Single line diagram of 2 incomer and 1 bus coupler LT panel showing 2-out-of-3 interlocking between INC-1, bus coupler and INC-2

2 Incomer + Bus Coupler Interlocking: 2-Out-of-3 Logic, Transfer Sequence & Design Guide

Bus coupler interlocking is the arrangement that decides which combination of the two incomer breakers and the bus coupler is permitted to be closed at any moment. In the standard scheme, known as 2-out-of-3 interlocking, any two of the three breakers may be closed together and all three must never be closed simultaneously — because all three closed parallels the two transformers through the LV bus.

The rule that follows is a design obligation, not a preference. If the installation has not been engineered for parallel transformer operation — matched voltage ratio, vector group, phase sequence and impedance, with breakers and busbar verified against the paralleled fault level — then the prohibited state must be blocked mechanically, not merely discouraged in software.

Introduction

Consider a typical industrial LT power control centre:

Transformer-1 → INC-1 → BUS-A ←[ BUS COUPLER ]→ BUS-B ← INC-2 ← Transformer-2

Three breakers. One busbar split into two sections. On a single line diagram it is one of the simplest arrangements in the plant.

Yet this arrangement controls the most consequential switching operations in the entire electrical network. Get the interlocking philosophy wrong and the failure modes include unintended transformer paralleling, exceeded breaker duty, an overloaded surviving transformer, damaged motor shafts, and a bus fault being re-fed from a second source.

The question that defines the whole design is not:

Can the bus coupler close?

It is:

Under exactly which system conditions should it be allowed to close, and what must be proven before that command is issued?

This guide covers the complete engineering philosophy — interlocking layers, transfer sequence, capacity verification, motor bus transfer, neutral and earthing, protection coordination, and the failure modes that separate a working scheme from a dangerous one.

1. What a Bus Coupler Actually Does

A bus coupler is a breaker connecting two sections of a common busbar.

Coupler open — Bus-A and Bus-B operate as two electrically independent switchboards, each fed by its own transformer.

Coupler closed — the two sections become one electrically continuous bus. What that means for the sources depends entirely on which incomer breakers are closed at that moment.

That last sentence is the entire subject of this article.

The common normal state

INC-1 Bus Coupler INC-2 Condition
CLOSED OPEN CLOSED Each transformer independently feeds its own bus section

This arrangement is widely used because it delivers several advantages simultaneously:

  • Each transformer carries only its allocated load
  • Bus sections are electrically separated, so a fault on one bus does not necessarily collapse the other
  • Prospective fault current on each section is limited to a single transformer's contribution (plus motor contribution on that section)
  • Maintenance isolation is straightforward
  • Loss of one transformer does not remove supply from the healthy bus

But this is a philosophy, not a rule. Some plants deliberately operate with the coupler normally closed and one incomer open — for example where one transformer is a standby unit, or where the load profile does not justify two transformers running simultaneously. The correct normal state must come from the system design, load study and redundancy requirement — not from habit.

2. 2-Out-of-3 Interlocking and the State Matrix

For three breakers — INC-1, Bus Coupler, INC-2 — the classical interlocking rule is:

Any two of the three breakers may be closed at the same time. All three must never be closed simultaneously.

Hence the name: 2-out-of-3 interlocking.

The principal permitted states:

INC-1 Coupler INC-2 Result
1 0 1 Normal — each source feeds its own bus
1 1 0 Transformer-1 supplies both bus sections
0 1 1 Transformer-2 supplies both bus sections
1 1 1 Prohibited in a standard 2-out-of-3 scheme

(1 = Closed, 0 = Open)

The complete operating matrix, including partial states:

Mode INC-1 Coupler INC-2 Allowed?
Normal operation Closed Open Closed Yes
T1 feeds both buses Closed Closed Open Yes
T2 feeds both buses Open Closed Closed Yes
Complete shutdown Open Open Open Yes
Bus-A only energised Closed Open Open Yes
Bus-B only energised Open Open Closed Yes
All three closed Closed Closed Closed No — unless engineered for parallel operation

This matrix is not a design note. It is a contractual deliverable. It must appear explicitly in:

If it is not written down before manufacturing starts, it will be interpreted differently by the design engineer, the panel builder, the commissioning engineer and the plant operator. That is how switchgear gets damaged.

3. Why "All Three Closed" Changes the Entire System

If INC-1, INC-2 and the coupler are all closed, Transformer-1 and Transformer-2 are electrically paralleled through the LV bus.

This is no longer a changeover arrangement. It is parallel transformer operation, and it is a fundamentally different electrical system.

What changes

Load sharing becomes impedance-driven. Two paralleled transformers share load in inverse proportion to their percentage impedance, not their kVA rating. Two 1000 kVA transformers with 5.0% and 6.0% impedance will not share 50/50 — the lower-impedance unit takes a disproportionate share and can reach its thermal limit while the other is still under-loaded. The usable combined capacity is therefore less than the arithmetic sum.

