There are no pro's for the relay .
Depending on the type of CVT , the transient introduced by the fault has varying effects.
A CVT with a passive damping circuit is normally not much of a problem for numerical relays as the transient is a decaying d.c with RC time constant of the circuit.
But a CVT with an active damping circuit will produce transients which are nearly the same frequency as the fundamental and very hard to filter , might as well say impossible to filter out. So different approaches have to be adopted for that.
Aakhrinaam is correct in the comment on active versus passive ferroresonance protection circuitry. In North America, passive type circuits are generally the only accepted designs.
The issue with distance relaying is called the transient response (TR) of the CCVT, which is the delay in secondary voltage changes, cooresponding to changes in the primary voltage.
One note to consider...TR of a CCVT is a function of the applied burden. The lower the applied burden, the better the TR. Since modern electronic relays and meters have a very high input impedance (i.e. low burden), the TR of CCVTs in modern applications is quite good.
Even though there are significant progress modern CVT’s design and performances of static relays, still there is valid concerns regarding unwanted noise produced by CVT that could cause malfunction or slow down relay tripping decision during transient conditions, particularly in systems with high source impedance ratio.
The apparent impedance determined by numerical relays may be smaller than the actual impedance do to the reduction of the fundamental component of the fault voltage by the CVT internal components (capacitance, burden and ferroresonance suppression circuit elements).
Relay manufacturers offering a variety of filtering schemes to mitigate this potential problem. However, the filtering logic adds time delays to the relay tripping function.
The enclose papers have good discussion and recommendations about the effect of CVT on distance relays during transient conditions.
Suggestion: Reference:
C. Russell Mason, "The Art and Science of Protective Relaying," John Wiley & Sons, 1956,
Section "Comparison of Instrument Potential Transformers and Capacitance Potential Devices" on page 144
I take some serious points with both of the papers you posted (which I am familiar with).
The GE paper (I believe) has some wrong/misleading statements:
1) It says CVTs perform better (transient-wise)when the applied burden is higher, which is completely backwards. They perform better as the applied burden is lower.
2) Is lists 2 types of ferroresonance suppression circuit types, active and passive. The 2 main CCVT suppliers in North America (Ritz and Trench) both use passive circuits, which perform much better than active circuits. Most of the info and simulations they show arew based on active supression devices, which are generally not accepted in the US and Canada.
3) The values they are using for the stack capacitances are way off for 500 kV applications. They are correct in saying that higher capacitances CAN lead to better TR performance, but they leave out the fact that the tap point (tap voltage) makes a big difference too!
The SEL paper is better, but it fails to consider the affect of actual applied burdens on the TR performance of CCVTs.
jbartos-
The GE link for the VTs you posted seems well outdated...still referencing the Class A type CCVTs, which is an old term from the old CCVT NEMA standard. Modern CCVTs are not rated like that any more and there have been many strides in the design in the last few decades.
