Comparison between conventional current transformers in Holmgreen formation and cable-type current transformers in an isolated grid

The increasing loads in distribution grids lead to higher currents. To measure these, the transformation ratio of the current transformers (CTs) needs to be increased. This can be done by replacing the existing CTs with new ones or by changing the transformation ratio of the existing CTs to a higher ratio. The second option comes with less costs but with a higher measurement error. One of the sub-grids that is affected has an isolated start point. To detect the earth fault direction, the protection devices currently use the earth fault current 3I0 by summing up the three secondary phase currents in Holmgreen formation. The motivation of this paper is to find out whether the accuracy of the existing CTs with higher transformation ratio is still enough to reliably detect the earth fault direction. To do so, three different measurement setups were compared: 3 conventional CTs (CCT) in Holmgreen connection, a cable-type CT (CA), and a calibrated cable-type CT (CCA) as reference. With a specific test setup, it was possible to inject realistic primary currents from a network calculation into the different CT configurations. After analysing the measurement data, it was possible to confirm that using existing CTs with a higher transformation ratio is still sufficient.