7. Summary of False Positive (FP) and False Negative (FN) Rates with O2a
Test
Number
8
9
11
12
14
15
Test Description
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
BW GasAlert
Micro 5
FP
0
100
0
NT
0
100
FN
0
0
0
NT
0
0
Drager X-am
7000
FP
0
0
0
NT
0
0
FN
0
0
0
NT
0
0
Industrial Scientific
iBRID MX6
FP
0
100
0
NT
0
100
FN
0
0
0
NT
0
0
RAE Systems
MultiRAE Pro
FP
100
100
100
100
100
100
FN
0
0
0
0
0
0
RKI Instruments
Eagle 2
FP
0
0
0
0
0
0
FN
0
0
0
0
0
0
Sperian PHD6
FP
0
100
0
80
0
0
FN
0
0
0
0
0
0
(a) Entries are FP and FN rates in percent in each test with O2 and an interferent.
NT Not Tested
Table 5.8-8. Summary of False Positive (FP) and False Negative (FN) Rates with CH4
Test
Number
8
9
11
12
14
15
Test Description
Paint vapors
Gasoline exhaust
Ammonia cleaner
Diesel exhaust
Air freshener
DEAE
BW GasAlert
Micro 5
FP
0
0
0
NT
0
0
FN
0
0
0
NT
0
0
Drager X-am
7000
FP
NT
NT
NT
NT
NT
NT
FN
NT
NT
NT
NT
NT
NT
Industrial Scientific
iBRID MX6
FP
0
100
0
NT
0
0
FN
0
0
0
NT
0
0
RAE Systems
MultiRAE Pro
FP
100
100
100
NT
100
100
FN
0
0
0
NT
0
0
RKI Instruments
Eagle 2
FP
0
0
0
NT
0
0
FN
0
0
0
NT
0
0
Sperian PHD6
FP
0
100
0
NT
0
0
FN
0
0
0
NT
0
0
(a) Entries are FP and FN rates in percent in each test with CH4 and an interferent.
NT Not Tested.
-------
5.9 Battery Life
The battery life of all seven detectors was
tested by operating them continuously
starting from a fully charged state and
monitoring them until operation stopped due
to battery depletion. For this test fresh
batteries were installed in two units of the
RKI Instruments Eagle 2: Unit E2A505,
which contained sensors for SC>2, PHa, and
HCN, and Unit E2A410, which contained
sensors for O2, H2S, and CH4 (i.e., LEL).
These two units were tested because test
operators noted substantially shorter battery
life when using Unit E2A410, presumably
due to the power needs of the sensors in that
unit. The rechargeable batteries in the other
six detectors were fully charged before the
start of the battery life test. All the detectors
were started from room temperature and
placed into normal operation (including use
of their internal air sampling pumps)
between 5:57 and 6:03 am on August 31,
2011. Figure 5.9-1 shows the results of the
battery life test.
RKI Eagle 2 (E2A505)
Indus. Sci. iBRID MX6
RKI Eagle 2 (E2A410)
BWIGasAlertMicroS
RAE MultiRAE Pro
Sperian PHD6
DragerX-am 7000
ChemPro lOOi
10 20 30
Battery Life (hours]
40
50
Figure 5.9-1. Summary of battery life test results.
The battery life of the seven detectors
ranged from less than 10 hours for the
ChemPro lOOi and Drager X-am 7000 to
nearly 46 hours for the RKI Eagle 2 unit
E2A505. The two Eagle 2 units exhibited
the longest and third-longest periods of
battery life, but the battery life of Unit
E2A505 was more than twice as long as that
Unit E2A410. This difference is attributed
largely to the greater power demand of the
LEL sensor in Unit E2A410.
5.10 Operational Factors
The following summaries of operational
factors for each detector were drawn from
the observations and records of test
operators during the test.
76
-------
BW Technologies GasAlert Micro 5.
Contractor testing personnel found the
GasAlert Micro 5 to be small and
lightweight, and easy to operate. While the
overall size and the area of the display were
relatively small, the large numbers and type
on the display made it easy to read during
testing. Both audible and visual alarms were
clear and distinctive. The operating menus
were simple to follow, but the calibration
menus were not as clear due to the
requirement to scroll through three screens
to define the calibration options. The
startup/shutdown procedures were
straightforward, and this detector responded
quickly to the daily bump check (within
approximately 30 seconds), although the
bump check readings of the detector were
often relatively higher than the
concentration used for the bump test. There
were no maintenance issues with this
detector during testing. When test personnel
operated the GasAlert Micro 5 while
wearing heavy protective gloves, they had
no difficulties turning the detector on or off.
However, those personnel found it difficult
to access the detector's menus because of
the need to press and hold more than one
button at the same time. Multiple attempts
were needed to successfully access the
menus.
