Third-Party Interference: Catching Excavator Strikes Before Impact

Third-party interference remains one of the largest causes of pipeline damage, and the mechanism is almost always the same: somebody digs where they should not, hits a line nobody told them was there, and a routine excavation becomes an incident. The window between the first bucket in the ground and the strike is usually minutes. That window is long enough to act, if you know it has opened.

Why one-call systems are not enough

Notification schemes work when they are used. They do not help with unrecorded works, contractors who dig outside the marked area, or excavation on a weekend. Aerial patrols see a site once every few weeks. Pressure monitoring tells you after the pipe is already open.

None of these watch the pipeline continuously, and third-party interference is a continuous risk.

What the fiber hears when third-party interference starts

Distributed Acoustic Sensing reads vibration along the entire route from a single interrogator, covering up to 50 km per end and up to 100 km with optical repeaters. Ground works produce loud, distinctive, sustained vibration.

The sequence is recognizable well before contact. Vehicles arrive and park. An excavator tracks into position. Digging starts, with the rhythmic signature of a bucket cycle. Each of these registers at a known location, and each is separable from the background of road traffic and pumps.

Turning vibration into a usable alarm

Raw detection is the easy part. The difficulty is that a busy pipeline route is a noisy place, and an operator who receives fifty alarms a day will stop reading them.

Useful classification separates machinery working near the line from vehicles passing over it, and it does so using the characteristics of your route. That means a tuning period on the live asset, listening to what actually happens there and building the thresholds from it. There is no universal profile.

What happens after a third-party interference alarm

The technology only pays for itself if it plugs into a response. In practice that means three things, decided before commissioning rather than after the first alarm:

• Who receives it. A control room already staffed around the clock, not an inbox.
• What they can do. A phone number for the field team nearest that kilometer point, and the authority to stop work.
• What counts as a false alarm. Logged and fed back into the classification, so the system gets quieter over time rather than louder.

An alarm that reaches the right person with a location accurate to a meter turns a strike into a phone call. That is the entire value proposition, and it depends as much on the process as on the instrument.

The secondary benefits nobody buys it for

Once the fiber is in place and being read continuously, the same data covers other things: leak signatures, pig tracking, unauthorized taps, and ground movement along the route. Operators who install for third-party interference usually find within a year that they are using it for two or three other purposes.

What a realistic alarm timeline looks like

Operators evaluating this technology usually want to know how much warning they actually get. The honest answer depends on the work, but the pattern is consistent.

Vehicle arrival and setup register first, and on a quiet rural route they stand out clearly. Excavation itself produces a sustained rhythmic signature from the moment the bucket enters the ground. Depending on depth of cover and soil, contact with the pipe follows somewhere between several minutes and an hour later.

That is enough time to make a phone call, and phone calls are what prevent strikes. It is not enough time to drive across a region, which is why the response plan matters more than the detection sensitivity.

Building the third-party interference response before go-live

The most common way these projects disappoint is organizational rather than technical. The instrument works, the alarms arrive, and nobody has agreed what happens next.

Three decisions to make during commissioning:

• Routing. Alarms go to a staffed control room, not to email. An alarm nobody reads at 2 a.m. has no value.
• Escalation. The person receiving it needs a number for the field team covering that kilometer point and the authority to stop third-party work.
• Feedback. Every dismissed alarm gets logged with a reason. That log is what makes the classification quieter over the following months.

Coverage planning across a long route

A single interrogator reads up to 50 km from one end, or 100 km with optical repeaters, and an 80 km long-range option exists. On a route longer than that, interrogators are placed at intervals, usually co-located with existing block valve stations or pump stations where power and shelter already exist. Planning those locations around what the pipeline already has is what keeps installation cost proportionate.

More detail on pipeline monitoring, or see the instrument behind it, the Eagle DAS.

Talk to an Engineer and tell us what you need to monitor. We will tell you whether fiber sensing is the right fit, and how to deploy it.