Network Redundancy Switch

A Network Redundancy Switch (also known as a redundant Ethernet switch, high‑availability switch, or industrial redundancy switch) is a managed industrial Ethernet switch that incorporates advanced redundancy protocols to ensure uninterrupted network connectivity in the event of a link or device failure. In AGL applications, where communication between the ALCMS, CCRs, remote I/O, and field equipment is mission‑critical, network redundancy is essential to maintain control and monitoring capabilities during cable breaks, switch failures, or maintenance activities.

Key characteristics of the Network Redundancy Switch include:

• Redundancy Protocols: Supports industry‑standard redundancy protocols such as Ethernet Ring Protection Switching (ERPS, ITU‑T G.8032), Rapid Spanning Tree Protocol (RSTP), Multiple Spanning Tree Protocol (MSTP), Parallel Redundancy Protocol (PRP), and High‑availability Seamless Redundancy (HSR).
• Fast Failover: Provides sub‑second (typically 10‑50 ms) recovery time, ensuring minimal disruption to time‑sensitive control traffic.
• Ring Topology Support: Enables the construction of resilient ring networks with automatic loop detection and protection switching.
• Dual Homing: Supports dual‑homing connections to two separate switches or networks, providing redundant uplinks for critical nodes.
• Redundant Power Supplies: Includes dual, hot‑swappable power inputs (24 VDC) to eliminate single points of failure.
• Wide Operating Temperature: Industrial‑grade design for -40°C to +75°C operation, suitable for outdoor substations and equipment shelters.
• Comprehensive Management: Provides SNMP, web GUI, and CLI for configuration, monitoring, and diagnostics, including ring status, failover logs, and performance statistics.

In airfield lighting systems, the Network Redundancy Switch is deployed at critical aggregation points — such as the main control room, substations, and equipment shelters — to create a resilient fibre optic backbone. It ensures that a single cable cut or switch failure does not compromise AGL control and monitoring, supporting the high availability requirements of 24/7 airfield operations.

The redundancy switch exists because a single cable cut or switch failure should never be able to leave an operator unable to see or control part of the airfield lighting system — and on a live airfield, a backhoe strike, rodent damage, or a failed port is a matter of when, not if. Without redundancy protocols, that kind of routine failure would silently sever the ALCMS's connection to whatever equipment sat on the far side of it, and the operator might not even know they'd lost visibility until something else went visibly wrong. Sub-second failover is the specific number that matters here: it's fast enough that a brief link interruption resolves before it can cause a control action to be missed or a status update to go stale, effectively making the network as reliable as the fixtures it connects rather than a weak link beneath them.

Network Redundancy Switches are deployed in all critical points of the AGL communication infrastructure where high availability is required. Key applications include:

• ALCMS Core Network: Providing redundant connectivity between the ALCMS server, workstations, and the airfield fibre backbone.
• CCR Substation Aggregation: Aggregating multiple CCR communication links and providing a redundant fibre uplink to the control room.
• Runway Edge Cabinet Rings: Creating resilient ring networks along the runway to connect edge cabinets, ensuring no single cable break isolates a segment.
• Approach Lighting Systems: Providing redundant Ethernet connectivity to approach light sequencers and flasher units.
• RWSL (Runway Status Light) System: Ensuring reliable, low‑latency communication between RWSL processing units and field controllers.
• Inter‑building Links: Providing redundant fibre connections between terminal buildings, ATC towers, and maintenance facilities.
• Critical Sensor Networks: Ensuring uninterrupted data from weather stations, visibility sensors, and RVR systems.

Redundancy switches are mounted in 19-inch equipment racks or wall-mounted DIN-rail enclosures at each network aggregation point — the main control room, CCR substations, and field equipment shelters — rather than out in the open at individual fixtures. Fibre optic runs connect each switch to its neighbours around the ring, typically routed in dedicated ducting alongside but physically separate from the AGL power circuits to avoid electromagnetic interference and to keep a single trench excavation from risking both power and data continuity at once. Dual 24 VDC power inputs are fed from separate circuits or UPS-protected sources so a single power failure cannot take the switch down along with its data path, and the enclosure is selected for the ambient conditions of its location — sealed and temperature-rated for an unconditioned substation, standard rack-mount for a climate-controlled control room. Ring topology design means the physical cable route between switches is planned to avoid a single duct run or conduit carrying more than one segment of the same ring, so that one excavation or cable fault cannot break the ring in two places at once.

The Network Redundancy Switch operates by implementing redundancy protocols that automatically detect and recover from network failures. Key functional aspects include:

• Ring Protection (ERPS/RSTP): In a ring topology, the switch uses ERPS (G.8032) or RSTP to designate one switch as the "ring master" (for ERPS) or the root bridge (for RSTP). The protocol continuously monitors the ring integrity by sending and receiving health messages. If a link or switch fails, the protection mechanism re‑routes traffic around the ring, typically within 10‑50 ms (ERPS) or 1‑2 seconds (RSTP).
• Dual Homing / Link Aggregation: The switch supports multiple uplinks (e.g., two fibre links to different core switches) and can use Link Aggregation (LACP) or active‑standby failover. In the event of an uplink failure, traffic is automatically switched to the backup link.
• Redundant Power Inputs: The switch includes two independent power inputs (24 VDC) with diode‑or'ing or active load sharing. If one power supply fails, the other continues to power the switch without interruption.
• Fast Recovery Mechanisms: The switch monitors link status and protocol states, and upon detecting a failure, immediately updates the forwarding tables and re‑routes traffic. For ERPS, the recovery is triggered by the ring master, which sends a flush message to all switches to clear their MAC address tables and recompute the forwarding path.
• Diagnostics and Monitoring: The switch provides comprehensive monitoring of ring status, link state, power supply health, and traffic statistics via SNMP, web GUI, or CLI. Alarms (e.g., ring broken, power supply failure) can be sent to the ALCMS via SNMP traps or alarm contacts.
• Traffic Prioritization (QoS): To ensure that critical control traffic (e.g., PLC commands) is not delayed during failover, the switch supports IEEE 802.1p and DiffServ QoS, prioritising high‑priority frames.

The switch is typically configured via a web interface or command line, with redundancy settings tailored to the network topology (ring, dual‑homing, or hybrid). It is designed for hot‑swappable power supplies and fanless operation for high reliability.

The switch is built in a rugged extruded-aluminium or steel DIN-rail enclosure with a wide industrial temperature rating, using shielded RJ-45 connectors with integrated magnetics for the copper ports and soldered SFP cages for fibre uplinks, keeping copper and optical interfaces electrically isolated from one another. Internal electronics are conformally coated to resist dust, humidity, and airborne contaminants typical of unconditioned equipment cabinets, and the redundant 24 VDC power inputs terminate at screw-clamp terminals, with optional hot-swappable dual PSU modules available on higher-end models so a single power-path failure doesn't interrupt ring or chain communication. A dedicated alarm relay contact is brought out to a separate terminal so a ring break or switch fault can be wired directly into an external annunciator circuit independent of the network itself. The enclosure includes a chassis earth stud for bonding to the cabinet ground bus, and where deployed outside a conditioned cabinet, the unit is upgraded to a sealed field enclosure with cable glands for each cable entry point.