The written documentation provided for the
GasAlert Micro 5 contained the necessary
information, however staff reported that it
was difficult to read because the required
key sequences for most operations were not
located together on the same page or within
the same section. Testing staff also found it
necessary to consult the documentation
every time for some routine activities such
as calibration because the key sequences to
access the menus were not intuitive and
were difficult to recall. Overall testing staff
found the GasAlertMicro 5 to be one of the
most user-friendly of the detectors evaluated
and it survived the entire test matrix with no
sensor failures.
Drager X-am 7000. Testing personnel
found that the relatively heavy, boxy shape
of the X-am 7000 was uncomfortable to
hold by hand for more than a few minutes at
a time. The display area of the detector was
relatively large and included a feature that
would enlarge the readings of any sensor
giving an alarm. Testing personnel found
the visual alarms to be quite bright and the
audible alarm to be relatively loud and
immediately noticeable. Testing personnel
also reported that this detector was
reasonably easy to operate and had the most
available user-defined options. The menus
on this detector were easy to understand, and
the startup/shutdown procedures were easy
to follow. However, the manual provided
with the detector had omissions regarding
certain operations that are possible on the
unit. For example, the manual did not
define all of the options available on the
menus. This detector took relatively long to
stabilize during the daily bump checks (on
the order of three to four minutes), but
usually gave readings in agreement with the
bump check concentration. When operating
the X-am 7000 while wearing heavy
protective gloves, test personnel found no
difficulty in turning the detector on or off,
accessing all menus, or selecting settings.
One unexplained alarm was triggered in this
exercise, apparently when the operator's
gloved hand inadvertently depressed two
keys at once.
One maintenance issue was encountered
with the X-am 7000. When testing was
about to begin with CH4, it was found that
the CAT-CH4 sensor of the X-am 7000
could not be calibrated and the detector then
locked out that sensor. Since the CH4 sensor
would no longer display, it was necessary to
obtain a new CH4 sensor from the
manufacturer in order to conduct the CH4
tests. Although the manufacturer responded
promptly with a new sensor, several tests
with CH4 were not completed before the
77
-------
testing schedule came to an end. The failure
of the CH4 sensor is likely due to its
exposure to the several TICs during the
testing that preceded the CTLj tests.
Environics ChemPro 1001. Testing
personnel found the ChemPro lOOi detector
relatively easy to use. Its large control
buttons made it easy to operate even when
wearing heavy HAZMAT gloves. The
display had a strong backlight which made it
easy to read. The startup/shutdown
procedures for this detector were simple but
did take several minutes to complete.
Documentation provided for the ChemPro
lOOi detector did not need to be used, as the
menus were quite intuitive. This was the
only detector which did not require the
sensors to be calibrated (which was the
primary reason documentation was
necessary for other detectors). The
ChemPro lOOi was the only detector among
those tested which had a dedicated
confidence check. The confidence check
vial (a "test tube" source of 1-propanol and
diisopropylmethylphosphonate) was
relatively easy to use and the detector
generally responded quickly to the
confidence check. Both audible and visual
alarms for this detector were clear and sharp.
When operating the ChemPro lOOi while
wearing heavy protective gloves, test
personnel had no trouble powering the
instrument on, or accessing menus and
entering selections. However, the gloves
inhibited the action of turning the detector
off. Three trials were needed before the
detector was successfully turned off, rather
than reentering its scrolling menu.
The ChemPro lOOi was the only detector
among those tested that responded based on
built-in gas libraries. Testing was
performed using the First Responder library,
which indicated the presence of TICs with
responses such as "Toxic" or "Chemical
Hazard". To identify the TIC present with
this detector, the user must perform a
narrowing search by using additional
libraries in sequence. While Environics
recommended using the "Trend" mode, this
mode did not provide a distinguishable
alarm and the unit often would not clear
after a challenge. The "First Responder"
mode was also not always consistent in
terms of alarms on successive challenges
with the same gas under the same
conditions. For example during a series of
challenges its response would change from
"Chemical Hazard" to "Toxic." This
detector also showed the greatest variability
in response to environmental conditions. For
example, during the high temperature test
with H2S it alarmed "Blister."
One complication with testing the ChemPro
lOOi was that the two different units used
during testing behaved quite differently. The
original unit (S/N 06CPil03701538) had an
unrecoverable "functional exception
D08:2057" on July 25, 2011, and was
returned to the manufacturer. The
replacement unit (S/N 06CPi 102201497)
was then used from July 29, 2011 until the
end of testing on August 31,2011. The
original unit periodically continued alarming
for long periods on clean air, and usually
had to be cleared with the "Recalculate
Baseline" function. The replacement unit
did not show this behavior. The original
unit also sometimes exhibited a long
response time, and periodically required
multiple attempts to pass the confidence
check, giving the error message "No MOS
signal detected." This message may have
been referring to the detector's metal oxide
sensor, and failure of that sensor may have
been the ultimate cause of the detector's
failure. The replacement unit always passed
the confidence check on the first attempt.
