9.4 Communications and Interconnect Fault Diagnosis

9.4 Communications and Interconnect Fault Diagnosis

IMSA's current course descriptions for the Level II Field certification name communication installation and troubleshooting and basic networking explicitly. This reflects reality: an agency's traffic signal system is now a network, and a technician who cannot diagnose a link is limited to whatever can be fixed while standing at one cabinet.

Communication faults have a characteristic that makes them dangerous to ignore: the intersection keeps working. When a controller loses its link to the master or to central, it falls back to its internal time-of-day schedule by design. Nobody calls, nothing flashes, and the corridor quietly loses coordination, real-time visibility, and the ability to be re-timed remotely.

---

Interconnect Media

Legacy telemetry systems commonly ran frequency-shift keyed data at 1200 baud over a four-wire circuit on the controller's Port 3, and a great deal of that plant is still in service. Modern systems run Ethernet over fiber, and mixed corridors carrying both are routine.

---

Step 1: Determine the Scope

Before anything else, establish how much of the system is affected.

This single step routes the entire call. Driving to the reported intersection when six in a row are offline wastes the visit.

---

Step 2: Confirm the Controller Is Healthy Locally

A controller that is offline may be perfectly functional. Confirm before chasing the network:

• Is the controller running its correct time-of-day plan?
• Is its clock correct? A controller whose clock has drifted will run the right plan at the wrong time and may appear to be in coordination when it is not.
• Does the controller's own display report a communication alarm, and does its event log show when the link dropped?
• Are the communication port indicators active?

A controller that free-runs correctly on its internal schedule has isolated the problem to the communication path.

---

Step 3: Diagnose by Medium

Copper Interconnect

1. Open the pull boxes. Water and corroded splices cause the majority of copper interconnect faults, and the symptom is almost always weather-correlated.
2. Measure the pair — continuity, resistance, and insulation resistance to ground with a megohmmeter. A pair with degraded insulation will pass data in dry weather and fail when the box floods.
3. Locate the fault with a TDR or cable fault locator. A time-domain reflectometer reports the distance to an open or a short, which turns a mile of buried cable into a specific pull box.
4. Check the surge protection. A copper interconnect run is a surge collector, and a suppressor that operated is a suppressor that may now be failed. Read the status indicator.

Fiber Optic

The order matters and it is examinable:

1. Visual fault locator — inject visible red light and look for it at the far end and at every accessible splice tray and patch panel. Light escaping at a tray localizes the break instantly. This is the fastest test and requires no calibration.
2. Optical power meter with a matched light source — measure end-to-end loss and compare it against the as-built loss budget. A span that has degraded but is not broken shows up here and nowhere else.
3. OTDR — when the break is in a buried or aerial run with no accessible intermediate point, reflectometry reports the distance to the event so the crew digs in one place.
4. Inspect and clean connectors before concluding anything. A contaminated connector end face is the single most common cause of high loss on a working fiber, and it is fixed with a cleaning tool rather than a splice crew.

Wireless

1. Check received signal strength at both ends and compare it against the value recorded at commissioning.
2. Verify line of sight. Tree growth over a few seasons is the most common cause of a wireless link that degraded gradually.
3. Check antenna alignment and mounting — a radio bumped during other pole work loses aim.
4. Consider interference from newly installed equipment on the same band.

Ethernet

1. Link and activity indicators at both the switch port and the device.
2. Ping the device from the cabinet and from central. A device that answers locally but not from central places the fault on the path between.
3. Duplicate IP address is the classic field failure — two devices configured with the same address produce intermittent, alternating reachability that looks like a flaky cable. It is created by installing a spare that was pre-configured for a different intersection.
4. Speed and duplex mismatch produces a link that comes up but passes traffic poorly, with errors accumulating on the switch port.
5. VLAN assignment on a managed switch determines whether the device can reach central at all; a port reconfigured during unrelated work will isolate a working device.

---

Step 4: Verify and Document

Verification is not a single successful ping. Confirm that:

• Central shows the intersection online and is receiving status.
• The controller has re-entered coordination and its clock is synchronized.
• The link is stable across an observation period, not just at the moment of repair.

Document the medium, the route, the pull box or splice location, the measured values, and what was replaced. Communication plant is the least-documented part of most agencies' signal systems, and a corridor whose fiber route and splice locations exist only in one technician's memory is a corridor that will be very expensive to repair the day that technician is unavailable.

---

Why Fiber Changes the Risk Profile

A copper interconnect running between intersections is a long conductor in the open right-of-way. It collects induced energy from nearby lightning and delivers it straight into the controller's communication port and any network hardware attached to it. Surge protection on that path reduces the damage but does not eliminate it.

Fiber optic interconnect carries no current. A strike near the route cannot couple onto the glass, so the entire class of storm-induced communication-port failures disappears. When an agency's outage history shows repeated controller or switch replacements after storms on a copper-interconnected corridor, the communication medium is the finding — not the components that keep failing.