Content uploaded by Steven Emmerich
Author content
All content in this area was uploaded by Steven Emmerich on Oct 03, 2022
Content may be subject to copyright.
Carbon Monoxide and Portable Generators
Steven J. Emmerich
Andrew K. Persily
Liangzhu (Leon) Wang
Engineering Laboratory, National Institute of Standards and Technology
100 Bureau Drive Gaithersburg, MD 20899
Content submitted to and published by:
ASHRAE Journal
Volume 56; Issue 9, pp.92-98
U.S. Department of Commerce
Penny Pritzker, Secretary of Commerce
National Institute of Standards and Technology
Willie E May, Acting Director
DISCLAIMERS
Certain commercial entities, equipment, or materials may be identified in this document in order
to describe an experimental procedure or concept adequately. Such identification is not intended
to imply recommendation or endorsement by the National Institute of Standards and Technology,
nor is it intended to imply that the entities, materials, or equipment are necessarily the best
available for the purpose.
Any link(s) to website(s) in this document have been provided because they may have
information of interest to our readers. NIST does not necessarily endorse the views expressed or
the facts presented on these sites. Further, NIST does not endorse any commercial products that
may be advertised or available on these sites.
Carbon Monoxide and Portable Generators
Steven J. Emmerich, Andrew K. Persily, and Liangzhu (Leon) Wang
Serious concerns exist about the hazard of acute residential carbon monoxide (CO) exposures
from portable gasoline-powered generators, which can result in death or serious adverse
health effects. As of April 23, 2013 and as shown in Figure 1, the U.S. Consumer Product
Safety Commission (CPSC) databases contain records of at least 800 deaths (involving 597
incidents) from CO poisoning caused by consumer use of a generator in the period of 1999
through 2012 (Hnatov 2013). Typically, these deaths occur when consumers use a generator
in an enclosed or partially enclosed space or, less often, outdoors near a partially open door,
window or vent. While avoiding the operation of such generators in or near a home is
expected to reduce indoor CO exposures significantly, it may not be realistic to expect such
usage to be eliminated completely.
Figure 1. Increase in generator-related CO poisoning deaths since 1999 (Hnatov 2013)
Another means of reducing these exposures would be to decrease the amount of CO emitted
from these devices. The magnitude of such reductions needed to reduce exposures to a
specific level depends on the complex relationship between CO emissions from these
generators and occupant exposure. In order to better understand the CO emissions from
portable generators, the potential for reducing these emissions and the impacts on occupant
exposure, a multi-year research effort was conducted involving both experimental and
simulation studies (Emmerich et al. 2013 and Persily et al. 2013).
Measurements of CO emissions from portable generators
To better understand CO emission rates from both stock (currently available) and reduced-
emission prototype portable generators operating in an enclosed space under real weather
conditions, experiments were conducted in a single zone shed and in a three-bedroom test
house with an attached garage. This paper summarizes the measurements conducted in the
shed; the tests with the generator operating in the attached garage are described in Emmerich
et al. (2013).
The shed experiments were conducted in a 43 m3 single-walled, uninsulated timber structure
for the purpose of measuring the CO emission rate and O2 consumption rate of the generators.
Figure 2 shows a generator installed in the shed along with the load bank used to place an
electric load on the generator. The shed also had two operable windows at both sidewalls and
an exhaust fan, which were used to vary the air change rate during the tests from about 0.5 h-1
to 10 h-1. Tests were conducted with three different generators that were configured in
multiple ways. Two unmodified ‘stock’ (i.e., in their as-purchased condition) generators were
tested. The first generator has a full-load power rating of 5.5 kW with a 10 horsepower,
carbureted, single cylinder gasoline engine and no CO emission control technology. The
second generator is powered by a carbureted 11 horsepower single-cylinder gasoline and has
an advertised full-load electric power rating of 5.0 kW. This generator was tested in both its
stock, unmodified condition and modified as a low-CO emission prototype. The modifications
included an engine management system (EMS) with sensors and actuators for electronic fuel
injection (replacing the carburetor) and a muffler with a small catalytic converter. The third
generator was similar to the second, but with an output rating of 7 kW and a different model
EMS.