Testing personnel did note that the original
detector came with a five year warranty, but
that the replacement had only a one year
warranty, with no explanation provided.
78
-------
Maintenance issues with the ChemPro lOOi
during testing included the detector having
difficulty maintaining its baseline operation
condition, causing false alarms upon the
slightest movement such as being picked up
or moved from one location to another.
Environics recommended that the unit never
be turned off, so the detector ran
continuously on line power (except during
data transfer). This was the only detector in
which data transfer capability was evaluated.
However, the manufacturer initially failed to
supply a working data transfer cable. The
provided cable used a serial port which most
modern computers do not support and failed
to work through a USB-serial converter. It
was necessary for the testing crew to find an
older laptop computer with a serial port to
connect to the detector. The Environics UTP
software required to download data did not
run properly on this computer and assistance
was required from a Contractor Information
Management technician. Changing the
screen size enabled the software enough to
perform the data download, but all
operational windows remained
nonfunctional. There is a single interface
socket on the ChemPro lOOi for both power
and data download and testing personnel
found it inconvenient and cumbersome to
switch between those uses. The
manufacturer responded promptly to that
issue by sending an adapter designed to
interface both data and power cables to the
single port. Unfortunately, that adaptor did
not work, i.e. the adapter allowed data
transfer but did not transmit power to the
unit. The original ChemPro lOOi unit
required 30 minutes to download a single
nine hour interval of testing data , and the
file size was inordinately large,
approximately 12 Megabytes. The
replacement unit used a different software
version, which reduced download time to
about 10 minutes, and the file size was more
manageable at roughly 1 Megabyte. Testing
staff found that Environics staff was both
helpful and proactive in assisting with these
issues.
Industrial Scientific iBRID MX6. Testing
personnel found the iBRID MX6 generally
easy to use. The written documentation
provided sufficient instructions, but rarely
was needed since the menus on the detector
display were self-explanatory. Overall, the
menus needed for operation were logical and
easy to understand, but were difficult to
navigate because the control buttons on the
iBRID MX6 were so small. Those control
buttons consisted of a single small oval
arrangement of four keys (up, down, left and
right arrows) surrounding a central "enter"
button. Because of the close proximity of
the buttons, testing staff found it difficult to
press only one intended button, especially
while wearing HAZMAT gloves. Testing
staff also found that the detector display was
difficult to read. The backlight on the
display was insufficient so a flashlight was
needed to view the display during testing.
Furthermore, the font size on the display
was very small, and the display alternated
between reading the concentration and the
time-weighted average (TWA), which was
confusing to the operators. The audible and
visual alarms were sufficient. While the
startup/shutdown procedures were relatively
easy, testing personnel reported that this
detector took a long time to stabilize during
the daily bump checks, and often read
relatively higher than the concentration used
for the bump test. Test personnel found this
detector relatively difficult to operate when
wearing heavy protective gloves because of
the placement of all five of its control
buttons in a single close arrangement.
Multiple efforts were needed to navigate the
control menus because of the operator's
gloved hands repeatedly contacting more
than one button at a time.
The only maintenance issue encountered
with the iBRID MX6 during testing was that
the original PHa sensor could not be
79
-------
calibrated. This sensor was replaced early
and quickly enough that all testing could be
completed. Industrial Scientific supplied the
replacement sensor under the warranty
agreement.
RAE Systems MultiRAE Pro. Testing
personnel found the MultiRAE Pro
relatively easy to operate by following the
instructions provided in the manual.
However, a common problem with this
detector was that the sensor concentration
range information was not in the manual and
was otherwise not easy to locate without
seeking technical support or product
information online. The display was easy to
read and the menus were easy to follow.
When wearing HAZMAT gloves, staff noted
that it was difficult to tell by feel if the
detector buttons had been depressed. In all
instances, the button was successfully
depressed, but it was difficult for staff to tell
this. All alarms were understandable and
were easy to adjust within the detector
menu. Startup and shutdown procedures
were uncomplicated. The MultiRAE Pro
had a quick, simple calibration procedure,
but it did not display a reason for not
passing any failed span calibrations. The
sensors were generally easy to change out.
The sensors and sensor locations were
slotted in order to match up the correct
sensor with the correct location, however
multiple sensors could fit into the 62 sensor
location but would not work in that location.
If this misplacement happened, the operator
would not know that a sensor was in the
wrong location until the detector was
reassembled and powered back up. When
operating the MultiRAE Pro while wearing
heavy protective gloves, test personnel
turned the instrument on and off and
accessed all menus successfully. However,
these operations were awkward because it
was difficult to feel when a button had been
depressed.
Testing personnel noted that the quantitative
response of the MultiRAE Pro to CH4
seemed to decrease as the series of tests with
that gas progressed (this observation is noted
in Section 5.2). It is possible that the
performance of the CH4 sensor in the
MultiRAE Pro was affected by exposure to
the TICs during the testing that preceded the
CH4 tests. However, since testing was
completed no effort was made to obtain a
new CH4 sensor to investigate this
possibility.
