Document Type


Publication Date



The GE N60 relay evaluation project was conducted by Brenna Andrews with help from Richard Setterstrom and Daniel Kachmarik from NorthWestern Energy. The goal of the project was to determine if the GE N60 relay was a sufficient replacement candidate for the Acceleration Trend Relays (ATRs) currently being utilized on four stem-coal generators in Colstrip, Montana, to protect the power system. The ATRs protect the power system from events such as faults or equipment damage. The criteria used to evaluate the GE N60 relay’s capabilities included the following:

  1. The GE N60 relay must have a sampling frequency equal to or greater than that of the ATRs.
  2. The GE N60 relay must have minimal noise so as not to affect trip decisions in the relay.
  3. The GE N60 relay must be able to accurately detect events.
  4. The GE N60 relay must be able to create and store event records of adequate length.

The GE N60 relay was programmed to emulate the functions of the ATRs. Two Colstrip-Broadview transmission-line events were captured simultaneously by the ATRs and the GE N60 relay. Comparison of the speed deviation, acceleration, and power responses from these two events showed that the GE N60 relay is an adequate replacement candidate for the GE N60 relay. Refer to Figures 27 through 40.

The GE N60 relay proved to have a variable sampling frequency. The relay determined events based on the frequency deviation and frequency rate of change threshold values. The frequency deviation value was sampled at 60Hz in the GE N60 relay, whereas the ATRs could sample at 150Hz. The decision was made that, until further testing could be completed, the lower frequency deviation sampling rate would not make the collected data unusable. From qualitative analysis, the noise level was proved to be acceptable, since the GE N60 relay trip decisions were not affected by what noise was present in the event responses. The decision was made that additional filtering could be implemented so long as the trip decision speed is not compromised. Finally, the GE N60 relay proved to accurately detect and store events. The GE N60 relay was able to detect events simultaneously with the ATRs, showing that the ATR functions were correctly emulated in the GE N60 relay. The GE N60 relay event records were of acceptable length and exceptional quality. Based on these results, the GE N60 relay was found to be an adequate replacement candidate for the ATRs in Colstrip and should move on to the next stages of testing.