Fault level rises. With bus sections separated, a fault on Bus-A is fed by Transformer-1 plus the motors on Bus-A. With the sources paralleled, that same fault is fed by both transformers plus motors on both sections. Prospective short-circuit current can approach the sum of the individual contributions.

Breaker and busbar duty must be re-verified. A higher prospective fault current directly challenges:

Motor contribution is routinely forgotten

Running induction motors are rotating machines. During a bus fault they behave as generators for the first few cycles, contributing roughly 4 to 6 times their full-load current, decaying over approximately 3 to 5 cycles.

On an MCC-heavy bus this contribution is not trivial. With the coupler closed, motors on both sections contribute to a fault on either section. Any short-circuit study that models only transformer contribution understates the peak duty that the breakers and busbar must survive.

Is parallel operation always prohibited?

No. Parallel transformer operation is entirely legitimate — when the system has been designed for it.

Conditions that must be verified before two transformers can be operated in parallel:

Parameter Requirement
Voltage ratio Identical, including tap position
Vector group / phase displacement Compatible
Phase sequence Identical
Percentage impedance Closely matched — commonly within ±10%
kVA ratio Commonly kept within approximately 3:1
Polarity Identical
Neutral earthing Coordinated single-point or engineered arrangement
Protection Rated and coordinated for the paralleled fault level
Busbar and breaker rating Verified against the paralleled fault level

The critical distinction:

An installation designed only for non-parallel transfer must not end up operating in parallel simply because three breakers are physically capable of closing.

That is precisely what the interlock exists to prevent.

4. Three Layers of Interlocking

A robust scheme uses three complementary layers, each independent of the others.

Layer 1 — Mechanical interlocking

A physical interlock — cable-operated, rod-and-lever, or a proprietary kit — that mechanically prevents the prohibited breaker combination. Most major ACB manufacturers supply mechanical interlocking kits specifically configured for two-source-plus-coupler arrangements.

Mechanical interlocking is the foundation layer because it does not depend on:

  • PLC software or firmware
  • Relay logic
  • Control supply availability
  • Operator training or memory

It cannot be bypassed by a software change, and it survives a total loss of control voltage.

Layer 2 — Hardwired electrical interlocking

Breaker auxiliary contacts wired directly into the closing circuits.

For example, the coupler closing coil circuit is permitted only when INC-1 or INC-2 auxiliary contacts confirm the approved open condition. Similarly, an incomer closing circuit can be blocked based on coupler status and the opposite incomer's status.

This layer catches invalid commands before they reach the breaker, and it also functions when the PLC is out of service.

Layer 3 — PLC / controller supervisory logic

The PLC manages the intelligent functions: sequence, timing, permissive verification, transfer initiation, load shedding, alarming and return-to-normal.

PLC is excellent at sequence logic. But there is a design principle worth stating plainly:

The single most safety-critical prohibited combination should not depend on software alone when a mechanical or hardwired interlock can also enforce it.

PLC failure philosophy

Define explicitly what happens when the PLC fails or is taken out of service:

  • Existing breaker states are maintained — no spurious operation
  • Automatic transfer is disabled
  • Manual operation remains possible, subject to mechanical and hardwired interlocks
  • A distinct alarm is annunciated
  • Operating mode indication reflects the degraded state

This produces a far more resilient architecture than one in which the PLC is the only barrier to a prohibited state.

5. A Protection Trip Is Not a Source Failure

This is the single most important distinction in the entire transfer philosophy, and it is the one most often missing from control schemes.

INC-1 can open for two completely different reasons:

Case A — Upstream source loss. The HT supply has failed, or Transformer-1 has lost its incoming feed. The bus itself is healthy. Transferring Bus-A to Transformer-2 is correct and desirable.

Case B — Protection operation. INC-1 has tripped on overcurrent, earth fault, or busbar differential protection because there is a fault on Bus-A or on a feeder that did not clear. The bus is not healthy.

If the control scheme cannot distinguish these two cases, the sequence is catastrophic:

  1. A fault develops on Bus-A
  2. INC-1 protection operates and trips
  3. Auto-transfer logic sees "INC-1 open, Source-2 healthy"
  4. Coupler closes
  5. Transformer-2 is now feeding the same uncleared fault
  6. A second breaker operates, or the busbar fails

One transformer outage has become a switchboard loss.

The requirement

Transfer must be blocked following a protection trip. Implementation:

  • A lockout relay (86) or trip-cause discrimination that latches on protection operation
  • The lockout contact wired as a hardwired transfer inhibit, not only a PLC input
  • Manual reset required, after inspection, before transfer capability is restored
  • Distinct annunciation: "Transfer blocked — protection lockout"
  • Where busbar differential protection is provided, its operation must lock out both the incomer and the coupler

This also means the trip signal source matters. Undervoltage detection on the incoming supply and a protection trip contact are different inputs and must be treated differently by the logic. Building the scheme around a single "breaker open" status is not sufficient. This is closely related to protection selectivity and discrimination philosophy — the transfer scheme and the protection scheme must be designed together, not separately.

