Neutral Between Delta Windings

[That Man]

Mike Holt Community,

One of the senior electricians at my company called and asked me a question about a solar install.

First some background of the install: Between the inverter at the solar array and the loads being fed, there is a pair of step up and step down transformers. 480Y to 4160 Delta step up, 4160 Delta to 480Y step down.

The question: Where do you land the neutral between the transformers?

Simple Answer: I told him that there is no neutral between delta winding and delta winding.

His concern arises from the fact that he's been on several sites with the same scenario, and there has been neutrals present.

Because I don't know what I don't know, and I don't have much experience with solar installations, I'd like to know if anyone out there knows of some other reason to run a neutral in the scenario I described.

 

[topgone]

Mike Holt Community,

One of the senior electricians at my company called and asked me a question about a solar install.

First some background of the install: Between the inverter at the solar array and the loads being fed, there is a pair of step up and step down transformers. 480Y to 4160 Delta step up, 4160 Delta to 480Y step down.

The question: Where do you land the neutral between the transformers?

Simple Answer: I told him that there is no neutral between delta winding and delta winding.

His concern arises from the fact that he's been on several sites with the same scenario, and there has been neutrals present.

Because I don't know what I don't know, and I don't have much experience with solar installations, I'd like to know if anyone out there knows of some other reason to run a neutral in the scenario I described.

A simple sketch of your setup can help.

 

[Jraef]

Just speculation:
Probably not really a neutral, but they likely had one leg of the delta winding grounded (corner grounded delta) because if not, they would have needed to add ground fault monitoring to that circuit, and at 4160V, that would have been expensive. Corner grounded delta is cheap.

 

[That Man]

A simple sketch of your setup can help.

I received this question over the phone, so all I know is what I wrote. I suspect one of the following:

1) the previous installations observed were not T1 delta secondary to T2 delta primary

2) this installation is not T1 delta secondary to T2 delta primary

3) there is no need for a neutral, but an extra wire was pulled anyway

4) there is a special reason for solar installations that I don't know about

It's scenario 4 that I'm concerned about.

 

[That Man]

Just speculation:
Probably not really a neutral, but they likely had one leg of the delta winding grounded (corner grounded delta) because if not, they would have needed to add ground fault monitoring to that circuit, and at 4160V, that would have been expensive. Corner grounded delta is cheap.

That's some good speculation.

This would qualify as an ungrounded system, and there are requirements for ground fault protection in those scenarios. Thank you, I'm going to look into the ungrounded system portion of the code and see if there are any special rules for the MV scenario. But I speculate that you identified the issue here, and that this is a safeguard to force trip the transformer in the event of a ground fault, instead of adding a ground fault detector.

 

[topgone]

That's some good speculation.

This would qualify as an ungrounded system, and there are requirements for ground fault protection in those scenarios. Thank you, I'm going to look into the ungrounded system portion of the code and see if there are any special rules for the MV scenario. But I speculate that you identified the issue here, and that this is a safeguard to force trip the transformer in the event of a ground fault, instead of adding a ground fault detector.

IMO, it's better that you leave the MV side "ungrounded". If you have to providing ground fault protection at this level, there are a lot of ground fault protection schemes available at 4160V. E.g. install grounding PTs with secondaries in "broken-delta" configuration.

 

[Electric-Light]

I would guess the center tap of 4.2kV "base" of the utility side through a high resistance grounding resistor appropriate for HRG and through the current relay coil that is set to trip at a reasonable value.

Thread the NC output of the relay with the solenoid of a motor starter like switch.

The HRG pulls down the CT to ground and dissipates capacitance so the cable insulation only sees 2.1 and 3.6kV against conduit and reduce RFI created by the inverter. If it somehow manage to develop a second fault, current returns through the HRG, pulls the release, separate NC and drops the primary.
Anti icelanding should knock off the static converter off line then.

The "motor starter" like switch has to be manually reset after clearing fault.

 

[GoldDigger]

I usually don't comment on spelling, but it looks like autocorrect has turned "anti-islanding" into "anti Icelanding"

mobile

 

[NYSolar]

Could the neutral be for a production meter?

 

[Bugman1400]

If the 480Y of the step up is for the inverter side then it may not be grounded or connected. You need to read the inverter manual. Connecting the 480Y to ground or to another neutral may interfere with the ground detection system of the inverter. I believe SMA inverters require that the neutral float. Not sure about the GEs.

 

[mbrooke]

I need more info, it could be anything.

 

[electrofelon]

Mike Holt Community,

One of the senior electricians at my company called and asked me a question about a solar install.

First some background of the install: Between the inverter at the solar array and the loads being fed, there is a pair of step up and step down transformers. 480Y to 4160 Delta step up, 4160 Delta to 480Y step down.

The question: Where do you land the neutral between the transformers?

Simple Answer: I told him that there is no neutral between delta winding and delta winding.

His concern arises from the fact that he's been on several sites with the same scenario, and there has been neutrals present.

Because I don't know what I don't know, and I don't have much experience with solar installations, I'd like to know if anyone out there knows of some other reason to run a neutral in the scenario I described.

I suspect some error in terminology. If he said, "where is the neutral" than I would say there is no neutral between two delta windings. If he said, "where do you land the neutral" (which is the way you relayed it) this implies there is indeed some conductor there that he is calling a neutral. Jraef made a good call - its common jargon for people to call grounded conductors neutrals. Or, he could be referring to the MV cable shields which are used as the neutrals on WYE systems (usually). This being a delta system, the shields would just be bonded to the transformer cases and GES at each end.

I am curious: do you provide or have any thought about transformer protection per 450.3?

 

[big john]

Be careful when deciding to ground these deltas:

A box of donuts says this was planned with 133% 5kV cable. That will run fine ungrounded but is only designed to withstand line voltage for 1 hour.

If you corner ground one phase you are putting the remaining two at full line voltage to ground. Your cables would need to be 173% rated.

 

[ggunn]

If the inverter bank needs a neutral, why wouldn't you have a wye configuration on the inverter side of the transformer with the neutral grounded and an ungrounded delta on the MV side?

 

[GoldDigger]

If the inverter bank needs a neutral, why wouldn't you have a wye configuration on the inverter side of the transformer with the neutral grounded and an ungrounded delta on the MV side?

In which case you would need a local GES to connect the neutral to and possibly an EGC or very long bonding jumper connecting that GES back to the GES at the other end of the run?
Still not a neutral, but yet another form of conductor connected to ground.

 

[Electric-Light]

I think that's what he's in mind. Not where to land the neutral but rather where to ground.

That's some good speculation.
This would qualify as an ungrounded system, and there are requirements for ground fault protection in those scenarios. Thank you, I'm going to look into the ungrounded system portion of the code and see if there are any special rules for the MV scenario. But I speculate that you identified the issue here, and that this is a safeguard to force trip the transformer in the event of a ground fault, instead of adding a ground fault detector.

Ungrounded system won't trip until you have a phase to phase fault.
You can't force it to trip without detecting it.