Wye-Delta Transformer on PV/Battery interconnection system 2

[PrimalPete]

Hi All,

I'm working on a small microgrid system that is interconnected to the main grid of a medium size building (Average load of building is about 15KW).
The PV in 30KW and battery is about 115KWH. It is an AC coupled system at 480Vac fed to a transformer which steps down to 3 phase 208Vac.
The 208Vac side of the transformer is fed to the interconnection point, which is a 110 amp breaker downstream of the main grid breaker.

Transformer: 45KW Wye-Delta (208 side is Wye, 480V side is Delta). The model is HPS SG3A0045KB

Issue: While commissioning this system, I noticed an unusually high current going into the 208Vac side of the transformer, while none of the 480V equipment was active. So I opened the 480Vac disconnect blade on the 480V side of the transformer, making an open circuit on the 480V side. The current going into the 208V side remained.

The current is reading about 10 amps on each leg and 30 amps on the neutral going into the 208V side of the transformer.

1. The neutral is not bonded to ground inside the transformer.
2. When the neutral is disconnected from the transformer on the 208V Wye side, the current disappears.
3. The transformer is wired correctly.

I'm looking for help on what is causing the current to flow into the transformer. The grid-tied system has islanding capabilities in the event of a grid outage, so we need the neutral for islanding operations.

Any help is appreciated. I've spoken with HPS tech support but we have yet to figure out the issue.

 

[itsmoked]

Just to clarify without a diagram the facility is fed from the PoCo on the 208V side?

The 480 is only to match the solar supply?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[PrimalPete]

That's correct. 480 side is PV and battery.

Point of grid connection is 208V. Really I think the whole 480V side can be ignored (i could be wrong) since the issue occurs even when the 480 side is disconnected.

 

[davidbeach]

The transformer is acting as a grounding bank. Voltage unbalance on the 208V side will drive circulating current in the delta and out on the wye-side phases.

 

[itsmoked]

Thanks and I agree.

Have you removed all 208 and 120 loads leaving only the PoCO hooked to the transformer?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[PrimalPete]

I have not done this, but it is on my list, and a great suggestion.

The problem is that it is City owned building with emergency response personnel. It takes 2-3 weeks to get permission to shut power. So it might be some time before I can do it.

 

[PrimalPete]

@David,

Assuming this is the case, what are the dangers of connecting the system if these unbalanced phases are pushing current through the transformer?

 

[davidbeach]

You can burn up the transformer, or lots of other problems short of that. Worst case will be a ground fault on the utility between their tap fuse and your facility. The fuse will open but the fault will continue to exist, fed now entirely by your transformer. The phase currents may not cause the phase protection to operate but there will be lots of current in the neutral. Enough to burn open the neutral-ground connection.

 

[itsmoked]

david; So the delta winding is part of the grounding aspect? If you opened all delta windings this would stop?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[davidbeach]

It’s the delta itself, so you’d have to open a corner, but then it can’t be used to connect the PV. Delta on the 208, wye on the 480 could have been better as could have a wye-wye.

 

[PrimalPete]

David,

We need Wye on the 208 side so we can output power to the building in a grid outage.

Are you saying a Wye/Wye transformer would work fine to connect PV and battery, and just leave neutral disconnected on the 480 PV/battery side? This would get rid of this grounding bank phenomena?

Also, is there any way to describe in more detail how the Delta side of the transformer is causing it to act as a grounding bank? You mentioned imbalanced voltage on the 208 side, but can you elaborate on this? Just trying to wrap my head around all this.

 

[xnuke]

I'd really like to see a sketch of the system, including all transformers and their grounding status, because things are unclear to me. Please correct me where I'm wrong in the following statements. It sounds to me like:
[ul]
[li]the facility operates from a utility transformer, the secondary of which is 208Y/120 V, and the power distribution in the building is 208Y/120 V. That means the neutral is probably grounded at the service entrance panel.[/li]
[li]the PV and battery connect to a power conversion system (PCS) that produces power at 480 V AC, the 480 V is fed through a disconnect switch and then stepped down through a delta-ungrounded wye transformer to 208Y/120 V, and then connects to the building system through a 110 V circuit breaker.[/li]
[li]with the disconnect switch open between the PCS and the step-down transformer so that the 480 V winding of the transformer is deenergized, there is still 10 A per phase on all three phases and 30 A going into the neutral of the step-down transformer.[/li]
[/ul]

If that is the case:
[ol 1]
[li]the step-down transformer cannot be acting as a grounding bank on the 208 V side, because it is ungrounded.[/li]
[li]since positive and negative sequence currents sum to zero at the neutral, since the neutral current appears to be the sum of the phase currents, the phase currents must be zero sequence currents as they add together to make the neutral current.[/li]
[/ol]

Do you know what could be the source of your zero sequence current? It could be a type of ground fault or triplen harmonics. It doesn't appear to be a line-to-ground or line-to-line-to-ground fault, since the current distribution would be different in the phases. Do you have a significant amount of triplen harmonics coming from the building loads? Do you have a second neutral grounding point in the system other than at the service entrance panel that could allow circulating currents, e.g., at a distribution panel where the neutral is supposed to be ungrounded?

xnuke
"Live and act within the limit of your knowledge and keep expanding it to the limit of your life." Ayn Rand, Atlas Shrugged.
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[waross]

One of a number of threads on this issue.

