should be done using the minimum source impedance values for Z
A
and the
maximum source impedance values for Z
B
in order to get the maximum through
fault current from A to B.
~
~
Z
A
Z
B
Z
L
A
B
IED
I
fB
Fault
IEC09000022-1-en.vsd
IEC09000022 V1 EN-US
Figure 69:
Through fault current from A to B: I
fB
Then a fault in A has to be applied and the through fault current I
fA
has to be
calculated, Figure
. In order to get the maximum through fault current, the
minimum value for Z
B
and the maximum value for Z
A
have to be considered.
IEC09000023-1-en.vsd
~
~
Z
A
Z
B
Z
L
A
B
IED
I
fA
Fault
IEC09000023 V1 EN-US
Figure 70:
Through fault current from B to A: I
fA
The IED must not trip for any of the two through-fault currents. Hence the
minimum theoretical current setting (Imin) will be:
Imin MAX I
fA
I
fB
,
(
)
³
EQUATION78 V1 EN-US
(Equation 95)
A safety margin of 5% for the maximum protection static inaccuracy and a safety
margin of 5% for the maximum possible transient overreach have to be introduced.
An additional 20% is suggested due to the inaccuracy of the instrument
transformers under transient conditions and inaccuracy in the system data.
1MRK 505 393-UEN B
Section 8
Current protection
Line differential protection RED650 2.2 IEC
149
Application manual
Summary of Contents for RED650
Page 1: ...RELION 650 SERIES Line differential protection RED650 Version 2 2 Application manual ...
Page 2: ......
Page 22: ...16 ...
Page 32: ...26 ...
Page 82: ...76 ...
Page 112: ...106 ...
Page 152: ...146 ...
Page 208: ...202 ...
Page 214: ...208 ...
Page 282: ...276 ...
Page 356: ...350 ...
Page 404: ...398 ...
Page 408: ...402 ...
Page 442: ...436 ...
Page 452: ...446 ...
Page 453: ...447 ...