|
San Bruno pipeline test method seen as flawed
Experts say method can miss cracking,
fatigue
By Jaxon Van Derbeken
September 18th, 2010
San Francisco Chronicle
|
|
Pacific Gas and Electric Co. is relying on what many experts
call an inferior method to check for corrosion problems on
nearly three-fourths of the utility's natural gas pipelines
running through urban areas - including the part of the line
that ruptured in San Bruno, with deadly results.
Federal and
state regulators have reached no conclusions about what caused
the Sept. 9 explosion in the Crestmoor neighborhood that killed
at least four people and destroyed 37 homes. But one of the
issues they are looking at is whether the 30-inch line was
weakened by corrosion - a potential problem that PG&E was
concerned about in 2009.
The pipe that ran under the San Bruno neighborhood was part of
PG&E's network of 1,107 miles of lines in residential and urban
areas - lines that must be checked every seven years under a
2002 law that Congress passed in reaction to a string of
pipeline disasters.
According to PG&E's filings with the
California Public Utilities Commission, the utility intends to
inspect 72 percent of its urban lines by 2014 using what some
analysts say is a flawed and outmoded method.
That method starts with running an electric current through a
pipeline. Workers walk the pipeline route with devices that
resemble ski poles and insert them in the ground.
If everything is fine with the pipe, the device will register
an electronic signal. But corrosion in the metal can block that
signal, so if the device detects a weaker response in one area,
it points to something that could be wrong with the pipe.
Used in San BrunoThis electronic mapping method is how the San Bruno pipe was
tested in November 2009, and PG&E says it passed. At the time,
according to a PG&E filing with the state Public Utilities
Commission, the utility knew the line was at greater risk of
corrosion because compressor oil and other contaminants had
turned up in sections of the pipe running from San Francisco to
Milpitas.
However, the ski-pole method - known as "direct assessment"
in industry parlance - is far from foolproof, many experts say.
"Direct assessment is certainly not the gold standard," said
Carl Weimer, executive director of the Pipeline Safety Trust, a
nonprofit organization created out of a 1999 gasoline pipeline
rupture in Bellingham, Wash., that killed three people.
"We have always been pretty skeptical about direct
assessment," Weimer said, "but it was one of the compromises
that ended up in the regulations."
What it missesRichard Kuprewicz, a pipeline safety consultant in Redmond,
Wash., said the electronic mapping can only estimate the extent
of corrosion, and only in areas that the poles can reach.
Also, corrosion is not the only problem that can weaken
pipes. Cracking and fatigue due to age and elevated gas-line
pressures are also dangers, Kuprewicz said. And the direct
assessment method "is not a real good tool for stress corrosion
cracking," he said.
Jim Hall, a former chairman of the National Transportation
Safety Board who is now a private consultant, said the technique
is so antiquated that it should not be allowed in urban areas.
The safety board is the agency taking the lead in investigating
the San Bruno disaster.
"That's essentially an old technology," Hall said of direct
assessment. "I don't think is an effective method for a pipe
that old."
Older pipes, such as the line in San Bruno that dates back to
at least 1956, are more prone to cracks and fatigue that may not
be detected through electronic mapping, Hall said.
The pipeline safety advocates agree that the better method of
checking for corrosion, cracks and similar problems is to use an
automated probe known as a pipeline inspection gauge, or "pig."
Better technologyA pig is typically a cylindrical instrument as long as 25
feet that measures ultrasound vibrations or magnetic field waves
to check for corrosion or tiny cracking that would escape
direct-assessment techniques, the analysts said.
PG&E, however, has used pigs to inspect only a fraction of
its urban pipelines. In a filing with the PUC, the utility said
that by the end of 2010, it will have used the devices to check
just 69 miles of the 1,107 miles of lines under cities and towns
in Northern and Central California.
It told the commission that from 2011 through 2014, it
intends to increase its use of pigs to check an additional 235
miles - including a 32-mile stretch of the pipeline that
ruptured Sept. 9.
That stretch, however, will start several miles south of San
Bruno and extend to the South Bay. The rest of the San Bruno
line will be among the more than 800 miles of pipeline where
PG&E will continue to use the ski-pole method.
PG&E's reasonsPG&E spokeswoman Katie Romans said installing pig devices to
check lines can take as long as four years because pipes have to
be modified and infrastructure put in place.
And using pig systems is not feasible on some lines, she said,
because the diameter of the pipe is too small or because of
sharp turns in the line. In those cases, she said, direct
assessment is an accepted alternative.
"It is a federally approved method" to check for pipe
problems, Romans said. The utility has to consider "the shape
and diameter of the pipeline" in deciding which technology to
use, she said.
Mike Florio, a gas expert with the consumer group The
Utilities Reform Network, said direct assessment has at best a
mixed track record.
"I don't think there is any doubt that pigging is more
comprehensive and accurate than direct assessment," Florio said.
"That's why they (PG&E) have this program to convert to allow
pigging. The obvious question is, is direct assessment good
enough in the meantime, or do we have a bigger problem than
that?"
Hall, the former safety board chairman, said utilities are
"clearly driven for economic reasons to try to sell direct
assessment as an effective means of providing safe operation."
Advanced pig devices, he said, are "certainly superior and
particularly should be in all of the (urban) areas. Hopefully,
they will re-evaluate that in light of this tragedy."
E-mail
Jaxon Van Derbeken at
jvanderbeken@sfchronicle.com.
(C)
San Francisco Chronicle 2010
|
|