- Introduction
- 1. Why Network Redundancy Is Required in IEC 61850 Substations
- 2. Understanding the Main Types of Ethernet Redundancy
- 3. PRP – Parallel Redundancy Protocol
- 4. HSR – High-availability Seamless Redundancy
- 5. PRP versus HSR
- 6. MiCOM Relay Compatibility with PRP and HSR
- 7. Redundancy and IEC 61850 GOOSE Communication
- 8. Redundancy and MMS Communication
- 9. Network Redundancy and VLAN Configuration
- 10. Practical Example: PRP Network for a MiCOM-Based Substation
- 11. Practical Example: HSR Ring for Protection IEDs
- 12. Network Redundancy Testing and Failure Scenarios
- 13. Wireshark Analysis of PRP and HSR
- 14. Common Redundancy Problems and Troubleshooting
- 15. Best Practices for PRP and HSR Engineering
- 16. Commissioning Checklist
- 17. Conclusion
- Practical MiCOM Protection and IEC 61850 Training by Dr Saeed Roostaee – Elec-Engg
Introduction
Communication availability is a critical requirement in modern digital substations. Protection relays, bay control units, SCADA systems and other Intelligent Electronic Devices (IEDs) depend on reliable communication networks to exchange protection signals, measurements, alarms and control information.
In an IEC 61850-based substation, a communication failure can affect more than monitoring. It may interrupt GOOSE signals used for protection and interlocking applications, prevent SCADA systems from receiving important indications, or make engineering access unavailable.
To improve network availability, substation automation systems may use redundant Ethernet architectures.
Two important redundancy technologies defined by IEC 62439-3 are:
- PRP – Parallel Redundancy Protocol
- HSR – High-availability Seamless Redundancy
These technologies are designed to provide seamless communication recovery following certain network failures, without relying on conventional reconvergence mechanisms.
This article explains PRP and HSR, their network architectures, their application to MiCOM protection systems, configuration considerations and practical commissioning tests.
Important: PRP and HSR support is dependent on the exact MiCOM relay model, communication hardware and firmware. Some installations require external Redundancy Boxes (RedBoxes) or other compatible network equipment. Always verify the actual device capabilities before designing the network.
1. Why Network Redundancy Is Required in IEC 61850 Substations
Consider a substation in which several protection relays exchange GOOSE signals through a single Ethernet switch.
If that switch fails, communication between the connected relays may be interrupted.
Possible consequences include:
- Loss of GOOSE communication.
- Missing protection indications.
- Failure of communication-dependent interlocking.
- Loss of MMS communication with SCADA.
- Loss of engineering access.
- Delayed fault investigation.
A redundant network is designed to reduce the impact of these failures.
Example: Single-network architecture
In a single-switch architecture, the switch and its associated connections may represent single points of failure.
Example: Redundant network architecture
A redundant arrangement provides alternative communication paths, depending on the selected architecture and device capabilities.
2. Understanding the Main Types of Ethernet Redundancy
Several technologies can be used to improve Ethernet network availability.
| Technology | Basic principle | Typical application |
|---|---|---|
| PRP | Sends duplicate frames through two independent LANs | High-availability substation networks |
| HSR | Sends duplicate frames in both directions around a ring | Ring-based redundant networks |
| RSTP | Reconfigures a network after a topology change | General switched Ethernet redundancy |
| MRP | Ring redundancy with defined recovery behavior | Industrial Ethernet networks |
These methods are not interchangeable.
PRP and HSR are particularly important in substation automation because they are designed for seamless redundancy when implemented correctly.
RSTP and MRP have different recovery mechanisms and may introduce communication interruption during topology changes.
The selected protocol must be compatible with the protection application and the connected devices.
3. PRP – Parallel Redundancy Protocol
PRP is defined in IEC 62439-3.
It provides network redundancy by connecting devices to two independent Ethernet networks, commonly identified as:
- LAN A
- LAN B
A PRP-capable device transmits duplicate frames through both networks.
The receiving device accepts the first valid copy and discards the duplicate.
3.1 How PRP works
Assume a MiCOM protection relay is connected to two independent Ethernet networks.
When the relay sends a communication frame:
- The PRP device generates the frame.
