Associations, Fire Department, Mounting Equipment

NFPA’s 1989 Standard: A Breath of Fresh Air

Issue 5 and Volume 16.

As you strap on your self-contained breathing apparatus (SCBA), are you sure the air you will be breathing is safe?

The answer to that question depends on what has been done in the past year to service and test your breathing air compressor. If your breathing air was tested quarterly as recommended by National Fire Protection Association (NFPA) 1989, Standard on Breathing Air Quality for Fire and Emergency Services Respiratory Protection, 2008 Edition, the safety of your air would be one less thing to worry about.

Becoming savvy with the NFPA 1989 standard and breathing air testing can save lives and prevent injuries. Testing breathing air requires more than checking the level of oxygen in a tank. In addition to low oxygen levels, other concerns can be high levels of moisture, carbon monoxide, carbon dioxide, hydrocarbons, oils, and particulates.

The atmospheric air that we breathe is composed of approximately 78 percent nitrogen, 21 percent oxygen, 1 percent argon, and many other trace components. The NFPA 1989 standard is one of the most stringent in the industry. Table 1 highlights the allowable limits of the specification for each component. Should compressed air samples fail to meet the NFPA limits, one of three problems exists: The air that was initially compressed was substandard, the processes of compressing the air altered the component ratios, or the sample was taken incorrectly or compromised during shipment to the testing laboratory.

A compressor works by pulling in a quantity of surrounding air and reducing its volume, thereby increasing the pressure of the air itself. This process essentially fits more air in the same location, which means that with more air, there will be a higher number of impurities. It is this magnification effect of compressing air that creates concern for the quality of breathing air. This is especially true if the local air is substandard.

Substandard air can be created, for example, if a compressor intake is near a high concentration of building or vehicle exhaust, causing elevated carbon monoxide and hydrocarbon levels. Compressors on the market today include filters designed to reduce or eliminate these impurities completely. The compressed air needs to be tested to ensure that the filters are working properly and that the air is safe.

To better understand why quarterly testing is necessary, it helps to understand the components of breathing air.

Table 1. NFPA 1989 Requirements
Component Allowable Limits
Nitrogen (percent) 75-81
Oxygen (percent) 19.5-23.5
Carbon Dioxide (ppmv) 1,000 Maximum
Hydrocarbon Total (ppmv) N/A
Methane (ppmv) N/A
Hydrocarbons as Non-Methane Equivalents (ppmv) 25 Maximum
Carbon Monoxide (ppmv) 5 Maximum
Argon (percent) N/A
Water (ppmv) 24 Maximum
Dew Point (°F/°C) N/A
Oil & Particulates (mg/m3) 2.0 Maximum (500 Liter Sample Volume)
Odor Not a pronounced or unusual odor
Testing Frequency Quarterly
Source: The National Fire Protection Association

Breathing Air Components

Quantifying the amount of oxygen present in a tank is a vital task of testing compressed breathing air. NFPA 1989 requires an oxygen level between 19.5 and 23.5 percent. The typical atmospheric level of oxygen at sea level is 20.9 percent, although the atmospheric oxygen level can vary based on factors such as elevation. Oxygen content outside of the required range can be caused either by the precompressed air being outside of this range or changes in the level caused by the compressing process itself.

Monitoring the oxygen level is crucial. Not surprisingly, a low level of oxygen can cause asphyxiation and death. High levels of oxygen are particularly concerning for deep sea diving or underwater activities, where the partial pressure of oxygen is elevated, causing a range of symptoms ultimately leading to death.

Along with oxygen, a specified percentage of nitrogen is also required in NFPA 1989. Nitrogen is an inert gas, which is generally harmless except in high concentrations, thereby causing a relative reduction in oxygen. Nitrogen in the atmosphere averages around 78 percent, and NFPA 1989 requires the nitrogen content in breathing air to be between 75 and 81 percent. The nitrogen level can offer a possible explanation if the oxygen level is outside the specification.

Unlike oxygen and nitrogen, the percentage of argon is not specified or required for the NFPA specification. Argon is an inert gas that is found in the atmosphere, typically at concentrations near one percent. Testing for detection of argon generally indicates how the breathing air was collected. Manufactured mixed air generally does not contain argon, but air that is sampled from an intake with atmospheric air will contain argon.

Carbon Monoxide

Carbon monoxide, frequently abbreviated as CO, is one of the most well known contaminants of breathing air. Carbon monoxide is an odorless, colorless, and tasteless gas, which is formed during incomplete combustion from a lack of oxygen supply. CO is dangerous because it binds with hemoglobin in the body and disrupts the flow of oxygen to the body, resulting in death at high exposures.

NFPA 1989 states that 5 parts per million (ppm) of CO is the permissible upper limit. Typically, a compressor has a carbon monoxide alarm, which will indicate when the level of CO is beyond the safety limit, although the detector will not remove CO. Removal is typically done by a manganese-copper catalytic compound called hopcalite, which converts the CO to carbon dioxide.

Quarterly testing of compressed air is crucial to ensure that the carbon monoxide filtering and detection systems are functioning properly.

