"It appears the problem cannot be isolated to newer manufacturing," Boord said. "The issue can apply to all CSE units."

The SR-100 model uses a chemical process to generate the oxygen needed for a 60-minute supply of breathable air. Generally, the SR-100 units are started by pulling a large orange tab that activates an oxygen cylinder. The cylinder inflates a breathing bag. Once a miner starts breathing through the bag, the exhaled gases react with the unit's chemicals to generate more oxygen for the miner.

SR-100s can also be started manually if the oxygen cylinder fails to inflate the breathing bag. But that process involves breathing ambient air -- which could be full of smoke -- and exhaling into the mouthpiece to start the chemical reaction.

Over the years, coal miners have complained repeatedly about SCSRs not starting or appearing to start slowly. Government and industry officials have generally dismissed those complaints. They said miners were not properly trained and did not understand how their SCSRs worked.

Months after the Sago disaster, survivor Randal McCloy said the SR-100s of four of the 12 miners trapped by the Jan. 2, 2006, explosion would not start. McCloy described how he tried especially hard to start the rescuer that belonged to his mining partner, Jerry Groves, who eventually died of carbon monoxide poisoning.

"I fought with it for I don't know how long, trying to mess with that valve, blow air through it, or anything I could do, but nothing would work," McCloy told investigators.

In lawsuits filed after the disaster, Sago miners' families were investigating concerns that the SR-100 oxygen cylinders -- made by South African-based African Oxygen Ltd. -- somehow leaked, leaving them without enough oxygen to properly start the units. Most of the suits against CSE have been settled, court records show.

CSE said in February that company representatives planned to visit mines with affected units "to replace the units and provide additional training and outreach."

But Shearer said the company has focused only on encouraging mine operators to remind workers of how to perform the manual start. Shearer said NIOSH would not currently allow his company to replace any problem SR-100s with new units.

Boord said the decision not to ship new units is actually a voluntary one by CSE. Boord said there is nothing preventing CSE from taking units that might be affected by the problem out of the mines -- but that mine operators would then have to replace them with units from one of CSE's competitors. So far, neither NIOSH nor MSHA has made any move to require such action.

"That would be a potential remedy," Boord said. "But until the problem is identified and a corrective action is identified, we can't really say what should happen."

Reach Ken Ward Jr. at kw...@wvgazette.com or 304-348-1702.

Read the user notice

Federal regulators are concerned the defect could affect 70,000 to 90,000 self-contained self rescuers in use by miners across the nation's coalfields, according to Les Boord, director of personal protective technology at the National Institute for Occupational Safety and Health.

"We think it's a problem," Boord said in a phone interview Tuesday.

Earlier Tuesday, the U.S. Mine Safety and Health Administration disclosed a "user notice" in which CSE itself acknowledged the potentially broader scope of problems with its SR-100 model.

"Until the root cause can be identified, we must assume that the potential for start-up oxygen cylinders to fail may extend to any field-deployed unit, and not just the serial numbers that were previously identified," CSE said in the notice, dated Monday.

In late February, Monroeville, Pa.-based CSE announced what it said at the time was a "recall" of more than 4,000 units because of concerns about the oxygen starter mechanism.

But contrary to the company's press release, CSE has not actually recalled any of the potentially affected units. None have been taken out of service or replaced, company president Scott Shearer said in an interview.

Instead, CSE is warning miners that the units might not start initially and urging mine operators to update training on a backup manual startup procedure.

"If for any reason a unit does not inflate the breathing bag, the user should don another unit if one is readily available," CSE said in a user notice issued Monday. "If a second unit is not readily available, the manual start should be used."

CSE had originally said in February that it believed the oxygen starter problem involved only a small portion of one lot of 4,000 SR-100s manufactured in May 2009.

The company said Tuesday its concerns had expanded to two lots and involved as many as 11,000 units. "We feel it is a very small part of the population, but we've taken the added precaution to alert everybody," Shearer said.

But NIOSH officials revealed that their agency and CSE had been jointly investigating similar oxygen starter problems with previously manufactured SR-100s since December -- two months before the company's February announcement.

"It appears the problem cannot be isolated to newer manufacturing," Boord said. "The issue can apply to all CSE units."

The SR-100 model uses a chemical process to generate the oxygen needed for a 60-minute supply of breathable air. Generally, the SR-100 units are started by pulling a large orange tab that activates an oxygen cylinder. The cylinder inflates a breathing bag. Once a miner starts breathing through the bag, the exhaled gases react with the unit's chemicals to generate more oxygen for the miner.

SR-100s can also be started manually if the oxygen cylinder fails to inflate the breathing bag. But that process involves breathing ambient air -- which could be full of smoke -- and exhaling into the mouthpiece to start the chemical reaction.

Over the years, coal miners have complained repeatedly about SCSRs not starting or appearing to start slowly. Government and industry officials have generally dismissed those complaints. They said miners were not properly trained and did not understand how their SCSRs worked.

Months after the Sago disaster, survivor Randal McCloy said the SR-100s of four of the 12 miners trapped by the Jan. 2, 2006, explosion would not start. McCloy described how he tried especially hard to start the rescuer that belonged to his mining partner, Jerry Groves, who eventually died of carbon monoxide poisoning.

"I fought with it for I don't know how long, trying to mess with that valve, blow air through it, or anything I could do, but nothing would work," McCloy told investigators.

In lawsuits filed after the disaster, Sago miners' families were investigating concerns that the SR-100 oxygen cylinders -- made by South African-based African Oxygen Ltd. -- somehow leaked, leaving them without enough oxygen to properly start the units. Most of the suits against CSE have been settled, court records show.

CSE said in February that company representatives planned to visit mines with affected units "to replace the units and provide additional training and outreach."

But Shearer said the company has focused only on encouraging mine operators to remind workers of how to perform the manual start. Shearer said NIOSH would not currently allow his company to replace any problem SR-100s with new units.

Boord said the decision not to ship new units is actually a voluntary one by CSE. Boord said there is nothing preventing CSE from taking units that might be affected by the problem out of the mines -- but that mine operators would then have to replace them with units from one of CSE's competitors. So far, neither NIOSH nor MSHA has made any move to require such action.

"That would be a potential remedy," Boord said. "But until the problem is identified and a corrective action is identified, we can't really say what should happen."

Reach Ken Ward Jr. at kw...@wvgazette.com or 304-348-1702.