PRP/HSR RedBox: RUGGEDCOM RS950G, X204RNA, Moxa PT-G503

Short answer

A RedBox (redundancy box) connects single-port Ethernet devices to a PRP or HSR network per IEC 62439-3, so they get redundant access with zero recovery time. Devices include the Siemens RUGGEDCOM RS950G (PRP and HSR), the SCALANCE X204RNA (PRP and HSR versions), the Moxa PT-G503-PHR-PTP and Hirschmann RSP switches with a RedBox function. They are used in IEC 61850 substations and industry with redundant networks.

What it is and where it is used

A RedBox (redundancy box) is the device that lets a device with a single Ethernet port take part in a network with PRP (Parallel Redundancy Protocol) or HSR (High-availability Seamless Redundancy), defined in IEC 62439-3. Both protocols give zero-recovery-time redundancy: each frame travels over two paths and the receiver discards the duplicate. Modern relays come with PRP or HSR from the factory; the RedBox serves the devices that do not, such as an engineering PC, a clock, a meter or an older IED. To decide between PRP and HSR by network size and total cost, see PRP vs HSR.

Devices include the Siemens RUGGEDCOM RS950G, the SCALANCE X204RNA, the Moxa PT-G503-PHR-PTP and the Hirschmann RSP switches. For the principle of the protocols, see PRP and HSR (IEC 62439-3).

Variants: RS950G, X204RNA, PT-G503-PHR-PTP and RSP

Device	Application	Main difference
Siemens RUGGEDCOM RS950G	PRP and HSR RedBox	Two IEC 62439-3 ports (PRP or HSR) and three ports for devices or coupling
Siemens SCALANCE X204RNA	Industrial RedBox	PRP and HSR versions; PRP A and PRP B ports
Moxa PT-G503-PHR-PTP	Three-port RedBox	PRP and HSR, coupling and QuadBox, with IEEE 1588v2
Hirschmann RSP	Switch with PRP and HSR	RedBox function inside a DIN-rail switch, with IEEE 1588v2
IED with native PRP	Relays and controllers	Does not need a RedBox: connects directly to LAN A and LAN B

Several relays, such as the SEL-751 or the SIPROTEC 5, offer PRP in their dual Ethernet port models.

Main features

Does the device need a RedBox?

Does the device have two ports with native PRP or HSR?
- Yes: It connects directly to LAN A and LAN B, or to the HSR ring
- No: Is its function critical for protection or control?
  - Yes: It connects to the redundant network through a RedBox
  - No: It can go on a single LAN, without redundancy

PRP: two independent networks, each with its own topology; the RedBox sends each frame to both.
HSR: a ring in which each node forwards the frames; it needs no switches, but each node must support HSR.
QuadBox: couples two HSR rings; some RedBoxes, such as the Moxa PT-G503, do it.
PRP-HSR coupling: allows joining an HSR ring of the process bus with a PRP network of the station bus.
Synchronization: with IEEE 1588 the time travels over both networks; the RedBox must support the power profile. See PTP IEEE 1588.

Node types: who needs the RedBox

IEC 62439-3 classifies devices by how they connect, and the RedBox exists for those without a native dual connection:

DANP (doubly attached node PRP): a device with two ports that runs PRP and connects to networks A and B. It does not need a RedBox.
DANH: the equivalent in an HSR ring.
SAN (singly attached node): a device with a single Ethernet port. To take part with full redundancy it must connect through a RedBox.

In its SCALANCE X-200RNA documentation, Siemens notes that many devices, from controllers to control computers, communicate over TCP/IP without supporting PRP and some have a single interface; the RedBox is the solution for them. It also points out a practical design rule: for a SAN to keep full communication even if a transition point fails, two RedBoxes are used, one toward network A and one toward B. A SAN connected to a single LAN, without a RedBox, is an exception accepted only if the design provides for it.

Manufacturer data: RUGGEDCOM RS950G

The Siemens installation manual (edition 11/2022, in English) defines it as an IEC 62439-3 RedBox for PRP and HSR, with “Zero-Packet-Loss” and zero switchover time. Published data:

Mode: two LEDs on the front show whether the device works in HSR RedBox or PRP RedBox mode, in addition to power, alarm and per-port status LEDs.
Ports and management: fiber SFP transceivers and copper ports, RS-232 serial console, fail-safe alarm relay, web and Telnet access with SSH/SSL encryption.
Power supply: two DC inputs of 10 to 36 V and 37 to 72 V, or a universal input of 85 to 264 V AC or 88 to 300 V DC.
Environment: operation from -40 to +85 °C without fans, steel enclosure.
Capacity: the manual warns that the current HSR implementation limits link utilization to 74 % with multicast traffic. Since GOOSE and sampled values are multicast, this weighs on ring design.

