11kv GENERATORS NEUTRAL-GROUNDING 2

[genhead]

8 X 2MW 11kv generators will be connected to an 11kv switchboard, as the sole source of supply. Is it acceptable to leave the generator neutrals open, and use a zig-zag transformer and NGR connected to the switchboard, to ground the neutral ?

 

[ScottyUK]

Normally a generator neutral point would be grounded via the primary of a relatively small single-phase transformer whose secondary winding is loaded with a resisive burden. This method allows the generator stator ground fault to be kept to a very low level, perhaps 10A or so. The low current minimises damage to the stator core in the event of a ground fault.

Is there some reason why you want to use the more complex system? It appears to offer less protection to your machine, unless you are using a very high ohmic value of NGR.





-----------------------------------

Start each new day with a smile.

Get it over with.

 

[jbartos]

Comment: Usually, 11kV level source is medium resistance system grounded since the system grounding calculations reveal higher ground charging current than useful for the high-resistance systems grounding, ~10A.
It is better, less complicated and more frequently done to ground the generator neutral over an appropriate system grounding method, probably the medium resistance system grounding method.
Alternately, the cited approach in the original posting may be implemented since it is not theoretically wrong; it just is less practical in most cases.

 

[buzzp]

I am with Scotty. We have 13.8 kV generators and all of the ones at our plants that we handle (and most company wide) are high resistance grounded via a neutral transformer.

Most cases, this is rather easy to implement.

 

[rbulsara]

I beleive the question genhead is asking if only ONE grounding tranformer/NGR installed at the switchboard in lieu of 8 separate units, one at each generator, is acceptable!

I beleive this is acceptable, but you can not 'leave' generator neutrals open, but rather need to connect it to the switchboard neutral. You should however have a means of disconnceting each generator neutral separately, when the unit is under maintenance so that it is completely isolated from rest of the system.

 

[dpc]

I agree with rbulsara that your approach is fine, provided you have a means of isolating the neutral of each generator for maintenance.

With the proper relaying, a low-resistance grounding system can provide for selective coordination for ground faults, something that is not possible with high-resistance grounding. With eight generator connected to a common bus, this may be an important consideration.

However, the low-resistance grounding does not provide the level of protection against machine damage that high-resistance grounding provides.

The IEEE has developed new recommendations for grounding of medium-voltage generators in industrial facilities. It is a hybrid system, combining high-resistance and low-resistance grounding methods. You might want to check it out:

http://www.beckwithelectric.com/powerlines/articles/i40/hybrid.htm

 

[genhead]

Thanks guys.
My understanding is this (PLEASE correct me if I'm wrong):-
With only one system grounding transformer/NGR, (instead of 8 separate units), and the generator neutrals not connected, ground fault protection is acheived via a ground-fault relay located at the neutral-grounding point. This gives protection for the 11kv system, but obviously not individual generators. If a ground fault occured, all generators would have to be tripped.
Assuming the above is correct, can individual generator ground-fault protection be achieved via the existing generator over-current relays, using their residual (out-of-balance?) current function, co-ordinated with the system ground-fault relay ?
The reason for this rather round-about approach is driven by the devilish dollar, and the need to re-use some existing equipment

 

[digitrex]

I tend to agree with genhead's idea to a zig-zag transformer to provide single-point earthing. And it is better than ground all generators' neutral. Consider these scenarios:

1. If all generators neutrals are connected to a common grounding point(whether through a NGR or not), there is a risk(or in reality, there is) of circulation of currents from one neutral to other neutrals.

2. If all generators have its own individual NGR and grounding electrodes, grounding electrodes not electrically linked to each others, then it is not economical compared to one zig-zag transformer with neutral grounding.

3. I agree with genhead's comment that all generators' ground fault relays may trip together for a 11kV busbar or system fault. Therefore it is not a good idea to have all the generators' neutrals to be grounded.

