Three phase Transformer Failure 1

[jlschoen]

I am looking for a good explaination to what theoretically happens to a three phase transformer (gnd wye-gnd wye connected) supplying three phase load when two of the source phases are the same? (Basically a three phase transformer is supplied by two A phases and a B phase?)

 

[jghrist]

Nothing happens to the transformer. The same thing happens to the three phase load as would happen if two of its terminals were connected together and then connected to one phase and the third terminal were connected to another phase. How the load reacts to such voltages depends on the nature of the load.

 

[davidbeach]

You will get a lot of current in the neutral that wouldn't have been there in the 3-phase condition.

 

[waross]

Will there be a lot of leakage flux if the transformer has a three legged core? How about return flux in other construction types?

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

 

[jghrist]

You will get a lot of current in the neutral that wouldn't have been there in the 3-phase condition.

Only if there is a wye connected load. If the load is connected delta, there will be no neutral current.

Will there be a lot of leakage flux if the transformer has a three legged core? How about return flux in other construction types?

I hadn't considered a three-legged core. If there is zero-sequence, then there will be a lot of leakage flux in a three-legged core. Don't most wye-wye transformers have five-legged cores to allow a zero-sequence flux path?

 

[honglien]

I agree with jghrist and davidbeach's explanation, but I would explain more details. As you know, when the transformer is connected correctly to the system, if transformer's load is balance, there is zero current on the grounding wire. As jlschoen's problem, the C-phase current vector was absent, the A-phase current vector was double so the total vector which was the current vector on the grounding wire would be very high so be dangerous for the transformer and the load.
Another effect I could recommend is the power on the A-phase on the primary side of the transformer would be double, make the A-phase overload, because the power line carries full capacity, cause protective devices trip.
You know, a power line on 3-phase system can operate with lossing one phase, and so the power is lossing too.

 

[waross]

When the poster says grounded wye primary, I have to assume that this is also a primary neutral connection.
With one phase missing, the neutral current will be the vector sum of the two phase currents that are present. If two transformer primary windings are now connected to the same phase, I am not sure what the neutral current will be, but I doubt that it will go to zero. I suspect that non-regenerative loads will increase the neutral current to about 173% of phase current.
Motor loads at rest fed from this supply will not start but will draw heavy currents and the neutral current will be about 173% of the line currents, but this current will be almost 90 degrees displaced from the resistive neutral currents.
HOWEVER, for this to happen in real life, I visualize one phase conductor on a pole line breaking and dropping onto another phase conductor. Any motors running will now have one regenerated phase 120 degrees out of phase with one supply phase. Expect heavy, unbalanced currents, and expect the unbalanced portion to return on the neutral.
This suggestion is based on three single phase transformers. With a three phase transformer, flux path complications may result in saturation of the flux return paths. This may cause reduced voltages and correspondingly reduced currents on the secondaries, and may result in higher exciting currents on the primaries.
I understand that an unbalanced load on the secondary of a wye-wye bank will cause a primary neutral current even though there is no secondary neutral current.
A question regarding three legged cores. If two phases are energized, will the vector sum of the flux, returning through the third leg, generate the missing phase? If so connecting one of the other phases to this winding may result in greater current flows than would be present in three single phase transformers.
Energizing two windings from the same phase on a five legged core may cause saturation of the return path. If so, then again, currents on both line and neutral may be greater than they would be with a three transformer bank.
Some of the excess flux may be forced to return through the third, normally energized core, leading to saturation and greater than anticipated current.
Comments Gentlemen?

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

 

[jghrist]

Assuming a path for all fluxes, and neglecting magnetizing currents, the current in the primary of one phase of a wye-wye transformer will be the turns ratio times the current in the secondary. Same thing for all the phases.

If the load on the secondary is connected between phases, with no connection to neutral, then there will be no neutral current in the secondary, no matter what voltage is put on each phase. If there is no neutral current in the secondary, there will be no neutral current in the primary.

The OP did not indicate anything about the load. If all of the load were connected between the two phases that were energized with the same voltage, then there would be no current in either the secondary or the primary.

 

[waross]

Thank you, jghrist;
I see the error in my ways. I did not consider that the power factor in the line to line current. The current will be leading one phase voltage and lagging the other.
If this is a real world application, then I suspect that it was a broken primary in contact with another phase.
In the real world I would expect a normal plant load of mostly motors and some resistive loads.
There will be magnetizing currents that may increase due to saturation and may not. My transformer book states that most 5 legged cores do not have full size outside legs, but have only enough iron in the outside legs to carry anticipated harmonic and unbalanced fluxes. Possible saturation will depend on a combination of the transformer design and the working voltage at the time of the incident.
Yes, I agree, the load current will not show up on the neutral but magnetizing currents will. Saturation is possible and that will result in excessive neutral currents.
If motors are running at the time of the incident they will continue running on the phase supplied, but will be drawing excessive phase currents. I expect the heavy phase currents to continue until either the motor protection or the transformer protection tripped out.

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

 

[jlschoen]

Thank you for all the replies.

The event occured when a (B phase) jumper from an air-break switch failed and welded to the C phase conductor. All of the three phase transformers connected downstream of this switch were effectively fed from an A phase conductor and two C phase conductors. The outcome was that the three phase transformers (serving three phase load, grounded primary and grounded secondary wye-wye connected) had damaged primaries and were exposed to a tremendous amount of heat.

 

[waross]

Thanks for the feedback.
I think that your description is consistent with magnetic saturation. Motor back feed would have overloaded the secondaries also. This does not rule out a combination of the two effects. Where there any user motors or other equipment damaged?

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

 

[apowerengr]

I saw a similar event happen. It was difficult to troubleshoot as all the single phase customers had power but the three-phase ones did not and were calling in outages. Since single-phasing protection is recommended as part of the utility service rules most customers did not suffer damaged equipment nor do I recall us as the utility having to replace any damaged transformers. I do not recall the transformer construction specification at that particular utility thought.