Generator transformer connection: delta vs wye

[radug]

Hello,

I have mainly seen MV synchronous generators connected to a generator transformer where the transformer generator side is in delta and the HV side in wye.

Why not put the generator side of the transformer in grounded wye?

Does anybody know of books or papers that discuss this topic?


Thanks.

 

[jghrist]

http://www.beckwithelectric.com/docs/tech-papers/DG_1547_Paper.pdf

has a section V. MAJOR IMPACT OF INTERCONNECTION TRANSFORMER CONNECTIONS ON INTERCONNECTION PROTECTION

 

[inpran]

Radug,
I do not know any book or paper. But my logic says that it is done to isolate the generator from the rest of the system which is prone to more faults. For the zero sequence current to flow you must have your neutral grounded. By having a delta on the generator side you are ensuring that these zero sequence currents are not reflected on your generator circuit. I think that is why it is done, may be others in the forum might give you a better /right answer.

 

[Overvoltage]

In addition, a single ground may not take the system offline (depending on how the relaying is set up). This will allow personnel to solve the problem without generator downtime in some cases.

Also, the generator is usually grounded through a resistance ground, reducing the available fault current in the generator to a low level. This reduces the amount of damage in the event of a fault as compared to a solidly grounded system, in which the available fault current would be in the thousands of amps and may completely destroy the generator.

 

[prc]

The reason why generator transformer secondary is in delta is to avoid the zerosequence currents (consequent on any line to earth fault on HV)entering in to stator windings of the generator.Rotating machines cannot withstand zero sequence currents due to severe vibration and throwing out of windings from stator slots.

 

[radug]

Thank you for your comments,

It is a 18 MVA generator, the project is still in the conceptual phase. I have to decide generator voltage, 6,3 kV or 11 kV, as well as generator transformer connection. There will be no auxiliary services transformer, so the distribution transformers will be connected to the same busbar as the generator.

"Transmission" voltage will be 25 kV, as there is an existing substation next to the plant.

If a go with a 25/11 kV YND type transformer, I think that unit transformer cost will also be lower than the DYN option, but I will have to add a zig-zag transformer to have a MV ground when the generator is offline but the auxiliary services are on.

 

[ScottyUK]

Overvoltage has hinted as to the main reason: the generator core and stator is a very high value asset with a lead-time for a replacement core being many months, perhaps years, depending on the size of the core. That is a very long time to have a generator out of service and not producing revenue. By limiting the fault current to a value of perhaps 10A or so it is possible to almost guarantee that damage resulting from a stator earth fault will be repairable, while still being large enough for the protection to reliably detect.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[waross]

Another factor is sharing of unbalanced loads. A line to neutral load on A phase will be supplied 50% by A phase of the delta and 50% by B phase and C phase of the delta. Sharing of unbalanced phase to phase loads is more complicated but still better than a wye connection. This is more of an issue with smaller sets, below about 1 megaWatt.

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

 

[odlanor]

1 - current short-to-ground withstand by generator
Stator core damage from arcing fault is responsable by vast majority of internal generator faults.
Arcing at the point of failure cause significant burning of the stator core laminations, the cost and duration of the outage will increase dramatically.
Typical values of arc current are:
< 20 A - negligible arc burning
20-40A - slight burning
> 40 A - severe damage.

2 - How to reduce the fault-to-ground
2.1 - adopt the delta winding of the stepup transformer low voltage delta / high voltage star grounded.
With this it is guaranteed that during ground fault in the generator, there is no contribution from the current high-voltage side.
Note that ground fault on the high voltage is seen as a two-phase fault by the side of generator.

2.2-grounding the generator neutral
Ungrounded neutral would eliminate concerns related to excessive ground fault current, because the current flow would be only minimal current sustained by the system shunt capacitance.
Unfortunately, this solution is unacceptable because this capacitance can cause damaging overvoltage.

The primary requirement of any grounding scheme is that it limits overvoltage to non-damaging levels.
Steady-state overvoltages will appear on the unfaulted phases during ground fault.
Transient overvoltages resulting from restriking currents are the most dangerous. Current restrike occur when the voltage across breaker contacts or other gaps increase faster than the dielectric gap. An ungrounded system will allow cumulative voltage increases with each restrike.

The popularity of high-impedance grounding schemes is directly related to the fact that these schemes limit ground fault current to values below 15A.
(adaptation from Donald Reimert book)

 

[radug]

What I am thinking is:
-Generator high resistance grounding.
-Unit transformer with the star in the generator side, connected to a resistor through a grounding switch, so that when the generator is online, there is only a grounded wye (the generator) and when the generator is offline and I want to feed the auxiliary loads I ground the wye of the transformer.