Neutral required on 480 VAC system 1

[sparkview]

Hi all,

We are currently installing a 1 MW Battery Energy Storage System which the inverters consists of 3 phase+ Neutral 480 VAC. This system is to be connected on a 3 phase 480 VAC system which doesn't use Neutral. The supplier thought that the Neutral of their system could be connected to the ground of the power plant. However this is not a normal practice. The main Transformer of the switchgear at the power plant us a WYE system where the neutral is connected through a High resistance grounding equipment and afterwards to earth. Could this problem be resolved with an Delta- WYE isolation transformer? Attached is a sketch of the system mentioned above.

In advance many thanks for your cooperation and support.

Jairo

 

[waross]

Some codes require the neutral to be solidly grounded when there are any line to neutral loads served.
This is a grey area as the inverter is not really a load.

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

 

[buzzp]

I would argue that an inverter is a load since it consumes electrical power. Yes it provides power too but only after consuming it (not a power generator). There is probably some room to argue otherwise. If it is a 'load', then a high impedance grounded system can't serve any L-N loads. This is per the NEC.

 

[GroovyGuy]

Howdy Ya'll,
I think it goes beyond saying; if you have a NGR, then you cannot have any L-G loads. There that's taken care of.
I'd also be surprised of the inverter OEM would be terrible pleased with a NGR at all. I suspect that they would be happier with a solidly grounded system. These guys don't get out much.
GG

"I have not failed. I've just found 10,000 ways that won't work." Thomas Alva Edison (1847-1931)

 

[davidbeach]

I'm pretty sure that in NEC land any connection to the inverter's N terminal would be a code violation on an NGR system and not connecting could also be a code violation if the installation instructions say to connect the neutral and don't address an NGR system. The isolation transformer or the removal of the NGR seem to be the two alternatives.

 

[JensenDrive]

1) NEC Violation
The specific section that applies is NEC 250.36. This article says high-impedance grounded neutral systems cannot have any line to neutral loads, and I would well argue that this inverter would be considered a line to neutral load (think of it as a negative impedance load). Another concern is if the neutral has a ground reference internally, this could be construed as a second ground reference where NEC says the neutral can be grounded only through the impedance.

2) Using Isolation Xfmr to Reduce Harmonics
2a) Three phase inverters tend to carry a lot of triplen (3rd, 9th, 15th...) harmonics over the neutral path, and in your 1200A application, I might expect in the 10s of amps of triplen harmonics flowing in the system neutral. An isolation xfmr would keep the triplen harmonics out of the system and also reduce other harmonics in the inverter output. Ideally, you would use specialized xfmrs for inverter applications that are derated for harmonic loading and might have heavier insulation.
2b) It is mentioned that there are 2x500kVA inverters, so this tells me you should use 2 independent isolation xfmrs. Ideally, each inverter would be on its own isolation xfmr. This allows you get into custom xfmrs with custom phase shifts (phase shifts other than +/- 30 degrees are available) for each xfmr. This allows very improved harmonic reduction. If for instance the 7th harmonic is shifted +15 degrees on one xfmr, and +30 degrees on the other, the harmonics cancel out and the 7th harmonic is not seen on the output. (I am making up this example; I do not know the correct angular differences to create the best harmonic cancellation.) The theory of the right phase shift and isolation xfmr design is something that engineers for your supplier should be able to assist with.

 

[GroovyGuy]

Howdy Jensen,
I suspect that the inverter(s) in question here have active front-ends, therefore harmonics should not be an issue.
GG


"I have not failed. I've just found 10,000 ways that won't work." Thomas Alva Edison (1847-1931)