RKI Instruments Eagle 2. A limitation of
the Eagle 2 in this testing was that the
needed sensors could not all be substituted
into a single unit, and consequently three
different units of the detector had to be
purchased to carry out the testing with all
target analytes. The Eagle 2 was relatively
large in size and relatively heavy, but its
design (including the built-in handle) made
it relatively easy to use. Use of this detector
with HAZMAT gloves was manageable, but
it was often difficult to tell by feel whether
detector buttons had actually been
depressed. All alarms were easy to
understand and easy to adjust within the
detector menus. Startup and shutdown
procedures were simple and easy to follow.
Testing personnel reported that overall the
Eagle 2 detector was easy to operate by
following the instruction manual. Those
personnel noted that the manual gave good
instructions on how to enter the main menu
on the unit, but that those instructions were
listed only once in the manual and took a bit
of time to locate. Many other instructions
required the use of the main menu and
testing personnel felt that it would have been
useful to reference the page number where
the main menu information was located. The
detector's display was easy to view and
understand, but did not display the
remaining battery status in the normal
display. The battery status could only be
viewed by toggling through a series of
displays. The Eagle 2 was unique among
80
-------
the detectors tested in using replaceable
rather than rechargeable batteries. Testing
staff noticed during testing that the battery
life of the Eagle 2 unit used for H^S, CJL;,
and O2 tests was noticeably shorter than it
had been for the Eagle 2 units used for other
testing (see Section 5.9). This difference
was attributed to greater power consumption
of the sensors used in that unit (likely
specifically the CH4 sensor). Test personnel
found operating the Eagle 2 while wearing
heavy protective gloves to be awkward, as it
was difficult to feel when a button had been
properly engaged. This was especially an
issue for actions such as accessing a menu
that required a button to be engaged twice in
rapid succession.
Sperian PHD6. Testing personnel reported
that the PHD6 detector display was easy to
view and understand. The alarms were easy
to understand; however, the unit's Short
Term Exposure Limit (STEL) and TWA
alarms would change the unit's display and
then the current concentrations could not be
viewed. Testing staff could not locate how
to disable the STEL and TWA alarms
through the menu and were only able to get
around this problem by adjusting the alarm
values so that they would not be triggered.
Testing staff also reported that selecting the
sensor of interest on the span menu was a
little complicated. Multiple choices had to
be toggled through to get to the desired
sensor and then any remaining sensors
would have to be toggled through to escape
the span menu. Otherwise, startup/shutdown
operations were easy and menus were easy
to navigate. The instruction manual was
complete and easy to follow. When wearing
HAZMAT gloves, staff noted that it was
difficult to tell by feel if the detector buttons
had been depressed. In all instances, the
intended button was successfully depressed,
but it was difficult for staff to tell this.
Additionally, this detector requires a pump
test by requesting the operator to block the
pump port. With gloves on, this process
became more difficult and time-consuming,
but could be accomplished with a little extra
effort.
Testing personnel also noted that it was
somewhat difficult to align the PHD6's
inner cover after installing the sensors, and it
was not initially clear whether access to the
sensors needed to be obtained through the
front or the back of the detector. A feature
peculiar to the PHD6 was that when using
its internal sample pump, as in this testing,
its sample intake port was at the bottom of
the unit, i.e., pointing toward the operator
when held in the hand. This arrangement
would seem to risk accidentally tangling or
pinching off the sample intake line while
using the detector. Staff also had difficulty
with the detector's charging system, which
makes an electrical connection solely by
gravity. On at least three occasions the
PHD6 detector failed to charge due to poor
contact between the unit and the charging
system.
81
-------
6.0 Summary
The testing reported here involved seven
handheld detectors, eight target gases, six
interferents, and six different
temperature/RH conditions, as well as
specific tests involving three cold start
conditions and two levels of reduced 62.
That testing showed a wide range of
performance of the handheld detectors, with
each detector performing well in some tests
and less well in others. This section
provides a summary of the test results on
each performance parameter. It should be
noted that the Environics ChemPro lOOi
used a different detection principle than the
BW Technologies GasAlert Micro 5, Drager
X-am 7000, Industrial Scientific iBRTD
MX6, RAE Systems MultiRAE Pro, RKI
Instruments Eagle 2, and Sperian PHD6,
which used similar detection technology.
The ChemPro lOOi also differed from the
other six detectors in that it did not provide
quantitative concentration readings for the
TICs, and was not equipped to indicate 62
or LEL. Consequently, certain performance
parameters were not determined for the
ChemPro lOOi, or are summarized
separately from the results for the other six
detectors.
