Busbar differential protection, device 87B, guards the busbar - the common node where many circuits connect together in a substation - by summing the currents of every circuit attached to it. Because all those circuits meet at one electrical node, in a healthy bus the currents flowing in must equal the currents flowing out and their sum is zero. A fault on the bus itself breaks that balance, and the differential relay trips every breaker feeding the bus. A bus fault is one of the most severe events in a substation, feeding energy from many sources at once, so 87B is engineered to clear it in one to two cycles.
Busbar Differential (87B) in one line: Busbar differential protection (87B) sums the currents of all circuits connected to a busbar. In a healthy bus these sum to zero; a bus fault unbalances them, so the relay trips every breaker feeding the bus. It must operate in one or two cycles and stay stable during heavy external faults that can saturate CTs.
A busbar is the meeting point of feeders, transformers, and lines within a substation, and electrically it is a single node. The current balance that all differential protection relies on applies to it directly: whatever current flows into the node from some circuits must flow out through the others, so the vector sum of all circuit currents at a healthy bus is zero. Busbar differential protection instruments every one of those circuits with a CT and adds the currents together. As long as the sum stays near zero, the bus is healthy.
When a fault occurs on the bus itself - a flashover across insulation, a failed support, a dropped tool - current pours in from every connected source and none of it leaves through a circuit, because it is escaping into the fault at the node. The sum of circuit currents is no longer zero, and that residual differential current is the relay's signal to trip. Since the fault is on the common node, there is no single feeder to isolate; the relay must open all the breakers connecting sources to that bus.
Speed is paramount. A bus fault is fed by the combined fault-current contribution of every source connected to it, so it is extremely severe both thermally and mechanically, and it stresses the whole station. It can also threaten system stability by depressing voltage across a wide area. For these reasons busbar protection is designed to detect and trip in one to two cycles, faster than almost any other scheme, and to be absolutely selective so it clears only the faulted bus and no more of the network than necessary.
The classic implementation is the high-impedance scheme. All the bus CTs, which must have matched ratios, are paralleled and connected to a high-impedance voltage-operated relay, usually with a nonlinear resistor to limit peak voltage. During an external fault the CT secondary currents circulate among themselves and little voltage appears across the relay, keeping it stable; during a bus fault the imbalance drives current through the relay's high impedance and develops a voltage that operates it. High-impedance protection is simple, fast, and inherently robust against CT saturation on through faults, but it requires dedicated matched CTs and does not easily accommodate flexible bus arrangements where circuits switch between buses.
The low-impedance, or numerical, scheme digitizes each CT current independently and computes the differential and restraint quantities in software. This flexibility is its great strength. It can apply a bias characteristic tuned to reject CT-error currents during heavy external faults, it can share CTs with other protection, and it can handle differing CT ratios by scaling in the algorithm. Crucially, it can track a substation's switching state so that circuits are assigned to the correct differential zone as isolators open and close, which high-impedance schemes cannot do gracefully.
That switching-aware capability is the dynamic bus replica. In a substation with multiple buses and bus-coupler and bus-section switches, a given feeder may be connected to one bus or another depending on isolator positions. The low-impedance relay reads the isolator status and dynamically includes each circuit's current in the zone for whichever bus it is presently connected to, so the differential sum is always taken over the right set of circuits. Get the replica wrong and the relay either fails to protect part of the bus or misoperates; get it right and it protects a complex, reconfigurable station accurately.
Because tripping an entire bus is so consequential and because a security failure of the dynamic replica could trip the wrong bus, low-impedance schemes add a check zone as a second, independent confirmation. The check zone sums the currents of all circuits across the whole substation regardless of which bus they are on. A true bus fault raises differential current in both a specific zone and the check zone, and the relay only trips when both agree. A wiring or replica error that fools one zone is very unlikely to fool the overall check zone, so requiring both to operate dramatically improves security against false trips.
CT saturation is the central technical challenge for any bus scheme. A severe external fault, drawing the full station's contribution through the CTs of the outgoing circuit, can saturate that CT so its secondary output collapses partway through each cycle, creating a spurious differential current that mimics an internal fault. High-impedance schemes resist this by design; low-impedance schemes counter it with saturation-detection algorithms that recognize the characteristic distortion of a saturated CT and restrain during the saturation intervals, so the scheme stays stable through-fault while remaining fast for genuine bus faults.
For a control room and a cloud SCADA platform, the value lies in what the relay reports rather than the differential mathematics, which stays inside the device. An 87B operation trips a whole bus and takes multiple circuits out at once, so the events, breaker positions, and zone-selection status the relay publishes over DNP3 or IEC 60870-5-104 are exactly what operators need to understand a major outage quickly. A platform like Merobix, aggregating those events across sites, lets engineers see which bus tripped, confirm the check zone corroborated it, and correlate the loss of every affected feeder in one timeline - turning a severe substation event into a clear, reviewable picture for field and control staff.
A bus fault is fed by the combined fault current of every source connected to the bus, making it one of the most severe faults in a substation both thermally and mechanically, and it can depress voltage across a wide area and threaten stability. To limit that damage and preserve stability, busbar differential is designed to detect and trip in one to two cycles, faster than most other protection schemes.
A check zone is a second, independent differential measurement that sums the currents of all circuits across the whole substation, regardless of which bus each is on. A genuine bus fault appears in both a specific bus zone and the check zone, and the relay only trips when both agree. This guards against a wiring or dynamic-replica error tripping the wrong bus, because such an error is very unlikely to fool the overall check zone as well.
In substations where feeders can switch between buses via isolators, a dynamic bus replica is the relay's live model of which circuit is connected to which bus, based on isolator status. The low-impedance differential relay uses it to include each circuit's current in the correct zone as switching changes, so the differential sum is always taken over the right set of circuits. It lets one scheme protect a reconfigurable multi-bus station accurately.
This page references the protocol specifications published by the organizations below. Editions, product capabilities, and documentation change over time - confirm current requirements and specifications directly with the source.
Last reviewed: July 27, 2026. Merobix is not affiliated with, endorsed by, or sponsored by these organizations; their names are used only to identify the standards and products discussed.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.