Jump to content


Photo

Relay settings for 15 MVa transformer.


  • Please log in to reply
9 replies to this topic

#1 chaterpilar

chaterpilar

    Member

  • Full Member
  • PipPipPipPip
  • 171 posts

Posted 04 November 2005 - 10:03 AM

WE have a 15 Mva 33/13.8 kv 60 Hz transformer on which the load is 7 Mva.

What should be 50/51 and 50n/51n setting on the secondary side.

Overcurrent... ? ( amps)

OC Curve..............? ( SIT or definite)

Delay...............? mS

Earth fault.......... ( Amps)

EF Curve.................

Delay.....................


The load is entirely motors and the biggest motor is 250 HP ( soft start).

The primary impedance voltage is 9.06%


thanks


chaterpilar

#2 chaterpilar

chaterpilar

    Member

  • Full Member
  • PipPipPipPip
  • 171 posts

Posted 05 November 2005 - 10:03 AM

Marke,


I am waiting for your valuable opinion.


regards,

Chaterpilar

#3 chaterpilar

chaterpilar

    Member

  • Full Member
  • PipPipPipPip
  • 171 posts

Posted 05 November 2005 - 10:03 AM

Marke,


I am waiting for your valuable opinion.


regards,

Chaterpilar

#4 RalphChristie

RalphChristie

    Intermediate Member

  • Members
  • PipPipPip
  • 30 posts
  • Location:Northern Cape, South Africa

Posted 08 November 2005 - 09:13 PM

chaterpillar:

This is a very difficult question to answer. All your settings would depend on how many levels of protection you have downstream, and how much time you are allowed upstream. It would also be dependant on the kind of relays you are using (electromechanical, electronic, etc) and what kind of protective schemes you are using on the transformer (REF, differential, etc)

Normally your residual O/C and E/F curves would be Normal Inverse Curves.
You E/F settings would depend on your transformer earthing arrangement. (solidly, high/low resistance, reactance, ungrounded, etc.)

Without your total system lay-out it would be impossible to give any setting recommendations.

Regards
Ralph

#5 jraef

jraef

    Posting Freak

  • Moderator
  • PipPipPipPipPipPip
  • 683 posts
  • Gender:Male
  • Location:USA, California

Posted 09 November 2005 - 02:40 AM

Marke is in a particularly remote part of the world just now, he does not have regular internet access.
"He's not dead, he's just pinin' for the fjords!"

#6 chaterpilar

chaterpilar

    Member

  • Full Member
  • PipPipPipPip
  • 171 posts

Posted 10 November 2005 - 08:58 AM

Thanks ralph,

The downstream equipment is protected my individual MCB's and the Feeders are protected by MCCB's.

The earthing is solid grounded.and we are using SEPAm 1000 ( Schneider electric ) protectivve system.


For a typical load of 6mva on a 15mva x-mer what should be the OC settings and Ef settings.

Currently OC is (1A) at set for SIT curve and 330 amps and 300msec delay, and (1B) definite at 9.9ka

EF is (1A)definite 40 Amps with delay of 100msec. and (1B) is 9ka with 300ms delay.

hope this info will be of help.

thanks again

chaterpilar

#7 RalphChristie

RalphChristie

    Intermediate Member

  • Members
  • PipPipPip
  • 30 posts
  • Location:Northern Cape, South Africa

Posted 10 November 2005 - 12:41 PM

chaterpillar:

You have to have coordination between your relays downstream, the transformer relay, and the relays upstream. You do not want your main breaker to trip for a fault at a downstream motor's terminals. Therefore you have to do a coordination study to ensure that your downstream relays will trip before your upstream relays. That is called protective coordination.
Thus, you have to set the transformer's secondary feeder on such a way that it does not trip before a downstream relay. If the downstream relay do not clear the fault for any reason, then the upstream relay has to operate.

Now for the relay curves:
You can use either a direct time or a IDMTL-curve (Inverse definite minimum time lag)
For a direct time: your relay will operate in the set time if the set current is exceeded.
For the IDMT-curves: The bigger the current into a relay, the faster will the relay operate. There are a lot of curves in use, but the most common are:

IEC-world:
standard/normal inverse
very inverse
extremely inverse
long-time inverse
short-time inverse

ANSI/IEEE-world
Moderately inverse
inverse
very inverse
extremely inverse
short-time inverse etc.

The IEC and ANSI-curves are not the same, but uses the same principle.

These curves have two setpoints: Current and Time
Note that it is not a specific time, but a time-multiplier.
The triptime of a relay would be dependant on the current factor (fault-current/set-current) and the time multiplier.

Download the NPAG-Book at Areva's site (free download) and read especially through chapter 9 (Overcurrent and earth fault protection):

http://www.areva-td....089880316992 p;rid=&lid=en&pid=1017999014820&tab=Chapters&id=1056536208254

(copy and paste whole address)


A few comments on some of your settings:

For a infinite bus, your max. three phase short circuit current will be almost 7kA.
Assuming you are using a standard/normal inverse curve from the IEC-world:
Your O/C setting of 330A is almost 8MVA. That is where your relay will begin to pick-up.
The time multiplier is set on 0.3 (not 300msec - it is a factor, not a time)
For a fault of 7kA your relay will trip theoretically in 0.67sec
For a fault of 1kA your relay will trip theoretically in 1.87sec
Your O/C instantaneous element will never operate, because you will never see such a high current.

For your earth-faults (remember with a solid earth your earth-fault currents can be very high)
For any earth-fault above 40A your relay will operate in 0.1second due to the fact that you use a definite time setting

Hope this will help you
Regards
Ralph

#8 RalphChristie

RalphChristie

    Intermediate Member

  • Members
  • PipPipPip
  • 30 posts
  • Location:Northern Cape, South Africa

Posted 10 November 2005 - 02:21 PM

Just an extra comment:

Overcurrent Protection is not the same as Overload Protection.

There is a general misconception that an IDMT O/C relay is there and to be set to cover overloading of a transformer. If the thermal characteristic of a transformer is plotted against the normal inverse curve, (which is on log-log graph scale) it will be seen as a straight line crossing the IDMT characteristic at some point. For small overloads the relay will trip before the transformer heats up to its limit. For sudden heavy overloads the transformer will cook before the relay trips. The NI IDMT-relay is therefore NOT suitable for overload duty - it is for fault protection. Some degree of overload protection may be obtained with these curves.

For overload monitoring of a transformer use:
1.) a relay with a suitable thermal characteristic.
2.) an EI characteristic in the IDMT range, the time being approx inverse proportional to the square of the current.
3.) Winding temp protection.

#9 chaterpilar

chaterpilar

    Member

  • Full Member
  • PipPipPipPip
  • 171 posts

Posted 11 November 2005 - 10:07 AM

Ralph,

You have aanswered all my questions.

thanks again.


Chaterpilar



" If you tell the truth, you don't have to remember your lies !"

#10 chaterpilar

chaterpilar

    Member

  • Full Member
  • PipPipPipPip
  • 171 posts

Posted 11 November 2005 - 10:07 AM

Ralph,

You have answered all my questions.

thanks again.


Chaterpilar



" If you tell the truth, you don't have to remember your lies !"




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users