Monday, May 01, 2006

Sago Update: Malfunctioning Alarms, Missing Lightning Arrestors, Weak Seals and an "Improbably Agile Bolt of Lightning"

Workplace accidents generally have many causes. There are the "direct" causes -- someone pushed the wrong button, didn't follow written procedures or ignored an alarm.

Then there are the root, or systemic causes: the pressure to speed productoin that causes workers to rush and neglect safe work procedures, written procedures that don't match the current configuration of the workplace and that no one is trained to follow, alarms that are always malfunctioning, so everyone ignores them.

So when you seen something like this, you need to suspect that there's something bad going on in this workplace.
At 6:10 a.m. on Jan. 2, a light on Sago Mine dispatcher Bill Chisolm’s computer screen changed from green to red.

The red light was an alarm. It warned Chisolm of an increase in carbon monoxide levels along the conveyor belt in Sago’s 1 Left section.

Under mine policy and federal rules, the 26 parts per million of carbon monoxide detected should have prompted an evacuation. Workers should have been cleared from areas deeper underground, probably including a crew headed to work in the next section over, called 2 Left.

But Chisolm ignored the alarm. He was sure it was a malfunction, and not a real signal of any problems underground, according to a sworn statement given to government investigators.

About 20 minutes later, shortly after 6:30 a.m., an explosion ripped through the Sago Mine.
First reaction is "Ah ha! Stupid worker mistake. Fire the bastard and all will be well."

But not so fast.

First, that alarm may not have been warning of the conditions that led to the explosion and fire that killed 12 Sago miners. But more important, further investigation reveals that Bill Chisolm’s failure to heed the alarm points not to an individual mistake, but to serious systemic safety problems at the mine.
Under federal rules, mine dispatchers are required to closely investigate the cause of any carbon monoxide alarm that indicates more than 10 parts per million of the gas. Miners who are not investigating the problem are to be evacuated from the area, the rules state.

At Sago, several mine dispatchers testified that they received very limited training on how to respond to mine carbon monoxide alarms. Several dispatchers were unable to answer detailed questions investigators asked about logs from the mine’s carbon monoxide monitoring system.

Dispatcher Nathan Eye testified that when he took the job, it “was supposed to be a temporary position over there, and I kind of got stuck with it.”

Eye said he did not know what concentration of carbon monoxide would require a mine evacuation.

“It had never really been discussed, but I would figure anything about 20 parts per million would be way too much to leave anybody [inside],” Eye said.

Dispatchers told investigators that the carbon monoxide alarms frequently malfunctioned at the Sago Mine.

“I’ve had CO monitors malfunction for no apparent reason,” dispatcher Vernon Hofer said during a Jan. 23 interview. “They just malfunction.

“And I don’t know the cause of the malfunction or why they — and when they show an alarm, if at the point in time that I check them, everything appears to be OK,” Hofer said.

Dispatchers also indicated they used the alarm system for purposes other than keeping an eye on carbon monoxide levels.

Chisolm told investigators that dispatchers would set off audible carbon monoxide alarms in the Sago Mine, “if you’re having trouble getting a hold of a section, it could be maybe your mom called, she’s in the hospital or anything.”

In interviews with investigators, Sago Mine managers also have revealed that they regularly failed to keep accurate records of the operations of the mine fan that was meant to keep clean air flowing through the underground workings.
Meanwhile the ongoing investigation of the Sago disaster is focusing on what caused the methane explosion in a closed off part of the mine. Sago's owner, International Coal Group (ICG), issued a report saying that a lightning strike caused the explosion. But investigators can't figure out how what the Pittsburgh Post Gazette calls "an improbably agile bolt of lightning" traveled 1-1/2 miles from the mouth of the mine, across the Buckhannon River, then another 13,000 feet to the sealed portion of the mine.

And blaming the explosion on lightning -- even if it's true -- may not get ICG off the hook. Ken Ward of the Charleston Gazette also reports that
The Sago Mine violated basic electrical safety rules by not installing equipment to prevent lightning from sending a charge into underground mine workings, U.S. and West Virginia investigators have learned.

At least two electrical systems at Sago were not equipped with lightning arresters similar to surge protectors, the mine’s chief electrician told investigators in a sworn statement.


Sago Mine owner International Coal Group has pushed the theory that lightning caused the explosion. But company press releases have not mentioned the serious electrical violations related to the lack of lightning-protection devices.


Under federal mine safety rules, all power lines and phone cables that lead into underground mines must be equipped with lightning arresters.

Lightning arresters are protective devices that limit surges of electricity from lightning strikes or equipment failures. They prevent damage to electrical equipment and, in the case of underground coal mines, help to prevent lightning from sparking fires or explosions.

The investigation is also looking into the blocks that were used to seal the closed-off part of the mine:
The seals were constructed from Omega Block, a cement-and-fiber foam block favored by many mine operators because they are lighter than the traditional cement blocks used to seal abandoned areas of mines.

The Mine Safety and Health Administration first approved the blocks for use nearly a decade ago, but more recently allowed the installation of the blocks without the traditional "hitching" -- the practice of digging a notch into the mine wall and ceiling to secure the seal. Unhitched Omega Block walls were approved after one such wall withstood the minimum 20 pounds per square inch blast pressure during a test of seals meant to be erected during mine emergencies.

But testimony by the men who installed the seals at Sago suggests the wall did not follow the approved plan in all instances and did not match the construction of the Omega Block wall that passed a 20 psi test.

Notably, they testified that they leveled out the floor by laying dry mortar into gaps in the mine floor and then setting the wall atop it. Plans called for all sides of the block floor to be mortared with a special product called BlocBond.
Kathy Snyder at Minesafety Watch reports that a two-day public hearing into the Sago tragedy starts tomorrow in Buckhannon, W.Va. MSHA and the state will be running the hearings jointly. Davitt McAteer, former head of MSHA, is chairing.