- The frame is transmitted through LAN A.
- A duplicate frame is transmitted through LAN B.
- The receiving PRP device receives both copies.
- It accepts the first valid frame.
- It discards the duplicate.
The receiving device therefore does not need to wait for a network recovery process if one LAN fails.
3.2 PRP network components
A typical PRP network may contain:
- PRP-capable IEDs.
- LAN A switches.
- LAN B switches.
- Redundant Ethernet connections.
- RedBoxes for connecting non-PRP devices, where required.
The two LANs should be designed to avoid common points of failure.
For example, if LAN A and LAN B use the same physical switch or power supply, the expected redundancy may not be achieved.
3.3 PRP operating modes
PRP defines different node types.
| Node type | Description |
|---|---|
| DANP | Doubly Attached Node implementing PRP |
| SAN | Single Attached Node connected to one LAN |
| RedBox | Connects single-attached devices to a PRP network |
A DANP has access to both LANs and can transmit and receive duplicate frames.
An SAN is connected to only one LAN and does not independently provide PRP redundancy.
A RedBox can provide PRP network access for devices that do not natively support PRP.
4. HSR – High-availability Seamless Redundancy
HSR is another redundancy protocol defined in IEC 62439-3.
Unlike PRP, which uses two separate LANs, HSR typically uses a ring topology.
Each HSR node forwards frames through the ring in both directions.
The destination node accepts the first valid copy and discards the duplicate.
4.1 How HSR works
Consider four HSR-capable IEDs connected in a ring.
When one IED sends a frame:
- It transmits copies in both ring directions.
- Intermediate nodes forward the frames.
- The destination receives copies through the two directions.
- It accepts the first valid frame.
- The duplicate is discarded.
If one ring connection fails, communication may continue through the remaining direction.
4.2 HSR network components
A typical HSR network includes:
- HSR-capable IEDs.
- Redundant Ethernet ports.
- Ring connections.
- HSR-capable devices or compatible network equipment.
- RedBoxes for connecting suitable non-HSR devices, where required.
4.3 HSR considerations
HSR is useful when the system architecture supports a ring arrangement.
However, engineers must consider:
- Number of nodes.
- Ring topology.
- Network traffic.
- Device forwarding capacity.
- Supported frame types.
- RedBox requirements.
- Network diagnostics.
- Compatibility between connected equipment.
An ordinary Ethernet device cannot automatically be connected to an HSR ring and be expected to participate correctly in HSR redundancy.
5. PRP versus HSR
PRP and HSR provide seamless redundancy but use different physical network arrangements.
| Characteristic | PRP | HSR |
|---|---|---|
| Standard | IEC 62439-3 | IEC 62439-3 |
| Topology | Two independent LANs | Ring |
| Frame transmission | Duplicate frames over two LANs | Duplicate frames in both ring directions |
| Redundancy mechanism | Parallel network paths | Redundant ring paths |
| Typical infrastructure | Two separate switched networks | HSR-capable ring |
| RedBox use | For single-attached devices | For non-HSR devices |
| Network planning | Two independent LANs | Ring topology and node forwarding |
| Recovery behavior | Seamless for supported single failures | Seamless for supported single failures |
Neither method is automatically better in every installation.
The selection depends on:
- Substation architecture.
- Number of IEDs.
- Device compatibility.
- Network infrastructure.
- Maintenance requirements.
- Project availability requirements.
- Lifecycle and expansion plans.
6. MiCOM Relay Compatibility with PRP and HSR
A critical engineering consideration is whether the installed MiCOM relay supports the selected redundancy technology.
Not all MiCOM relays have the same Ethernet hardware or redundancy capabilities.
Some devices may support native redundancy, while others may require external equipment.
Before selecting PRP or HSR, verify:
- Exact relay model.
- Product order code.
- Communication module.
- Firmware version.
- Number and type of Ethernet ports.
- Supported redundancy protocols.
- Supported IEC 61850 communication services.
- Compatibility with the proposed switches or RedBoxes.
Native redundancy
If a relay supports PRP or HSR natively, it can participate in the corresponding redundancy architecture using its supported communication interfaces.
External redundancy equipment
If a relay does not support the required protocol natively, a compatible external RedBox may be used where permitted by the project design.