4. Without a zig-zag transformer and if one of the generators' neutrals is selected for grounding through a switch, then someone needs to remember which generator. In case this generator shuts down, he has to changeover grounding switch to another running generator.

5. With a zig-zag transformer as grounding, the generators' ground fault relays will serve as restricted earth fault relay and settings can be at minimum.

 

[jbartos]

Comment: Supposing that the system grounding calculation reveals the ground charging current 30A instead of 10A that is considered limit for the high-resistance system grounding. Then,
the size of the single phase distribution transformer "would be used for the high-resistance grounding" is:
40A x 11.8kV ~ 500kVA and associated resistor on the secondary side may have power rating ~200kW. These are large pieces of hardware that may happen to be aligned with a relatively small generator. Then, the practicality of the medium resistance grounding may prevail.

 

[rbulsara]

Besides stating the obvious that you will do well to consult (hire) some exprerience engineer(s) in such installtions, here are some thoughts:

1. There are more than one ways to skin a cat, final scheme depends on your business objectives.

2. You are correct that with one NGR the system GFP works but individual gen GFP gets tricky. For one thing you need not trip all gens upon a ground fault on the system. Because a feeder breaker (of the faulted circuit) should trip, if the fault is downstream of the gen bus. Any fault withing the generator should trip its differential protection.

2. Hopefully you will not have a fault right on the gen bus. If you really want to protect the bus, in theory a bus differentiacl protection scheme will be required. Very expensive.

3. You can set up the individual GFP of generators to coordiate with feeder breaker's GFP. This gets tricky, it depends in how many generator (minimum) will be on line at the time of the fault. For example, if the feeder GFP is set at 80A, you do not want each gen GFP to be set at 10A. The feeder breaker need to trip before any generator breaker. Assuming at least six generator will always be on line, and feeder GFP is set at 80A, I would set the gen GFP no less than say 20A (more than 80A/6).

4. I beg to differ some views that a zigzag xfmr is better than bringing neutrals. The very reason zigzag xfmrs are used is to creat a neutral when none is available. Why use a zigzag xfmrs when one is available?.

5. Circulating currents can easily be overcome by having identical generator pitch and imepedances.

6.There may be many other aspects may have to be considered.

 

[jbartos]

Comment: Whether there is high-resistance system grounding or medium-resistance system grounding, the trip on ground fault is often set for 10seconds or less at 100% insulation level. The full advantage of the high-resistance grounding is at 173% insulation level at which the fault can stay for an indefinite period of time; therefore, it can potentially be located and removed without any outage. This may mean "money in the pocket."

 

[digitrex]

I have some reservation about manufacturing all generators identical properties(pitch & impedances). If I am a client(end user) I would doubt that. Evenif identical design exist, different lengths of the neutral cables/conductors from each genset to grounding point will make the differences. If the generator neutrals are connected together first and then to a NGR, I think it will create even higher circulation currents. It will be a high risk of tripping if a ground fault relay is sensing generator's neutral current.

Situation is different if each geneartor has its own unit(step-up) transformer(delta-wye configuration). For this case every generator should be individually grounded.

 

[jbartos]

Comment on the previous posting: The generator neutrals can be tied together when there is solidly grounded neutral system grounding. If there is high-resistance system grounding, only one generator neutral to ground via high-resistance grounding is used. Other generator neutrals can be ungrounded.

 

[bhalfweeg]

Low resistance earthing is characterized by placing one or more NERs between the system neutral and earth, restricting potential E/F current to a medium magnitude in the order of 200 A, but typically no more than 1000 A.