6.1 Response and Recovery Time
The response and recovery times of the
seven handheld detectors in determination of
TICs are summarized in Figures 6.1-1 and
6.1-2, respectively. Each figure shows the
mean, median, and ±1 SD range of all the
response times recorded for each detector in
all testing with the six TICs. In compiling
these figures, response and recovery times
that were recorded as "greater than" (>)
values (see Tables 5.1-1 to 5.1-8) were
assigned their numerical value (i.e., the >
sign was dropped). Thus, the calculated
means, medians, and standard deviations
shown in Figures 6.1-1 and 6.1-2 must be
recognized as underestimates of these
parameters.
Figure 6.1-1 shows that the ChemPro lOOi
exhibited the fastest response overall in
testing with the six TICs, and the iBRTD
MX6 exhibited the slowest response overall
with those TICs. Median response times in
the TIC testing ranged from approximately
20 seconds with the ChemPro lOOi to
approximately 100 seconds with the iBRTD
MX6. The other five detectors exhibited
response times in TIC testing that were
closely similar and intermediate between
those of the ChemPro lOOi and the iBRTD
MX6, e.g., median TIC response times of
approximately 40 to 50 seconds. In testing
of six detectors with O2 and CH4 (not shown
in Figure 6.1-1), relatively faster response
was observed as compared to the TIC
responses. With C>2, response times for all
six detectors were typically < 30 seconds,
and the Eagle 2 often responded in less than
10 seconds. With CH/i, response times for
most of the six detectors were < 30 seconds,
with the GasAlert Micro 5 always
responding within 20 seconds and the Eagle
2 often responding in 10 seconds or less.
The X-am 7000 response times for CFLi
ranged from about 30 to nearly 50 seconds.
Figure 6.1-2 shows that the GasAlert Micro
5, ChemPro lOOi, Eagle 2, and PHD6
exhibited the fastest recovery overall in
testing with the six TICs, and the iBRTD
MX6 exhibited the slowest recovery overall
with those TICs. Median recovery times in
the TIC testing ranged from approximately
50 seconds with the GasAlert Micro 5 to
approximately 360 seconds with the iBRTD
MX6. In testing of six detectors with C>2 and
CH4 (not shown in Figure 6.1-2), relatively
faster recovery was observed as compared to
the TIC recoveries. With C>2, recovery times
for most of the six detectors were typically <
82
-------
30 seconds, and the MultiRAE Pro and
Eagle 2 often recovered in approximately 10
seconds or less. However, the recovery
times for the Sperian PHD6 with C>2 were
usually > 40 seconds and ranged up to more
than 250 seconds. With CH4, recovery
times for the six detectors were usually < 25
seconds, but the GasAlert Micro 5, X-am
7000, iBRID MX6, MultiRAE Pro, and
PHD6 all showed recovery times for CH4
that exceeded 280 seconds in testing
conducted at 35 °C.
180
i/i
•a
c
o
u
QJ
fcO
aT
E
QJ
I/I
C
O
Q.
i/l
QJ
or
140
120
100
80
Mean Response Time
Median Response Time
+ Mean + Std. Dev.
"Mean - Std. Dev.
GasAlert X-am 7000 ChemPro iBRID MultiRAE Eagle 2 PHD6
Figure 6.1-1. Summary of response time results in TIC testing.
83
-------
I/I
•a
c
o
u
QJ
i/l
H
£•
o
u
800
700
600
500
400
300
200
100
• Mean Recovery Time
O Median Recovery Time
+ Mean + Std. Dev.
— Mean -Std. Dev.
GasAlert X-am 7000 ChemPro iBRID MultiRAE Eagle 2 PHD6
Figure 6.1-2. Summary of recovery time results in TIC testing.
6.2 Accuracy
Quantitative accuracy was determined for all
detectors except the Environics ChemPro
lOOi. Figure 6.2-1 summarizes the QUA
results determined for the other six detectors
in all testing with the six TICs, C>2, and CH/i.
That figure shows the mean, median, and ±1
SD range of all the QUA values recorded for
each detector in all testing, excluding any
readings that resulted from a pegged (i.e.,
quantitative but unvarying) overrange
response on a detector. Thus, for example,
Figure 6.2-1 does not include values such as
the 111 percent QUA recorded for the
MultiRAE Pro with H2S in Table 5.2-1,
which resulted from the monitor pegging at
a reading of 99.9 ppm when challenged with
90 ppm of H2S.
Figure 6.2-1 shows that over all the target
gases the mean QUA values for the six
detectors ranged from 91% for the
MultiRAE Pro to 125% for the iBRID MX6,
and the median QUA values ranged from
95% for the MultiRAE Pro to 113% for the
iBRID MX6. However, Figure ES-3 is
based on only about two-thirds of the
possible QUA results for the X-am 7000 due
to non-quantitative overrange indications by
that detector in some tests. The same is true
for the MultiRAE Pro and Eagle 2 due to
exclusion of fixed quantitative readings
exhibited during overrange conditions on
those detectors. The exclusion of these
results indicates that QUA values for those
three detectors might be significantly higher
if quantitative readings above the
84
-------
1 fin
1 An
ItU
1 9n
l/U
s* -i on
re
u
< 80 -
0)
•S fin -
c
re
3 /in
a 4U
90
n .
