Gnd Wye Transformer connected to Gnd Wye Generator

[steelerfan28655]

I have been asked to consult on a new generator installation that is having startup problems. They have already fried two DC displays.
The system is installed at a primary metered facility that is fed 12470 from the utility. The feed has been re-routed to feed thru a 1500 kVA, 12470 GY to 480 GY transformer with the secondary connected to the output of the generator.
The problem they are having is this. The transformer secondary was bonded to ground, and the electrician also bonded the generator neutral inside the generator. The two ground grids were bonded together. When they started the generator, the circulating current in the ground-neutral loop was enough to fry the electronics in the generator.
After frying two units they called me. My suggestion was to remove the bonding strap in the genset and leave the bond in the transformer.They decided to follow the advice of another engineer with more experience, and removed the bonds at both ends and the neutral connection between the transformer and the generator.
My concern is the single-phase load at the facility now creating current on the downstream transformers(8 of them) with only the concentric on the underground cable as a return to the utility ground. Each downstream transformer is bonded to ground.
When the generator is in island mode, neutral current will only have the generator ground(through earth), which is not bonded to the neutral.
I have read some of the GY-GY connection threads on this site, and am still not sure what to suggest.

Happiness is a way of travel, not a destination.

 

[jghrist]

Is the single phase load on the 12 kV or the 480 volt system?

 

[steelerfan28655]

The single phase load is fed from the down stream transformers.
Here is a basic One line diagram.


Happiness is a way of travel, not a destination.

 

[dpc]

If I understand the system correctly, there should be a neutral conductor from both the transformer and the generator and these need to be bonded to ground at one place. Bonding both the transformer and generator neutrals to ground at different locations is not a good idea, since it can cause problems with ground fault protection. The generator frame can be locally grounded, but the neutral should be brought out and grounded at the same point as the transformer.



"The more the universe seems comprehensible, the more it also seems pointless." -- Steven Weinberg

 

[jghrist]

I can't tell from the one-line whether the "down stream" transformers are connected to the 12 kV or to the 480V.

You need to have the neutral bonded to ground at one point. Otherwise, there is no metallic path for ground fault current. This will not provide reliable fault clearing and is a violation of the NEC.

 

[steelerfan28655]

Sorry about that. The customers transformers are fed 12470 either from the utility, or the generator. The generator output is 480 to the first transformer. Then the voltage is stepped up to 12470 and fed to the customer.

There is a breaker in the utility feed to the high side of the first transformer controlled by the generator for utility isolation.

Here is a new drawing.

Happiness is a way of travel, not a destination.

 

[rbulsara]

The real issue probably is not the grounding but pitch of the generator, the real cause of the circulating currents. A 2/3 pitch gen would have just worked fine the way you had it in first place.

I do not see any need to connect the neutral of the xfmr to the N of the gen. In a Y-Y xfmrs, generally the both neutrals are internally bonded and secondary side (load side) neutral is bonded to earth.

You may want to impedance ground the generator to limit the circulating current as long as there are no single phase loads on the gen bus (480V).

 

[jghrist]

I do not see any need to connect the neutral of the xfmr to the N of the gen. In a Y-Y xfmrs, generally the both neutrals are internally bonded and secondary side (load side) neutral is bonded to earth.

If the xfmr neutral is not connected to the generator neutral, then 12 kV ground fault current and 12 kV load unbalance current will have to flow through the earth when the generator is the source. This is not acceptable.

 

[steelerfan28655]

Quote:
I do not see any need to connect the neutral of the xfmr to the N of the gen. In a Y-Y xfmrs, generally the both neutrals are internally bonded and secondary side (load side) neutral is bonded to earth.
If the xfmr neutral is not connected to the generator neutral, then 12 kV ground fault current and 12 kV load unbalance current will have to flow through the earth when the generator is the source. This is not acceptable.

That is the reason for this thread. The engineer that was consulted said that the downstream transformers would create their own neutrals, or would be able to "absorb" any neutral current that exists. He stated that we may see some heating in the first transformer because of this, but it should be ok.

I guess I am a sceptic. I would be more comfortable not relying on the ability of a transformer to "absorb" anything.

Happiness is a way of travel, not a destination.

 

[davidbeach]

A wye-wye transformer is not a source for a neutral. It can have a neutral connected to both sides, that's fine, but you don't have a separately derived source. Usable neutrals can be derived from wye windings of transformers that include a delta winding, zig-zag grounding transformers, or from solidly grounded machine windings. No other sources.

