Esta página en Español

cover image of NIOSH Alert  96-118


Preventing Carbon Monoxide Poisoning from Small Gasoline-Powered Engines and Tools

DHHS (NIOSH) Publication No. 96-118



Do not use equipment and tools powered by gasoline engines inside buildings or other partially enclosed spaces unless the gasoline engine can be placed outdoors and away from air intakes.


This ALERT is the joint product of a combined effort among the following agencies:

Each agency has a unique role in protecting workers, consumers, or the general public from safety and health hazards. Because of their common interest in prevention of carbon monoxide (CO) poisonings resulting from widespread use of small gasoline-powered engines and tools in enclosed or confined spaces, the agencies elected to work together to produce a joint document to address this problem and provide recommendations for prevention. Such a combined effort avoids duplication and confusion from multiple documents and promotes efficient use of government resources.


Principal contributors to the development of this joint ALERT include:

Preventing Carbon Monoxide Poisoning from
Small Gasoline-Powered Engines and Tools

Hundreds of people performing many different tasks have been poisoned because small gasoline-powered engines and tools produced hazardous concentrations of carbon monoxide (CO) even in relatively open buildings:

These are examples of the many situations in which people have been poisoned because they did not recognize the danger of using small gasoline-powered engines indoors. These poisonings can occur quickly, even in the presence of what many would consider "adequate ventilation" and in areas that many would define as relatively open spaces, such as parking garages.


CO is a lethal poison that is produced when fuels such as gasoline are burned. It is one of many chemicals found in engine exhaust and can rapidly accumulate even in areas that might appear to be well ventilated. Because CO is colorless, tasteless, odorless, and nonirritating, it can overcome the exposed person without warning. It produces weakness and confusion, depriving the person of the ability to seek safety.

CO poisons primarily by tightly binding to hemoglobin in the blood (forming carboxyhemoglobin), replacing oxygen, and reducing the oxygen-carrying capacity of the blood. CO may also poison by binding to tissues and cells of the human body and interfering with their normal function. Persons with pre-existing heart disease are at increased risk. Fetuses of pregnant women are also at increased risk-especially when mothers are exposed to high CO levels. Recognizing early warning signs of CO poisoning is sometimes difficult because early symptoms of CO exposure (headache, dizziness, and nausea) are nonspecific and may be mistaken for symptoms of other illnesses such as colds, flu, or food poisoning. Confusion and weakness can inhibit a person's ability to escape the hazardous environment.

The severity of symptoms of CO exposure is influenced by three main factors: (1) the concentration of CO in the environment; (2) how long the exposure lasts, and (3) work-load and breathing rate. In general, assuming that users of gasoline-powered engines are engaged in at least a moderate level of activity, exposure to CO concentrations of 80 to100 parts per million (ppm) for 1 to 2 hours can result in decreased exercise tolerance and, in persons who are at risk, may bring on chest pain and cause irregular heartbeat [EPA 1991a]. Symptoms associated with CO exposure concentrations of 100 to 200 ppm include headache, nausea, and mental impairment. More serious central nervous system effects, coma, and death are associated with CO exposure concentrations of 700 ppm or greater for an hour or more [Ilano and Raffin 1990; Forbes et al. 1945]. Symptoms of nervous system effects include staggering, confusion, changes in personality, and muscle aches. These symptoms may continue to occur for several days to several weeks after the exposure stops and the poisoned person has apparently recovered. Victims of CO poisoning should be immediately removed from the exposure site and given 100% oxygen. Hyperbaric chambers provide oxygen under pressure and are sometimes necessary in cases of serious CO poisoning.


Organizations set standards or make recommendations for exposure to hazardous substances based on assumptions inherent to their regulatory oversight or authority.  Differences in the stated values reflect variations in the place, duration, characteristics of the population, or proposed use.


The current Occupational Safety and Health Administration (OSHA) permissible exposure limit (PEL) for CO is 50 ppm as an 8-hour time-weighted average (TWA) [29 CFR 1910.1000*]. The NIOSH recommended exposure limit (REL) for CO is 35 ppm as an 8-hour TWA and a ceiling limit (CL) of 200 ppm [NIOSH 1992]. The NIOSH recommended immediately dangerous to life and health concentration (IDLH) for CO is 1,200 ppm. The IDLH is the concentration that could result in death or irreversible health effects, or prevent escape from the contaminated environment within 30 minutes. The American Conference of Governmental Industrial Hygienists (ACGIH) has adopted a threshold limit value (TLV) for CO of 25 ppm as an 8-hour TWA [ACGIH 1992a].
**Code of Federal Regulations. See CFR in references.

