Short circuit withstand – thermal and dynamic, attributed by the manufacturers of electrical equipment intended for use in power systems – must be confirmed by performing relevant tests in specialised laboratories. Thermal short-circuit withstand is tested using symmetrical current over a time specified by relevant standards, while dynamic short-circuit withstand is testes using asymmetrical current with a decaying transient constant component with a time constant , depending on the installation location of equipment. In practice, the value of the time constant is in the 20–250 ms range. High current laboratories also conduct tests of heating equipment with rated symmetrical current. Symmetrical and asymmetrical currents, with a constant transient component, should be carefully measured to remove doubts about the credibility of the short-circuit tests. The values of symmetrical currents used to test electrical equipment are very different – from several dozen amperes to several dozen kiloamperes. The values of short circuit currents used to test dynamic short-circuit withstand are 2.5 times greater than the value of symmetrical currents. Problems occur in the measurement of asymmetrical currents, when the measured current contains a decaying constant component with a time constant . The constant component of current causes the current transformer core saturation and the formation of very large transformation errors. These errors grow very rapidly with an increase of the time constant .
Since the short-circuit tests and heating tests are often carried out in the same laboratory, it is convenient to use universal current transformers with multiple ranges for measuring short circuit currents.
2. Design assumptions
Design assumptions for the current transformers have been developed at the Institute of Power Engineering, taking into account the current needs, e.g. the inability to measure currents using resistive shunts in circuits where, due to their topologies, one side of the shunt cannot be earthed. Assumptions include technical parameters as well as design and operational requirements placed on current transformers:
- Rated insulation voltage: 30 kV
- Rated primary currents: 1–2–5–10–20–50 kA
- Rated secondary currents: 5 A and 2 A
- Rated power: 20 VA
- Accuracy class: 0.2
- Decay time constant of the short circuit current’s constant component: s1 = 200 ms
- Continuous operation: up to 20 kA
- Primary winding: single bus
- Switching rated ranges between primary and secondary current: on the insulation terminal plate, on the secondary side of the current transformer
- Design suitable for transport with a forklift and a hoist with ropes fitted with hooks
- Design adapted for attachment to the floor
3. Choosing the type of design
Selection of the type of design is not obvious due to the large range of the rated values of primary currents, from 1000 to 50,000 A, and the required „pass-through” design, in which the primary winding is in the form of a single bus and thus the rated primary current corresponds to the rated ampere- turns of the current transformer. Cascade designs are used for large current values, greater than 2000 A. In the case of a ratio of 50,000/5 A/A, for single-core designs, it would be necessary to wind secondary winding, with 25,000 turns, which would be very labour-intensive, resulting in a significant extension of the dimensions of the current transformer and would cause secondary winding to induct very high voltage in the case of inadvertently opening the secondary terminals. Therefore, for rated primary currents of 2000-50,000 A, a cascade (double-core) design has been adopted.
The linking winding is wound on the primary core with a window for the current circuit. It is selectable for different ranges of the primary current (2–5–10–20–50 kA) and connected to the primary winding of the secondary core, wound with taps selected so that the nominal ampere- turns of the secondary core always amounts to 2000 ampere- turns for all measurement ranges of the current transformer (2–5–10–20–50 kA). Secondary winding of the secondary core is wound for rated secondary currents of 5 A and 2 A.
For primary current of 1000 A, the cascade design is very unfavourable since for the rated ampere- turns of 1000, when the secondary core loads
on the primary core, the accuracy of transformation would be small, and improving it would require enlarging the section of the primary core, which would increase the weight and cost of the current transformer. Therefore, a single-core design has been assumed for the primary current of 1000 A. For the ratio of 1000/5 A/A, additional secondary winding was wound on the primary core, and only the primary core works with this ratio as a single-core transformer (secondary terminals 1S1-1S2, Fig. 1).
Fig. 1. Current transformer winding connections diagram
For the ratio of 1000/2 A/A, secondary winding of a single-core transformer is in the form of four sections of linking winding connected in series, wound on the primary core (secondary terminals 1S1-1S3, Fig. 1).
Photo. 1. PPZw-30 current transformers, general view
Currrent transformer winding connections diagram is shown on Fig. 1. (Patent application P-419 196 of 13.10.2016)
4. Description of the transformer’s design
The transformer is built in the form of two cores:
- Primary core, consisting of a toroidal core made of high grade magnetic steel with the link winding and additional secondary winding wound on it, whose window has an insulating tubular passageway, for passing a bus which is a part of the current circuit
- Secondary core, placed in the upper part of the transformer, with primary windings constituting the linking winding of the cascade current transformer and secondary winding serving as secondary winding of the cascade current transformer (for primary currents 2–5–10–20–50 kA).
The two cores are mounted in the metal frame support structure with insulated walls and cover. One of the walls is used as a terminal plate with terminals to which the connections of all link windings and secondary windings are connected. Switching between the rated primary current ranges takes place by means of special connectors that constitute the current transformer’s elements. Choosing the value of rated secondary current is accomplished by connecting the current recorders to the suitable, labelled secondary terminals.
The current transformer was subjected to metrological verification consisting of the following tests:
- Product tests performed at TRANSFORMEX sp. z o.o.