Figure 2. Generator in test shed
Figure 3 shows the measured CO and O2 concentrations for two of the experiments with the
first, unmodified generator (see Emmerich et al. 2013 for an explanation of the test
conditions). The patterns of CO concentrations in both tests are almost an inverse to the O2
levels for this unmodified generator. The CO level is low at the beginning of generator startup
and increases steadily as the O2 level drops. As the O2 drops further, causing a very rich fuel
mixture in the engine, CO generation reaches a maximum level. Test 13 shows an extreme
case in which the generator eventually produces a zero electrical load when the O2 drops to
around 16.4 %, although it was set at a full load and the crankshaft was still rotating.
16
16.5
17
17.5
18
18.5
19
19.5
20
20.5
21
0
5000
10000
15000
20000
25000
30000
35000
0 10 20 30 40 50 60 70 80 90 100
O2(%)
CO (µL/L)
Time (min)
CO( Tes t 1 )
CO(Test 13)
O2(Test 1)
O2(Test 13)
Generator off(Test 1)
Generator on(Test 13)
Generator on(Test 1)
Zero load output(Test 13)
Generator off(Test 13)
Figure 3. Measured CO and O2 concentrations of Tests 1 and 13.
In order to generalize these test results to other conditions beyond these particular tests, it is
necessary to convert the results into CO emission and O2 consumption rates. Figure 4 shows
5-min average CO emission rates as a function of O2 levels in the thirteen shed tests of the
first, unmodified generator. For both full and half load settings, CO emission rates increase
with decreasing O2, reaching maximum values when O2 drop to about 17 % to 18 %, and then
decline at lower O2 levels. Under the extreme case of Test 13 (5.0kw-CW-LA), the CO rate
decreases dramatically as the O2 level reaches around 16.4 % with an electrical output of
zero. The solid points in Figure 4 are data points for a half-load setting (2.5 kW) and the
hollow ones for a full load setting (5.0 kW).
0
500
1,000
1,500
2,000
2,500
3,000
3,500
4,000
4,500
16 16.5 17 17.5 18 18.5 19 19.5 20 20.5 21
CO Emission Rate (g/h)
O2(%)
2.5kw-OW-1
2.5kw-OW-2
2.5kw-CW-1
2.5kw-CW-2
2.5kw-CW-3
2.5kw-CW-S1
5.0kw-CW-1
5.0kw-CW-2
5.0kw-OW-1
5.0kw-OW-2
5.0kw-CW-S1
5.0kw-CW-S2
5.0kw-CW-LA
Figure 4. Five-minute averaged CO emission rates at different O2 levels.
The second generator was tested in both unmodified and modified (low CO emission)
configurations. Figure 5 presents the CO emission rates as a function of O2 levels for the
unmodified generator, while Figure 6 presents the CO emission rates as a function of O2
levels for the modified generator. Although the modified generator was not tested as many
times as the unmodified version, these figures show the dramatic reduction in CO emission
rates due to the low CO emission modifications included on the prototype. Most of the
modified generator’s emission rates were well below 500 g/h.
0
500
1,000
1,500
2,000
2,500
3,000
3,500
4,000
16 17 18 19 20 21
CO Emission Rate (g/h)
O
2
Level (%)
Idle(CO)
0.5kW(CO)
1.5kW(CO)
2.5kW(CO)
4.0kW(CO)
5.5kW(CO)
5.5kw-2(CO)
4.5kw(CO)
3.0kW(CO)
5.5kW-2-NOVA2(CO)
1.5kW-2(CO)
0.5kW-2(CO)
Idle-2(CO)
4.5kW-2(CO)
3.0kW-2(CO)
5.5kW-3(CO)
0.5kW-3(CO)
3.0kW-3(CO)
5.5kW-4(CO)
Idle-4(CO)
0.5kW-4(CO)
1.5kW-4(CO)
3.0kW-4(CO)
3.0kW-4(CO)
4.5kW-5(CO)
5.5kW-4b(CO)
5.5kW-5(CO)
5.5kW-6(CO)
Figure 5. CO emission rates at different O2 levels for unmodified Generator X.
