Converting MV generators from solid ground to LRG

[rockman7892]

We are looking to convert (4) 4.16kV 1650kW diesel generators from having solidly grounded neutrals to having LRG neutrals. All four generators are currently connected to a paralleling switchgear lineup with each generator neutral brought to a common ground bus in the switchgear where it is solidly grounded.

In determining the neutral resistor size and ground current limit are there any particular approach or rules of thumb? I have modeled both 100A and 200A values for each generator and ran fault cases with one gen running (minimum case) and all four running (max case) in order to evaluate system relay response and coordination. Limiting each generator to 100A provides enough fault current to get the relay response needed for system coordination. The normal system which the downstream equipment is connected to is limited to 400A (I do not believe Gens cannot parallel with normal system)

Is there anything else to consider with NGR values other than relay response and coordination?

Also is using (4) separate NGR's the preferred method as opposed to having a single common NGR with each generator neutral connected to it? Paralleling gear is located several hundred feed from generators so individual NGR's located at the generator may be a more difficult installation approach unless for connecting to gen neutrals and running CT wiring. Another option would be to locate individual NGR's at paralleling gear location and pull generator neutrals to these NGR's at that location which would not require penetrating existing generator enclosures or trenching for new CT wiring.

 

[RRaghunath]

It can be one resistor switched to one of the running generator neutrals at any given point of time.
Typically, 100A rating is recommended for industrial generators.
In your case, since you are converting exiting solidly earthed system to LRG system, you need to see the protection aspect if you do not have Core balance or zero sequence CTs (CBCTs) in outgoing feeders.

Rompicherla Raghunath

 

[xnuke]

It can simply be one resistor, and I think the switching of it has fallen out of favor with most people since you may be changing the running machines and would then have to switch the neutrals.

Also, everything connected to the LRG generators , e.g., VTs, SPDs, etc., must be rated for line-to-line voltage. 100% rated conductors are okay as long relaying relaying trips on a ground fault. 133% rated conductors are required if you want up to an hour before tripping. Is your desire to change to resistance grounding to increase up-time or to reduce chance of damage to generators on a ground fault, or reduce arc flash current, or reduce circulating currents, or a combination of these? The answer affects how and why you ground your generators.

I recommend you talk to a grounding resistor manufacturer representative. They help people with this sort of thing regularly.

xnuke
"Live and act within the limit of your knowledge and keep expanding it to the limit of your life." Ayn Rand, Atlas Shrugged.
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[GroovyGuy]

Howdy Rockman,
100A should be just fine. I would also consider 50A as well.
Have you considered a single-ground resistor connected via a zig-zag transformer to the main bus?
I trust that your entire downstream protection (on the 5kV side) is via CBs, and not fuses, correct? (ie to maintain coordination , you will need 51G relays on every CB. 51G's cannot be fitted to 5kv LISw's).

GG

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

 

[rockman7892]

xnuke - The desire change to resistance grounding is primarily to reduce damage to generators and large motors from ground faults occurring when system is operating from emergency generators. Also the normal source is LGR at 400A so currently downstream system loads operate from LRG system under normal conditions and then solidly grounded system during emergency conditions which throws off ground fault relaying and coordination.

Is a single resistor with disconnect switches for the generators still an acceptable practice? Can the existing generators neutrals which are terminated on the paralleling switchgear ground bus simply be de-terminated from the ground bus and connected to disconnect switches which are then in turn connected to a single 100A NGR? This would certainly be easier than trying to install an NGR at each generator and penetrate the generator enclosures to connect to generator neutrals.

GroovyGuy - Can you elaborate further with the single ground resistor connected via zig-zag transformer to main bus? I'm not sure I understand why a zig-zag transformer would be needed here.

Yes all downstream protection is via CB's with 51G relays and zero sequence CT's. There is only one 3000:5 residual CT on a main breaker downstream that will not see 100A ground fault current however the feeder breaker feeding this main has 50:5 CT and thus will see the 100A and coordinate upstream.

 

[MatthewDB]

The zig-zag transformer acts as a grounding transformer. It gives you another neutral independent of the generators. Another common option is to use a wye-delta transformer. With wye-delta, the resistor can be either from the transformer wye to earth or the wye is solidly grounded and the delta is broken by a resistor. The advantage to this method is a LV resistor can be used.

I would see the biggest downside of the grounding transformer is that it becomes a single point of failure.

Wikipedia: Grounding Transformer

 

[MatthewDB]

I have been in several large hydropower stations where they used a single grounding resistor per machine group. (A group of several main generators, a station service generator and a step up transformer all on one bus, repeated several times.) All of them were USACE so I suspect this is a "standard" design for them.

All of the generators would normally be connected to the neutral. The neutral then had a single resistor to ground. Each generator had hotstick operated disconnects inside a locked cabinet for disconnecting the neutral when the generator was undergoing maintenance. The phases were connected / disconnected by each generator's circuit breaker, but the neutral remained connected except when manually disconnected.

It was memorable for me because I was going through the LOTO checklist for a generator, and one of the items was to hang another lock on the neutral switch blade, on a few feet from the neutral bar which interconnected to several operating 13.8kV generators.

 

[waross]

That sounds like a good plan MatthewDB.
If I was designing new, I would add a Kirk type interlock so that the generator could not be placed in service when the neutral switch was open.
The Kirk interlock would also prevent opening the neutral when the generator was in service.

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

 

[MatthewDB]

I honestly don't know if they used a kirk key interlock or used procedural checklists before opening the cabinet. I was a 2nd tier subcontractor so I wasn't familiar with any of the operating procedures.

Making it a kirk key with the generator breaker inter-locked would be a really good idea. I just wasn't a fan of the design of the switch. The blade of the switch prevented the door from being closed when it was in the ground position. That meant the neutral jaw was exposed at the top of the cabinet. If the door could have been closed again, LOTO locks could just go on the door after verifying the switch was down through a viewing window.

Then again, making it impossible to close the door makes it obvious the switch was in a position unsuitable for operating the generator.