- Checking the accuracy of processing short circuit current waveforms with a non-periodic component having large time constants, at the Distribution Equipment Laboratory of the Institute of Power Engineering in Warsaw
- Calibration at the Laboratory of Low Frequency Electrical Values of the Electrical Department at the Central Office of Measures, in regard to the processing of steady-state symmetrical currents.
Product tests and calibration refer to the final device, which the current transformer is, and the accuracy of transformation of the actual short-circuit current waveforms with an aperiodic component was checked throughout the process of developing the final design of the current transformer. This is due to the fact that meeting the condition of faithfully reconstructing the waveform of asymmetrical short circuit current with a long decay time of the constant component was the most difficult to implement and required designing, building and testing many current transformer prototypes. They were performed at the Distribution Equipment Laboratory of the Institute of Power Engineering in Warsaw, in a special, single-phase, double-loop test circuit. In a typical short-circuit circuit, it is not possible to obtain such large decay time components of the aperiodic constant.
The measuring part of the test circuit consists of the tested prototype of a current transformer and a resistive shunt regarded as a reference measuring device as well as a digital recorder for recording and comparing the two current waveforms. According to the plans, the main parameters of the short-circuit current (effective value of the periodic component, peak factor and the decay time constant of the aperiodic component) were determined for each short-circuit test and recorded using a resistive shunt. Two sets of the results of verifications are presented to illustrate the method used to verify the correctness of processing asymmetrical short-circuit current waveforms. The first is an example of a negative result for one of the working prototypes of the current transformer, while the other illustrates a positive result of checking the final version of the current transformer.
Fig. 2 shows the oscillograms from the failed attempt to check the processing accuracy of short-circuit current with a time constant τ = 235 ms. Two current waveforms shown on the same time axis: from the resistive shunt, treated as a reference – in red, and from the tests prototype of a current transformer – in blue.
Fig. 2. Oscillogram from test checking the processing of short-circuit current with τ = 235 ms for the current transformer prototype with a negative result of the verification a) Oscillogram for the entire test, b) Oscillogram of first 16 peaksof current
The lower oscillogram is the initial part of the entire test, which the upper oscillogram contains, stretched over time. Inaccuracy of processing is visible to the naked eye.
Fig. 3 shows a graph of amplitude errors on selected 8 peaks of the current waveforms being compared. For the selected 8 peaks of the larger half-waves of short circuit current (peaks no. 1, 2, 5, 10, 15, 20, 25 and 30), the instantaneous values of the current waveforms being compared from the shunt (reference) and the tested prototype of a current transformer were determined, and then the relative percentage amplitude errors for each of these current peaks. The values of these errors and the shapes of current waveforms being compared were a determinant of the processing accuracy of the prototype current transformer being tested. The maximum relative error in processing (in a set of 8 designated value) of 48.3% occurred for the verification test in the current’s 10th peak. This is a definite negative result.
Fig. 3. Graph of relative amplitude errors on selected peaks of current waveforms from the test verifying the processing of short-circuit current with τ = 235 ms for the current transformer prototype with a negative result of the verification
It was decided that for this original, non-normative method for checking the accuracy of processing an asymmetrical short circuit current waveform, successful verification occurs when the amplitude error on any of the 8 analysed current peaks will not exceed 1%, and the superimposed waveforms will not show any obvious phase shifts. For such large values (~200 ms) of the time constants, this criterion is very severe. Below is an example of a short-circuit test that meets these criteria.
Fig. 4 shows oscillograms from the test checking the transformation fidelity of short-circuit current with a time constant of τ = 193 ms by a current transformer in its final design. As before, two current waveforms are shown on the same time axis: reference waveform from the resistive shunt – in red, and the waveform from the current transformer tested – in blue. The lower oscillogram is a portion of the upper oscillogram of the entire test stretched over time. Visually, the two current waveforms practically overlap. There is no significant phase shift between the waveforms (oscillogram 4b).
Fig. 4. Oscillogram from the verification of processing of short-circuit current with τ = 193 ms for a current transformer with a positive result of the verification / a) oscillogram for the entire test, b) oscillogram of the first 16 peaks of current
It should be noted that the verification test presented as an example was the fourth short-circuit test with the same parameters (especially the same peak polarity), with small intervals of several minutes between the tests. This represents a significant increase in the working conditions of the current transformer.
Fig. 5 shows a graph of relative amplitude errors for the selected 8 peaks of current waveforms. The highest error value occurs on peak 15 and amounts to 0.92%. According to the accepted criterion for assessing the accuracy of processing asymmetrical short-circuit current, the current transformer verification result is positive.
Fig. 5. Graph of relative amplitude errors on selected peaks of current waveforms from the test verifying the processing of short-circuit current with τ = 193 ms for a transformer with a positive verification result
After checking the processing of asymmetrical short-circuit currents at the Institute of Power Engineering in Warsaw (and after product testing at TRANSFORMEX sp. z o.o. – the manufacturer), the current transformer was subjected to a verification of the measurement accuracy of symmetrical currents at the Laboratory of Low Frequency Electrical Values of the Electrical Department at the Central Office of Measures. The obtained calibration certificate confirmed the assigned accuracy class 0.2 assigned to the transformer.
Based on the results of the tests conducted, it should be stated that the presented universal current transformer for accurate measurement of short-circuit current meets the metrological assumptions.