ABB REL670 2.2 IEC, Руководство по применению

Страница: 553 / 908

the permissible time as a function of the level of over-excitation. When the
transformer is loaded, the induced voltage and hence the flux density in the core
can not be read off directly from the transformer terminal voltage. Normally, the
leakage reactance of each separate winding is not known and the flux density in the
transformer core can then not be calculated. In two-winding transformers, the low
voltage winding is normally located close to the core and the voltage across this
winding reflects the flux density in the core. However, depending on the design,
the flux flowing in the yoke may be critical for the ability of the transformer to
handle excess flux.
The Overexcitation protection (OEXPVPH) has current inputs to allow calculation
of the load influence on the induced voltage. This gives a more exact measurement
of the magnetizing flow. For power transformers with unidirectional load flow, the
voltage to OEXPVPH should therefore be taken from the feeder side.
Heat accumulated in critical parts during a period of overexcitation will be reduced
gradually when the excitation returns to the normal value. If a new period of
overexcitation occurs after a short time interval, the heating will start from a higher
level, therefore, OEXPVPH must have thermal memory. A fixed cooling time
constant is settable within a wide range.
The general experience is that the overexcitation characteristics for a number of
power transformers are not in accordance with standard inverse time curves. In
order to make optimal settings possible, a transformer adapted characteristic is
available in the IED. The operate characteristic of the protection function can be set
to correspond quite well with any characteristic by setting the operate time for six
different figures of overexcitation in the range from 100% to 180% of rated V/Hz.
When configured to a single phase-to-phase voltage input, a corresponding phase-
to-phase current is calculated which has the same phase angle relative the phase-to-
phase voltage as the phase currents have relative the phase voltages in a
symmetrical system. The function should preferably be configured to use a three-
phase voltage input if available. It then uses the positive sequence quantities of
voltages and currents.
Analog measurements shall not be taken from any winding where a
load tap changer is located.
Some different connection alternatives are shown in figure 289 .
1MRK 506 369-UEN B
Section 10
Voltage protection
Line distance protection REL670 2.2 IEC 547
Application manual