0
500
1000
1500
2000
2500
16 17 18 19 20 21
CO Emission Rate (g/h)
O
2
Level (%)
Idle -GenXmod(CO)
0.5kW -GenXmod(CO)
1.5kW-GenXmod(CO)
3.0kW -GenXmod(CO)
4.5kW -GenXmod(CO)
5.5kW -GenXmod(CO)
Figure 6. CO emission rates at different O2 levels for modified generator
Simulations of CO Exposure from Portable Generators
To address the CO exposure associated with portable generators and to support potential
control strategies such as reduced emissions, a better understanding of the relationship
between CO emission rates and occupant exposure is needed. This relationship involves the
interaction between generator location and operation, house characteristics, occupant location
and activities, and weather conditions. In order to support life-safety based analyses of
potential CO emission limits for generators, a computer simulation study was conducted to
evaluate indoor CO exposures as a function of generator source location and CO emission
rate. Simulations were performed using the multizone airflow and contaminant transport
model CONTAM (Walton and Dols 2005), which was applied to 87 single-family, detached
dwellings that are representative of the U.S. housing stock. Using these homes, indoor CO
concentrations were calculated over a range of generator locations, CO emission rates, and
weather conditions. These simulations yielded CO concentrations in the rooms of each house
as a function of time during the 24-h analysis interval. In order to compare the results for
different cases, the concentrations from each simulation were used to calculate COHb values
in each occupied room. The maximum COHb value among the occupied rooms was used as a
metric of CO exposure for each combination of house, source, and weather.
The results of the simulations constitute a large amount of data, which can be interpreted by
considering the percentage of cases simulated that meet a specific criterion for the target value
of maxCOHb. Determination of such criteria was beyond the scope of this project but for
comparison purposes, the maximum source strength was estimated for which 80 % of the
cases simulated are below 30 % maxCOHb for each of the source locations considered. The
values of 80 % below 30 % maxCOHb are used only for illustrative purposes and are not
presented as life-safety based limits to support any policy or regulatory decisions.
Considering all the constant source results, the maximum source strength corresponding to
80 % of the cases having a value of maxCOHb below 30 % is 27 g/h. Note that the CO
emission rates measured in unmodified generators mentioned earlier tended to be well above
this value, but that the modified generators tested were in this range.
In 2006, CPSC issued an Advance Notice of Proposed Rulemaking; Request for Comments
and Information describing its strategy to reduce generator engine CO emission rates.
Additionally, Underwriters Laboratories, Inc. has formed a working group to develop a
specific proposal for requirements for portable engine-generator sets that fall under the scope
of UL 2201, Portable Engine-Generator Assemblies to reduce the risk of death and injury due
to CO poisoning.
Acknowledgement
This research was supported by the U.S. Consumer Product Safety Commission (CPSC)
under interagency agreement CPSC-I-06-0012.
References
Emmerich SJ, Persily AK, and Wang L 2013. Modeling and Measuring the Effects of
Portable Gasoline Powered Generator Exhaust on Indoor Carbon Monoxide Level. NIST
Technical Note 1781. National Institute of Standards and Technology
Hnatov MV. 2013. Incidents, Deaths, and In-Depth Investigations Associated with Non-Fire
Carbon Monoxide from Engine-Driven Generators and Other Engine-Driven Tools, 1999-
2012; U.S. Consumer Product Safety Commission: Bethesda, MD.
Persily, A.K., Wang, Y., Polidoro, B. and Emmerich, S.J. 2013. Residential Carbon
Monoxide Exposure due to Indoor Generator Operation: Effects of Source Location and
Emission Rate. NIST Technical Note 1782. National Institute of Standards and Technology.
Walton, G.N. and W.S. Dols. 2005 CONTAMW 2.4 User Guide and Program Documentation.
NISTIR 7251, National Institute of Standards and Technology.
