51G protection 1

[tulum]

Hello,

I am working in a facility trying to refurbish a fan/motor control center. I noticed that regardless of the load(50A,150A,225A,400A) the ct selected has been 200:5A for the 51G protection for each feeder. I was told this is standard practice.

What are the pit-falls in doing so (if any)? My concern is the ct primary being smaller than the FLC for the application.

Regards,
TULUM

 

[dpc]

I assume the CT goes around all three phases - a flux summation or zero sequence CT.

The load current is not really an issue, since the CT will only see the imbalanced current resulting from a line-to-ground fault.

The CT ratio depends on several factors including the type of system grounding (solid or resistance) and the type of relay being used.

In most cases, there is nothing downstream of this relay to coordinate with, so the ground relay can be quite sensitive. In fact, many people use a 50G relay - instantaneous trip with 50/5 zero sequence CT.

One concern can be saturation of a low ratio CT on solidly-grounded system. If you are trying to selectively coordinate with something downstream this could be an issue. When an instantaneous relay is used, the thinking is that the relay will still operate even if the CT saturates.

 

[GOTWW]

The NEC says that for feeders 800A and above need GFP. It does not require it be designed well.

 

[tulum]

Thanks folks.

Impedance grounded.. and end of line (no downstream protection to coordinate with)

I think I need to "edumacate" myself a bit more on the subject.

As always, insightful and appreciated.

TULUM

 

[stevenal]

Regarding edumacakeshun, Check out Stan Zocholl's papers at:

http://www.selinc.com/techpprs.htm

Good stuff on CTs and motors.

 

[dpc]

tulum,

With low impedance grounding, the 200:5 CT should be no problem - the only problem might making sure you have enough ground fault current to detect and relay reliably.

If it's high impedance grounding, the fault current will probably be too low for this setup to work reliably.

 

[tulum]

Dpc,

Thanks again. The NGR is a 10A, 347V resistor (on the secondary of a 1500kva 4160v-600v dy transformer)

Regards,
TULUM

 

[busbar]

 
One paper on the subject is: John P. Nelson, Pankaj K. Sen, High-Resistance Grounding of Low-Voltage Systems: A Standard for the Petroleum and Chemical Industry, IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 35, NO. 4, JULY/AUGUST 1999 at

http://www.neiengineering.com/papers/paper1JN.pdf

 

[dpc]

Well, on 10A of ground fault current, the 200/5 CT would only be putting out .25A of secondary current. You will probably need some other method for detecting ground faults, such as a zero sequence voltage detection scheme.

Is the resistor rated for 10A continuously?

 

[tulum]

DPC,

The resistor is continuosly rated. The protective device is a startco SE-105 (on feeders) and an se-325 on the NGR itself. I wish I could add more... I installed and worked on many of these, but I am not to familiar with the engineering behind them.

I understand the basics behind the zero sequence current ct (this is what the feeder ct's are), but what is a zero sequence voltage detection scheme?

Regards,
TULUM

 

[busbar]

 
A limitation of zero-sequence overcurrent in low-voltage, high-resistance-grounded systems is that it can be hard to coordinate with various levels of overcurrent protection—given the comparatively low currents during faults. Negative-sequence voltage, instead of a CT in the grounding-resistor lead, typcially uses a ø-ø rated VT in parallel with the resistor.

The VT serves a low-pickup, high-continuous-withstand voltage relay, often restrained for third-harmonic voltage. However sensitive, it is more of a non-directional “all-or-none” method of fault detection or indication.

 

[busbar]

 
Negative-sequence overvoltage relays are typically referred to as ‘59G’ or ‘59N’ devices. Single-function examples with inverse-time characteristics are Baser BE1-59N [www.basler.com/html/pcs59-87.htm#BE1-59N] or General Electric IAV-51D [www.geindustrial.com/products/brochures/iav.pdf]

 

[davidbeach]

Busbar - 59N and 59G are zero-sequence, not negative sequence. Negative-sequence would be 59Q.

 

[busbar]

Thank you, David. I goofed.

 

[dpc]

tulum,

The zero-sequence voltage detection is just three PTs - grounded wye on the primary and broken delta on the secondary. You put a voltage relay (in parallel with a resistor to prevent ferro-resonance) across one corner of the secondary delta. The voltage that shows up there is 3Vo, so it is good indicator of a ground fault. There can be some harmonic voltages present, so the 59G relay often is equipped with some filtering.

Blackburn's Protective Relaying or the old Westinghouse Applied Protective Relaying books have better explanations.

 

[tulum]

Much thanks to all.

Regards,
TULUM