6. The Automatic Transfer Sequence

Assume the normal state — INC-1 closed, coupler open, INC-2 closed — and Transformer-1 supply fails.

A typical open-transition sequence:

Step Action Purpose
1 Detect Source-1 failure (undervoltage / phase loss / reverse power as applicable) Initiate
2 Confirm after approved time delay Avoid transfer on a momentary dip
3 Verify no protection lockout is present Prevent transfer into a fault
4 Open INC-1 Isolate the failed source
5 Confirm INC-1 physically open via auxiliary contact State-based, not command-based
6 Verify Source-2 and Bus-B healthy Confirm the destination
7 Verify Bus-A dead / residual voltage decayed See Section 7
8 Check all transfer permissives See below
9 Shed non-priority load if required See Section 8
10 Close bus coupler Transfer
11 Confirm coupler closed; annunciate transfer complete Verify

Steps 3, 5, 7, 8 and 9 are exactly where poorly engineered schemes fail.

Transfer permissives — the full list

Source and bus status

  • Healthy source voltage on the incoming side
  • Frequency within limits where applicable
  • Bus-A residual voltage below the approved threshold
  • Correct status confirmed for both incomers and the coupler

Breaker readiness

  • Breaker healthy, no trip circuit fault
  • Closing spring charged
  • Breaker in service (racked-in) position — not test or disconnected
  • No mechanical or electrical interlock inhibit active

Protection and safety

  • No protection lockout on either incomer
  • No busbar differential operation
  • No emergency stop active
  • No maintenance inhibit or permit-to-work lock applied

Operational

  • Auto mode selected
  • Surviving transformer loading within permitted limit after transfer
  • Load shedding available and confirmed where required
  • Control supply healthy

For sophisticated installations, add priority load status, bus voltage confirmation, and DG or BESS source status.

7. Motor Residual Voltage and Transfer Dead Time

This section addresses the most commonly omitted requirement in bus coupler design, and it matters enormously on any board feeding significant motor load.

The problem

When a bus feeding running induction motors is disconnected from its source, the motors do not stop. They continue rotating, driven by connected load inertia, and their trapped rotor flux causes them to act as generators. The bus retains a residual voltage that:

  • Decays over a period governed by the motors' open-circuit time constant and the load inertia
  • Slips in phase relative to the incoming alternative source, because the motors are decelerating and their frequency is falling

If the coupler is closed while the residual voltage is still substantial and significantly out of phase with Transformer-2, the vector difference between the two voltages is applied across the motor. The result can be a transient current and torque far exceeding a normal direct-on-line start — sufficient to damage shafts, couplings, gearboxes and motor windings, and to trip the incoming breaker on the transient inrush.

The three recognised transfer methods

Fast transfer. The transfer is completed within a very short interval — typically under about 10 cycles — before the residual voltage has slipped far in phase. A synchronism-check or phase-angle element supervises the closing. Requires fast breakers and a well-engineered scheme.

In-phase transfer. A controller predicts the instant at which the decaying motor residual voltage passes through phase coincidence with the incoming source and issues the close command with breaker-time compensation so contact make occurs at that instant.

Residual voltage transfer. The simplest and most common approach on general industrial LT boards. A deliberate dead time is enforced, and the coupler is not permitted to close until the bus residual voltage has decayed below a defined threshold — commonly in the region of 25% to 33% of rated voltage, confirmed by an undervoltage relay measuring Bus-A. Accepted industry guidance for motor bus transfer expresses the criterion as a limit on the resultant volts-per-hertz of the vector difference between the incoming source and the motor residual voltage.

What this means for the control philosophy

Step 10 of the transfer sequence cannot simply read "close bus coupler." It must read:

Close bus coupler after Bus-A residual voltage has been confirmed below the approved threshold by a dedicated undervoltage element, or after the approved dead time has elapsed, whichever criterion the project has adopted.

The threshold, the measuring device and the time delay must be specified in the control philosophy and verified during FAT with the actual relay settings.

The consequence nobody plans for: contactor drop-out

If a residual-voltage dead time is used, the transfer is not instantaneous. Feeder contactors held closed by AC coils will release when bus voltage falls to roughly 35% to 70% of rated — typically within a cycle or two.

So even a perfectly executed transfer leaves the motors stopped.

The design must decide, explicitly, how to handle this:

  • Accept it — motors stop, and an automatic or manual restart sequence follows. Acceptable for many processes.
  • Latched contactors — mechanically latched or DC-held contactors ride through the dead time. The motors then experience the re-energisation transient, so residual voltage checking becomes even more critical.
  • Undervoltage timers — timed UV relays permit ride-through of short interruptions with staggered auto-restart to limit simultaneous starting current.
  • Sequential restart — a defined restart order with time steps, so the surviving transformer is not hit with the simultaneous starting current of every motor on both buses.

