DTR2011
Electrical
- Oct 12, 2006
- 682
I have been asked to investigate an issue for an industrial customer. There was an event where a 480V Main - Tie - Main Switchgear tripped both mains.
I performed a site visit today, gathered information. The 480V SWGR feeds motor loads only, a combination of VFD's and across the line motors, the largest being 250 HP.
There is a Basler BE1-47N Relay that performs dual duties of (27/59 - For auto transfer purposes) and 47 for what it is designed for. The trip of both mains was attributed to the 47 target on the Basler 47N Relay. Settings on the relay are 6% of Nom (120). This unit does not have time delay available and according to the specs, operates in less than 50 ms.
Both sources for the M-T-M Swgr are fed from the same single (69kV) utility source. There is no event (fault - waveform) recording equipment in this installation. I asked about any other events at the same time. Later an answer came that there was a motor ground fault that occurred (ONE hour difference) to the trip. I have have no idea how the plants historian SW works, but could imagine daylight savings, etc to be the source of time stamp problems.
The motor ground fault that occurred is fed from a separate 69kV (Delta) to 4.16kV (Wye) transformer, but tied to the same source.
The Basler 47N Relay (at 480V Level) has an Open Delta (2 VT) Connection feeding it.
The 4.16 kV - 480 V transformers are Delta / Wye, with a high resistance ground connection.
The setting on the Basler Relay (BE1-47N) is 6% of nominal (120 V), which is 7.2 V secondary. This is an instantaneous unit.
I have done some research and the 6% appears to be adequate *5% preferred) for motor bus protection, however it appears to me that some time delay is necessary. I feel the easiest (and cost effective) approach is to order (2) new relays with a time delay available. Basler offers definite and inverse time delay options. The next question is definite time or inverse, when ordering a replacement relay?
So my question really is can a ground fault on an adjacent feeder (through a separate transformer) cause a significant amount of Negative Sequence Voltage to trigger a 6% Negative Sequence Voltage? The 69kV(D) -4.16kV(Y) Xfmrs are ~5-6%Z. The distribution transformers (4.16 (D)/ .480 kV(Y) are ~ 5-6%Z. The plant Basler relays have 480V/120 (USA Open Delta) Sensing.
Thanks for the comments, input and wisdom.
I performed a site visit today, gathered information. The 480V SWGR feeds motor loads only, a combination of VFD's and across the line motors, the largest being 250 HP.
There is a Basler BE1-47N Relay that performs dual duties of (27/59 - For auto transfer purposes) and 47 for what it is designed for. The trip of both mains was attributed to the 47 target on the Basler 47N Relay. Settings on the relay are 6% of Nom (120). This unit does not have time delay available and according to the specs, operates in less than 50 ms.
Both sources for the M-T-M Swgr are fed from the same single (69kV) utility source. There is no event (fault - waveform) recording equipment in this installation. I asked about any other events at the same time. Later an answer came that there was a motor ground fault that occurred (ONE hour difference) to the trip. I have have no idea how the plants historian SW works, but could imagine daylight savings, etc to be the source of time stamp problems.
The motor ground fault that occurred is fed from a separate 69kV (Delta) to 4.16kV (Wye) transformer, but tied to the same source.
The Basler 47N Relay (at 480V Level) has an Open Delta (2 VT) Connection feeding it.
The 4.16 kV - 480 V transformers are Delta / Wye, with a high resistance ground connection.
The setting on the Basler Relay (BE1-47N) is 6% of nominal (120 V), which is 7.2 V secondary. This is an instantaneous unit.
I have done some research and the 6% appears to be adequate *5% preferred) for motor bus protection, however it appears to me that some time delay is necessary. I feel the easiest (and cost effective) approach is to order (2) new relays with a time delay available. Basler offers definite and inverse time delay options. The next question is definite time or inverse, when ordering a replacement relay?
So my question really is can a ground fault on an adjacent feeder (through a separate transformer) cause a significant amount of Negative Sequence Voltage to trigger a 6% Negative Sequence Voltage? The 69kV(D) -4.16kV(Y) Xfmrs are ~5-6%Z. The distribution transformers (4.16 (D)/ .480 kV(Y) are ~ 5-6%Z. The plant Basler relays have 480V/120 (USA Open Delta) Sensing.
Thanks for the comments, input and wisdom.