Pipeline isolation

Last week there was an interesting article in the US press which among other things focused on the provision (or otherwise) of remotely operated valves as a measure that can reduce the impact of pipeline failures.  It was written in the context of the San Bruno incident, in which the pipeline isolation valves had to be operated locally and it took nearly an hour and a half before PG&E operators could close them.  It seems plausible that the impact of the fire (possibly including the fatalities) would have been greatly reduced if the valves had been closed by remote control within a few minutes of the rupture.

I was surprised that remote operation capability is not more widespread in the USA.  For a long time it seems to have been the practice in Australia that mainline valves are remotely operable by default.  There are exceptions, mostly on older pipelines but sometimes on pipelines in very isolated areas.  I recall making a design decision to include no intermediate mainline valves, and to not automate the valve at the midline scraper station, on 300 km of DN 150 pipeline that didn’t pass within kilometres of any occupied location.  There was no safety reason whatsoever to require prompt closure of valves, and the value of gas that might be lost in the highly unlikely event of a rupture was greater than the cost of providing the remote control (not just valve actuators but also power supply, telemetry infrastructure, etc, etc).  But that was a fairly unusual case.

AS 2885 doesn’t mandate remotely operable valves anywhere.  It does however have comprehensive requirements for pipeline isolation (Section 4.6).  Those requirements include some fairly broad hints about the desirability of considering remote operation of valves, especially in T1 and T2 location classes.  Interestingly, for a liquids pipeline it requires a leak detection system (4.6.2(ix)) but doesn’t require that the isolation valves be remotely operable.  For all pipelines a formal isolation plan is required and must be approved.

From time to time I turn over in my mind the question of whether an incident like San Bruno could occur in Australia, or whether our practices (largely encoded in AS 2885) adequately mitigate that possibility.  Putting aside the root cause of the San Bruno failure (likely defective pipe manufacture), if we had a full bore rupture of a large diameter pipeline in an urban area I think that in most Australian cities the damaged pipeline section could be isolated rapidly and the flow-on consequences should be less severe than San Bruno.

Of course, a full bore pipeline rupture in a populated area can result in almost immediate destruction and deaths.  Actuated mainline valves won’t stop that but they should limit the ongoing damage.  Fortunately no-one here seems to quibble about the cost of remote operation capability for pipeline valves, unlike the cost arguments that have apparently been marshalled in the US.

This entry was posted in External interference, Standards. Bookmark the permalink.

4 Responses to Pipeline isolation

  1. Chris Hughes says:

    If I remember correctly, the SEAGas pipeline runs 160km of 14″ between valves in its centre section. I usually find it difficult to justify mid-line MLVs on the basis of inventory conservation and only install them where it is necessary to limit the burn time in the (rare) event of linebreak and ignition.

    The MAP has MLVs every 30km, because that’s what the US codes required in 1968, but only the ones at the compressor stations are remotely operated.

    • petertuft says:

      Yes, most older pipelines seem to have large numbers of manually operated valves, presumably for the reason you note, and I’ve always wondered what the point was. By the time someone can close a manual valve the horse has bolted, so to speak.

  2. Wan Fauzi says:

    Can remote operated valves considered as emergency shutdown valves? If so, will be treated like any other ESD valves in the plant?

    • petertuft says:

      Hard to respond without knowing more about the situation and the applicable code. Pipeline valves are almost always “fail in last position” rather than “fail closed” so might not meet some requirements for ESD service.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s