The RedBox connects the device to the redundant network and manages the redundancy-related communication behavior.
The exact arrangement must be validated for the required traffic types and performance.
7. Redundancy and IEC 61850 GOOSE Communication
GOOSE is commonly used for time-critical protection and control communication.
A redundant network can improve the availability of GOOSE communication.
Consider a feeder relay publishing a trip signal to another protection relay.
In a non-redundant network, failure of a critical switch or link may prevent the subscriber from receiving the message.
In a properly engineered PRP or HSR network, duplicate frames may continue to reach the subscriber through the remaining network path.
7.1 GOOSE communication through PRP
In a PRP architecture:
- The publisher transmits duplicate frames through LAN A and LAN B.
- The subscriber receives the frames through both LANs.
- The first valid copy is processed.
- The duplicate is discarded.
This allows the subscriber to continue receiving GOOSE messages after certain single-network failures.
7.2 GOOSE communication through HSR
In an HSR architecture:
- The publisher sends frames in both ring directions.
- Intermediate nodes forward the frames.
- The subscriber receives copies through the ring.
- The first valid copy is accepted.
If one ring connection fails, the other direction can maintain communication, provided the network is correctly designed.
7.3 Important testing requirement
A GOOSE message visible in one network path does not prove that the complete protection application is operating correctly.
Testing must verify:
- Publisher operation.
- Subscriber reception.
- Dataset values.
- Communication supervision.
- Internal signal mapping.
- Intended protection or control response.
8. Redundancy and MMS Communication
MMS is used for client-server communication between IEC 61850 clients and IED servers.
For example, a SCADA server may communicate with a MiCOM relay through MMS.
A redundant Ethernet architecture can improve the availability of this communication path.
However, engineers must distinguish between:
- Network redundancy.
- Application connection redundancy.
- MMS association recovery.
- SCADA client reconnection behavior.
PRP or HSR can maintain Ethernet frame delivery across supported network failures, but this does not automatically guarantee that every application-level session will remain uninterrupted.
The behavior depends on the devices, network interfaces, redundancy implementation and client/server software.
MMS commissioning checks
- Confirm normal MMS association.
- Verify data-model access.
- Test report reception.
- Monitor communication during network failure.
- Check client reconnection behavior.
- Verify event and measurement continuity.
- Confirm that communication returns to normal after recovery.
9. Network Redundancy and VLAN Configuration
Redundancy does not remove the need for correct VLAN engineering.
VLANs may still be used to logically separate:
- Protection communication.
- Station-level SCADA.
- Engineering access.
- Network management.
The network designer must ensure that the required VLANs are correctly configured across both redundant paths or throughout the HSR ring.
Example PRP VLAN arrangement
| Network | VLAN | Purpose |
|---|---|---|
| LAN A | 10 | Protection communication |
| LAN A | 20 | Station-level communication |
| LAN B | 10 | Protection communication |
| LAN B | 20 | Station-level communication |
The values are illustrative.
The VLAN configuration must be consistent with the project’s architecture and the capabilities of the installed devices.
Priority handling
GOOSE traffic may require appropriate priority treatment.
Engineers should verify:
- VLAN tagging.
- Priority Code Point.
- Switch forwarding behavior.
- Multicast handling.
- Traffic loading.
- Consistency between the two PRP LANs or across the HSR ring.
A redundancy protocol does not compensate for incorrect VLAN or multicast configuration.
10. Practical Example: PRP Network for a MiCOM-Based Substation
Consider a substation with the following equipment:
- Three MiCOM feeder protection relays.
- One MiCOM transformer protection relay.
- One MiCOM busbar protection relay.
- One SCADA server.
- Two independent Ethernet LANs.
- PRP-compatible switches.
The objective is to maintain communication availability if one LAN fails.
Step 1 – Define the network architecture
Create two physically independent networks:
- LAN A.
- LAN B.
Where practical, use separate switches, power supplies and cable routes to reduce common failure points.
Step 2 – Verify relay capabilities
Confirm whether each MiCOM relay supports native PRP.
For relays without native PRP support, evaluate a compatible RedBox arrangement.