PRP and HSR overhead the design must count

Longer frame: PRP adds 6 octets of redundancy control trailer (RCT) to the Ethernet frame; supervision frames also use bandwidth.
HSR needs less infrastructure but yields less bandwidth: according to Siemens, HSR needs about half the network infrastructure of PRP. But the bandwidth of an HSR ring is about half that of an RSTP-based ring, because each frame circulates in both directions.
All nodes of the HSR ring must support HSR; those that do not connect through a RedBox.
Recovery times of other technologies: Siemens cites as an example 200 ms for MRP with 50 nodes in the ring and 300 ms for High Speed Redundancy with 50 nodes, and concludes they are not suitable for high-availability applications such as the substation. PRP and HSR have no reconfiguration time.
Duplicate filter: on the X-200RNA the duplicate aging time is 400 ms per the standard and is not configurable. The filter window (32 by default) can be widened when there are large time differences between networks A and B because of different topologies. In exchange, fewer source addresses can be filtered. The values for your device depend on the manual for your version.
Rate limiting on coupling links: the X-200RNA allows limiting throughput between 50 % and 80 %; above the set value it drops frames regardless of priority or VLAN. It is a useful function and at the same time a risk if it drops GOOSE traffic, so it is not enabled without analysis.

Coupling between HSR and PRP

Siemens documentation describes coupling two HSR rings through a PRP network, and two PRP networks through an HSR ring. It is the usual architecture when the process bus is an HSR ring and the station bus is a PRP network. The traffic crossing the coupling must be planned, because bandwidth concentrates there. See IEC 61850 digital substation and IEC 61850 transfer time.

Communications and integration

The RedBox is transparent to MMS, GOOSE and sampled values, but it changes the frame size: PRP and HSR add a trailer or tag to each one. Siemens warns, for example, that the X204RNA extends frames by 6 bytes, something to keep in mind with devices or firewalls that do not accept frames larger than the standard size. Also check that the two PRP networks have the same speed and duplex mode on the RedBox ports. See substation communications network and IEC 61850 GOOSE.

Frequent technical questions

Dedicated RedBox or switch with a PRP function: some RedBoxes are dedicated devices (X204RNA, PT-G503) and others are switches with the function included (RS950G, RSP).
RedBox in PRP or HSR mode: confirm the configured mode; a RedBox in PRP mode does not work in an HSR ring.
Hirschmann RedBox: the function is in switches such as the RSP family; confirm that the model and the version of its operating system include the function.
A device behind the RedBox shows up twice or loses packets: check the RedBox node table and that there are no parallel paths outside it.
Supervision: PRP and HSR send supervision frames; taking their alarms to SCADA lets you know that one of the networks failed before the other is lost.

How we help

At Tiza Group we design and configure substation networks with PRP, HSR, RedBox, VLAN and time synchronization, and test them with the relays and SCADA connected. We configure the IEDs, switches and gateways remotely with remote relay and communications configuration, with a backup before and after each change. See control, protection and communications configuration and IEC 61850. For the design of a new network, see communications engineering. Write to us through contact.

References

IEC 62439-3, industrial communication networks, high availability automation networks: PRP and HSR.
IEC 61850-90-4, network engineering guidelines for substations.
IEC 61850-9-3 / IEEE C37.238, PTP power profile.
Siemens, RUGGEDCOM RS950G installation manual and SCALANCE X-200RNA manual; Moxa, PT-G503-PHR-PTP data sheet.

Frequently asked questions

What is a PRP RedBox?

It is a device that connects one or more single-port devices (without their own PRP) to the two independent PRP networks, LAN A and LAN B. The RedBox duplicates each frame onto both networks and discards the duplicate that comes back, so the connected device gets redundancy without knowing it.

What is the difference between a PRP and an HSR RedBox?

In PRP, the RedBox connects to two separate networks (LAN A and LAN B), which can have any topology. In HSR, the RedBox is a node of a ring and sends each frame in both directions. Devices such as the RUGGEDCOM RS950G or the Moxa PT-G503-PHR-PTP are configured to work in one mode or the other.

Is the SCALANCE X204RNA suitable for PRP and HSR?

Siemens sells it in versions for PRP and for HSR, and its manual says that depending on the version it is configured as a PRP or HSR device. When planning the network, keep in mind that it adds bytes to each frame, which matters with devices that do not accept frames larger than the standard size.