Unit connected industrial generators.

[electrix]

With several unit connected generators synchronised on to a bus through DY transformers, should just one transformer neutral be grounded or all transformer neutrals?The bus feeds several uniground wood pole distribution feeders. What are the pros and cons of both arrangements? Any reference material would be appreciated.Thanks.

 

[peebee]

What voltage are you talking about? Not that it'd make too much difference.

On a 3-wire system, I think you'd want to bond all of them, in fact I think you'd be required to by code. On a 4-wire system, where each tx had a neutral conductor bonded to a neutral bus bar, you could provide a single bond.

Perhaps consider bonding them using neutral-bonding switches. This will facilitate maintenance, avoiding dealing with an energized neutral.

The IEEE Green book has some information on this, although even after reading that you'll likely still be a bit confused. I think that's all covered in Chapter 1, which is available for free download from IEEE.

I don't think the generators are relevant to your question, meaning that you'd treat the transformer bonding the same as if they were supplied from a utility source.

 

[busbar]

The IEEE C62.92.* texts offer a decent perspective. C37.101 and C37.102 cover important protection aspects.

“Hardcopy versions of published non-Government standards… may be obtained from Global Engineering Documents, 15 Inverness Way East, Englewood, Colorado, 80112, 800.854.7179, Fax 303.397,2740,

http://global.ihs.com.

Electronic delivery of selected documents is available through ANSI at

http://webstore.ansi.org

Copies of… draft standards and the procedure for commenting on them may be obtained by contacting the standards developing organization… Additional information on ANSI activities and available non-Government standards can be found… through the National Standards System Network,

http://www.nssn.org

”

 

[peebee]

I am not familiar with C62.92, so take what I say here with a grain of salt. After looking at the table of contents on the IEEE website, it does look like it could be a great reference for this, and will likely provide a lot of information on the theory you are looking for. However, be careful of the following putting it into practice:

1. It is a standard for utilities, and therefore may not meet NEC requirements.

2. Utilities don't make much of a distinction between neutral and ground.

3. Utilities commonly use the planet for neutral and/or ground return paths, rather than installing a neutral or ground conductor. This would not only be against NEC, but could be a safety hazard.

Again, I have no idea what this text says, but it wouldn't surprise me if it said it was OK to install separate ground rods for each transformer, completely isolated from each other except for the conductance of the dirt between them. You probably couldn't get away with that, nor would you want too.

 

[jwerthman]

In my experience, all of the transformer neutrals would be solidly grounded. The only exception would be if the short circuit duty for ground faults was excessive, in which case some neutrals might be left ungrounded. The ground fault levels for faults close to the transformers will be higher than the 3-phase faults due to the delta-wye transformers. So it is possible that the breakers could be OK for 3-phase but overdutied for single-line-to-ground faults.

If you are going to leave the transformer neutrals ungrounded, you need to make sure that the neutral insulation is adequate. At higher voltages, it is quite common to used graded insulation for transformers that will be solidly grounded. If your transformers have graded insulation at the neutral, they must be grounded.

 

[electrix]

Thanks for the input so far guys. I have a copy of C62.92 and it is silent on the issue of multiple unit connected generation systems.So too is the Westinghouse T&D book. For clarification the bus is 11kv Y and the transformers neutrals are connected to the station ground grid through individual switches. The distribution is 3 wire uniground so that ground faults currents travel through the ground.

 

[peebee]

Sounds like a European system. In the US, you could do medium voltage any way you want, it could be a grounded, ungrounded, or impedance grounded system; I suspect other countries would handle this the same way. If you go with anything but a solidly grounded system, your insulation should be rated at 133% to handle overvoltages during fault conditions.

If you're going to ground the system (solid or impedance), it would probably make sense to make a single neutral-ground bond at your substation bus. If you have a 4-pole tie breaker or neutral bus link, you should provide two bonds, one on each side of the tie or link.

 

[busbar]

To be sure we’re on the same page, reference on this end is IEEE C62.92.2-1989 (R1993, 2001) Guide for the Application of Neutral Grounding in Electrical Utility Systems, Part II-Grounding of Synchronous Generator Systems and seem to deal exclusively with medium-voltage systems. I agree that the text can be considered to be the IEEE “party line” of sorts. It is authored by a committee through consensus of attending utility engineers, but for MV applications, there is no reason why an industrial application would be significantly different. Section 3—Generator Grounding Types—seems like it adequately parallels variations of the OP’s described system, but with GSU transformer higsides on a commoon bus. Tradeoffs considered, wouldn’t a common connection at all H0 wyepoint bushings to the yard ground grid be in order, except in the one case where zero-sequence fault-current contribution exceeded the switchgear rating? That would be the major determining factor in this case, right?

The poleline section is of no consequence, for fault magnitudes from earth return do not appreciably change regardless of the state of HO connections.

{As for utility-versus-industrial configuration and operating differences, for decentralized groups of low-voltage systems, agreed utilities treat the situation somewhat differently than industrials. There will be forseeable variation in understanding of how a localized low-voltage system should be grounded, but it’s a mistake to base lesser differences on the opinions of a line crew compared to that of an electrical contractor or inside wireman. One should not consequentially extend that admittedly quite different thinking mode to other electrical systems.}