Step 3 – Configure the Ethernet network
Configure the two networks according to the approved design.
Check:
- VLAN membership.
- Switch-port configuration.
- Multicast forwarding.
- Priority handling.
- IP addressing.
- Network management access.
Step 4 – Configure IED communication
Configure the relay’s IEC 61850 communication according to the supported engineering procedure.
Verify the GOOSE and MMS settings independently of the physical redundancy configuration.
Step 5 – Verify normal operation
Check that:
- Both LANs are operational.
- The devices are communicating.
- GOOSE messages are received.
- MMS communication operates.
- No abnormal network errors are reported.
Step 6 – Perform a network failure test
Under an approved commissioning procedure:
- Record the normal network condition.
- Disconnect the LAN A connection or perform the approved LAN A failure simulation.
- Monitor GOOSE communication.
- Verify subscriber operation.
- Check MMS communication.
- Restore LAN A.
- Repeat the procedure for LAN B.
The objective is to confirm that the remaining network path maintains the required communication.
11. Practical Example: HSR Ring for Protection IEDs
Consider four HSR-compatible devices arranged in a ring.
The ring contains:
- MiCOM Feeder Relay 1.
- MiCOM Feeder Relay 2.
- MiCOM Transformer Relay.
- A compatible HSR network device.
The ring is designed to maintain communication following a single link failure.
Step 1 – Confirm HSR compatibility
Check the HSR capability of every device that will participate in the ring.
Where non-HSR devices are required, assess the use of a suitable RedBox.
Step 2 – Design the ring
Prepare the ring topology and identify:
- Each node.
- Each ring connection.
- The direction of frame forwarding.
- RedBox locations, if applicable.
- Power supply arrangements.
Step 3 – Configure the devices
Apply the supported HSR configuration.
Verify the correct port connections and network settings.
Step 4 – Verify ring operation
Check the network diagnostics and confirm that the ring is operating normally.
Step 5 – Simulate a link failure
Under an approved procedure:
- Disconnect one ring connection.
- Confirm that communication continues through the remaining direction.
- Check GOOSE reception.
- Verify MMS communication.
- Restore the connection.
- Confirm that the ring returns to its normal state.
12. Network Redundancy Testing and Failure Scenarios
A complete redundancy test should examine more than simply disconnecting an Ethernet cable.
The test plan should include relevant failure scenarios.
| Test | Expected observation |
|---|---|
| Normal operation | Both redundant paths are healthy |
| LAN A failure in PRP | Communication continues through LAN B |
| LAN B failure in PRP | Communication continues through LAN A |
| Single HSR ring-link failure | Communication continues through the remaining ring path |
| Switch failure | Communication remains available if the architecture provides an independent alternative path |
| Network restoration | Redundant communication returns to normal |
| GOOSE test | Required dataset continues to reach subscribers |
| MMS test | SCADA communication behaves as specified |
| RedBox failure | Impact matches the approved redundancy design |
| Power supply failure | Redundant power arrangements operate as designed |
Do not assume that every possible failure can be tolerated.
The redundancy design should identify the failure scenarios it is intended to withstand.
13. Wireshark Analysis of PRP and HSR
Wireshark can help engineers analyze redundant Ethernet communication.
It can be used to inspect:
- GOOSE traffic.
- MMS traffic.
- VLAN tags.
- Ethernet frame duplication.
- Communication interruptions.
- Unexpected network traffic.
13.1 GOOSE filter
goose
13.2 GOOSE EtherType filter
eth.type == 0x88b8
13.3 VLAN filter
vlan
13.4 MMS filter
tcp.port == 102
13.5 PRP analysis
PRP-capable equipment uses redundancy-related frame information, including a Redundancy Control Trailer (RCT).
A packet capture can help identify duplicate frames and inspect redundancy-related information.
The capture point matters: a capture taken on only one LAN will not show the complete behavior of both paths.
13.6 HSR analysis
HSR uses redundancy-related information in its Ethernet frames.
Engineers should examine the relevant frame information and network forwarding behavior using a capture method suitable for the HSR topology.
A standard capture on one port may not reveal every frame transmitted in both ring directions.
