Power fuses have been used for many years to provide transformer fault protection. Generally it is recommended that transformers sized larger than 10 MVA be protected with more sensitive devices such as the differential relay discussed later in this section.
Fuses provide a low maintenance, economical solution for protection. Protection and control devices, circuit breakers, and station batteries are not required.
There are some drawbacks. Fuses provide limited protection for some internal transformer faults. A fuse is also a single phase device. Certain system faults may only operate one fuse. This will result in single phase service to connected three phase customers.
Fuse selection criteria include: adequate interrupting capability, calculating load currents during peak and emergency conditions, performing coordination studies that include source and low side protection equipment, and expected transformer size and winding configuration (ANSI=IEEE, 1985).
Overcurrent relays generally provide the same level of protection as power fuses. Higher sensitivity and fault clearing times can be achieved in some instances by using an overcurrent relay connected to measure residual current.
This application allows pick up settings to be lower than expected maximum load current. It is also possible to apply an instantaneous overcurrent relay set to respond only to faults within the first 75% of the transformer.
This solution, for which careful fault current calculations are needed, does not require coordination with low side protective devices. Overcurrent relays do not have the same maintenance and cost advantages found with power fuses.
Protection and control devices, circuit breakers, and station batteries are required. The overcurrent relays are a small part of the total cost and when this alternative is chosen, differential relays are generally added to enhance transformer protection. In this instance, the overcurrent relays will provide backup protection for the differentials.
The most widely accepted device for transformer protection is called a restrained differential relay. This relay compares current values flowing into and out of the transformer windings.
To assure protection under varying conditions, the main protection element has a multislope restrained characteristic. The initial slope ensures sensitivity for internal faults while allowing for up to 15% mismatch when the power transformer is at the limit of its tap range (if supplied with a load tap changer). At currents above rated transformer capacity, extra errors may be gradually introduced as a result of CT saturation.
However, misoperation of the differential element is possible during transformer energization. High inrush currents may occur, depending on the point on wave of switching as well as the magnetic state of the transformer core. Since the inrush current flows only in the energized winding, differential current results.
The use of traditional second harmonic restraint to block the relay during inrush conditions may result in a significant slowing of the relay during heavy internal faults due to the possible presence of second harmonics as a result of saturation of the line current transformers.
To overcome this, some relays use a waveform recognition technique to detect the inrush condition. The differential current waveform associated with magnetizing inrush is characterized by a period of each cycle where its magnitude is very small.
By measuring the time of this period of low current, an inrush condition can be identified. The detection of inrush current in the differential current is used to inhibit that phase of the low set restrained differential algorithm.
Overexcitation can also be caused by an increase in system voltage or a reduction in frequency. It follows, therefore, that transformers can withstand an increase in voltage with a corresponding increase in frequency but not an increase in voltage with a decrease in frequency.
Operation cannot be sustained when the ratio of voltage to frequency exceeds more than a small amount. Protection against overflux conditions does not require high-speed tripping. In fact, instantaneous tripping is undesirable, as it would cause tripping for transient system disturbances, which are not damaging to the transformer.
An alarm is triggered at a lower level than the trip setting and is used to initiate corrective action. The alarm has a definite time delay, while the trip characteristic generally has a choice of definite time delay or inverse time characteristic.