, • Mean QUA
+ • Median QUA
+ Mean QUA + SD
— Mean QUA- SD
H "^
* ft ft _
— _ _
GasAlert X-am 7000
iBRID
MultiRAE
Ea?le 2
PHD6
Figure 6.2-1. Summary of QUA results in TIC, Oi, and CELt testing (QUA not determined
for ChemPro lOOi). Data shown exclude any readings indicating a constant overrange
condition of a detector.
nominal full scale value could be obtained
from the detectors. In contrast, the iBRID
MX6 and Sperian PHD6 never reported an
overrange condition in any test. The PHD6
in particular achieved mean and median
QUA values near 100% and a relatively
narrow range of QUA results around 100%,
as indicated by the ±1 SD range in Figure
6.2-1.
Identification accuracy was 100% (i.e., the
detectors correctly identified the gas
challenge in all trials) in almost all tests.
Other than in tests at the lowest challenge
concentrations, the only cases of IA less
than 100% were with the ChemPro lOOi,
which failed to respond in some tests with
SO2, NH3, C12, and HCN that involved
interferent vapors or temperature and RH
conditions other than 22°C and 50% RH.
6.3 Repeatability
For the six detectors other than the ChemPro
lOOi, repeatability was consistently within
5% RSD in detection of H2S, SO2, PH3,
HCN, O2, and CH/i. A few exceptions of
repeatability up to approximately 10% RSD
occurred with the Eagle 2 with HCN and
with the PHD6 with CH4. Repeatability
results were substantially higher (usually
within 10% RSD, with occasional values of
20% or more) for all six detectors with NH?
and C12. Repeatability for these six
detectors was not affected by interferent
vapors or by test conditions of temperature
andRH.
85
-------
Repeatability values for the ChemPro lOOi
were constrained by the detector's l-to-3-
bar intensity indication, and in most cases
the ChemPro lOOi gave the same intensity
response with all five challenges in a test
(i.e., repeatability = 0% RSD). However,
the presence of interferent vapors and test
conditions other than room temperature and
50% RH sometimes degraded the
repeatability of ChemPro lOOi response.
6.4 Response Threshold
With few exceptions, all detectors tested
exhibited response thresholds of < 3 ppm for
H2S and NH3, < 5 ppm for SO2 and HCN, <
1 ppm for C12 and PH3, and < 0.2% by
volume (i.e., < 4% of the LEL) for CH4.
The exceptions were that the BW GasAlert
Micro 5 showed a response threshold in the
range of 1 to 3 ppm for C12, the RAE
MultiRAE Pro showed a response threshold
in the range of 0.2 to 0.5% for CH4, and the
Environics ChemPro lOOi showed response
thresholds in the range of 20 to 50 ppm for
SO2, 10 to 50 ppm for NH3, and 3 to 10 ppm
for C12. It is possible that the response
threshold of the RAE MultiRAE Pro for
CH4 was affected by the suspected
progressive failure of the LEL sensor in that
detector, which was noted in Section 5.2.
6.5 Effect of Operating Conditions
With all seven detectors the performance
factors most affected by variations in
temperature and RH conditions were
response and recovery times, which were
usually lengthened by conditions other than
normal room temperature and 50% RH.
Effects of temperature and RH on response
and recovery times were seen less frequently
with the ChemPro lOOi than with the other
six detectors. The performance factors least
affected by variations in temperature and
RH were QUA, IA, and repeatability.
Effects on QUA occurred with several
detectors (this performance parameter was
not determined for the ChemPro lOOi),
whereas the majority of effects on IA and
repeatability occurred with the ChemPro
lOOi.
6.6 Effect of O2 Deficiency on TIC
Response
The RKI Eagle 2 showed no significant
differences in any performance parameter
for H2S with reduced O2 levels, and none of
the detectors showed any significant
differences in IA for H2S at reduced O2
levels. Significant effects of O2 level on
response time, recovery time, and QUA for
H2S were seen with some detectors. The
response time for H2S was shortened at the
16% O2 level with both the BW GasAlert
Micro 5 and Industrial Scientific iBRID
MX6, but was increased (i.e., nearly
doubled) with the Drager X-am 7000 at both
19% and 16% O2. The recovery time for
H2S was greatly increased at 16% O2 for the
Environics ChemPro lOOi and at both 19%
and 16% O2 for the Industrial Scientific
iBRID MX6. The QUA for H2S declined
consistently with reduced O2 levels for the
BW GasAlert Micro 5, Drager X-am 7000,
and Industrial Scientific iBRID MX6.
