Ground Fault Detection on secondary of wye-wye transformer 1

[rockman7892]

In looking at the zero sequence network of a ungrounded-wye / grounded-wye transformer it appears that this transformer would not produce zero sequence current on the secondary wye side even though it is solidly grounded.

Am I looking at the zero sequence circuit correctly in that the secondary of this transformer will not produce zero sequence current and will be the same as an ungrounded transformer?

How is ground fault current from this transformer usually detected? Similar to methods used for detecting ground fault on ungrounded systems?

I have seen several utilities use this configuration at the service entrance? Is there any particular reason for using this configuration?

Thanks

 

[davidbeach]

You are reading it correctly. You may see some phantom delta affect if it's a three legged core. Not sure why utilities would use it; are you sure they are or are they just using transformers with H0 and X0 bonded internally?

 

[HamburgerHelper]

I don't know why someone would want that configuration, too. I remember reading I think it was in the "Whys of Wye" that it wasn't common in the U.S. but was seen in asia. I think the kicker was that it produced bad harmonic overvoltages on the high capacitive grounding impedance on the ungrounded side.

 

[rockman7892]

Thank you for the responses.

I first noticed this configuration when I was reviewing a short circuit study completed by others. I noticed that the fault report was showing any L-G fault and when I looked closer I noticed that the utility transformer was modeled in this configuration.

As I go back now and look into the response letter from the utility I notice that they said the transformer is a "12.47kV:480/277 Wye-Wye with a Solid Neutral-ground Connection". Reading this I would have to assume that both the primary and secondary wye points are solidly grounded perhaps with the H0 and X0 bonded internally as mentioned. So based on this new information it would appear to me that both the primary and secondary wye's are grounded and the engineer completing the study made an error by modeling the primary of the transformer as an ungrounded wye.

From the zero sequence circuit it appears that with both the primary and secondary wye points grounded the transformer will pass zero sequence current from primary to secondary?

If for some reason the primary of the wye was ungrounded (not sure why anyone would want this as mentioned above) then this is essentially the same as having an ungrounded system on the secondary. Is this correct?

Thanks for the help!

 

[Kiribanda]

rockman7892,
1) I donot think any transformer manufacturer will deliver his transformer with H0 & X0 bonded & grounded internally.
It has to be done by the user after the transformer is installed.
2) If you do a simple system sudy by modeling Y(GND)/y(gnd) you will find that it is equivalent to a DELTA/wye(gnd) connection
for ground faults.

 

[HamburgerHelper]

wye-ground-why-ground is not the same as a delta-wye. Why-ground-wye-ground can't source zero sequence. It only passes it. Why-ground-wye-ground can only source ground current if it has a tertiary delta.

 

[stevenal]

A three phase wye/wye pad mount generally has no separate H0 bushing. H0 is simply connected to the transformer case internally. Of course no manufacturer can ground anything, they simply provide the lug for the installer to use.

 

[dfdt]

Ungrounded wye/grounded wye transformer will definitely provide path for zero sequence flux to flow for line to ground fault on secondary side.The flux path is via the tank of the transformet for core type transformets.Zero seq impedance is typically in the range of 50% to 100%.
Regards

 

[ShooterBooter]

Ungrounded wye/grounded wye transformer will definitely provide path for zero sequence flux to flow for line to ground fault on secondary side.The flux path is via the tank of the transformet for core type transformets.Zero seq impedance is typically in the range of 50% to 100%.

No,you don't have the required ampere turns on the primary to balance the ampere turns on the secondary during a secondary phase to ground fault

-SB

 

[davidbeach]

What bholas is referring to is the phantom delta affect I mentioned earlier. The flux will take a path that includes core, air gap, case, air gap, and back into the core. That phantom delta creates something neither fish not fowl; the grounded wye side isn't as ungrounded as a delta winding would be, but also not nearly as grounded as it would be if the other side was a delta or a grounded wye.

I don't recall seeing transformer sequence diagrams that include the phantom delta, but it's a recognized phenomena.

 

[Marmite]

Ungrounded Y/Grounded y transformers are in widespread use by distribution utilities in the UK at voltages typically 33/11kV, 66/11kV or similar. They make use of the tank phantom delta effect. They have been in common use since the 1970's. This design of transformer has also been applied as station transformers in coal power plants and supply the Unit switchboard when the unit is off, typically 132/11kV. The zero sequence impedance is typically 75-200% of the positive sequence impedance.

Regards
Marmite

 

[prc]

Marmite, why in UK Y/yn is used instead of YN/yn ? Any advantage?

 

[waross]

The zero sequence impedance is typically 75-200% of the positive sequence impedance.

Thank you for the information Marmite. I have wondered about this for years. This is the first time I have seen an actual impedance value assigned to the phantom delta effect. As broad as the range is, it will be useful to know what is a common range when trouble shooting.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ShooterBooter]

I can't really recall seeing any test records mentioning any tests done to verify this "feature".
Anybody else seen this?

-SB

 

[dfdt]

When ordering YN/yno transformer,it's good idea to specify your requirement with regards to zero seq impedance to begin with or else transformer manufacture will decide depending on their design.we normally specify a value of Zo impedance of around 75-100% so that phase to ground fault current on LV side does not exceed 1.5 times of the full load current.Also one can specify to conduct Zero seq test during factory acceptance testing in which manufactures will conduct following four tests
HV energised with LV open
HV energised with LV sorted
LV energised with HV open and
LV energised with HV shorted
Based on these result you can compute zero sequence impedance of the transformer.
Hope this helps.
Regards,

 

[ShooterBooter]

The standard zero sequence impedance measurment/test won't detect this since the primary neutral is used as return path.
The only way to test this should be by applying 3-phase voltage on the primary terminals with the primary neutral unconnected and one secondary phase shorted to neutral, the other secondary terminals left open.

The "use" of this effect can't be much more than the annoying need to add zero sequence filtering in the transformer differential protection?

Funny though, I've never actually paid this phenomenon much of a thought, less having to deal with it's negative side effects.
Anyhow, it's always nice to learn new things.

-SB

 

[rockman7892]

Thanks for all the responses they are very helpful

 

[GroovyGuy]

I understand that wye-wye connections are popular with Utilities simply because they are less expensive.
You will have (secondary) zero-sequence current flowing in any transformer with a grounded secondary.
The best way to detect zero-sequence current is with a zero-sequence CT and relay.
Any reason why you are not considering a neutral-ground resistor for limiting ground-fault current on the secondary side? Obviously a NGR system cannot facilitate any phase-to-neutral loads. I like NGRs a lot and would never consider not installing one, but then I'm tainted.
BTW, what size of transformer are we talking about here?