PG&E control-room operator: 'We're screwed'

San Francisco Chronicle ^ | 3/8/11 | Eric Nalder, Chronicle Staff Writer

[SmithL]

The San Bruno pipeline explosion and fire in September killed eight people and destroyed 38 homes.

[Egon]

Ping.

[scooby321]

How did this happen under all the California Regulations.

[archivist007]

Time to fire some people and for them to financially pay for it to those who were affected.

[dfwgator]

How did this happen under all the California Regulations.

Affirmative Action?

[Parley Baer]

Union affirmative action.

[Hodar]

Umm, just a naive question.

Why aren’t there safety precautions in place that can prevent this from happening?

I’ve designed Hydraulic testers for a living, every one of them had EMERGENCY precautions to prevent this sort of thing from happening, yet these testers dealt with very high pressures and flow rates. Yes, if left without safety bypass valves - this was a very dangerous piece of equipment.

There is an Operational Range of pressure - then there is the HOLY S**T range where automatic safeguards SHOULD have kicked in, shut down the pumps, and bled off the pressure before something like this happened.

What happened to very basic safety design reviews?

[nascarnation]

a system set up by PG&E automatically increased pressure on all three Peninsula pipelines

An interesting fail-safe response.

Now the media often screws up this kind of reporting but will be interesting to see follow up info.

[A. Patriot]

Another instance of the Pepsi Syndrome?

[MaxMax]

Looks like they need mooooore regulation /s

[PA Engineer]

I’ve designed Hydraulic testers for a living, every one of them had EMERGENCY precautions to prevent this sort of thing from happening, yet these testers dealt with very high pressures and flow rates. Yes, if left without safety bypass valves - this was a very dangerous piece of equipment.

I think this will be blamed as the precipitating reason for the failure, however I believe the line was already damaged from previous sewer line "burst" construction upgrades. From experience I believe this line would have failed anyways, but could not predict the severity. It may have started out with a slower detectible leak.

[SMARTY]

Nepotism is RAMPANT there. You know.

[BipolarBob]

This is what happens when a repair crew isolates the pressure gauges to the regulators, which thinks the pressure is dropping when in fact it is increasing. Bad data = bad results.

[dfwgator]

Another instance of the Pepsi Syndrome?

But, you spilled a Coke.

[AFreeBird]

Why no failsafe built into the system? If electrical power is a critical component, the absense of it should trigger a failsafe to avoid such a catastrophe.

Or am I missing something?

[org.whodat]

Union workers, probably on break.

[Peter from Rutland]

I work for a company that does control systems for these type of operations and this is inexusable. There should have been multiple fail-safes and some method of shutting down the lines in such an emergency. It also appears a pressure sensor or sensors were reading wrong. I’d love to know who manufactured that sensor.

[Hodar]

I agree, every safety device I have used was fully automatic. Never used a computer to respond to an emergency (computers can lock-up), never made a signal’s absense allow flow (wires get cut).

The default condition was zero flow. Things had to be perfect before flow was allowed - any failure shut the system down. The simple trick, is to make the safety system easy enough to understand, quick enough to matter, and robust enough to withstand an emergency.

[null and void]

ping

[dblshot]

Why no failsafe built into the system?
All systmes are subject to failure. When a fail safe system fails, it fails by failing to fail safely.