Troubleshooting SEL Relay Voltage Collapse Mis-Trips

SEL relay event data showed all three bus-voltage channels falling nearly to zero during a phase-to-phase fault, even though the remote substation remained energized. Follow the measurement path from the primary bus through the potential-transformer circuits and automatic throwover contacts before treating the oscillography as proof of a primary-system voltage collapse.

What signal path produced the recorded voltage?

The fault occurred on a 34.5 kV subtransmission loop. Fault current at the fault was about 11,000 A. A remote substation approximately 8 miles away contributed about 1,000 A through a different transmission line, and its sources remained connected.

The relay did not measure the primary bus directly. Each recorded voltage passed through several elements:

1. The energized primary bus applied phase-to-ground voltage to three PTs.
2. The grounded-wye PT bank converted the primary voltages to secondary-level signals.
3. Secondary wiring carried those signals through the bus-potential throwover scheme.
4. Transfer-relay contacts selected or interrupted the potential source.
5. The SEL relay voltage inputs sampled the selected circuits.
6. Protection logic combined voltage and current measurements, identified an apparent Zone 1 condition, and issued a trip.

Proof check: draw one continuous path for each phase from its PT secondary terminal to the corresponding relay input, including every fuse, test switch, terminal, and throwover contact.

Did the primary bus or only the measurement collapse?

A simultaneous near-zero indication on three relay voltage channels does not by itself prove that the three primary phase voltages collapsed. A phase-to-phase power-system fault can depress voltages across a looped network, but the observed combination requires separating a primary disturbance from a common-mode interruption in the voltage-measurement circuit.

The all-phase transition is the key discriminator. Correlate the first sample showing lost voltage with transfer-relay status, DC control voltage, and contact operation. If voltage exists at the PT secondary but disappears after the throwover contacts, the apparent collapse is in the instrument circuit.

Proof check: demonstrate whether all three voltages were present on the source side of the selection circuit at the instant the relay inputs lost them.

Are all three PT channels connected and configured?

Three installed PTs do not prove that the protection relay receives three independent phase voltages. Commissioning records must establish both the physical circuits and the relay configuration. Confirm that each phase-to-ground secondary reaches its intended input, that the grounded-wye neutral path is intact, and that the relay is configured for the actual three-voltage connection.

1. Measure the normal operating voltage at each PT secondary and record the phase relationship.
2. Repeat the measurement at the input and output sides of each intervening device.
3. Measure directly at all three relay voltage-input terminals.
4. Compare the terminal measurements with the relay metering display.
5. Review the relay configuration to confirm that it uses all three installed phase-voltage inputs and the correct connection type.
6. Check channel labels against the physical phase conductors; a correct magnitude on a mislabeled channel can still corrupt directional and distance calculations.

Normal relay metering is a necessary baseline because it detects an unused input, open conductor, mapping error, or incorrect connection setting before fault analysis begins. Record the baseline so post-event values can be compared with the same circuit and configuration.

Proof check: obtain matching, stable three-phase voltage indications at the PT terminals, selector output, relay terminals, and relay metering display.

Can the automatic throwover open every potential circuit?

The installation contained an automatic throwover between two separate bus-potential sets. During the event, the relay responsible for that transfer dropped out and effectively opened the potential circuits. That common switching point explains why all three measured voltages approached zero together while the power-system sources remained connected.

A transfer scheme can produce a break in measured voltage when the active source contacts open before the alternate source contacts establish a complete path, or when the transfer device drops out without completing either source path. The exact contact sequence must come from the AC schematic, DC control schematic, and device records rather than from the intended operating description.

1. Mark the normal current path through the selected set of bus PTs.
2. Mark the alternate path through the second PT set.
3. Identify every contact that can disconnect all three phases or remove the transfer relay’s control power.
4. Check the transfer relay’s pickup, dropout, and seal-in dependencies shown in the DC schematic.
5. Compare auxiliary-contact or digital-input transitions with the relay voltage waveform.
6. Test each allowed transfer state and verify that at least one complete potential path exists whenever protection remains enabled.

A transfer scheme may be electrically functional in steady states yet create an unacceptable interruption during transition. Test the transition itself, not only the two final selector positions.

