The zero-sequence impedance of a transformer depends on whether the zero-sequence current of one winding has ampere-turns balancing it in another. If it does (YNyn, YNd, Yzn), the impedance is the short-circuit impedance between windings. If not (YNy, Yyn), it is a high magnetizing impedance that depends on the core. A delta lowers it and increases the ground fault current.
What zero-sequence impedance measures
The zero-sequence impedance of a transformer is the impedance that the zero-sequence current meets. That current is the same in all three phases and returns through the neutral or through the delta. This guide summarizes how the value changes with the connection (IEC 60076-8, clause 4.7). More on the zero-sequence diagram of a transformer in symmetrical components and the zero-sequence diagram of a transformer.
There is a single rule. If the zero-sequence current of one winding has ampere-turns balancing it in another winding, the impedance is equal or close to the short-circuit impedance between those windings. If it has none, the current acts as magnetizing current and the impedance is high or very high, depending on the magnetic circuit.
The role of the magnetic circuit
In the three-limb core, the zero-sequence flux leaves the iron and closes outside. The magnetizing impedance is moderate and nonlinear. In the five-limb core, the shell-form core and a bank of single-phase units, the flux returns through the iron and the impedance is very high (clause 4.4). More on what changes between a three-limb and a five-limb core in three-limb and five-limb core.
If the study depends on this value, the measurement is requested as a special test (IEC 60076-1, clause 10.7). The zero-sequence impedance test describes it.
Delta winding
The phase-to-phase voltages of a delta add up to zero because of the closed connection. For zero-sequence voltage, the delta behaves as a short circuit (clause 4.5).
A zero-sequence current cannot be exchanged between the three terminals of the delta and an external system. But another wye winding with a grounded neutral can induce a short-circuit current that circulates inside the delta. Seen from the other winding, the transformer then presents an impedance with the character of a short circuit between that winding and the delta.
Zigzag winding
In a zigzag, each limb carries sections of two phases wound in opposite directions. The zero-sequence ampere-turns cancel on each limb and there is no magnetization. The current meets only the low impedance of the leakage flux between the winding sections (clause 4.6).
Summary by connection
The standard summarizes typical values in its table 1 for concentric windings, numbered from (1) the outermost to (3) the innermost. The percentages refer to the reference impedance U²/S. The transformers marked with an asterisk in the table are those in which the zero-sequence current of the excited winding is not balanced by that of any other.
| Connection | Zero-sequence impedance seen from the winding with grounded neutral | Clause |
|---|---|---|
| YNyn or YNauto without delta | The current passes between the two systems if both neutrals are grounded. Normal short-circuit impedance. With no transfer possible, it is the magnetizing impedance. | 4.7.1 |
| YNd, Dyn, YNynd, YNyn + d | Low impedance, with the character of a short circuit. A current circulates in the delta and compensates the ampere-turns. | 4.7.2 |
| Yzn or ZNy | Low impedance from the zigzag side. The zigzag balances itself. | 4.7.3 |
| YNy or Yyn, three-limb core | Magnetizing impedance. In table 1, on the order of 50 % to 70 % of the reference impedance. | 4.3, 4.4 |
| YNy or Yyn, five-limb, shell-form or bank | Very high magnetizing impedance. | 4.3, 4.4 |
For the combinations with ampere-turn balance, table 1 uses factors a1, a2 and a3 applied to the short-circuit impedances. Those factors are generally between 0.8 and 1.
YNyn or YNauto without delta
The transformer receives and transfers zero-sequence current between the two systems, provided the neutrals are grounded. If the neutral of an autotransformer is not grounded, transfer is still possible, but with a different impedance. If the opposite system does not accept the current, the transformer presents the magnetizing impedance, which is very high in five-limb, shell-form and single-phase banks (clause 4.7.1).
YNd, Dyn, YNynd
The transformer presents a low impedance, with the character of a short circuit, to the zero-sequence current of an effectively grounded system entering through a yn winding. The current circulating in the delta provides the compensating ampere-turns.
That is why an additional delta winding in a Yy, or in a bank of three single-phase units, lowers the zero-sequence impedance of the connected system and its earth fault factor. The consequence is that the prospective ground fault current increases (clause 4.7.2).
Verify that the short-circuit withstand of the tertiary or stabilizing delta is sufficient for the maximum zero-sequence current induced in a ground fault in any of the connected systems. Built-in limiting reactors can also be connected inside the delta to bring that current to a tolerable value.
Yzn or ZNy
The transformer presents a low impedance to the zero-sequence current of the system on the zigzag side. That is why a transformer with a ZN winding can be used to create a neutral point where a grounding impedance is connected, when the winding of the main transformer is a delta. It is called a grounding transformer or neutral coupler (clause 4.7.3, referring to IEC 60289, section 6). The same function is obtained with a YNd combination in the grounding transformer. For the advantages of wye, delta and zigzag windings, see wye, delta and zigzag windings.
If the Y side has its neutral grounded (YNzn), the transformer presents a magnetizing impedance to the zero-sequence current of that side, of a value similar to that of a YNyn. The zigzag is balanced in itself, but it cannot compensate the ampere-turns of the zero-sequence current of the opposite Y winding.
Two connections that should not be mixed
IEC 60076-8, clause 6.1, does not recommend parallel operation of a Dyn with a Yzn because their zero-sequence impedance properties are different. The guide on parallel operation covers the other conditions.
We can help with your transformer
At Tiza Group we carry out power transformer testing and commissioning in Colombia, or elsewhere on request, including zero-sequence impedance measurement, and the protection coordination studies that depend on that data, remotely and worldwide. If you need the value for your unit, we will quote the work.
Related guides
- Zero-sequence impedance test on power transformers
- Grounding transformer
- Zero-sequence current
- Neutral grounding resistor
- Restricted earth fault protection (87N / 64)
- Three-Limb vs Five-Limb Transformer Core: What Changes
- Wye, Delta and Zigzag Transformer Windings Explained
- Transformer Symmetrical Components: Zero-Sequence Diagram
References
- IEC 60076-8, power transformers: application guide, clauses 4.4 to 4.7 and table 1.
- IEC 60076-1, power transformers: general requirements, clause 10.7.
Frequently asked questions
Why does a YNd transformer have a low zero-sequence impedance?
Because the zero-sequence current of the wye winding causes a compensating current that circulates in the delta. The ampere-turns are balanced and the current meets only the short-circuit impedance between the two windings (IEC 60076-8, clauses 4.5 and 4.7.2).
What changes when a delta stabilizing winding is added to a Yy?
The zero-sequence impedance of the connected system drops, and with it its earth fault factor. The consequence is that the prospective ground fault current increases (IEC 60076-8, clause 4.7.2).
Is the zero-sequence impedance always equal to the positive-sequence one?
No. When ampere-turns are balanced it is equal or close to the short-circuit impedance. The standard indicates that differences are usually between 10 % and 15 %, and for the two main windings with a tertiary, between 10 % and 20 % (IEC 60076-8, clauses 4.3.2 and 5.3).