Ambient Air/Residential Settings

The U.S. Environmental Protection Agency (EPA) has established an ambient (outdoor) CO air quality Federal standard of 9 ppm for an 8-hour exposure and 25 ppm for a short-term (1-hr) exposure [EPA 1991a]. The Consumer Product Safety Commission (CPSC) staff recommends that long-term exposures to CO in indoor environments be limited to less than 15 ppm as an 8-hour TWA and 25 ppm for 1 hour, but product-specific recommendations for CO may vary depending on expected usage patterns and exposure.


Illness related to CO exposure is probably underestimated because workers with mild symptoms may go untreated or medical providers may not recognize their symptoms as CO poisoning. In addition, persons may not recognize the cause of their symptoms unless coworkers and other persons become ill at the same time.

Two surveys assessing individuals' beliefs, knowledge, and risk perceptions regarding CO suggest that many people are unaware of the hazards associated with CO. In 1993, NIOSH assessed flood victims' risk perceptions associated with CO poisoning from using equipment powered by small engines (e.g., gasoline-powered pressure washers indoors to clean up flood-related debris) [Greife et al. 1995]. Many of the 416 respondents (26%) incorrectly believed that with only a window open, the use of a gasoline-powered engine indoors would be safe. A majority of respondents (54%) and 92% of respondents between the ages of 12 and 20, incorrectly believed that it was safe to operate a gasoline-powered engine indoors with windows and doors open and an exhaust fan running. In a second survey, during follow-up investigations of nonfatal, unintentional CO poisoning in residential settings in Connecticut between November 1993 and March 1994, investigators interviewed 36 victims or their adult representatives [CDC 1995b]. Many of the victims of CO poisoning (poisoning that was related to heating systems, gas appliances, and fireplaces) still demonstrated a lack of knowledge about prevention strategies. When asked to list prevention methods, 14% were unable to list any method, 44% listed appropriate maintenance of appliances, 39% listed the use of a CO detector, and 14% listed proper ventilation.

Reports from a number of sources show that CO poisoning from the use of gasoline-powered tools indoors happens frequently:


Three of the above groups measured CO concentrations after the CO poisoning incidents in the same or similar exposure situations to estimate how quickly dangerous CO concentrations developed. A fourth group modeled the time to reach dangerous CO concentrations.

First, CDPHE attempted to estimate the CO exposure of the drywall texturizer discussed earlier by sampling the air at another construction site where he was doing similar work. On the day of the air sampling, the gasoline-powered compressor was placed just outside the garage door. Because of the way the equipment was designed and oriented, exhaust from the engine on the compressor went directly into the house when the garage door was open. As is usual for this operation, all windows and external doors in the home had been closed and sealed with tape and paper to protect the surfaces from the texturing material and to maintain the proper conditions for drying. The concentration of CO at the tailpipe of the compressor engine was substantially greater than 1,000 ppm (this was the upper limit of the testing equipment). Within the first 20 minutes of the operation, CO concentrations as high as 410 ppm were measured in the basement of the home, and concentrations as high as 322 ppm were measured where the worker was standing. CDPHE asked the worker to open the windows and external doors on the upper floor of the duplex because of concerns about this concentration of exposure. CO concentrations within the house dropped to approximately 30 ppm when that was done, but this is not the way the process is usually carried out.

In response to the second incident (another CO poisoning related to the use of an 8-horsepower pressure washer in a 30,000-cubic-foot room of a municipal construction project), CDPHE asked to run the same pressure washer in the same room a few days later. There was no mechanical ventilation in this room because the facility was not yet operational. The pressure washer was placed approximately 15 feet from one corner of the room (the same place the worker had placed it on the day of the poisoning). The power unit was an integral part of the washer. Again, the CO concentration at the engine exhaust pipe was greater than 1,000 ppm, the highest concentration CDPHE could measure at the time. CDPHE measured CO concentrations as high as 450 ppm at several locations in the room within 20 minutes of activation of the washer engine, and concentrations as high as 546 ppm approximately 50 minutes after the washer engine was activated. The test was then terminated.

In a third incident, CDPHE asked the managers of the enclosed municipal water treatment plant to recreate the exposure situation encountered by the worker mentioned before who was using the 8-horsepower pump in the 59,000-cubic-foot room (48 × 88 × 14 feet). This room was only partially enclosed so that employees could observe operations in the room from the level above. Outside air was introduced into the area through a forced-air heating system that was running on the day of the poisoning and on the day of air sampling. External doors to the treatment plant were opened on both days as well. Ten minutes after the pump engine was started, CO concentrations as high as 395 ppm were measured within 7 feet of the pump, near the location where the employee was standing for much of the time on the day of the poisoning. CO concentrations 25 feet from the water pump rose to as high as 193 ppm during the 20-minute test. CDPHE returned to the room 1 hour after the water pump was stopped and measured 40 ppm of CO.