Proof check: reproduce the voltage interruption at the selector output by operating or dropping out the throwover relay under controlled test conditions.

Why did lost voltage resemble a Zone 1 fault?

Distance protection derives an apparent impedance from measured voltage and fault-loop current. In simplified form, Zapp = Vloop / Iloop. The exact voltage, current, compensation, polarization, and supervision quantities depend on the relay model and configured element.

During this external phase-to-phase fault, the remote terminal supplied substantial fault current while the opened potential circuit drove the relay’s measured voltages toward zero. With current present and voltage artificially reduced, the calculated apparent impedance can become very small. A small apparent impedance may enter the configured Zone 1 characteristic even though the physical fault lies outside that zone.

Proof check: use time-aligned event quantities to show that measured voltage disappeared while fault current remained and that the apparent impedance entered the enabled Zone 1 characteristic.

Why did loss-of-potential supervision not prevent the trip?

The relay was expected to block directional tripping following loss of potential, but correct analysis requires the sequence of individual logic assertions. “Loss of potential” and “trip” are not simultaneous abstract states. Each is produced by measurements, pickup criteria, logic processing, and any configured timing or latching behavior.

Two sequences require investigation. First, the distance or directional condition may have asserted before the voltage-loss function recognized the open potential circuit. Second, the protection element may have picked up before the block and then continued through its logic after the block asserted. Which sequence applies is decided by the event report and the model-specific logic equations.

1. Display raw phase voltages and relevant loop currents on one time axis.
2. Add the voltage-loss indication, directional or distance pickup, Zone 1 pickup, trip equation, and breaker-output indication.
3. Identify the first state change; do not infer order from a summary target list.
4. Review whether the blocking signal inhibits pickup, interrupts an active timing path, resets a latched state, or only gates the final trip equation.
5. Check whether the potential interruption was too brief or had the wrong measurement signature for the configured voltage-loss logic.
6. Use the exact relay model’s settings and logic documentation to resolve behavior not visible in the waveform.

Proof check: produce a chronological logic trace showing whether the blocking condition preceded pickup and whether it actually gated the final trip equation.

How should the potential-transfer circuit be corrected?

The correction must prevent an automatic-transfer dropout from leaving the protection relay with fault current but no valid voltage. The selected design depends on the existing schematics and operating requirements, so verify the final contact arrangement through a formal protection and control review.

1. Document the discovered throwover logic on both AC and DC drawings.
2. Correct drawing omissions or unclear contact cross-references that concealed the common interruption point.
3. Modify the transfer sequence or protective interlocking so an invalid potential state cannot feed unsupervised voltage-dependent tripping.
4. Map transfer status and voltage-valid status into recorded relay inputs where suitable points exist.
5. Review the loss-of-potential block against every trip path that depends on reliable voltage, including any seal-in or timing behavior.
6. Repeat three-phase voltage-injection and transfer tests for normal selection, alternate selection, transfer, and control-relay dropout.
7. Save the final settings, logic equations, drawings, and test records under configuration control.

Proof check: during every selector transition and dropout test, either valid voltage remains at the relay inputs or the applicable voltage-dependent trip path is blocked before it can operate.

How is the complete repair verified?

End-to-end verification must exercise the measurement circuit and protection logic together. A steady-state meter check cannot prove the throwover transition, and a relay logic test alone cannot expose an opening contact in the PT circuit.

1. Confirm normal three-phase voltage at both PT sets and at the relay inputs.
2. Confirm that relay metering matches the injected or measured terminal values.
3. Operate the throwover through every permitted transition while recording all three voltages and transfer indications.
4. Apply a test representing external-fault current while controlling the voltage inputs through the abnormal transfer condition.
5. Verify the chronological order of voltage-valid indication, loss-of-potential supervision, distance or directional pickup, trip equation, and output contact.
6. Confirm that a valid in-zone test still produces the intended trip after the corrective logic is applied.
7. Download the final event record and compare it with the acceptance criteria, rather than relying only on front-panel targets.

Proof check: retain an event record showing that an abnormal potential-transfer state cannot issue the unwanted trip and that a valid protection test still operates correctly.

FAQ: How Do I Confirm the Repair?