480V delta/wye PV inverter question

[ggunn]

We have a PV inverter that has 480 3 phase output with no neutral (delta). The service is 480V with a neutral (wye). Can we just connect the three output buses of the inverter to the service and leave the neutral open? Please forgive the elementary question; I am an engineer, not an electrician.

 

[bluesmoke...oops]

I would equate it to a 3 phase balanced load that doesn't require a neutral and say yes you can.

 

[hurk27]

We have a PV inverter that has 480 3 phase output with no neutral (delta). The service is 480V with a neutral (wye). Can we just connect the three output buses of the inverter to the service and leave the neutral open? Please forgive the elementary question; I am an engineer, not an electrician.

Depends on if there are any 277 volt loads served by the service, also if you have a 480/277 service with the neutral bonded, even if there are no neutral loads, you could not bond the PV supply as it would result in a short circuit if you bonded a phase conductor, I would say that the right inverter needs to be obtained.

 

[ron]

If the service is 4 wire, then how will the inverter react if there is an imbalanced load draw?
I would think of this like any additional source to a service like a generator. If the loads are 4 wire, then the source needs to be 4 wire as well.
Maybe an isolation transformer to derive a neutral?

 

[ggunn]

Thanks for the replies so far. The documentation for the inverter does not mention delta or wye, it just says connect the three output bus bars to the three phases of the service. There is no neutral bus. Also, if it makes any difference, there are no loads at the connection point.

I have found documentation that says that in general equipment that is designed to connect to delta can connect to wye, but not the other way around, since the phase to phase voltages are the same. That would indicate to me that we can just connect the phases.

 

[gar]

101215-1422 EST

ggunn:

I have found documentation that says that in general equipment that is designed to connect to delta can connect to wye, but not the other way around, since the phase to phase voltages are the same. That would indicate to me that we can just connect the phases.

I would tend to agree with that.

We can assume that the system you are feeding into is stiff, a low source impedance. If this is correct, then even if loads were present and varying, then this should not matter either. A PV system inverter is sort of like a constant current source within some range.

.

 

[LarryFine]

The service wye system will act to derive the neutral as it does now, so there is no issue.

 

[dbuckley]

The service wye system will act to derive the neutral as it does now, so there is no issue.

And there is the right answer.

As long as... this is the common case of the inverter requires the presence of the utility to operate. If the inverter can run stand-alone (which requires a utility disconnect) AND there are P/N loads then that wont work as there is then no neutral reference.

 

[hurk27]

The service wye system will act to derive the neutral as it does now, so there is no issue.

My question is how does the output of the inverter get bonded, if depending upon the service neutral bond, what happens when the PV disconnect is opened? you can't bond a phase as in a grounded "B" delta since the utility will have the neutral bonded, so when the PV disconnect is open it will be an ungrounded system, is this allowed?

I agree the utility transformer will derive a neutral for the PV system, but will they allow this added current on their transformer knowing how they under rate them.

Also I can see where there might be a problem if the PV supplied a neutral, as if this PV was set up to supply back to the grid there would be problems with the two X/O connections if the utility transformer was a common core?

 

[LarryFine]

My question is how does the output of the inverter get bonded, if depending upon the service neutral bond, what happens when the PV disconnect is opened? you can't bond a phase as in a grounded "B" delta since the utility will have the neutral bonded, so when the PV disconnect is open it will be an ungrounded system, is this allowed?

The output, having no load conductors connected, would simply float without an internal reference. I don't know if an unconnected source still requires bonding.

I agree the utility transformer will derive a neutral for the PV system, but will they allow this added current on their transformer knowing how they under rate them.

Isn't cogeneration design already required to consider contributed current?

Also I can see where there might be a problem if the PV supplied a neutral, as if this PV was set up to supply back to the grid there would be problems with the two X/O connections if the utility transformer was a common core?

I don't follow you. (Please understand I have no first-hand knowledge here.)

 

[hurk27]

I don't follow you. (Please understand I have no first-hand knowledge here.)

Me either, trying to learn myself

The last question was we know how supplying a 3 phase common core transformer from a Y source will produce extra heating of the transformer and over load the neutral or grounding that gets connected to the X/0 of the supplied transformer, from any unbalancing on the load side right?

Why would a 3 phase 4 wire PV system be any different feeding a common core utility transformer if it also connected to the X/0 which it would, as it will now be supplied instead of being the supplier. maybe why it's a delta?

 

[iwire]

The service wye system will act to derive the neutral as it does now, so there is no issue.

Bingo!


