UNDER-VOLTAGE RELAY SETTING

[ozziemick]

Hi, i'm an electrician at a pilot hydrocarbon processing plant in Australia. I'm trying to find a rule of thumb setting for the site incoming power supply under-voltage relay time delay. The reason is that the supply to the site was interrupted which caused equipment to stop. The result of this could have very serious consequences due to the nature of the process. The relay is set to trip the incomer circuit breaker at 70% of normal voltage in a time of 0.01 secs. The time delay was set by the high voltage switch gear vendor, but i'm wondering if the time was set too short, due to the effect a trip has on the safety of the plant? My understanding of under-voltage protection is to prevent motors etc from gettig to hot because of increased current. I can't see why, if the time delay was say, 0.5 or 1 sec, the equipment wouldn't still be protected from prolonged "brownout's", but site power could be maintained if a very quick dip in power occured. I'm not sure of the duration of the dip we had, but we have two sub's with identical u/v relay's and setting's, but only one tripped.

 

[SooryaShrestha]

I think you are right. You need not have very low time setting, if you are sure of the characteristics of the motors installed in your plant. But if you are not sure you can install the UV relay for the breaker supplying the critical motors only, for which you may require some design alteration to the wiring design of the plant.
Regarding one UV functioning and the other not tripping, it could be a) the relay malfunction for which you can test the relay or b) time setting not exactly the same in which case after the under voltage situation one with lower time setting tripped and the voltage improved, in which case it did not trip the other. Such things could happen in case of small generating stations with appreciable load.

 

[peterb]

I agree with you, SooryaShrestha. 0.01 sec is 1/2 cycle on a 50 Hz basis and is much too sensitive for direct tripping by an undervoltage relay on an industrial plant circuit. That kind of time delay is probably only warranted where there is an immediate danger of loss of synchronism on a high-capacity utility circuit, and the undervoltage should not be the only trip criterion even then.

A 0.5 sec delay sounds much more reasonable for this case. Do you have any transient recording equipment on the circuit, Ozziemick? If yes, this could provide data to guide you in developing a "good" setting, based on actual circuit behaviour during external system disturbances.

 

[jbartos]

Suggestion: Visit industry standard Reference:
1. ANSI/IEEE Std 242-1986 "American National Standard IEEE Recommended Practice for Protection and Coordination of Industrial and Commercial Power Systems" (Buff Book)
1a. Page 144 Sect. 4.5 "Undervoltage Relay - Device No. 27" Describes various applications; e.g. Bus Undervoltage Protection for momentary dips, etc.; Source Transfer Scheme; Permissive Functions; Backup Functions; Timing Applications, etc.
The solid state instantaneous relays are probably with the shortest setting, e.g. less than 1.5 cycle=0.025 seconds. They would respond to a reverse phase sequence condition in addition to other undervoltage conditions.
Figure 48 on page 146 shows typical time-voltage characteristics of an Undervoltage Relay. For 70% of tap setting, the time ranges from about 0.8 second to about 9 seconds, which is a reasonably wide range for a wide area of applications. The time dial goes from 0.5 to 10, i.e. there are 11 time-voltage inverse curves.
Section 11.4.7 on page 460 describes an application of 27 relay related to generator operational applications.
Section 13.5.4 on page 522 describes an application of 27 relay related to automatic transfer, which should be delayed to coordinate for faults.