IEC 61850 Test Mode: Mod, Beh, Sim and the Test Bit

Why IEC 61850 has a test mode

In a conventional panel, isolating a protection to test it means opening the test blocks: trips are cut and the current and voltage circuits are separated. In an IEC 61850 substation, trips and interlocks travel over the network in GOOSE messages and measurements can arrive as sampled values. There is no terminal to open in the path of those messages. Isolation becomes logical: the IED or the function is put into a mode that tells it how to treat its outputs and the data it receives.

Specifications describe it as bay or function isolation for testing.

Mod and Beh

Each logical node has a Mod data object (the requested mode) and a Beh data object (the resulting behavior). The values are the same in both:

Beh does not always match Mod: it is derived from the mode of the node itself and the mode of the LLN0 node of its logical device. If the logical device is in test, all its nodes behave as test even if each one says on. That way a whole bay can be put in test with a single change.

The test bit in the quality

Every IEC 61850 data item travels with its quality, and one of the quality fields is test. A function in test or test/blocked mode flags its data with that bit. The receiver decides according to its own behavior:

• A function with behavior on treats data with the test bit as not valid: it does not act on them.
• A function in test or test/blocked processes data with or without the test bit.

This behavior is what allows testing a function that publishes a trip over GOOSE without the real breaker opening: the sender in test/blocked flags its trip as test, the receiver in service ignores it, and a receiver also put in test processes it with its outputs blocked. The standard leaves parts of this handling to each manufacturer’s declaration, so the exact behavior is confirmed in the IED documentation.

Simulation with LPHD.Sim

Edition 2 added simulation. The LPHD node (physical device) has the Sim data object. When it is active, the IED accepts GOOSE and sampled value messages that arrive with the simulation flag and uses them instead of the real messages from the same publishers. A test set then publishes the simulated messages, and the IED under test reacts to them while the other IEDs, without simulation active, ignore them.

An IED does not publish simulated messages: the test set generates them. The usual combination for testing a complete function is the IED in test/blocked with Sim active, fed by the test set, with its physical outputs blocked and its publications flagged as test.

Differences between editions

In Edition 1, the GOOSE header had a flag called test. Edition 2 renamed it as simulation and clarified the model of modes and behaviors. Edition 1 devices do not always interpret these flags the same way, and some manufacturers implement only part of the modes. That is why testing with logical isolation in a substation with mixed devices is not guaranteed by the standard: it is designed and verified for each combination of devices. See IEC 61850 Editions 1, 2 and 2.1 for the other differences.

What changes in bay isolation

With test modes, the isolation procedure stops being a sequence of test blocks and becomes a sequence of mode changes, verifiable from the HMI. That brings advantages (no terminals are touched, everything is logged) and its own risks:

• Leaving an IED or a logical device in test or blocked when the work is done: the protection looks in service but its outputs do not operate, or its trips arrive flagged as test and the others ignore them.
• Simulation left active: the IED keeps listening to simulated messages and stops using the real ones.
• Mode changes without visible indication on the HMI or at the control center.
• Edition assumptions that do not hold in a device of another edition.

In a digital substation with a process bus, the same principle extends to the merging unit: putting the merging unit in test flags the sampled values it publishes as test, and the protections in service stop using them. A mistake in that sequence leaves a protection without valid measurements. When a subscriber ignores messages it should process, the check starts with the modes.

That is why specifications usually require the mode of each IED and the simulation state to be shown on the HMI and reported as an alarm while not in normal service.

How it is verified

During commissioning, for each IED and each function that will be isolated in operation, the test checks that each mode produces the expected behavior on its outputs and on the quality of what it publishes, that receivers in service ignore test data, that simulation replaces the real messages only when it is active, and that every mode change is indicated. In substations with devices of different editions, the verification is done per publisher and subscriber pair. A Wireshark capture shows the test and simulation flags of each message.

Are you defining the isolation procedure of a digital substation, or testing GOOSE functions without touching the primary? At Tiza Group we configure test modes in the IEDs remotely through our remote relay configuration service and test them with the complete system in Colombia, or elsewhere on request. Contact us.

Related guides

• Interlock testing: what is checked at FAT and SAT
• IEC 61850 Editions 1, 2 and 2.1: differences and compatibility
• IEC 61850 traffic in Wireshark: what a capture reveals
• Substation interlocking: types, levels and logic

References

• IEC 61850-7-4, Compatible logical node classes and data object classes (Mod, Beh, LLN0, LPHD.Sim; annex on modes and quality).
• IEC 61850-7-3, Common data classes (quality: test field).
• IEC 61850-7-1, Principles and models.
• IEC 61850-8-1, Specific communication service mapping (simulation flag in GOOSE).
• UCA International Users Group, technical issues (tissues) on the handling of test quality in GOOSE.

Frequently asked questions