That last point is a genuine hazard: if forty motors on two bus sections all attempt to restart simultaneously on a single transformer, the starting kVA can exceed the transformer's short-time capability and collapse the bus voltage — turning a successful transfer into a failed one.

In our FAT experience, this is where the approved control philosophy and the plant's actual expectation diverge most often. The scheme is signed off with a residual-voltage dead time of a few hundred milliseconds, and nobody has told the process team that every AC-coil contactor on the transferred section will drop out during that window. On a chemical plant PMCC we tested, the transfer logic was faultless and the operators still called it a failed transfer, because eleven pump feeders needed manual restart afterwards. Our recommendation: decide the post-transfer restart philosophy at the same time as the dead time, classify which feeders are auto-restart and which are manual, and stagger the auto-restart steps so the surviving transformer never sees more than a defined block of starting kVA at once.

8. "Source Healthy" Does Not Mean "Source Sufficient"

A worked example makes this immediate.

Installation

  • Transformer-1: 1000 kVA
  • Transformer-2: 1000 kVA

Normal loading

  • Bus-A: 750 kVA
  • Bus-B: 700 kVA

Under normal operation both transformers are comfortably within rating. The scheme looks correct.

Transformer-1 fails. INC-1 opens. The coupler closes.

Transformer-2 is now asked to supply 750 + 700 = 1450 kVA — 145% of its rating.

The interlocking was electrically correct. Every permissive was satisfied. The scheme still failed, because it verified breaker status and never verified capacity.

Bus coupler transfer logic must be coordinated with surviving-source capacity, not only with breaker status.

Firm capacity and the N+1 question

The engineering term for what matters here is firm capacity: the load that can be supplied with one source unavailable.

Two transformers do not automatically deliver redundancy. For full N+1 continuity without load shedding, each transformer must normally be loaded such that either one can carry the entire combined load — which in practice means neither exceeds approximately 50% loading, after allowing for ambient temperature, harmonic derating and overload capability.

If both transformers routinely run at 70–90%, the installation has two transformers but no redundancy. Closing the coupler after a failure will overload the survivor. The design must then include load shedding as an integral part of the transfer sequence, not as an optional extra.

Load shedding within the transfer sequence

Where firm capacity is insufficient, shedding must occur in the correct order:

  1. INC-1 opens
  2. Non-priority feeders on both buses are shed
  3. Residual voltage / dead-time criterion satisfied
  4. Coupler closes
  5. Transformer-2 supplies priority loads on both sections
  6. Sequential restart of shed loads only if capacity permits

Typical priority classification — always project-specific:

Priority Load category Action on transfer
1 Critical process, safety systems, emergency lighting Never shed
2 Essential utilities — instrument air, cooling water, control supply Retain if capacity permits
3 Comfort HVAC, non-critical services Shed
4 Production loads tolerant of interruption Shed first

The shedding must be fast — it has to complete before the coupler closes, otherwise the transformer sees the full combined load, even briefly, and its protection may operate.

9. Neutral, Earthing and Circulating Current

This section explains why the 3-pole versus 4-pole coupler decision cannot be made from the ampere rating.

The hidden parallel path

Consider the standard arrangement: two transformers with solidly earthed star points, 3-pole incomers, a 3-pole coupler, and a continuous neutral busbar running the full length of the switchboard.

Both transformer neutrals are connected to a common neutral bus. Both star points are earthed at their respective transformers.

This creates a closed loop:

T1 neutral → common neutral bus → T2 neutral → T2 earth → earth grid → T1 earth → T1 neutral

The two transformers are therefore already paralleled on the neutral — even in the "normal" state where the coupler is open and the phases are completely separated.

What flows in that loop

Triplen harmonics. Third, ninth and fifteenth harmonic currents are zero-sequence. They do not cancel in the neutral; they add. In an installation with significant single-phase non-linear load — LED lighting, switch-mode power supplies, UPS front ends — or with VFD-generated harmonics present, these currents find the low-impedance loop between the two transformer neutrals and circulate continuously.

Load unbalance currents. Zero-sequence current from unbalanced single-phase loading divides between the two neutral paths according to their relative impedance, rather than returning entirely to its own transformer.

Consequences:

  • Neutral conductor and neutral busbar carry current that no load calculation predicted
  • Additional transformer heating and losses in both units
  • Circulating current present even when the plant believes the sources are fully separated
  • Earth-fault protection measurement is corrupted — residual CT or core-balance CT measurement includes circulating current that is not a fault, causing nuisance trips or, worse, desensitisation
  • Restricted earth fault schemes can maloperate

The design responses

Four-pole coupler with switched neutral. The neutral is broken along with the phases, and each bus section retains its own neutral referenced to its own transformer. This is the cleanest solution where the sources are separately derived.

Sectionalised neutral busbar. The neutral bus is split at the same point as the phase bus, with a neutral link that follows the coupler.

