Why Transformer zero sequence impedance differ from (+)&(-) sequence impedance 5

[TS Liew]

Most instances during fault analysis, I assume the transformer zero seq impedance = given positive sequence impedance in the transformer manual, but sometimes actual zero sequence impedance is quite different from the given (+) seq impedance. Hope to get some conceptual explanation as to why the difference occurs and why industry recommend to assume (+),(-) & (0) to be the same.

TS Liew (Electrical testing, Commissioning, Power system studies)

 

[dpc]

Zero sequence impedance of transformer depends on the transformer connection, grounding, and which side of the transformer you are looking at. You'll need to be more specific. Most textbooks on fault analysis will have diagrams of the various transformer connections and the resulting zero sequence impedance. Conceptually, zero sequence currents are all in phase with each other so there needs to be a current path available for this current.

Cheers,

Dave

 

[Kiribanda]

Zero seq impedance (Z0) depends on the core configuration, winding connection, 2 or 3-wdg etc etc.
Therefore, Z0 is not always equal to positive sequence impedance (Z1). Z0 test is an additional
test per IEEE/ IEC. Therefore, unless you pay extra for testing cost upfront, the manufacturer
will not give you the actual Z0 value. But you can always refer to any hand book/ manual and
get the ratio of Z0/Z1. Since the modeling is different and complex, for a 3-wdg transformer
it is always advisable to include the Z0 test into the FAT test list.

 

[jwatty]

The transformer impedance is dictated by leakage flux, so if the flux paths are identical for positive and zero sequence currents, you will end up with the same leakage impedance. This is clearly not true in 3 limb core form transformers. Under a zero sequence impedance test, you will short one side of the transformer, and tie all three phases together on the other side and energize together. Since the windings surround all core limbs, the return path of flux must be through air. Compare this to a positive sequence test: the return path for A-phase flux will be through B and C phase limbs (mostly). For a four or five limb core form transformer, the zero sequence impedance will tend to match the positive sequence because the zero sequence flux has a low reluctance path through the 4th and 5th limb instead of the air gap.

 

[cuky2000]

Considering the leakage flux associated with the path for zero sequence current for different forms of transformer cores, the following rule of thump could be applied:

a) X0=X1 for 5 legs (limbs)or shell form core
b) X0 < X1 for 3 leg core.

Some of the reasons why X0 ≠(X1=X2) for case (b) is because the zero sequence flux forced out the core and interact with the transformer steel tank resulting in a low exciting inductance path of the magnetizing circuit.

For case (a), there is a path for the zero sequence with minimum impact difference with the positive sequence impedance.

[sup]NOTE: Additional related information can be found on the following thread: thread238-89502[/sup]

 

[TS Liew]

Thank you fellas for your valuable posts. I totally get it now. The transformer zero sequence impedance test circuit shows exactly the point you mentioned in your posts. I now understand why the (0) seq test uses a single phase current injecting into all 3 branches of the transformer, forcing the "zero" seq flux to take the "outside" path for return. Whereby unprovided path (3 limb core type), low inductance causes lower Z(o). As for provided path (shell type), all 3 (+),(-),(0) has almost entirely same magnetic path thus Z(0)≈Z(+)=Z(-).

TS Liew (Electrical testing, Commissioning, Power system studies)

 

[prc]

Zero sequence impedance characteristics of transformers (for various connections and various core constructions) are given in Clause 4.0 of IEC 60076-8 -1997 Transformer Application Guide. Typical Z0 for each case is given in Table 1 of this IEC

Please see attached sketch. Zero sequence impedance three terminal star equivalent circuit is shown. ZA+ ZB is the positive sequence impedance and ZC forms the zero sequence magnetizing impedance, characteristic of which will depend on magnetic circuit construction.

With a YNy connection, Z0 will be 50 % for 3limbed core and 1000 % for 5 limbed core. Only ZA+ZC will be there in the picture.
With YNyn or YNd connection, Zo will be 80-90% of Positive sequence impedance ( 3limb core- Zc comes down) and same as positive sequence impedance ( 5limbed core- ZC,very high)
Magnetization of the core limb is also shown. With 3limb ZC comes down and net zero sequence impedance also comes down.

For a thorough analysis, refer to seminal paper-ZERO-PHASE-SEQUENCE CHARACTERISTICS OF TRANSFORMERS, A. N. Garin. General Electric Re-view, Schenectady, N. Y., vol. 43, 1940, pp. 131-136 and 174-179.