6.6kV system - Resistance Earthed_Current magnitude 2

[RRaghunath]

Resistance earthing of MV system is quite common. But the magnitude of allowable earth fault current seems to be a subjective issue.
I have come across 300A as a typical value.
Shell standard prescribes that the current should be equal to the transformer full load current.
Basically, the requirement is to minimise the damage at the point of fault without sacrificing the detectability. Thus, both practices seem OK.
Could experts tell me which of the practices is more common and why.

Thanks in anticipation.

 

[dpc]

This is really application-specific. If this is an industrial facility, the lower the current the better, in terms of limiting damage. If this is a utility distribution feeder, much lower resistances are generally used to make sure the fault is detectable.

Assuming an industrial facility, the main goal is protection of motors, generators and transformer cores. The lower the current, the less chance of damage to the core steel and repairs are limited to replacing the winding. If the core is damaged, the repairs are a much bigger problem and much more expensive. But if the current is too low, there is not enough current to allow selective coordination using inverse time overcurrent relays. At the extreme, high-resistance grounding is used, and selective coordination is sacrificed.

With modern digital relays, ground current can often be limited to 100 A or even 50 A and still allow two to three layers of selective coordination.

 

[rbulsara]

Refer to IEEE Green book for various practices.

Short of that the Shell guidelines as you described seems most appropriate for application at 6.6kV voltage level, which will meet your basic intent as you correctly described.

The 300A (200A-400A) is generally for large utility grade generating plants and distribution systems at higher voltages, where rated or fault currents can be much greater than those. They also need larger current to allow coordinate with many levels below the source equipment. You may not have that many downstream devices to coordinate.

 

[slavag]

Hi.
STARS to Dave and RBulsara.
I think for the 6.6kV a Dave's:
"With modern digital relays, ground current can often be limited to 100 A or even 50 A and still allow two to three layers of selective coordination." is very applicable today.
I would like add:
1. With toroid CT.
2. With blocking scheme as option.
Best Regards.
Slava

 

[TestBeforeTouch]

The 6.6kv systems at my plant are limited to 10A. The Multilin 469 with the 50:.025 zero sequence CT does fine at tripping at only 2-3 amps.

 

[REM76]

300 A sounds appropriate. The CT usually sized at 50 % of the resistor rated amps i.e. 150:5

TestBeforeTouch: The resistor amp value should be higher than the capacitive charging current of the system Ic. Otherwise, you will end up with escalating voltages to ground during ground faults. If Ic is higher than 10 Amps, then you will need to go with low resistance grounding scheme

 

[TestBeforeTouch]

Rem76: I understand the issue of capacitive charging, however, the very large plant I work at has been using 10 amp high resistance grounding at 13.8KV and 6.9KV for over 40 years. The plant has had many faults over the years and is alarm only on a ground fault. In this time there has not been any failures due to escalting voltages. I believe that by putting the ground resistor in that it limits the overvoltage to a moderate value and allows an orderly shutdown of production if necessary.

 

[lansford]

When I worked in the strip mining industry, we were required by MSHA (gov regulatory agency) to use 25 amp resistance grounding for the portable and mobile Medium Voltage electric shovels and draglines). JIM

 

[REM76]

TestBeforeTouch: I agree with you and I believe the 10 amps value that you have for your resistors is greater than the charging current of the system, and that is why you are not experiencing voltage escalation during ground faults.