I have hooked up a 50 KW delta inverter to a wye service, no problem.

 

[iwire]

I agree the utility transformer will derive a neutral for the PV system, but will they allow this added current on their transformer knowing how they under rate them.

From my experience when applying for the cogeneration approval from the utility you provided enough information so that they can look at their equipment to see if this new installation may hurt it.

For instance we looking into putting about 500 KW of PV on a building that only had a 600 amp 208 volt service. That was going to be an issue with the utility and the amount of PV on that building was reduced.

 

[Smart $]

Thanks for the replies so far. The documentation for the inverter does not mention delta or wye, it just says connect the three output bus bars to the three phases of the service. There is no neutral bus. Also, if it makes any difference, there are no loads at the connection point.

...

I believe everyone here is assuming this is a utility-interactive inverter system. Is this correct?

Is there any means for standby power when utility is down?

If this is a utility-interactive inverter system, the inverter is required to disconnect (or otherwise cease power generation) automatically on utility loss. If there is a bypass for standby power, you may have a problem... but no point in getting into that if there is none.

 

[broadgage]

An inverter is inherently a single phase device, unlike say a 3 phase alternator.
A "3 phase" inverter is actually 3 single phase inverters assembled in the same enclosure, and with common control and protection circuits.
In this case it would appear that we have three inverters each for 480 volt service.
No neutral needed, such an inverter could not of course supply 277 volt loads, but this is a grid tied system with such loads supplied by the grid.

Wire the inverter like any other 3 phase 3 wire appliance with 3 hots and (probably) an EGC.

In case of grid failure the inverter will automaticly shut down.

With no connection to the service entrance neutral, I dont believe that any extra current can be imposed on the neutral conductor.

 

[ggunn]

I believe everyone here is assuming this is a utility-interactive inverter system. Is this correct?

Yes.

Is there any means for standby power when utility is down?

No.

 

[ggunn]

Thanks for all the insight. A followup question: My guy in the field has called with some concerns about it being a separately derived system and therefore being required by the NEC to connect to the neutral. I've looked at the NEC and I can't find anything to address his concerns. Can one of you code warriors speak to this?

 

[ggunn]

Followup: I talked to the guy in the field, and he told me that the concern was coming from the definition section of the Code about separately derived systems. The Code defines it as "A premises wiring system whose power is derived from a source of electric energy or equipment other than a service. Such systems have no direct electrical connection, including a solidly connected grounded circuit conductor, to supply conductors originating in another system."

Huh? The language about having "no direct electrical connection, including a solidly connected grounded circuit conductor" makes no sense to me. Should the grounding conductors be connected, or not? In our case, should the ECG of the inverter be connected to the grounded center of the Wye?

It seems to me that our inverter is not a separately derived system, anyway, because its output IS directly connected to the service. Am I missing something?

 

[Smart $]

Followup: I talked to the guy in the field, and he told me that the concern was coming from the definition section of the Code about separately derived systems. The Code defines it as "A premises wiring system whose power is derived from a source of electric energy or equipment other than a service. Such systems have no direct electrical connection, including a solidly connected grounded circuit conductor, to supply conductors originating in another system."

Huh? The language about having "no direct electrical connection, including a solidly connected grounded circuit conductor" makes no sense to me. Should the grounding conductors be connected, or not? In our case, should the ECG of the inverter be connected to the grounded center of the Wye?

It seems to me that our inverter is not a separately derived system, anyway, because its output IS directly connected to the service. Am I missing something?

Not missing anything. A grid-tied PV system is not a separately-derived system (SDS). It is directly connected to whichever system it supplies.

It could tie in to a premises SDS and still be considered grid-tied, but it still wouldn't be an SDS itself... it would be one source of a dual-source system... but the NEC doesn't use this terminology... it vaguely refers to one, or each, as an alternate source, while the system as a whole is SDS.

EDIT to add:

FWIW, the definition changes in 2011 NEC...

Separately Derived System. A premises wiring system
whose power is derived from a source of electric energy or
equipment other than a service. Such systems have no direct
connection from circuit conductors of one system to
circuit conductors of another system, other than connections
through the earth, metal enclosures, metallic raceways,
or equipment grounding conductors.

 

[ggunn]

Thanks for all the replies. Let me see if I am getting it. If, for example, you have a high voltage supply coming to a location, and you have two transformers connected to it, the outputs of those transformers are separately derived systems, even if they produce the same voltages, are they not? They are both derived from the same high voltage feed, but since no two transformers are exactly alike, it makes sense that you wouldn't want them to connect in any way other than equipment grounding.