BOSTON - A gathering of experts on carbon monoxide (CO) testing and measurement spoke with attendees of the ISH-NA trade show in mid-October.

The group included Jim Davis, senior instructor/technical development, National Comfort Institute Inc., Sheffield Lake, Ohio; Rudy Leatherman, training and technical support, Bacharach Inc., Athens, Ohio; Tim McElwain, president of Gas Appliance Service Training and Consulting & Gas Training Institute, Warren, R.I.; Mark Hunt, owner of Comfortable Home Technologies, Ballston Lake, N.Y.; and Bill Spohn, product manager/HVAC, Testo Inc., Flanders, N.J.

Davis spoke first about some of the "problems" associated with detecting and preventing CO poisoning. "Somehow we have standards in our industry that say we will not inconvenience our customers while we kill them," he said. "Every person who died of CO poisoning was warm and clean when they died. We keep them warm and clean.

"Our job is to inconvenience them when it comes to safety issues. It's funny - customers actually get mad at us for inconveniencing them."

Davis pointed out that there are 20,000 to 50,000 deaths each year from CO poisoning-related in-juries. He said the people attending the ISH show may be facing health risks, too.

"Some of you are being poisoned by your own hotel rooms," he stated, noting that he carries his CO test instruments everywhere he goes to test the air quality.

Davis said it is a mistake to suggest that licensing home inspectors and service techs will solve the problem. "A license does not give us the ability to do things right. It gives us an excuse when we do things wrong," he noted. "Licensing is not the cure - good education is."

He said it is the contractor's responsibility to ensure that equipment operates safely and efficiently, not the responsibility of anyone else, including building code officials. "Not one code official has ever been held liable for a bad installation," Davis said.

Tim McElwain


McElwain said his reason for starting in CO testing began after seeing a news report about a family of eight - two adults and six children - who all died from accidental CO poisoning. "I had to know about CO," he said. "I had been through a lot of gas appliance service training, but CO was not talked about very much."

McElwain believes training goes far beyond learning how to correct equipment problems. "My motivation for [being here today] is to express the moral obligation that we have to protect our customers and ourselves," he added. "It is not unusual for a technician to put himself in danger because he didn't know what he was walking into."

He added that low-level CO poisoning can go undetected when the proper testing equipment is not used, and that technicians "may never know they are getting poisoned."

He emphasized, "If you don't test, you don't know. A clean, blue flame does not mean that the appliance is burning cleanly. You can't use visual observation to confirm a problem. Make sure you have a process of testing the equipment and that the equipment is burning properly."

McElwain took a moment to ask if those in the audience had CO detectors in their homes and if they tested them. "How do you know they work?" he asked. "The test button only tells you if the batteries are good.

"You may be using an alarm that only sounds at CO levels that are higher than acceptable levels for certain people, like the young, elderly, or those with breathing problems."


Hunt said that his original training in testing gas and oil-burning equipment was rather basic. "I was taught to drill a hole in a pipe, take a whole bunch of air and put it into a bottle, turn the bottle upside down, turn it right side up again, and get a number," he said. "That number would tell me if the equipment was burning cleanly."

Hunt said that attending a CO seminar convinced him to change. He felt guilty that he had put people at risk because he failed to test the equipment after he had installed it. "CO poisoning is the No. 1 cause of accidental poisoning in the U.S.," he said.

He said a lot of injuries can be prevented if service techs are properly trained and if they carry test equipment with them wherever they go. That includes being able to test the equipment after they have installed it.

"When you start a fire [in the equipment] and flip the switch, I contend that the most important part of our job begins," Hunt said. "People trust me to go in their basements and set things on fire. They trust that when I put it in their vents, it will go up the chimney."

Hunt gave the example of a homeowner who had remodeled his kitchen and installed a 1,000-cfm exhaust fan. The fan was venting out the air, which had to be replaced by outside air coming down the chimney and pulling exhaust gases from the boiler and hot water heater.

"And the report said that a faulty water heater caused a buildup of CO gas," he joked. "There was nothing wrong with the water heater."

Low Levels Of CO

Leatherman agreed that methods of testing and analyzing equipment are often wrong and lead to incorrect diagnosis.

"I've seen technicians test for CO by sticking a probe into a heating vent," he said. "If CO is present they say it is probably a cracked heat exchanger. That's probably the last thing it is - the CO could be coming from the garage!"

Leatherman believes that store-bought CO detectors may do more harm than good - because most will not display or alarm until there are potentially high levels of CO in the air.

"Thirty ppm [parts per million] is the lowest level that a [United Laboratories] U.L.-approved CO alarm is allowed to display - not alarm," he said. "Seventy ppm is the lowest level for a U.L.-approved alarm to sound."

The problem is that at those levels, certain people, such as infants, the elderly, and people with chronic breathing problems, may be in danger. Leatherman said it is the low-level CO poisoning that is slowly injuring and killing people and not just the high levels which occur during fires or malfunctioning equipment.

He told the story of a television news station that conducted a test of eight store-bought CO detectors. Each was placed in a sealed container and subjected to 100 ppm of calibrated CO. None of the alarms sounded. The eight alarms were later placed by a car's exhaust system and only five sounded.

"The case on one was starting to melt and it hadn't even alarmed," he noted. Leatherman added that the U.L. tests only check to see if the alarm works, not the sensor.

Leatherman also pointed out that some CO detectors have an optional wall plug built in, as well as a battery backup, in lieu of hard wiring the alarm to the home's electrical system.

"The problem is, most outlets are near the floor, where there is no CO," he said. "CO is lighter than air and travels upward. And batteries don't last very long, and need constant replacing."

He asked attendees if they remembered when smoke detectors first came out, they used to alarm when the stove was used or someone was smoking. He then asked if the latest ones they bought had the same problem. The answer was no.

"That's because the manufacturers are dumbing up the alarms," Leatherman added. "The same thing is happening to CO alarms."

Spohn said that false alarms used to be big problems in CO detectors. In fact, he helped form the Carbon Monoxide Awareness Coalition in Pittsburgh back in the early 1990s because "we found that everyone's alarm was going off back then.

False Alarms

"A lot of learning has gone into CO detectors because of the false alarming," Spohn said. "The standards have been loosened so much that the threshold of alarms to sound has gotten very high."

He added false alarms are sometimes called "nuisance" alarms, and standards have been changed to reduce the number of "nuisances."

As a result of looser standards, Spohn noted that there are now 29 different brands of CO detectors in the United States and Canada, up from 19 in 2001.

He talked about U.L. standard 2034, which tests CO alarms, and offered this interpretation of one part of the standard: "According to U.L. 2034, alarms must ignore 70 ppm for at least 60 minutes and could ignore 70 ppm for up to 239 minutes."

Publication date: 11/01/2004