Single-point earthing. Only one transformer neutral is earthed at any time, with the earthing arrangement switched together with the source configuration. This requires careful engineering — an unearthed system during transfer is its own hazard.

The correct choice depends on the earthing system (TN-S, TN-C-S, TT, IT), the earth-fault protection philosophy, whether the sources are separately derived, and whether a DG or inverter source is also connected. The same reasoning that governs 3-pole versus 4-pole selection in automatic transfer switches applies here in full.

The neutral and earthing architecture decides the coupler pole configuration. The ampere rating does not.

10. Rating the Coupler, the Incomer and the Busbar

A common specification error:

"The incomers are 2500 A, so specify a 2500 A coupler."

That is a coincidence, not a calculation. And it addresses only one of the three components whose rating changes during transfer.

The bus coupler

Size it on:

  • Maximum transferred current — the full load current of the bus section being transferred, at its worst-case loading, plus any load growth allowance
  • Prospective fault level in every permitted operating configuration
  • Icu / Ics appropriate to the operating philosophy — Ics matters where the board must remain serviceable after a fault
  • Icw where short-time delayed tripping is used for selectivity
  • Icm against peak asymmetrical current, including motor contribution
  • Duty cycle — a coupler in a frequently transferring installation has a different mechanical endurance requirement than one operated twice a year
  • Pole configuration per Section 9

The coupler may be required to carry the entire load of one bus section indefinitely. It is not a "transfer-only" device sized for a brief period.

The healthy incomer and its bus section

In the transferred state — say INC-2 closed, coupler closed, INC-1 open — the following carry the combined load of both bus sections:

  • Incomer-2 breaker
  • Bus-B busbar, along its full length
  • The bus-B side busbar run up to the coupler

If Incomer-2 and the Bus-B busbar were sized only for Bus-B's normal load, the switchboard is under-rated in exactly the condition it exists to handle. This must be checked against the busbar current-carrying and temperature-rise calculation for the transfer state, not the normal state.

Does the coupler need protection?

It depends on the architecture, and the answer must be deliberate.

A coupler ACB with an electronic trip unit gives an additional protection zone, but its settings must coordinate with both incomers, all downstream feeders, and the busbar withstand curve. Note that the coupler frequently needs different behaviour in different operating modes — its role during normal operation is not the same as during emergency transfer. Some schemes therefore use a second settings group activated by transfer status.

Whichever approach is chosen, it must come out of a protection coordination study covering all credible configurations — not from a default trip unit setting.

11. Protection Coordination Changes With Coupler Position

Consider a fault on Bus-A.

Coupler open, INC-1 closed: the fault is fed through Transformer-1 and by motors on Bus-A. Current flows in one direction through INC-1.

Coupler closed, INC-1 open: the same fault is fed through Transformer-2 and the coupler, plus motors on both sections. The current path, magnitude and direction have all changed.

This affects:

  • Fault magnitude at each protective device
  • Grading margins between coupler, incomer and feeder protection
  • Direction of power flow through the coupler
  • Which device should operate first
  • Arc flash incident energy at each work location

The protection study must therefore cover every permitted operating configuration, and where the settings that give correct selectivity in one configuration are wrong in another, a settings group changeover linked to breaker status is required. A single fixed setting for a board that operates in three distinct configurations is an accepted-by-default compromise, not a design.

Where the coupler is expected to provide selectivity against downstream devices, the principles of selectivity, discrimination and cascading apply directly, and manufacturer coordination tables must be checked for the specific device combinations used.

12. Return to Normal, and the Closed Transition Question

Return-to-normal sequence

Transformer-1 has returned. Current state: INC-1 open, coupler closed, INC-2 closed.

An open-transition return sequence:

  1. Confirm Transformer-1 supply healthy — correct voltage, frequency, phase sequence
  2. Wait the approved stabilisation delay — this prevents oscillation on an unstable supply
  3. Confirm no protection lockout on INC-1
  4. Open the bus coupler
  5. Confirm coupler physically open
  6. Confirm Bus-A residual voltage criterion satisfied
  7. Close INC-1
  8. Restore any shed load in the defined sequence

Bus-A experiences a second interruption. This is inherent to open-transition transfer, and it is why return-to-normal is frequently specified as manual only — a planned operation at a convenient time, rather than an automatic one at an arbitrary moment.

Whether return is automatic or manual is a business decision, not a technical one. It depends on the cost of the second interruption versus the risk of prolonged operation on a single transformer.

Can the return be seamless?

Only with a different architecture.

A genuinely uninterrupted return requires closed transition — both sources briefly paralleled, then one opened. That requires:

  • Synchronism check across the open coupler
  • Voltage magnitude matching
  • Frequency matching
  • Phase angle within limits
  • A defined maximum parallel time, typically fractions of a second, with a fail-safe trip
  • Protection rated for the momentary paralleled fault level
  • Explicit permission for parallel operation, including all the transformer paralleling conditions in Section 3

And critically:

A mechanical interlock designed never to permit all three breakers closed will physically prevent closed transition.

