A current-transformer core’s excitation curve, an a.c. hysteresis-loop reconstruction and a remanence acceptance result describe different evidence. Join the specified remanence quantity, controlled magnetic history, uncertainty and tested construction before deciding whether a low-remanence requirement is met.
IEC 61869-2 is the relevant current-transformer standard [1]. Its official edition and scope have been checked here. The purchase specification must identify the applicable class and controlled test procedure; a public scope or contents page does not supply the detailed procedure.
Start with the specified remanence quantity
ABB’s application guide defines remanent flux as the flux remaining three minutes after interruption of an excitation current sufficient to induce saturation. It gives a maximum remanence factor of 0.1 for TPY and PR cores on that basis [2]. This is the guide’s class-specific statement, rather than a universal condition for every CT core.
The ratio is KR = |ψr|/|ψsat| × 100%, using linked flux from the same winding and a consistently defined saturation reference. ABB describes that reference at the transition where a 10% increase in B requires a 50% increase in H [2]. Retain the chosen criterion and its measured basis. Turns cancel in the ratio; changing the winding, flux baseline or magnetic history can still invalidate the comparison.
Keep the a.c. loop diagnostic separate
The winding relation is ψ(t) = ψ(t0) + ∫[u(t) − Rct i(t)]dt. It reconstructs a linked-flux change after correction for winding resistance and instrument effects. An unknown initial flux cannot be established by resetting a software integrator.
For a separate assumed 50 Hz loop record at 20 °C, let its established integration baseline be zero. At the positive peak, terminal-voltage and winding-drop integrals are +0.820 and +0.020 V·s. At a later descending-branch current zero, they are +0.080 and +0.024 V·s. Corrected values are 0.800 and 0.056 V·s. Their loop-point ratio is 7.0%; the resistance-drop integral increases by 0.004 V·s consistently with the stipulated nonnegative current before that crossing.
This calculation checks the specified waveform reconstruction. The instantaneous current-zero value has not been established as the three-minute remanent quantity. Keep it labelled as a loop-point ratio and do not compare it directly with the guide’s class limit. An applicable documented method would have to establish the evaluation points, conditioning and equivalence before that substitution could support acceptance.
Apply an acceptance rule to an identified record
Consider a different, explicitly assumed acceptance record at 20 °C. Its agreed procedure identifies ψsat = 0.800 V·s and ψr,3min = 0.072 V·s after the specified saturating excitation and three-minute interval. These values are record inputs, not outputs inferred from the preceding a.c. loop. Assume the purchase basis adopts a 10.0% maximum for its specified TPY/PR application.
The nominal ratio is 100 × 0.072/0.800 = 9.0%. A nominal comparison alone would leave 1.0 percentage point below the adopted limit. Now assume independent standard uncertainties ur = 0.004 V·s and usat = 0.010 V·s. First-order propagation gives u(KR) = 100√[(ur/ψsat)² + (ψr usat/ψsat²)²] = 0.5125 percentage point [3].
For an assumed project decision rule KR + U ≤ 10.0%, with U = 2u(KR), the upper result is 9.0 + 1.025 = 10.025%. That rule therefore withholds acceptance despite the nominal 9.0% ratio. The coverage factor and guard band are project assumptions, not a rule attributed to ABB or an unread IEC clause. Obtain the actual covariance and measurement budget before applying them to a real record.
Keep the tested core’s geometry in the record
Use catalogue nominal sheet gauge 0.23 mm [4]. For an assumed 200-sheet, 20 mm-wide GOES core, the steel-only section is 920 mm². With an assumed 500-turn test winding, the separate acceptance record’s 0.800 V·s saturation reference corresponds to mean B = ψ/(NA) = 1.739 T. This conversion checks section and winding identity; it does not prove that the selected point meets the saturation criterion.
Ten missing sheets reduce nominal steel section to 874 mm², a 5% change. At the same linked flux and turns, mean B rises to 1.831 T, or 5.26%. A ratio with cancelling turns can therefore conceal a wrong build if section and gap records are absent. Retain actual gauge/stack measurement, gap spacers, winding identity, restraint and manufacturing stage with the magnetic result.
Release against the method and the final construction
ABB identifies controlled air-gap width as important to low-remanence construction [2]. Reconcile the final gap and clamp state after work capable of changing them. Record each test’s conditioning, interruption or waveform sequence, evaluation time, saturation criterion and uncertainty. A demagnetization step must occur at the stage required by the selected procedure; erasing the retained state before its observation changes the measurand.
The example supports two decisions: the loop-point ratio stays a diagnostic quantity, and the separate assumed three-minute record fails its stated uncertainty-aware acceptance rule. Actual release additionally requires the applicable procedure and traceable raw evidence. Obtain those records before treating a smooth curve or a nominal percentage as a completed acceptance test.
References
- IEC, IEC 61869-2:2012, Instrument transformers — Part 2: Additional requirements for current transformers.
- ABB, Instrument Transformers Application Guide, publication 1HSM 9543 40-00en, Edition 4, February 2015, sections 3.4 and 4.9–4.12.
- B. N. Taylor and C. E. Kuyatt, NIST Technical Note 1297, Guidelines for Evaluating and Expressing the Uncertainty of NIST Measurement Results, 1994, sections 5 and 6.
- Nippon Steel Corporation, Grain-Oriented Electrical Steel Sheets, catalogue D004je, April 2020, specification tables, size tolerances and Annealing by Customers.