Table 2. Air Impurities and Removal Mechanisms in a Compressor
Component Danger Removal Mechanism
Carbon Monoxide Binds with hemoglobin in the body and disrupts the flow of oxygen to the body, resulting in death at high exposure. Hopcalite
Water/High Moisture Level Moisture inside compressed breathing air can freeze, causing damage to the cylinder or regulator equipment. Can degrade the hopcalite filter system, resulting in elevated carbon monoxide levels. Desiccant
Carbon Dioxide Hypercapnia, also known as carbon dioxide poisoning-can cause shortness of breath, drowsiness, headaches, and ultimately unconsciousness. Activated Carbon
Hydrocarbons, Oils, and Odor Carcinogenic and the oils may build up over time in the lungs. Activated Carbon
Source: Dyne Technologies

Moisture Content

A leading cause of sample failure is high moisture content. NFPA 1989 states that a moisture reading above 24 ppm on a volume basis or a dew point warmer than -650F is considered elevated. Elevated moisture levels can be a serious concern, especially in climates with temperatures that can dip below freezing.

When exposed to cold temperatures, moisture inside compressed breathing air can freeze, causing damage to the cylinder or regulator equipment. Additionally, high moisture content can degrade the hopcalite filter system, which can in turn cause elevated carbon monoxide levels.

To remove moisture, most compressors contain a dryer cartridge filled with a desiccant material, which dries the air by collecting the water present. The desiccant acts like a sponge, soaking up the water. However, once the desiccant is fully saturated, it can no longer absorb any remaining water from the air stream and must be replaced. As such, it is important to run a quarterly test to ensure that the dryer cartridge is not saturated with water.

Typically, moisture content is measured on site with a colorimetric moisture tube. Or a sample can be sent to a laboratory for analysis.

Frequently, sampling errors occur because the compressor has not run long enough before taking the breathing air sample. It is recommended that a compressor run at least 10 minutes before taking a sample. In high-humidity or marine environments, 15 minutes or more is preferred. The reason is because water can accumulate inside the fill hose; if this water is not blown through the hose before testing, an elevated moisture level reading will occur.

Carbon Dioxide

Similar to carbon monoxide, carbon dioxide can be dangerous when present in compressed air. Carbon dioxide, abbreviated as CO2 , is also present in atmospheric air; depending on location, typical concentrations can range anywhere from 300 to 400 ppm.

The maximum acceptable level of carbon dioxide for compressed breathing air in NFPA 1989 is 1,000 ppm. High levels of CO2 may cause hypercapnia, also known as carbon dioxide poisoning, resulting in shortness of breath, drowsiness, headaches, and ultimately unconsciousness.

Activated carbon is a highly porous charcoal that is used to absorb carbon dioxide and hydrocarbons from the air stream.


An adage states that oil and water do not mix. The same is true with hydrocarbons and water. Hydrocarbons and oils present in compressed air may originate from the air intake, especially if that intake is near a vehicle or building exhaust location.

Hydrocarbons compromise a variety of compounds; some examples are methane, ethane and ethylene. NFPA 1989 dictates that a maximum of 25 ppm of nonmethane hydrocarbons as methane equivalents may be present in the compressed air.

Measurement of hydrocarbons as methane equivalents is a confusing concept but is actually used to make the quantification of total hydrocarbons simpler. Because of the large number of hydrocarbons that exist in the environment, it would be impossible to measure each compound and give a result. Therefore, testing is completed by defaulting to the measurement of hydrocarbons based on methane, which is the most abundant hydrocarbon found in the atmosphere.

Hydrocarbons are generally trapped in the compressor by a cartridge filled with a charcoal material called activated carbon. Based on their similar chemical nature, hydrocarbons remain on the activated carbon as the air passes through the cartridge. Many hydrocarbons along with oils and particulates have been proven to be carcinogenic, and the oils may build up over time in the lungs, making removal necessary.

Oil and particulates are generally large weight, high-boiling compounds. The major difference between hydrocarbons and oil and particulates is in their physical properties. Typically, hydrocarbons are lighter compounds and remain suspended in air and can therefore be tested in air along with oxygen and carbon monoxide. Oil and particulates, on the other hand, usually consist of heavy, high-boiling compounds that like to adhere to surfaces.

Testing for oil and particulates in any collection flask would be difficult because the oil would remain on the sampling container as opposed to the air stream, resulting in a false lower reading when sampling. Therefore, testing for oil and particulates is done with a filter or membrane to trap the oils and particles in the air stream.

NFPA 1989 requires that the oil and particulates be below a certain level, and a minimum of 500 liters of air must go through the filter for the testing to be valid. It is imperative to accurately quantify the volume of air that has flowed through the filter to calculate the oil and particulates in the sample. The quantity of air can be measured by either a flow meter or a pressure gauge. Be sure to check with your air-testing provider to ensure that the 500-liter requirement is met in your testing setup.

Testing compressed breathing air does not have to be a difficult process if patience and education are employed. When taking an air sample, read through the testing instructions thoroughly. Rushing through the sampling process will most likely cause errors that may require retesting.

NFPA 1989 says testing should be completed quarterly, especially since the hydrocarbon trap and dryer cartridge need to be replaced over time. This ensures safety for all personnel, and it improves familiarity with the compressor and air sampling equipment.

RITA SILBERNAGEL is the senior analytical chemist at Dyne Technologies LLC, a compressed breathing air and firefighting foam compliance testing laboratory in Minnesota.

More Fire Apparatus Current Issue Articles
More Fire Apparatus Archives Issue Articles