Your cart is currently empty!
Introduction
In industrial automation, downtime isn’t just inconvenient—it’s expensive. Whether you’re running a bottling line, an oil processing system, or a semiconductor fab, even a few seconds of lost communication between controllers, drives, and I/O can halt production, damage equipment, or compromise safety.
That’s why network redundancy and link aggregation are essential features of managed industrial switches. They ensure communication continues seamlessly even if a cable, port, or switch fails—keeping critical operations online 24/7.
Why Redundancy Matters in Industrial Automation
Industrial ethernet networks are increasingly used for real-time control, where PLCs exchange I/O data and motion commands with strict timing requirements. If a single connection in that data path fails, the PLC may lose contact with I/O modules or drives, forcing an emergency stop or system reboot.
Redundancy eliminates this single point of failure by providing alternate communication paths that activate automatically when a fault occurs. The key is to design for fast network recovery time—ideally below 300 milliseconds for most automation applications, and under 50 milliseconds for motion control.
Rapid Spanning Tree Protocol (RSTP)
RSTP (IEEE 802.1w) is one of the most widely supported redundancy methods in managed switches. It automatically detects and disables network loops while maintaining standby paths for recovery.
• Normal Operation: RSTP blocks redundant links to prevent loops but keeps them ready for use.
• Failure Scenario: If an active link fails, RSTP immediately unblocks an alternate link.
Typical convergence time in industrial networks is <300 ms, depending on network size.
Example:
A series of managed switches form a ring connecting PLCs, drives, and HMIs. If a single switch or cable goes down, RSTP reroutes traffic around the break—operators see no interruption.
RSTP is simple to configure and compatible across most vendors, making it a solid general-purpose redundancy choice for mixed industrial environments.
Media Redundancy Protocol (MRP)
For PROFINET-based automation systems, MRP (IEC 62439-2) provides even faster recovery—typically <200 ms for a ring with up to 50 switches.
• A Media Redundancy Manager (MRM) controls the ring topology.
• All other switches act as Media Redundancy Clients (MRCs).
• When a fault occurs, the MRM instantly switches the network to the alternate path.
Example:
A robotic assembly cell uses an MRP ring connecting a Siemens PLC, IO-Link masters, and safety I/O blocks. When a cable is damaged, MRP detects the failure and reroutes communication so quickly that the PLC scan cycle continues unaffected.
MRP is deterministic, predictable, and purpose-built for industrial control systems—ideal when real-time performance is non-negotiable.
Device-Level Ring (DLR)
DLR is another ring redundancy protocol, specific to EtherNet/IP networks. DLR is implemented directly within end devices such as PLCs, drives, and I/O modules—meaning no additional switches are required in simple topologies.
When one device detects a link break, it automatically reroutes data in the opposite direction of the ring, with recovery times under 3 ms.
Example: A CompactLogix PLC and PowerFlex drives form a DLR ring. If one connection fails, control and status packets continue flowing seamlessly around the alternate path. This makes DLR especially useful for smaller machine-level networks.
Link Aggregation (LACP)
Redundancy isn’t only about alternate paths—it’s also about increasing bandwidth and balancing traffic loads.
Link Aggregation Control Protocol (LACP, IEEE 802.3ad) allows multiple physical links between switches (or between a switch and a server/PLC) to be combined into one logical “aggregated” connection.
Benefits include:
• Higher throughput: Aggregate bandwidth of all member links.
• Automatic failover: If one link fails, traffic continues on the remaining links.
• Simplified management: The bundle appears as a single interface.
Example:
A control network backbone connecting two managed switches uses four 1 Gbps Ethernet cables combined into an LACP trunk. If one cable disconnects, the link remains active with 3 Gbps of bandwidth and zero downtime.
Combining Redundancy and Aggregation
In large-scale industrial networks, redundancy and aggregation are often used together:
• Aggregation (LACP) for high-bandwidth, redundant uplinks between core switches.
• Ring protocols (MRP/RSTP) for resilient field-level device networks.
This hybrid approach ensures both availability and performance, from the control room to the plant floor.
Best Practices
- Design for recovery time. Match redundancy protocol performance to your application’s timing needs (e.g., DLR for motion, RSTP for supervisory).
- Avoid loops with unmanaged switches. Always place managed switches at loop junctions to prevent broadcast storms.
- Label and document network topology. Clear diagrams help technicians locate redundant paths and troubleshoot faster.
- Test redundancy during commissioning. Simulate link failures and verify recovery times before going live.
- Monitor link status. Use SNMP or Syslog alerts to detect failover events automatically.
Real-World Impact
Imagine a bottling plant running three parallel production lines, each controlled by a PLC and several I/O racks. When one fiber link between switches fails, redundancy protocols reroute communication in milliseconds—no downtime, no alarms, no product loss. Operators may never even notice the failure.
In contrast, a network without redundancy would see communication dropouts, drive faults, and potentially thousands of dollars in wasted material before maintenance could intervene.
The Role of Redundancy for Industry 4.0
As industrial automation converges with IT systems and cloud connectivity, networks must support both mission-critical control and data-intensive analytics. Redundancy and aggregation ensure that increased connectivity doesn’t come at the cost of reliability.
In Industry 4.0-ready plants, resilient network design isn’t optional—it’s the backbone of continuous, data-driven production.
Learn more with Maple Systems
Whether building complex industrial networks or setting up your first network switch, Maple Systems has an expansive library of technical and educational resources to support your success. Explore our support pages for tutorials, technical notes, and sample projects on a variety of topics.
Contact us
Contact our sales team for pricing, availability, and support.
Email [email protected] or call (425) 745-3229.
About the Author