Performance Measurements for IEC 61850 IEDs and Systems

Authors: Fred Steinhauser, Thomas Schossig, Andreas Klien, Stephan Geiger, OMICRON electronics, Austria

A common starting point for discussions about performance is the meanwhile famous Figure 16 in IEC 61850-5. (Figure 1) It outlines the issue as it shows that there are several times involved in a transfer of information from one device to another. But as simple as the figure is, it is not straightforward to implement a test that uses the indicated signal flow and delivers the times shown.

IEC 61850-10 proposes tests with combined measurements with physical I/Os.

It is not guaranteed that these physical I/Os are always provided and the test setup and instrumentation becomes more complicated. And if these physical I/Os are provided, they should be implemented with minimal delays (solid state outputs), as the errors incurred by normal relay contacts would be too large for a useful measurement.

The Testing Subcommittee of the UCA International Users Group (UCA IUG) has published test procedures that allow the assessment of an IED's GOOSE performance. When discussing communication performance ratings for IEDs, it is often commented that this may not be useful as these figures do not guarantee a system performance, which is important in practical applications. While this is true, it might be argued why these performance figures are questioned when it comes to communication and not in other cases.

A comparable case is with rated pick up times for distance relays. As well, these ratings do not guarantee the performance of a protection scheme, since several other factors (e.g. the performance of the signal channel) have significant effects. Nevertheless, no protection engineer would accept if this rating is waived, as it gives important hints for the applicability of the relay.

This is similar with the communication performance of IEDs. Such ratings are important for assessing the applicability of an IED for an intended system function.

If an IED already consumes most or all of the time allowed for the system function, it is obvious that it is not suited for this purpose.

Round-Trip Test

The round-trip test is also sometimes called a "ping-pong" test. A stimulus is sent to a device under test (DUT). The DUT is set up to reply with a response as fast as possible. The time between sending the stimulus and receiving the response is the round-trip time tRT.

The test procedures published by the UCA IUG are widely based on round-trip tests. The scenario in the figure from IEC 61850-5 has to be adapted for round-trip tests. Compared to Figure 1, the round-trip scenario contains two transfers. The ta corresponds to tout,TS and tout,DUT. The tc corresponds to tin,DUT and tin,TS, and tb corresponds to tNet. (Figure 2).

The test method is a compromise. On the one hand it is easy to implement, but it also has some shortcomings.

The actual measurement tRT is the total of seven individual times, thus introducing some averaging, and some assumptions have to be made to come to useful results.

The assumption which is easiest to implement is to neglect tNet. Such scenarios can be created in a test setup by a direct point-to-point link between test set and DUT or by only using one high performance switch in between. In such cases, the tNet can be assumed to be one order of magnitude smaller than the other times involved.

Another assumption is the symmetry or (isotropy) of tin and tout of a device. If this cannot be assumed, the evaluation will only deliver the sum of the two times.

The application time tApp is explicitly mentioned to account for processing delays in the DUT, such as cycle times of a programmable logic. Most protective relays utilize optimized signal paths for important protection signals (e.g. trip signals), while the construction of a receive/send feedback for such a test would require the use of a programmable logic. These programmable logics typically have scan cycles that essentially contribute in the test result. If a vendor can make a specification of the cycle time, it can be accordingly deducted from the round-trip times to evaluate the (essentially smaller) communication delays, which will then be effective for the protection applications. The details of the calculation of the times for the DUT from the round trip time are described in.

The same methods can also be used for calibrating a test set.

Relion advanced protection & control.
Protecting your electrical assets? today and tomorrow