87B Trips During Disconnector Switching: Causes

Short answer

In double busbar substations the 87B decides which zone each bay belongs to by reading the position of its disconnectors. If that replica does not match reality during a switching operation, one bay's current is added to the wrong zone and a differential appears under normal load. The usual causes are misadjusted or damaged auxiliary contacts, intermediate positions misread, and the switching sequence.

Symptom

In a double busbar substation, when transferring a bay from one bus to the other, opening or closing the bus coupler or operating a bus disconnector, one of two things can happen. The bus differential 87B trips a zone with no fault, or it locks out with a disconnector discrepancy alarm. Outside switching operations the protection works fine.

On a single-zone bus, all the bays always sum into the same differential. On a double bus they do not: each bay can be on bus 1, on bus 2 or on both during a transfer. The 87B decides which zone each current is summed into according to the position of the bus disconnectors, the so-called replica or bus image, which is read with auxiliary contacts or received by GOOSE. If the replica does not match the real state of the primary, the protection sums wrongly and sees a differential that does not exist.

The problem is not exclusive to double busbar. It appears in any arrangement in which the differential zone changes with the topology: buses with transfer, sectionalized buses with bus-section disconnectors, and schemes in which a bay can move to the transfer bus. In breaker-and-a-half, by contrast, the bus zones are fixed and the replica does not take part in summing currents. That is why this symptom concentrates in substations where daily operation includes transferring bays between buses.

Possible causes, from most to least likely

Cause	Why it happens	What gives it away
Misadjusted auxiliary contact	The contact that says “closed” changes before or after the blade makes electrical contact	Differential during the disconnector’s travel, not before or after
Damaged auxiliary contact or reversed wiring	Wear, moisture, loose terminal, disconnector replaced without checking the wiring	Replica that does not match the real state at rest
Intermediate position misread	Logic that takes “neither open nor closed” as open	Trip or block during the disconnector’s transit time
Wrong switching sequence	Transfer with the coupler open or without closing the second disconnector first	The bay changed bus under load with no parallel path
Open CT or test block not properly closed	A bay does not deliver current to the 87B	Block by CT supervision, differential equal to the bay current
Loss of communication with a bay unit	In distributed systems the replica or the current arrives over the process bus	Communication alarm and zone block
Check zone disabled or wrongly applied	The zone trip does not require the overall confirmation	Zone trip without operation of the check zone

What to check first

87B event log: which zone, which phase, which bays, and in what order the disconnector positions changed relative to the instant of the differential.
Differential current per zone and of the check zone in the oscillography. If the check zone saw no differential, the fault was not on the buses: the problem is the replica.
Auxiliary contact timing against the real movement of the blade. See motorized disconnector drive.
Switching sequence applied against the approved one, and the state of the coupler at that moment.
CT supervision and test blocks of all the bays.

First check on an 87B trip during switching

Did the check zone see differential current?
- Yes: Possible real bus fault: inspect before re-energizing
- No: Does the replica match the real state of the disconnectors at rest?
  - No: Damaged auxiliary contact or reversed wiring
  - Yes: Did the differential appear during the disconnector's travel?
    - Yes: Auxiliary contact that changes before or after electrical contact, or intermediate position mishandled
    - No: Wrong switching sequence or an open CT on a bay

How to confirm

Replica test bay by bay: each disconnector is operated and the 87B is checked to confirm the bay changes zone at the right moment.
Adjustment of the disconnector auxiliary contacts: the contact that indicates “closed” for the differential must act consistently with the instant the blade conducts, per the protection manufacturer’s criterion.
Logic test with secondary injection simulating disconnector positions, including intermediate ones and bays on both buses. See secondary injection.
Differential measurement under load in each zone after every operation: it should stay close to zero.

How to keep it from coming back

Use the check zone as a trip condition, if the scheme allows, so that a replica error is not enough.
Disconnector discrepancy supervision with an alarm and, per the asset owner’s practice, selective blocking.
Disconnector maintenance that includes the auxiliary contacts the 87B uses, not just the blade.
Written switching sequences and, if possible, interlocked in the bay control.
Replica tests every time a disconnector or an auxiliary contact is replaced. See substation disconnector.

What not to do

Do not disable the 87B to carry out switching: it is the fastest protection of the bus and switching is when it is most needed.
Do not force positions in the replica without a written order and a deadline to remove them.
Do not re-energize the bus without reviewing the check zone and the bay protections.
Do not replace a disconnector without repeating the replica test.

Why disconnector switching is a critical moment for the 87B

In a double busbar substation, the bus differential protection decides which circuits belong to each zone from the position of the bus disconnectors. That information, called the replica, reassigns currents between zones and lets the differential compute the correct balance. When a circuit is transferred from one bus to another, there is a moment when both disconnectors are closed: the buses are tied through the circuit and current can split between them. If the replica does not reflect that state at the precise instant, the balance of one zone stops closing and the differential sees an operating current that corresponds to no fault.

The IEEE C37.234 guide covers the application of relays to buses and describes those configurations. What matters in the field is the criterion each manufacturer uses to implement the replica and the disconnector supervision:

Some relays use two contacts per disconnector (one for closed and one for open) and declare an indeterminate state when they do not agree, with block logic or holding the last valid state.
Others use a single contact and assume the opposite by absence; there a broken wire or a misadjusted contact reads as an open disconnector.
Behavior during the transition (both disconnectors closed, one opening, another closing) varies between manufacturers: it can merge the zones, block the differential or keep the previous assignment. This decision must be documented in the protection study, not left to the relay’s default value.

What leads most often to a trip is the timing difference between the disconnector auxiliary contact and the real position of its blades. The contact can indicate “closed” while the blades are not yet touching, or “open” when current is still flowing through the arc of the operation. The current of a loop or of a circuit in parallel with the transfer disconnector changes path for a few instants, and the replica must follow it.

When investigating a trip, compare the time at which each disconnector contact changed with the time at which the current of each zone changed. If disconnector status is received by GOOSE or from a field unit, also check the quality of those signals and their subscription. A trip repeated in the same operation points to the interlock, the contact or the wiring of that disconnector, and not to the protection in general. The guides on disconnector replica for the 87B and on interlocking in double busbar complete this analysis.

When to call us

If your bus differential trips or blocks during switching, send us the 87B events and oscillography: we quote the trip analysis, a remote service available worldwide, and tell you whether the problem is in the replica, the logic or the sequence. Our team corrects the logic and the replica through remote relay configuration; if auxiliary contacts and replica must be tested bay by bay, we do it on site with protection testing, in Colombia or elsewhere on request. Write to us. On this point, read case of the 87B that tripped the bus during a bay test.

References

IEEE C37.234-2021, Guide for Protective Relay Applications to Power System Buses.
Blackburn and Domin, Protective Relaying: Principles and Applications.

Frequently asked questions

Why did the 87B trip when a bay was changed from one bus to the other?

Because during the transfer the disconnector replica did not match the real state: the bay current was assigned to a zone it was not flowing through, or left the zone it was flowing through. That difference shows up as differential current under load. The check zone, if present and properly applied, prevents the trip.

What is the check zone of a bus differential?

It is an overall differential that sums all the bays of the substation without regard to the disconnectors. It sees all faults on any bus and does not depend on the replica. If the trip of each zone also requires the check zone to operate, a replica error is not enough to trip.

What happens to the 87B when a bay has both buses tied through its disconnectors?

The two buses are connected in parallel through that bay and, in most schemes, the protection treats them as a single zone for as long as that condition lasts. If the logic does not recognize it, there can be differential current in both zones or a trip of both buses for a fault on one.