Finally, in January 1996, two Colorado workers were poisoned as a result of operating a gasoline-powered, 5-horsepower, walk-behind concrete saw during a remodeling project. The machine was 3 years old and was used two to three times per year. The workers operated the saw for about an hour and a half inside what had previously been two bathrooms (the dividing wall had been removed and the volume of the room was 2,332 cubic feet). The workers were cutting a hole in the floor to allow access to pipes below the floor. The two doors to the room were open and the bathroom ventilation system was operating when these poisonings occurred. The day after the poisonings occurred, the work in this bathroom was continued with two differences: A cooling fan was used this time in an effort to better move CO from the room, and the saw was operated for shorter periods (the periods of operation were not clearly defined but were thought to be 15 to 30 minutes). CDPHE recreated the second day's operating conditions to measure CO concentrations in the room. The NIOSH ceiling limit of 200 ppm was exceeded within the first minute of operation. Within 5 minutes of operation, the CO concentration within the room reached 842 ppm, at which point the demonstration was discontinued (see Figure 1).

 CO Concentration versus time - 2,332 ft3 room chart

Figure 1. Actual CO concentrations measured inside a 2,332-cubic-foot bathroom with a gasoline-powered, 5-horsepower concrete saw operating (doors open, cooling fan and ventilation running).

 CO Concentration versus time - 1,000 ft3 room chart

Figure 2. Calculated CO concentrations generated by a 5-horsepower, 4-cycle gasoline-powered engine in a 1,000-cubic-foot room with various air changes per hour.

 CO Concentration versus time - 10,000 ft3 room chart

Figure 3. Calculated CO concentrations generated by a 5-horsepower, 4-cycle gasoline-powered engine in a 10,000-cubic-foot room with various air changes per hour.

 CO Concentration versus time - 100,000 ft3 room chart

Figure 4. Calculated CO concentrations generated by a 5-horsepower, 4-cycle gasoline-powered engine in a 100,000-cubic-foot room with various air changes per hour.


It is not widely known that small gasoline-powered engines and tools present a serious health hazard. They produce high concentrations of carbon monoxide, a poisonous gas that can cause illness, permanent neurological damage, and death. Because it is colorless, odorless, and nonirritating, CO can overcome exposed persons without warning. Often there is little time before they experience symptoms that inhibit their ability to seek safety. Prior use of equipment without incident has sometimes given users a false sense of safety; such users have been poisoned on subsequent occasions. Recommendations for preventing CO poisoning are provided below for employers, equipment users, tool rental agencies, and tool manufacturers.

All Employers and Equipment Users Should:

Employers Should Also:

Equipment Users Should Also:

Tool Rental Agencies Should:

Tool Manufacturers Should:


NIOSH, CDPHE, CPSC, OSHA, and EPA request that the information in this ALERT be brought to the attention of (1) all employers and workers who use small gasoline-powered engines and tools in their jobs and trades (e.g., building, construction, agriculture, and maintenance and cleaning operations), (2) tool rental agencies and equipment sellers and users, (3) tool manufacturers, and (4) editors of appropriate trade journals.


ACGIH [1992a]. 1992-1993 threshold limit values for chemical substances and physical agents and biological exposure indices. Cincinnati, OH: American Conference of Governmental Industrial Hygienists.

ACGIH [1992b]. Industrial ventilation--a manual of recommended practice. 21st ed. Cincinnati, OH: American Conference of Governmental Industrial Hygienists, Committee on Industrial Ventilation, pp. 2-1 to 2-16.

BLS [1992a]. Census of fatal occupational injuries. Washington, DC: U.S. Department of Labor, Bureau of Labor Statistics. Unpublished data.

BLS [1992b]. Survey of occupational injuries and illnesses. Washington, DC: U.S. Department of Labor, Bureau of Labor Statistics. Unpublished data.

CDC (Centers for Disease Control and Prevention) [1993]. Unintentional carbon monoxide poisoning from indoor use of pressure washers--Iowa, January 1992-January 1993. MMWR 42(40):777-779, 785.

CDC (Centers for Disease Control and Prevention) [1995a]. Carbon monoxide poisoning from use of gasoline-fueled powered washers in an underground parking garage--District of Columbia, 1994. MMWR 44(18):356-357, 363-364.