Open-transition 2-out-of-3 interlocking and closed-transition paralleling are mutually exclusive architectures. This decision must be made at the specification stage, before the interlocking kit is ordered. It cannot be retrofitted by changing the PLC program. The relationship between these transfer philosophies is covered further in AMF vs ATS vs DG synchronisation.

13. Bus Status, Multiple Sources and Control Modes

Never assume an open incomer means a dead bus

The logic "INC-1 is open, therefore Bus-A is dead" is unsafe in any modern installation.

Bus-A may remain energised from:

  • A diesel generator connected to that section
  • A UPS with a backfeed path
  • A solar inverter or BESS connection
  • Motor residual voltage (Section 7)
  • Another interconnection elsewhere in the plant

Two consequences follow. First, bus voltage must be measured, not inferred. Second, if both sides of the coupler are energised from different sources, closing the coupler creates unintended paralleling of those sources — with all the consequences of Section 3.

Define the complete source matrix

Modern plants may contain transformers, DG sets, solar inverters and battery storage on the same board. Before finalising the interlocking, document every source that can energise each bus section, and define the permitted and prohibited combinations for all of them. The 2-out-of-3 matrix is then a subset of a larger source matrix.

Control modes

Define behaviour explicitly for:

Mode Automatic transfer Interlocks Command source
Auto / Remote Enabled Fully active PLC / SCADA
Auto / Local Enabled Fully active PLC, local initiate
Manual / Local Disabled Fully active Panel front
Maintenance Disabled Fully active + permit lock Restricted

The essential rule:

Selecting MANUAL must never disable the safety interlocks. Manual means the operator chooses the sequence — not that the prohibited states become permitted.

Feedback and failure handling

The PLC must never assume that a command was executed. Use actual breaker feedback: open status, closed status, trip status, racked position (service / test / disconnected), spring charged, ready-to-close. Transfer logic must be state-based, driven by confirmed positions.

Failure-to-close: if the coupler is commanded closed and does not close, the system must detect the timeout, annunciate, abort or hold the sequence, and await operator intervention. It must never repeat close commands indefinitely — repeated close attempts onto a fault destroy the breaker.

Failure-to-open: if INC-1 is commanded open and does not open, the coupler must be blocked from closing in a non-parallel scheme. This is the exact condition the interlock exists for, and the PLC must reinforce it rather than override it.

14. Standards and Compliance

Where the arrangement forms part of a low-voltage switchgear and controlgear assembly, the governing framework is:

Standard Scope
IEC 61439-1 General rules for LV switchgear and controlgear assemblies
IEC 61439-2 Power switchgear and controlgear assemblies (PSC-assemblies)
IS/IEC 61439 Part 1 and Part 2 Indian adoptions for projects in India
IEC 60947-2 Circuit breakers — Icu, Ics, Icw, Icm definitions and testing
IEC 60947-4-1 Contactors and motor starters
IEEE C50.41 Referenced guidance for motor bus transfer criteria

For assemblies claiming compliance, the distinction between design verification and routine verification is essential — an interlocking scheme is verified by routine test on every panel, while the assembly's temperature rise and short-circuit withstand are established by design verification. This is covered in detail in IEC 61439 design verification vs routine verification.

Standard edition dates should be confirmed against the current published versions applicable to your project at the time of specification.

15. Common Design Mistakes

1. PLC-only interlocking. Software can be modified, corrupted or bypassed. Provide independent mechanical and hardwired layers for the critical prohibited state.

2. Permitting all three breakers to close accidentally. Creates unintended transformer paralleling with all the fault-level and load-sharing consequences.

3. No surviving-transformer capacity check. The transfer succeeds and immediately overloads the healthy transformer.

4. Assuming an open incomer means a dead bus. Other sources and motor residual voltage may energise it.

5. Ignoring motor residual voltage. Out-of-phase reconnection damages motors, shafts and couplings.

6. Ignoring motor contribution to fault level. Short-circuit study understates peak duty on breakers and busbar.

7. Treating a protection trip as a source failure. Transfers a healthy source into an uncleared fault.

8. Same protection settings for every operating state. Selectivity that is correct in one configuration is wrong in another.

9. Neutral and earthing decided from ampere rating. Creates circulating current and corrupts earth-fault protection.

10. Sizing only the coupler for transfer. The healthy incomer and its busbar carry the combined load too.

11. No failure-to-close / failure-to-open logic. Automatic sequences require abnormal-condition handling.

12. Manual mode bypassing interlocks. Manual must never mean unsafe.

13. No engineered return-to-normal sequence. Transfer back deserves the same rigour as transfer out.

14. Treating closed transition as simple bus coupling. Temporary paralleling is a different architecture requiring synchronism check and explicit permission.

16. FAT Checklist for 2 Incomers + Bus Coupler

A meaningful Factory Acceptance Test verifies every valid state and attempts every invalid one.