86
-------
6.7 Cold/Hot Start Behavior
In most cases, response times, QUA, IA, and
repeatability for detection of H2S were
affected only minimally by rapid startup
after storage overnight at room, cold, or hot
temperature. The delay times between
powering up each detector and being ready
to begin monitoring similarly showed little
impact from the storage condition before
startup. However, recovery times were
lengthened with several detectors, especially
after rapid startup from room temperature or
cold conditions. Repeatability was degraded
with the ChemPro lOOi after cold starts from
all three storage conditions.
6.8 Interference Effects
All of the seven detectors showed FP
responses in some tests when sampling an
interferent vapor in otherwise clean air.
Gasoline and diesel exhaust hydrocarbons
and paint vapors were the interferents that
most frequently caused FP responses. The
MultiRAE Pro was the detector most subject
to interference effects, showing FP
responses with all six interferents in testing
with H2S, O2, and CFLj, and FP responses
with at least one interferent with every target
gas. The ChemPro lOOi and iBRTD MX6
also showed FP responses with at least one
interferent with every target gas with which
they were tested. The X-am 7000 and
Gas Alert Micro 5 were the detectors least
subject to FP responses. The X-am 7000
showed no FP responses at all in testing with
H2S, PH3, HCN, and O2. The GasAlert
Micro 5 showed no FP responses at all in
testing with H2S, C12, PH3, HCN, and CH4.
The FN rates that resulted from the
interferents were almost always zero. In fact,
for six of the seven detectors (i.e., the
GasAlert Micro 5, X-am 7000, iBRID MX6,
MultiRAE Pro, Eagle 2, and PHD6) the FN
rate was zero with every interferent in every
test. False negatives were observed with the
ChemPro lOOi in tests with SO2, NH3, C12,
and HCN. Gasoline engine exhaust
hydrocarbons caused FN with the ChemPro
lOOi with all four of these TICs, and
ammonia cleaner, air freshener, and diesel
exhaust also caused FN responses in a few
tests with the ChemPro lOOi.
6.9 Battery Life
The battery life of the seven detectors is
illustrated in Figure 6.9-1, and ranged from
less than 10 hours for the ChemPro lOOi and
Drager X-am 7000 to nearly 46 hours for the
RKI Eagle 2 unit E2A505. The two Eagle 2
units exhibited the longest and third-longest
periods of battery life, but the battery life of
Unit E2A505 was more than twice as long
as that Unit E2A410. This difference is
attributed largely to the greater power
demand of the LEL sensor in Unit E2A410.
6.10 Operational Factors
The following are brief summaries of key
positive and negative operational factors
reported by the test operators for each
handheld detector.
BW Technologies GasAlert Micro 5. This
detector was small, lightweight, and easy to
use, and large font on the display made it
easy to read. Operating menus were easy to
understand, calibration menus less so. The
operating manual was troublesome because
required key sequences were sometimes not
located together on the same page.
87
-------
RKI Eagle 2 (E2A505)
Indus. Sci. iBRID MX6
RKI Eagle 2 (E2A410)
BWIGasAlertMicroS
RAE MultiRAE Pro
Sperian PHD6
DragerX-am 7000
ChemPro lOOi
10 20 30
Battery Life (hours)
40
50
Figure 6.9-1. Summary of battery life test results.
Drager X-am 7000. This detector was
relatively heavy and boxy in shape, making
it uncomfortable to hold in the hand for
more than a few minutes. The display area
was large and easily readable. Operating
menus were easy to understand and the
detector was easy to use and had numerous
user-defined options. However, the
operating manual did not appear to cover all
of the features or operations of the unit.
Environics ChemPro lOOi. This detector
was easy to operate, with intuitive menus,
and had large control buttons that could be
manipulated correctly even when wearing
heavy gloves. The ChemPro lOOi required
confidence checks with a chemical vapor
source provided with the detector. Those
checks were simple to perform and the
detector responded quickly to the confidence
check. The ChemPro lOOi was relatively
sensitive to the test conditions (temperature
and RH) and occasionally had difficulty
maintaining its baseline operating condition
when moved during testing, causing false
alarms and requiring that the operator reset
the baseline. The MOS sensor in the first
ChemPro lOOi unit failed during testing, and
a replacement ChemPro lOOi unit was
provided by the manufacturer.
Industrial Scientific iBRID MX6. This
detector had logical and self-explanatory
menus, but the menus were difficult to
navigate because the buttons on this detector
were small and clustered tightly together.
This was especially a problem when wearing
heavy gloves. The display of the iBRID
MX6 was weakly backlit and the display
font was small, making readings difficult to
discern. This detector also responded
relatively slowly to daily bump checks.