Circulating currents in banks of transformers with star-delta connection 3
thread238-444324

Measure your phase to phase voltages on the 208 Volt side with the wye/delta transformer disconnected.
Your phase to neutral voltages may be equal but a phase shift may be causing unequal phase to phase voltages.
This phase shift is more common on long circuits with voltage regulators in the field. Issues are very often found on rural circuits.
If you measure different phase to phase voltages then reconnect the wye/delta transformer and check again, very often you will see a difference as the transformer bank uses the delta winding to try to correct the phase shift.

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

 

[PrimalPete]

@xnuke,

Your three bullet point assumption are spot on, except that the transformer is grounded. What I meant in my original post was that the neutral feeding into the 208 Wye side is not bonded to the ground inside the transformer ( HPS tech support asked me to check for this, and I have confirmed it is not).

I don't know what triplen harmonics are but I have something to research now! Thanks for the patience, I'm learning a lot.
Regarding the neutral being grounded at a second point, I'll be looking for this.

My first plan of action is to measure the current flowing through the neutral coming from the grid with the transformer connected VS not connected. I want to see if the neutral currents will match. If the current in the Neutral goes away with the transformer is disconnected from the grid, is it safe to assume it is not an issue with neutral wiring on the building load side?

If neutral current doesn't go away when transformer is disconnected, I plan to measure all neutral paths that go to different distribution panels, to see if one of them has excessive current. This will at least tell me which load panel could be causing the problem. If something weird is found, an inspection of the wiring will ensue.

Any thoughts on this plan? Anything else I can check for? I am not able to disconnect loads to the building as mentioned earlier.

Thanks!

 

[PrimalPete]

Thanks Waross,

I will do this today.

I do know that the phases are imbalanced in terms of current, I can report a snapshot of loads in a bit. I will check phase to phase voltages with transformer connected vs disconnected and report back soon.

 

[PrimalPete]

Transformer NOT connected:
A-B 213.0V
A-C 214.5V
B-C 214.1V

Transformer Connected:
A-B 213.5V
A-C 214.5V
B-C 214.5V

Currents measured at main utility breaker while transformer is connected
A: 62.8A
B: 63.0A
C: 45.2A
N: Not Measured

Couldn't get current readings with transformer disconnected. There are CT's on main grid lines that connect to a meter, but meter will lose power if I open 208V breaker to the PV system. I'll have to open the main panels and use a clamp on for that. Also, I have no CT around the neutral incoming from grid.
I'll be doing those measurements on Monday or Tuesday of next week.

 

[waross]

The delta is taking power from the A-C phase to raise the voltages on the A-B and B-C phases.

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

 

[itsmoked]

Nice Bill.

Is this a problem pushed down the wires from the PoCo or is it an in-house problem like heavy single phase non-linear loads on two phases (lighting for-instance)?

If this is a big facility it would likely have it's own transformer.
If instead, the transformer is shared it could be something next door.

If it's a single transformer can this problem even come from the PoCo?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

Hi Keith. Good question. It has started me thinking again about the issues with a wye/delta and this may be an issue that I haven't encountered before. Thank you for prodding my grey cells.
In the event that heavy unequal loading causes one or two phases of the 120/208 Volt supply to be unequal due to transformer regulation of the local supply transformer, then the delta will circulate current in an attempt to balance the voltages.
Basically, with the wye/delta bank or single three phase transformer, any unbalance of voltage and/or phase angle on the wye side will cause circulating currents in the delta.
Yes, you are correct in that even with a perfectly balanced PoCo supply uneven loading on the local transformer may have an impact on the wye/delta.
In my previous experience with wye/delta systems the customer's load has always been on the delta. Single phase loading on the delta is supported by all three phases equally, so uneven loading on the delta generally makes no difference.

If this is a big facility it would likely have it's own transformer.
If instead, the transformer is shared it could be something next door.

This situation is somewhat unique in that the customer may have control of the transformer feeding the wye/delta and may be responsible for uneven voltages. On a shared transformer, yes, his neighbour may be responsible.

If it's a single transformer can this problem even come from the PoCo?

Definitely YES. We recently saw an instance where an entire country is coming to grips with the issues caused by PoCo unbalances on wye/delta systems.

Solution:
The delta must be broken or the wye/delta must go. In some instances it is possible to break the delta, convert to open delta or float the neutral, but unfortunately it is doubtfull if any of those solutions are possible here.
While balancing the local loading may seem like a solution, the other effects of the wye/delta bank are possibly more serious. You may not be aware of these other issues until the transformer fails as noted by DavidBeach.
WARNING:
A transformer with a three legged core will exhibit a "Phantom Delta" effect. Avoid replacing the existing transformer with a three legged transformer.
lps for your insightful question, Keith.

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

 

[stevenal]

What's the nature of the power company service transformer? At 208, the secondary would be wye. Is the primary wye or delta?