Interlocking verification

  • INC-1 and INC-2 close, coupler remains open — permitted
  • INC-1 and coupler close, INC-2 blocked — permitted state, third breaker blocked
  • INC-2 and coupler close, INC-1 blocked — permitted state, third breaker blocked
  • Attempt all three closed from every possible order of operation — third breaker must not close
  • Verify mechanical interlock independently, with control supply removed
  • Verify hardwired electrical interlock independently, with PLC in stop

Transfer sequence

  • Simulate Source-1 failure — verify complete sequence including time delays
  • Simulate Source-2 failure — verify reverse sequence
  • Verify residual voltage / dead time criterion is enforced before coupler closes
  • Verify load shedding operates before coupler closure, in the correct priority order
  • Verify return-to-normal sequence and stabilisation delay

Failure modes

  • Failure to open INC-1 — coupler close must be blocked
  • Failure to close coupler — timeout, alarm, sequence abort, no repeat commands
  • Protection trip on INC-1 — transfer must be blocked and lockout annunciated
  • Busbar differential operation — both incomer and coupler locked out
  • PLC power removed — verify approved fail-safe behaviour and that manual operation remains interlocked
  • Control supply failure — verify breaker states maintained

Modes and indication

  • Auto / Manual / Local / Remote — verify command ownership and that interlocks remain active in all modes
  • Maintenance mode — verify inhibits
  • Verify all indications: breaker open, closed, tripped, racked position, spring charged, source healthy, bus live, auto/manual, transfer in progress, transfer failed, protection lockout

Every item in this list should be recorded with a pass/fail result and witnessed. The complete methodology is set out in the electrical panel FAT and routine test checklist.

How Wisdom Techno Solutions Engineers a 2 Incomer + Bus Coupler PCC

At Wisdom Techno Solutions, a 2-incomer-plus-coupler board is not three electrically operated ACBs wired into a common enclosure. It is a power transfer system, engineered around an approved operating philosophy that is documented before manufacturing begins.

Wisdom Techno Solutions is a licensed partner for three design-verified switchgear platforms: Rittal Ri4Power (Rittal Solutions Partner since 2021 — the Ri4Power system carries Intertek ASTA design verification to IEC 61439-1 and -2 Edition 3, with copper and aluminium busbar systems up to 6300 A), C&S CX (Licensed Partner since 2023, up to Form 4B Type II) and Siemens SIEPAN Elite/8PU (IEC 61439 Solutions Partner since 2024). We build to each system's verified configuration and routine-verify every assembly in-house at our Vadodara, Gujarat facility, for industrial, EPC and infrastructure projects across India. That places the transfer-state busbar loading inside a verified design envelope.

What we build into these boards

  • Sectionalised busbar assemblies engineered and verified per IEC 61439, with the transfer-state loading verified in the busbar calculation — not only the normal state
  • Mechanical interlocking kits for two-source-plus-coupler arrangements, supplied and set up for the approved state matrix
  • Hardwired electrical interlocking as an independent second layer, functional with the PLC out of service
  • PLC and HMI-based automatic transfer with full permissive logic, residual voltage supervision, load shedding, transfer-failure alarming and defined PLC-failure behaviour
  • Protection lockout logic that blocks transfer on protection operation — hardwired, not software-dependent
  • Bus voltage sensing and complete breaker status monitoring including racked position
  • Neutral and earthing arrangements engineered from the system earthing philosophy, with 3-pole or 4-pole coupler selection justified
  • SCADA, BMS and Modbus/Profibus integration for remote monitoring and event logging
  • Complete interlock matrix, control philosophy and FAT documentation as project deliverables

What we need from you

To engineer the scheme correctly rather than assume it, we work from:

  • Single line diagram
  • Transformer ratings, impedances, vector groups and tap arrangements
  • System fault level and short-circuit study
  • Normal load per bus section and emergency/priority load list
  • Motor list where significant motor load is connected
  • Parallel operation permitted or prohibited
  • Earthing system and earth-fault protection philosophy
  • Required transfer type — open or closed transition, auto or manual return
  • Load shedding requirement and priority classification
  • Control interface requirements

We manufacture LT and HT control panels — PCC, MCC, APFC, VFD and PLC automation panels — from our facility at Kamrol, Vadodara, Gujarat, working to an engineer-to-order model for industrial, EPC and infrastructure clients.

Send us your single line diagram and transformer data. We will review your bus coupler control philosophy, identify the capacity, protection and interlocking gaps before they reach the shop floor, and quote a panel engineered to the operating states your plant actually needs.

If you are still at the specification stage, our LT panel technical specification and RFQ guide sets out the complete input list to define before floating the enquiry.

Conclusion

A 2 incomer plus bus coupler arrangement occupies three symbols on a single line diagram and controls the most consequential switching in the plant.

