I have 2 questions, 1.we have an i

[144x]

I have 2 questions,
1.we have an indoor 72.5 kV substation with cable infeeds.
the length of the cable is between 3 to 4 Km's.do we have
to use surge arresters for the line bays?

2.we want to have a restricted earth fault protection for
a transformer in a one and half scheme substation.
the upperhand and lowerhand CT's have the same ratios
but the upper hand CT'S have a knee point of 1500 V and
the neutral CT of the transformer has a Knee point of
850 V .does REF works properly?
thanks and regards.

 

[jbartos]

Suggestions:
1.
1a. More info about the cable runs and the other end transient behavior are needed
1b. Too simply: If there is a record of surges or a source of surges anticipated, then use surge arresters
2.
2a. Please, more info is needed:
e.g. what are the turn ratios?
2b. See ANSI/IEEE Std C57.13-1978 (or more recent) "IEEE Standard Requirements for Instrument Transformers"
2c.

http://www.geindustrial.com/industrialsystems/meter/notes/GET-97D.pdf

for more info

 

[144x]

1.it is a 72.5/11 KV H scheme substation with 30 MVA transformers and 12 ougoing feeders and 2 capacitor feeders
each of 2.4 MVAR capacity.
well I'm actually more concerned about how cables can have
an effect on reducing the surges.is there any rule of thumb?.any useful formula.?

2.the CT specification is 1500-1000-500/1 5P20 Vk=1500
Rct<4 Ohm for the upperhand CT's and for the neutral bushing CT of power transformer it is the same except for Knee point voltage which is 850 V.
the transformer is 125 MVA .I heard that all the
specification must match for an REF protection so the above scheme will not work is that correct?
thank you for your consideration

 

144x,

1. Yes, you need surge arresters. You've identified all the items necessary to justify them. HV cable, nearby cap banks, and transformers. Either get out your books or find an engineering firm very experienced with overvoltage protection and get them to specify the arrester ratings required. Also, you want the arresters to protect the cables and xfmr windings from reflected transient overvoltages, not the other way around as implied in your earlier message. (For good info on overvoltage protection, visit the cooperpower.com Systems Engineering web site. No I don't work for them.)

2. Generally, yes the CT knee points must be matched. But, assuming you are using high-impedance differential (REF) relaying, if the maximum through-fault current produces less than 850V at your relay, the ones you have will work. Just be sure to calculate the necessary stabilizing resistance for the relay circuit.
For low-impedance differential relaying, the checks are much the same -- calculate whether the neutral ct saturates before the phase cts on through faults.

 

[jbartos]

Suggestions:
1. Reference:
Fink D.G., Beaty H.W., &quot;Standard Handbook for Electrical Engineers,&quot; 14th Ed., McGraw-Hill, 2000
Reference 1 paragraph 15.10.17 includes useful reference covering arresters, e.g.
Sakshoug E.C., Kresge J.S., Miske S.A., &quot;A new concept in station arrester design,&quot; Power Apparatus and Systems, Vol. PAS-96, No. 2, March-April 1977
Reference 1 Section 27 &quot;Lightning and Overvoltage Protection&quot; includes 27.6 Overvoltage Protection Devices.
Potentially, if you have several subconductors per each phase line, then the subconductor insulation can partially offset surges, however, the proper way to face surges is by suitably design arrestors.
Also, visit

http://www.thomasregister.com

and type Arresters: Electric, which will return 29 Companies to choose from.
2. Notice that 850V rms x sqrt3 = 1472V rms. The normal practice is to consider line to line voltage in neutral. Therefore, 1500V rms CT would be more appropriate. However, there are some savings and obviously the increased risk in 850V rms.

 

[peterb]

Re the REF protection, I agree fully with what aggieinalgeria says. I would add the following notes, applicable where the two sets of CTs are not identical -
- Calculate the required setting voltage by the normal formula, i.e. Vs >= If(sec'y)*(wiring loop resistance from relay to furthest CT + CT secondary resistance) {Where If(sec'y) = max external earth fault current, secondary value}
- Evaluate the excitation current of both sets of CTs at the selected setting voltage
- Evaluate the scheme setting current. For a voltage based relay, such as GE type PVD or Alstom type MFAC, this will be in the order of milliamps - add the current of the non-linear resistance (Metrosil) at the setting voltage, if appropriate
- The setting current must be above the difference in excitation currents, otherwise the scheme could misoperate.

If needed, set the scheme current higher (current based relay, such as Alstom MCAG with external stabilizing resistor) or increase the effective current of a voltage based relay by installing a shunt (parallel) resistor - see the related post re vitreous enamelled resistors in this forum

In passing I will note that the knee point voltage rating of the CTs is in no way related to the CT primary circuit voltage rating, it is a function of the number of secondary turns and the flux density at saturation of the CT iron core. This is most likely a bushing CT with a rated primary voltage of the order of 600V. Of course, the transformer neutral bushing itself could be rated for either L-G voltage or L-L voltage, depending on whether the transformer winding has reduced or full insulation at the neutral end.