14. Common Redundancy Problems and Troubleshooting
| Problem | Possible cause | Recommended action |
|---|---|---|
| PRP communication does not operate | Incorrect LAN connections or incompatible devices | Verify DANP, SAN and RedBox configuration |
| One LAN is unavailable | Switch, cable or power supply failure | Inspect LAN-specific diagnostics |
| Duplicate frames are not handled correctly | Redundancy configuration or compatibility problem | Check device and network protocol support |
| HSR ring is not closed | Incorrect port connections | Inspect ring topology and node configuration |
| GOOSE fails after network redundancy changes | VLAN or multicast forwarding problem | Verify frame forwarding and VLAN settings |
| MMS reconnects after a network failure | Application-level session recovery | Check client/server reconnection behavior |
| RedBox is not forwarding traffic | Configuration or compatibility issue | Check RedBox status and supported traffic |
| Unexpected communication interruption | Common-mode failure | Investigate shared switches, power and cable routes |
| Excessive network traffic | Incorrect topology or forwarding | Analyze network statistics and packet captures |
| Communication does not recover after restoration | Device or network configuration issue | Check redundancy diagnostics and restore sequence |
15. Best Practices for PRP and HSR Engineering
For reliable redundant communication in MiCOM-based substations:
- Select the redundancy protocol during the initial network design.
- Verify the exact MiCOM hardware and firmware compatibility.
- Avoid common points of failure between redundant paths.
- Use compatible managed switches and RedBoxes.
- Configure VLANs consistently.
- Check multicast forwarding and GOOSE priority.
- Verify redundancy diagnostics.
- Test both normal and failure conditions.
- Monitor communication after network changes.
- Maintain network configuration backups.
- Document all tested failure scenarios.
- Revalidate communication after relay firmware or switch upgrades.
Redundancy must be engineered as a complete system. Two Ethernet cables do not automatically provide a reliable redundant network.
16. Commissioning Checklist
Architecture and compatibility
- Confirm the approved network architecture.
- Identify PRP or HSR requirements.
- Verify MiCOM relay compatibility.
- Verify switch and RedBox compatibility.
- Confirm the redundancy protocol implementation.
- Identify potential common points of failure.
Network configuration
- Verify LAN A and LAN B connections, where applicable.
- Verify HSR ring connections, where applicable.
- Check VLAN configuration.
- Confirm multicast forwarding.
- Verify QoS and traffic priority.
- Check switch power and network status.
- Confirm network diagnostics.
IEC 61850 communication
- Verify GOOSE publisher operation.
- Verify GOOSE subscriber reception.
- Check dataset values.
- Verify MMS association.
- Test report reception.
- Confirm communication supervision.
Redundancy testing
- Test normal network operation.
- Simulate LAN A failure.
- Simulate LAN B failure.
- Test the approved HSR ring failure scenario.
- Verify communication continuity.
- Restore the network.
- Confirm normal communication.
- Save test records and final configurations.
17. Conclusion
Network redundancy is an important part of reliable IEC 61850 substation automation.
PRP provides two independent Ethernet LANs, while HSR uses a redundant ring topology. Both are designed to provide seamless communication availability for supported devices and network architectures.
For MiCOM protection systems, the most important engineering step is to verify compatibility before selecting the redundancy technology. Native relay support, communication hardware, firmware, switches and RedBoxes must all be considered.
Commissioning must also include real failure simulations to confirm that GOOSE and MMS communication behave as required when network components fail.
A properly engineered redundant network improves the availability of protection communication and supports reliable operation of modern digital substations.
Practical MiCOM Protection and IEC 61850 Training by Dr Saeed Roostaee – Elec-Engg
Elec-Engg provides practical training for protection, commissioning and substation automation engineers.
Relevant training topics include:
- MiCOM protection relay configuration.
- IEC 61850 communication engineering.
- GOOSE publisher and subscriber configuration.
- MMS communication and SCADA integration.
- PRP and HSR redundancy fundamentals.
- Ethernet switches and VLAN configuration.
- Wireshark network analysis.
- IEC 61850 testing and commissioning.
- Communication troubleshooting and practical applications.
Training is available through video courses, structured self-paced training, face-to-face courses and customized in-house programs.
Visit Elec-Engg for practical power system protection and substation automation training.