RAE Systems MultiRAE Pro. This
detector was easy to operate by following
the instruction manual, the menus were
clearly understandable, and the display was
easy to read. However, it was difficult to
determine the full-scale ranges of the
sensors installed in the MultiRAE Pro
88
-------
without seeking technical support or online
information from the manufacturer. The use
of heavy gloves made it difficult to feel
when the control buttons had been
successfully pressed. Multiple EC sensors
could fit into the 62 sensor location of this
detector, but would not work in that
location. The operator would not know that
the sensor was not working until the detector
had been reassembled and powered up.
RKI Instruments Eagle 2. Three separate
units of this detector had to be purchased to
conduct testing, because the needed sensors
could not be interchanged within a single
unit. The Eagle 2 was relatively large and
heavy, but its design and built-in handle
made it comfortable to use. The display was
clear and legible but did not indicate the
status of the batteries. Operation of this
detector while wearing heavy gloves was
difficult, as it was hard to feel when the
control buttons had been successfully
pressed.
Sperian PHD6. This detector's display was
easy to read, but the detector's alarms would
change the display, interfering with
concentration readings. Testing staff
adjusted the alarm values to avoid this issue
during testing. Selection of a particular
sensor on the calibration menu required
toggling through multiple menu steps.
Operation of the detector's control buttons
and performance of the pump test were
difficult when wearing heavy gloves. The
sample inlet tubing of the PHD6 connects at
the bottom of the detector, and thus the
connection point is directed toward the user
when the detector is held in the hand,
potentially leading to pinching or snagging
of the inlet tubing. The battery charger of
the PHD6 makes electrical contact by
gravity and sometimes did not make proper
contact.
89
-------
APPENDIX A
NOMINAL UPPER RANGE LIMITS OF THE TESTED DETECTORS
FOR EACH TARGET GAS
Gas
02
LEL
H2S
SO2
NH3
CI2
PH3
HCN
NA: Not
BW
Technol.
Micro 5
30%
100%
500 ppm
150 ppm
100 ppm
50 ppm
5 ppm
30 ppm
applicable.
Drager
X-am 7000
25%
100%
100 ppm
100 ppm
300 ppm
20 ppm
1000 ppm
50 ppm
Environics
ChemPro
100i
NA
NA
100 ppm
100 ppm
100 ppm
10 ppm
50 ppm
20 ppm
Industrial
Scientific
iBRID MX6
30%
100%
500 ppm
100 ppm
100 ppm
100 ppm
5 ppm
30 ppm
RAE
Systems
MultiRAE
Pro
30%
100%
100 ppm
20 ppm
100 ppm
10 ppm
20 ppm
100 ppm
RKI
Instrum.
Eagle 2
40%
100%
100 ppm
6 ppm
75 ppm
3 ppm
1 ppm
15 ppm
Sperian
PHD6
30%
100%
200 ppm
25 ppm
100 ppm
50 ppm
20 ppm
100 ppm
90
-------
APPENDIX B
EXAMPLE OF LABORATORY DATA RECORDING SHEETS
DATA SHEETS FROM TESTING OF BW TECHNOLOGIES
GAS ALERT MICRO 5 WITH HYDROGEN CYANIDE
AT TARGET CONDITIONS OF 35 °C AND 80% RH
(i.e., Test #20 with HCN)
91
-------
EPA TESTING DATA SHEET - Detector TestingJ 00000638-03TESTING
OjtL - ' ' '_ Sheet ID^JjldJ^Time: Start j^J^___ Finish lf^' "'
sum •* - - __~ ___
i esl ( hanihci ID ___ - J^__^L.
IX'tottoi hem.; tested (»as Alert Senior ( on^
Test being conducted: /j (_, — r^i
f~* f" — '••*
Environmental Chamber ID: *~~' '"" ""•—
Stopwatch ID:
Stopwatch Calibration Expiration Daie:..__
Actual Test Conditions at Start of Testing:
"') ,-"1 ... «, -~r -- ' ~,
^ ( I " .^ ^..^ ^ y ^ y-
Temperature: '--> ^ x < L- Relative Humidity: ^_.0 _/ /_. .^ _ % O2
Notes/Comments:
92
-------
Reference Method OC".-/- f /J
HP A TESTING DATA SHEET - Detector Testin&_100000638-03TESTlNG
Sheet 1D#.A.C"J- C/f.'- -2-
Detector being tested: Gas Alert
Method Zeroed (HHMM): ,A//lif
Calibration Standard:,^,- //.•.,",,..} /, t^rff-^
Zero Reading (ppm):
Calibration Formulation:
Time
ID
TIC
Target
Cone.
(ppm)
Actual
Cone.
(ppm)
File Name/Comment
Time
ID
TIC/I"
Cone.
I (ppm)
File Name/Comment
Date
7/7
93
-------
EPA TESTING DATA SHEET - Detector Testing 10
Time_ Concentration _ J%TTEl
/ : .
1 Time to i Time to Notes / Obser\ations
! Alarm i Clear '^^",^""0
i c >-,
94
-------
United States
Environmental Protection
Agency
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT NO. G-35
Office of Research and Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
------- |