CDC (Centers for Disease Control and Prevention) [1995b]. Unintentional carbon monoxide poisonings in residential settings--Connecticut, November 1993-March 1994. MMWR 44(41):765-767.

CDHS [1993]. Causes of unintentional deaths from carbon monoxide poisonings in California. Sacramento, CA: California Health Services Department.

CDPHE [1996]. Occupational carbon monoxide poisonings in Colorado. Denver, CO: Colorado Department of Public Health and Environment. Unpublished data.

CFR. Code of Federal regulations. Washington, DC: U.S. Government Printing Office, Office of the Federal Register.

CPSC [1994]. National electronic injury surveillance system. Washington, DC: Consumer Product Safety Commission.

Ehlers J [1994]. Carbon monoxide poisoning among Iowa farmers while using gasoline-powered washers--a case series. Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health. Unpublished report.

EPA [1991a]. Air quality criteria for carbon monoxide. Washington, DC: U.S. Environmental Protection Agency, Office of Research and Development, Publication No. EPA-600/8-90/045F.

EPA [1991b]. Nonroad engine and vehicle emission study-report. Washington, DC: U.S. Environmental Protection Agency, Office of Air and Radiation, Publication No. EPA 21A-2001.

Forbes WH, Sargent F, Foughton FJW [1945]. The rate of CO uptake by normal man. Am J Physiol 143:594-608.

Greife A, Goldenhar LM, Freund E, Stock A, Hornung R, Connon C, et al [1995]. Risk perception of carbon monoxide poisoning from gasoline-powered engines among midwest flood victims. Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health. Unpublished report.

Ilano A, Raffin T [1990]. Management of carbon monoxide poisoning. Chest 97:165-169.

NCHS/CPSC [1992]. Death certificate file. Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Center for Health Statistics; and U.S. Consumer Product Safety Commission.

NIOSH [1992]. NIOSH recommendations for occupational safety and health: compendium of policy documents and statements. Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH) Publication No. 92-100.

Venable H, Wallingford K, Roberts D, Booher D [1995]. Simulated carbon monoxide exposure in an enclosed structure from a gasoline-powered pressure washer. Appl Occup Environ Hyg 10(7):581-584.


Carbon Monoxide Monitors and Detectors

Detectors for carbon monoxide (CO) are manufactured and marketed for use in either the home or occupational industrial settings. The detectors for home use are devices that will sound an alarm before CO concentrations in the home become hazardous. There is an Underwriters Laboratories, Inc., performance standard (UL 2034) for residential CO detectors. Detectors currently available on the market are battery-powered, plug-in, or hard-wired. Some models incorporate a visual display of the parts per million (ppm ) concentration of CO present in the home. For more information on CO detectors for home use, call the Consumer Product Safety Commission Hotline at 1-800-638-2772.

CO detectors for use in residential settings are not designed for use in typical workplace settings. Monitoring requirements in an occupational setting are different from monitoring requirements in the home. In the workplace, it is frequently necessary to monitor a worker's exposure to carbon monoxide over an entire work shift and determine the time-weighted average (TWA) concentration of the exposure. It may also be necessary to have carbon monoxide monitors with alarm capabilities in the workplace. CO in the workplace can be detected using detector tubes, direct-reading passive badges, dosimeter tubes and direct-reading instruments. These badges, tubes, and instruments operate on a variety of principles including colorimetric reaction, potentiometry, coulometry, infrared spectrometry, fluorescence, thermal conductivity, and heat of combustion. The direct-reading instruments are frequently equipped with audio and/or visual alarms and may be used for area and/or personal exposure monitoring. Some have microprocessors and memory for storing CO concentration readings taken during the day. It is significant to note that some of the devices mentioned for workplace CO monitoring are not capable of monitoring TWAs, and not all are equipped with alarms. The appropriate monitor must be chosen on an application-by-application basis. For more information on the availability of workplace CO monitors or their application, call the National Institute for Occupational Safety and Health at 1-800-35-NIOSH (1-800-356-4674).


Mention of any company or product does not constitute endorsement by the National Institute for Occupational Safety and Health, the Colorado Department of Public Health and Environment, the U.S. Consumer Product Safety Commission, the Occupational Safety and Health Administration, or the U.S. Environmental Protection Agency.

This document is in the public domain and may be freely copied or reprinted.

Copies of this and other documents are available from:

To receive other information about occupational safety and health problems, call 1-800-35-NIOSH (1-800-356-4674).

DHHS (NIOSH) Publication No. 96-118

This page was last updated : December 19, 1996

Go back to the NIOSH home page NIOSH logo or to the CDC home page. CDC logo