Selective Protection

The selective protection was developed in the 19th century. The Niagara plants needed a selective line protection for their 11-kV-grid. The technical director Stillwell created it in an innovative manner. He used common elements like overcurrent and time-relays. He used a GE fan motor as a directional unit. The stator field  was interfused through the voltage, the rotor was flown through the current. The reverse-current-relay for the double line was invented (Fig. 20). An automatic operating circuit breaker, using overcurrent and reverse-current relays was proposed at the Paris fair in 1900. The two-pole relay operated an oil circuit breaker with compressed air.

The first reverse-current-relays for alternating current (AC) had the same name as those for DC. H. Probst invented the AC reverse-current-relay, influenced by the use of small shunt-motors. He said that there were a lot of objections against this idea; some even said that he did not know the first thing about electrical engineering. Nevertheless he invented the reverse-power-protection. Before claiming the patent by AEG Gerhard published in 1903 a paper in the ETZ magazine describing a reverse-power protection. The relays are supplied with current and voltage by the measuring transformers; closing a parallel circuit which operates the tripping devices. This was the first reference of transformer-operated tripping devices. Contrary to later devices the relay was supplied by the voltage transformer in parallel despite the fact that the voltage is not available during a fault condition. Current transformer-operated relays are supplied by current transformers in series.The main use of this directional relay was to trip the faulty generator in the case of parallel operating transformers using the criteria "change of power direction". Auxiliary power from the grid was required to operate both the secondary relays and the shunt trip. Henri Owen Tudor proposed the accumulator n a patent in 1886 (Fig. 16).

Transformer operated tripping devices were used in less important stations. Please note that in the first stations two current transformers were used one for the relay and one for the tripping device. In the Cleveland-protection, developed in 1908, the current transformer was connected with the tripping device of the breaker mechanism. To achieve a delayed tripping; a time-fusible-link was connected in parallel; unblocking the current after the link melted. To achieve selectivity, a time delay was sometimes necessary. The term "selectivity" was already used in telephone systems to differ between intermittent and permanent currents. The introduction of overcurrent protection in 1901 was without time grading. Later they used clocks as timing elements (Fig. 21; Fig. 18)

Merz und Price proposed the longitudinal current differential protection in 1903. This will be described in a later article. The use of a hinged armature in an overcurrent relay was published in 1902 in by SIEMENS and Brown,C.E.L.However BBC claim the patent on a current-dependent relay (DRP No. 143556). AEG supplied overcurrent relays with two phases on one Ferraris disk one year later and published a patent with "common maximum and reverse-current-relay" for generator protection. It was an aluminum disk driven by a three-limb magnetic core. The outer limbs were excited by the voltage, the center limb by the current (Fig. 10).  Under normal current direction the relay had a huge delay at high overcurrents but tripped immediately under reverse current conditions.

Krämer,Chr., F&G patented a "relay for automatic opening of alternating current circuits" in 1904 (DRP 174218). They used a Ferraris disk as a time-element of the voltage drop relay as selective short-circuit protection. The principle of distance protection is published in a general summary in this paper; this is the invention of the distance protection.

ASEA designed and built their first protection relay in 1905, the overcurrent-time-protection type TCB. The need for protection for  ring system operation was becoming more important in 1908. The first ideas used voltage drop relays without directional elements. An AEG patentin 1908 (DRP 210 152) described a solution that used a short circuit voltage setting to block the timing element. This was of course only possible if all relays had the same time settingand the stations inside the ring had only one circuit breaker.

The idea for an automatic recloser was patented by Bollinger, Emag, in 1913. In the 1930's the dead time was 3 minutes to allow motor starters to reset or to switch off motors manually. V&H provided in 1925 an "automatically on-off-switch" which switches on the station 3 minutes after tripping automatically. Two or three reclosings were common, sometimes operated by dropping weights.

Schrader,W., V&H, patented a relay for automatically working overcurrent tripping devices in 1914. Contrary to known extra-high current relays this device did not wait until a certain level of current was reached; it reacted after certain increases of amplitude in a certain time - the first di/dt- relay.The first time grading in opposite directions was used in 1918 although the principle had been  known since 1900. Today it is known as directional overcurrent protection. The time for protection operation close to power plants was very high up to 12 seconds. Biermanns, AEG, used dependent overcurrent protection as line protection already 1919. The circuit breakers and relays used at this time required a selective time interval in the range of 0.5 up to 1.5 seconds (Fig. 12).

Pfannkuch.W of AEG developed a cable protection relay using a pilot wire in the main wire - the patent "Pfannkuch-Protection" (DRP 398482). Höchstädter,M. proposed another cable protection ("Lypro") in the same year. These relays could detect two-phase short-circuits and earth-faults. This  split-conductor protection caused a stir, later they were substituted by distance relays with advanced features.  Another approach to achieve fastest tripping times with rugged devices (not  high accuracy) was to observe that the fault current has the same direction on both sides. Common operating times of 0.5 s were already achieved in the US in the 1920's; in Germany 2 s and more were quite common.

Further initiatives in distance, differential and generator protection will be covered in later articles.

walter.schossig@pacw.org
www.walter-schossig.de