The technical report IEC TR 61869-100 explains that today the oversizing of a protection CT is set by the relay manufacturer, who publishes its requirements in the manual. The designer applies them to each case and gives the class parameters to the CT manufacturer. The TPX, TPY and TPZ classes, which moved the calculation to the CT manufacturer, did not become widespread.
Short answer
Sizing a protection CT means deciding how much the core must be able to deliver before it saturates. That decision depends on the fault current, its DC component and what the relay algorithm tolerates. The technical report IEC TR 61869-100 (2017 edition) explains how the task is divided today: the relay manufacturer sets the oversizing, the designer applies it to their case and the CT manufacturer builds the core with the parameters it receives.
How this division came about
The technical report summarizes the history in its clause 4.1:
- The standard IEC 60044-6 was introduced in 1992 and was the first to consider the transient behavior of protection CTs.
- In the traditional class P, the oversizing for the DC component of the primary current is «hidden» in the accuracy limit factor or in the burden.
- The TPX, TPY and TPZ classes were defined as «duty-cycle oriented»: they specified all the parameters of a C-O cycle and of a C-O-C-O cycle. With them, the CT manufacturer calculated the oversizing, represented by the transient factor, which the report now calls the transient dimensioning factor.
The transient-performance classes were never widely accepted, for two reasons given in the report:
- Specifying them by duty cycles and time to accuracy limit is much more complex than using the 5P and 10P classes, originally intended for electromechanical relays.
- The definition of the cycle no longer reflects the real criterion with which the oversizing factors are defined.
Today it is common practice for relay developers to set the required oversizing of protection CTs, considering the waveform of the primary signal and their own protection requirements. When IEC 60044-6 was merged into IEC 61869-2, it was taken into account that the duty cycle loses weight. That is why the standard allows the transient behavior to be defined by directly stating the transient dimensioning factor, instead of the cycle parameters. It is a simple way of specifying, similar to that of the P classes.
A note on names: IEC 61869-2 writes these factors with the letter K (Ktd, Ktf, Kssc). The technical report uses the letter F (Ftd, Ftf). They are the same concepts.
The chain of responsibilities
Clause 4.2 of the report recommends this order:
- Relay manufacturer: analytical formulas. First estimate of the oversizing
- Relay manufacturer: simulation. Tests its algorithms with a model of the CT core
- Relay manufacturer: publishes the requirements. Simple parameters for the most demanding fault cases, in the relay manual
- Designer: applies the requirements. Carries them over to the specific cases of the substation
- CT manufacturer. Receives the class parameters, already verified
The procedures by which the relay manufacturer determines the transient dimensioning factor from its protection rules are the own knowledge of relay specialists. The technical report does not include them. The formulas in the report are broader and more complex than those of the previous standard, but they are meant to be applied by few relay developers, who then give designers and CT manufacturers parameters that allow simpler sizing.
For the designer, this changes the starting point: the input is not a formula from a book, but the relay manufacturer’s requirements document.
The compatibility route
For compatibility with IEC 60044-6, the report allows a shortcut. The transient factor calculated analytically, from the time to accuracy limit, the duty cycle and other parameters, is applied directly as the transient dimensioning factor. No additional tests are made with the relay. The report declares it a route that is not recommended. It can be chosen in situations where the design responsibility is not fully delegated to the relay manufacturer.
IEC 61869-2 itself warns that the formula for the C-O-C-O cycle of the previous standard ignores saturation during the first cycle and in many cases leads to unnecessarily high sizing values (note 1 of 2B.1.3).
Why cores can be smaller
The technical report gives two reasons (4.2):
- In digital relays the decision time has decreased continuously because of higher sampling rates and more refined algorithms. So the saturation-free time required of the CT has also been reduced. This leads to smaller cores, in line with the development of compact gas-insulated substations.
- Some algorithms apply Fourier or r.m.s. filtering to the dynamic CT signal. That is why relays react more gently to high currents with possible saturation.
As a consequence, the performance required of the CT and the final core size cannot be expressed only with the initial time to saturation and a simple analytical formula.
What this means for your project
- Ask the relay manufacturer for the document with the CT requirements for your protection function: which parameters it demands and for which fault cases.
- Calculate, with your network data, the specific cases to which those requirements apply: fault current, X/R ratio, secondary burden and reclosing cycle.
- Turn the result into the CT specification, either by class (5P, PX, TP) or by direct transient dimensioning factor, and verify that the manufacturer’s offer meets it.
- Document what the relay requirement was and where it comes from. That record is what gets reviewed when the relay or the manufacturer changes.
What not to do
- Do not size the CT with a generic formula without consulting the relay manual.
- Do not mix in the same specification the duty cycle parameters and the direct transient dimensioning factor: the standard asks not to mix the methods.
- Do not reuse the sizing of one relay for another from a different manufacturer without checking its requirements.
Related topics
Who sizes the CT applies to the TP classes, the duty cycle, the 5P and 10P classes and Ktd. Settings depend on the setting criteria and on transformer differential protection. The general view is in high-voltage CT. Our free CT burden and saturation tool covers classes 5P, 10P and PX; the TP classes are calculated with the duty cycle.
When to call us
At Tiza Group we size protection CTs from the relay manufacturer’s requirements and your network data, and we verify on site that the installed cores meet them. See our engineering design and studies service, remote and worldwide, and our instrument transformer testing, in Colombia or elsewhere on request. Write to us for a quote.
Related guides
- TPX, TPY and TPZ CT Classes: Transient Error
- CT Duty Cycles: C-O and C-O-C-O
- 5P and 10P Protection CT Classes: Accuracy Limit Factor
- High-Voltage Current Transformer (CT): Types
- Current Transformer (CT): Parts, Functions and Tests
References
- IEC TR 61869-100, Instrument transformers - Part 100: Guidance for application of current transformers in power system protection, edition 1.0 (2017), clauses 1, 4.1 and 4.2.
- IEC 61869-2, Instrument transformers - Part 2: Additional requirements for current transformers, edition 1.0 (2012), clauses 3.4.233, 3.4.234, 5.6.202.5.3 and 2B.1.3.
Frequently asked questions
Who should calculate the transient dimensioning factor Ktd?
According to IEC TR 61869-100, the relay manufacturer is the one who knows how much saturation its algorithm tolerates. It first uses analytical formulas, then tests its algorithms with a core model and publishes the requirements for the most demanding fault cases. The designer applies those requirements and passes the class parameters to the CT manufacturer.
Can I use the factor Ktf calculated with formulas as if it were Ktd?
Yes, for compatibility with the previous standard, and without further relay testing. The technical report describes it as a path that is not recommended, which can be chosen when the design responsibility is not fully delegated to the relay manufacturer.
Why do digital relays allow smaller cores?
Because the decision time of digital relays has decreased with higher sampling rates and more refined algorithms. That reduces the time the CT must work without saturating, and cores can be smaller, in line with gas-insulated substations, which are compact.