If there are no loads on the 480V system produced by the generator, the transformer ought to have a delta winding on the 480V side, and the generator should be grounded through a resistor that limits ground fault current to 5-10A. A grounded wye winding on the 12470V side would produce a quite nice neutral.

 

[jghrist]

The engineer that was consulted said that the downstream transformers would create their own neutrals, or would be able to "absorb" any neutral current that exists.

You have good reason to be skeptical. This doesn't make any sense. If the downstream transformers are delta-wye connected, then no neutral current from unbalance will exist. You still will need a neutral connection between the transformer and the generator to carry fault current in the event of a ground fault. If the downstream transformers are single phase connected phase-to-neutral, or are grd wye - grd wye connected, then there is no neutral created. Unbalanced load will create neutral current which must flow between the transformer and the generator when the generator is the only source.

 

[rockman7892]

The wye-wye transformer is making it hard for me to visualize, but if it was a delta wye transformer like davidbeach suggested then would we need to worry about grounding the neutral generator together with the transformer neutral since the transformer would be a seperate source and all fault current would flow back to it?

What if the generator neutral was grounded and not tied to the transformer, but a seperate EGC was run from the generator out to the downstream transformers. Wouldn't fault current then return to the generator source? Again the whole wye-wye things is throwing me off, so I may be overlooking something in regards to this connection.

 

[waross]

Generators are much happier feeding a delta winding. Feeding a wye transformer without connecting the neutral is not a good idea.
Consider the single phase analogy. One 120:120 Volt transformer is feeding a 100 W load. Another 120:120 Volt transformer is feeding a 1000 W load.
The primaries of the transformers are connected in series and fed from a 240 volt circuit.
What happens?
The simple calculation would see 21.8 Volts across the 1000 W load and 218.2 Volts across the 100 W load.
In the real world, the over voltage on the 100 W loaded transformer would probably lead to saturation which would increase the IZ Voltage drop across the lightly loaded transformer so the voltage shifts may not be as severe as the calculations may indicate.
The calculation gets a little more complicated with a floating neutral on a wye primary transformer. You may also have a phantom delta effect that may stabilize the neutral point somewhat, but this effect is only present in some transformer designs, particularly three legged cores.
If the grounding/neutral/bonding confusion has caused burnout of DC devices, I suspect that you may be using a ground connection to replace a neutral conductor and the ground connections or path are compromised.
DON'T confuse a ground with a neutral even though they may be the same potential. Don't use grounding to replace a neutral in a circuit that needs a neutral.

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

 

[steelerfan28655]

The generator is being installed as an emergency back up system. It will also be used for paralelling with the utility for load management. During load management, I think that any phase imbalance can be handled by the utility feed, but in the event of a power outage, I am concerned about the effects of over-voltage on the single phase circuits. This customer has a lot of lab equipment that is fed through 12470 to 277/480 step downs, then through 480 to 120/208 transformers. I will have to check the 480 to 120/208 transformers to see if they are delta primary. That may take care of the problem.

Happiness is a way of travel, not a destination.

 

[jghrist]

If you are going to use a wye-wye generator transformer, ground fault current while operating in backup mode will have to return to the generator neutral. There must be connection from the transformer neutral to the generator neutral for ground fault current return or the ground fault may not trip the protection.

There may be a problem with the wye-wye connection during load management mode with a solidly grounded generator. Generators have low zero-sequence impedance compared to the utility source. Unbalanced load on the utility system may be fed from the generator, causing neutral overcurrent protection trips.

 

[steelerfan28655]

Well after two weeks, the engineer that they were consulting with came up with the idea to connect the generator neutral to the transformer neutral.
He is a brilliant man! I wonder which one of the electricians I have been telling this to for the past two weeks mentioned it to him.
They have not bonded the neutral in the trnsformer or the generator. I would prefer if they bonded in the transformer, at least that would satisfy the NEC. But this is a utility installation and just because the NEC makes good common sense some times, that doesn't mean they will follow it.

Happiness is a way of travel, not a destination.

 

[steelerfan28655]

I wanted to thank all of you for your input and ideas. Hopefully when they decide to install more of these systems (and I know they will), they will remember the problems they had with this one. Maybe they will even ask me for my input ahead of the installation. (Probably not)

Happiness is a way of travel, not a destination.

 

[waross]

Generators are happier feeding into a delta transformer winding. Hopefully they will go with a delta:wye transformer and a high impedance ground connection on the generator wye point the next time.

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