The reasons for limiting the current by resistance earthing include:

• Eliminate destructive transient over-voltages
• Reduce arcing/burning damage to equipment during E/Fs
• Reduce mechanical stress during E/Fs
• Minimize electric shock hazard to personnel from stray E/F currents
• Reduce arc blast/flash hazard to personnel

High resistance earthing is characterized by placing one or more NERs between the system neutral and earth, restricting potential E/F current to a low magnitude in the order of 10 A, but typically no greater than 100 A. In addition to the mutual benefits of low resistance earthing, by restricting current to such low levels, high resistance earthing has the following benefits:

• Does not require immediate clearing of an E/F
• Limits damage to generator windings due to E/Fs
• Limits circulating third harmonic currents between generators to negligible levels

The primary downside of high resistance earthing is the increased difficulty of E/F detection and selectivity. Although faults are self-revealing through over-current, the small magnitude of potential E/F current make the setting of earth protection difficult:

• High impedance nature of most E/Fs (reduces available E/F current)
• Star-connected winding faults close to the neutral point within generators and transformers generate less than the maximum E/F magnitude
• Capacitive current flows in un-faulted circuits during E/Fs (makes selectivity difficult)
• Summation CTs inaccuracies and saturation during transient over-currents (can cause spurious faults)

Summation CT inaccuracies and saturation can be eliminated by using core-balance CTs rather than summation CTs. Core balance CTs can only be installed on cables, not on busbar. E/F protection is further complicated by the potential for different maximum E/F current depending on the number of neutral earthing points connected. Protection must be set on the basis of the lowest possible maximum E/F current available.

With parallel generators, the requirement is to minimise harmful third-harmonic circulating current that can occur when:

• Generators of different design are paralleled
• Generators of identical design are paralleled with unequal loading

Circulating third harmonic current can adversely affect E/F protection and generator thermal capacity.

Circulating third harmonic current can be minimised or eliminated by using either one or a combination of the following methods:

• High resistance earthing
• Switched neutrals, such that only one generator neutral is connected to earth at any one time

High resistance earthing limits the magnitude of possible third harmonic circulating current to negligible levels. Switched neutrals completely prevents circulating third harmonic currents. A combination of high resistance earthing and a switched neutral gives the benefits of high resistance earthing without introducing undesirable third harmonic currents.

Despite the aforementioned benefits, switched neutrals are considered undesirable for two reasons:

• Additional switchgear requirements
• Potential for operational errors resulting in operating the HV system unearthed

For both these reasons, switched neutrals are not the preferred solution. The requirement for switched earths will be determined by the ability of the generator to handle the proposed NER restricted circulating third harmonic current.

 

[davidbeach]

One scheme for grounding multiple generators that has not yet been mentioned is the use of three transformers in a grounded wye - broken delta configuration with the grounding resistor in the broken delta. Refer to Blackburn Protective Relaying Principles and Applications, 2nd Edition 7.5.4 beginning on page 208.

 

[digitrex]

There are 2 good things about having a zig-zag transformer:

1. Provide simple neutral grounding, in place of complicated individual neutral groundings of multiple gensets.

2. Same zig-zag transformer can supply power to auxiliary switchboards in the power plant.

 

[jbartos]

Comment: The zig-zag transformer also has disadvantages, namely:
1. It is essentially design for very short alarm, and quick trips, e.g. in less than 10seconds. This corresponds to insulation level 100%. Supposing that there is a need to trace a ground fault without any tripping, then there is a need for 173% insulation level and low frequency pulses sent into the grounded circuit.
2. It can burn, if tripping is not fast enough

 

[Rodmcm]

I agree with ScottyUK, the other advantage is that the transformer alows the path for 3rd harmonics, the loss of which can be detected. This provides '100% stator ef' protection which has a better coverage than differential

 

[jbartos]

Suggestion: It is supposed to be assessed whether or not it is better to let 3rd harmonic pass through a type of transformer or trap it in delta winding. The appropriate grounding transformer for trapping 3rd harmonics would have a delta-wye connection.

 

[Rodmcm]

Comment to previous posting.
The method of 100% stator protection is by detection of the loss of 3rd harmonics which happens under faults,hence the single phase neutral relay allows you to measure the 3rds. Another advantage of this system is that a generator earth fault can be detected when excited, before it is on line.