The 2-out-of-3 principle is the foundation:

  • INC-1 + INC-2 closed, coupler open — permitted
  • INC-1 + coupler closed, INC-2 open — permitted
  • INC-2 + coupler closed, INC-1 open — permitted
  • All three closed — prohibited, unless the system is deliberately engineered for parallel operation

But correct interlocking is only the entry requirement. A complete design must additionally verify surviving-source capacity, motor residual voltage before reconnection, protection lockout before transfer, fault level in every configuration, neutral and earthing architecture, busbar and incomer rating in the transferred state, protection coordination across all operating states, and defined behaviour for every failure mode.

The design question is never "can the bus coupler close?"

It is: under exactly which system conditions should it be permitted to close, and what must be proven before that command is issued?

Answering that question completely is what turns a breaker interlock into a reliable power transfer system.

Related Guides

  • Wisdom Techno Solutions
  • Reviewed by Ankit Patel

Frequently Asked Questions

What is 2-out-of-3 interlocking in a bus coupler panel?

It permits any two of three breakers — two incomers and one bus coupler — to be closed simultaneously, while preventing all three from closing together. This stops the two transformers from being paralleled through the LV bus in a system not designed for parallel operation.

Why is the bus coupler normally kept open?

In the common operating philosophy, an open coupler allows two transformer-fed bus sections to operate independently, limits fault level on each section, and simplifies fault isolation and maintenance. Whether the coupler should normally be open or closed is a project-specific design decision, not a universal rule.

Can both incomers and the bus coupler be closed at the same time?

Only if the installation is deliberately designed for parallel source operation, with matched transformer voltage ratio, vector group, phase sequence and impedance, and with breakers, busbar and protection rated for the paralleled fault level. A standard 2-out-of-3 interlocking scheme prevents this state.

What happens if all three breakers close in a system not designed for it?

The transformers become paralleled through the common bus. Load sharing becomes impedance-driven rather than proportional to rating, prospective fault current rises, and breaker and busbar duty may be exceeded.

Can one transformer supply both bus sections?

Yes, provided the surviving transformer has sufficient firm capacity for the combined load, and the incomer and busbar on that section are rated for the transferred current. Two transformers alone do not create redundancy — each must be able to carry the required load with the other unavailable.

What if one transformer cannot carry the entire plant load?

Load shedding must be integrated into the transfer sequence, shedding non-priority feeders before the coupler closes. The alternative is accepting partial supply or a different redundancy architecture.

Should bus coupler interlocking be mechanical or PLC based?

A robust design uses all three layers: mechanical interlocking to physically prevent the prohibited combination, hardwired electrical interlocking to block invalid close commands, and PLC logic for sequence, timing and supervision. The critical prohibited state should not depend on software alone.

Can a PLC failure allow all three breakers to close?

Not in a properly engineered scheme. Mechanical and hardwired interlocking must remain effective with the PLC out of service. PLC failure should disable automatic transfer, maintain existing breaker states, annunciate, and permit only interlocked manual operation.

Does closing the bus coupler increase the fault level?

It changes fault current paths, and where sources become paralleled it increases the available fault contribution. Motor contribution from both bus sections also adds to the total. The actual value requires a short-circuit study covering every permitted configuration.

Why does motor residual voltage matter during bus transfer?

Running motors continue to generate a decaying residual voltage that slips in phase relative to the incoming source. Closing the coupler while that voltage is high and out of phase applies a large vector difference across the motors, producing severe transient torque and current capable of damaging shafts, couplings and windings.

Should a 3-pole or 4-pole bus coupler be used?

The decision is driven by the system earthing arrangement and earth-fault protection philosophy, not the ampere rating. A solidly linked neutral across a 3-pole coupler parallels the two transformer neutrals, allowing triplen harmonic and unbalance currents to circulate and corrupting earth-fault measurement.

Can bus transfer be made seamless with no interruption?

Not with a basic open-transition 2-out-of-3 scheme, which always produces an interruption. Uninterrupted transfer requires closed transition with synchronism check and temporary paralleling — a different architecture that a mechanical 2-out-of-3 interlock will physically prevent.

Does a bus coupler breaker need its own protection?

It depends on the architecture. Where provided, coupler protection settings must coordinate with both incomers, all downstream feeders and the busbar withstand curve, and may require different settings groups for normal and transfer operating states. This must come from a protection coordination study.

What must be tested during FAT for a bus coupler panel?

FAT must verify every permitted breaker state and must also attempt every prohibited state, from every possible order of operation, to confirm the third breaker cannot close. Beyond interlocking, the test schedule should cover the complete transfer sequence in both directions with actual time delays, enforcement of the residual voltage or dead-time criterion before coupler closure, load shedding in the correct priority order, protection lockout blocking transfer, failure-to-open and failure-to-close handling, and defined PLC-failure behaviour with manual operation still interlocked. The mechanical interlock should be proved independently with the control supply removed, and the hardwired electrical interlock proved with the PLC in stop — otherwise a scheme that is really PLC-only can pass as a three-layer design. Every item needs a recorded pass/fail result against the approved interlock matrix.