ABB Emax 2 Selectivity and Coordination with Downstream Breakers Guide

What is ABB Emax 2 selectivity? ABB Emax 2 selectivity is the coordinated protection strategy under IEC 60947-2 that ensures only the breaker closest to a fault clears it, relying on the Ekip trip unit's adjustable short-time delay (S-delay, typically 40–500 ms) and the Emax 2's rated breaking capacity up to 150 kA to maintain system continuity. Miscalculated time-current grading or ignored let-through energy between the Emax 2 and downstream MCCBs causes unwanted upstream tripping, extended fault exposure, and potential busbar damage. This guide covers IEC 60947-2 selectivity definitions, Ekip trip unit parameter configuration, step-by-step MCCB coordination calculations, ABB SACE coordination table interpretation, zone-selective interlocking for total selectivity, and S-delay versus downstream clearing time verification.

Published: 2026-04-27 | Last updated: 2026-04-27 | By Stoklink Engineering Team

Selectivity is one of those topics that looks straightforward on a one-line diagram and turns into a three-day argument the moment you open the trip-unit settings. We have seen consultants specify "full selectivity" on a tender drawing without ever calculating the I²t crossover point, then discover during commissioning that a 50 kA bolted fault on a downstream busbar trips the 4000 A main and dumps the entire MV transformer. The standards exist for a reason. So do the manufacturer coordination tables. This guide pulls them together.

What selectivity actually means under IEC 60947-2

Before diving into ABB Emax 2 settings, definitions matter. IEC 60947-2 Clause 2.5.23 defines two grades of selectivity that every engineer should be able to recite from memory.

In practice, total selectivity is rare in real LV networks above 25 kA prospective. What we typically see in the field is partial selectivity engineered to a Is value comfortably above the maximum credible fault at the downstream busbar — and that is perfectly acceptable, provided the calculation is documented.

Time-current selectivity vs. energy-based selectivity

There are two distinct mechanisms at work, and engineers often conflate them.

Time-current selectivity relies on the upstream device having a longer tripping time than the downstream device for the same fault current. This is what the trip-unit time-delay bands are designed to deliver. It works well for overload and low-level short-circuit currents where both devices are operating in their thermal or short-time delay regions.

Energy-based selectivity (also called dynamic or current-limiting selectivity) takes over at high fault currents where the downstream device — typically a current-limiting MCCB like a Tmax XT — clears the fault so quickly (often under 5 ms) that the upstream Emax 2 never sees enough let-through energy to trip. This is the regime where the I²t curves matter more than the time-current curves.

For complete trip unit settings, coordination tables, and verified selectivity limits applicable to the ABB Emax 2 range, refer to the manufacturer's SACE Emax 2 technical catalogue.

How the Emax 2 Ekip trip unit enables selectivity

The Ekip family of trip units on the ABB Emax 2 — Dip, Touch, Hi-Touch, and G versions — supports four protection functions referenced by their letters: L (long-time, thermal-equivalent overload), S (short-time delay), I (instantaneous), and G (earth fault). Selectivity engineering is essentially the art of choosing the right combination and tuning the pickup and delay parameters.

L, S, I, G functions and what they cost you

An LI trip unit (long-time + instantaneous) is the simplest and cheapest. It is also the worst for selectivity because the I function trips with no intentional delay above the I-pickup setting, typically 30–40 ms total clearing time. If a downstream MCCB has the same instantaneous clearing time, you cannot guarantee selectivity at high fault levels with time grading alone.

An LSI or LSIG trip unit adds the S function, which is a definite-time or I²t-shaped delay band that lets you insert 100, 200, 400 ms of intentional delay before the upstream breaker trips. This is what makes time-graded selectivity possible. The cost is that the Emax 2 itself must be rated for the prospective fault current for the duration of the delay — its Icw (rated short-time withstand current) must equal or exceed Ik for at least the S-delay time.

For applications where time-graded selectivity is essential, we typically specify the LSI variant. The ABB 1SDA070782R1 E1.2B 1000 Ekip Dip LSI at 1000 A 42 kA is a common choice for sub-main applications. For pure feeder protection where downstream current-limiting MCCBs handle the high-end faults, an LI version like the ABB 1SDA070701R1 E1.2B 630 Ekip Dip LI is often sufficient and saves cost.

Why Icw matters for the upstream breaker

This is where engineers often overlook a critical sizing constraint. If you set the S-delay on an Emax 2 to 400 ms to achieve selectivity with a downstream device, the breaker must withstand the full prospective fault current for those 400 ms without damage. Per IEC 60947-2 Clause 4.3.5.4, the Icw is verified at 0.5 s or 1 s for ACBs, and the manufacturer publishes both values.

The E1.2B family carries an Icw of 42 kA for 1 s. The E2.2B is 50 kA for 1 s. The E4.2 and E6.2 frames go up to 100 kA for 1 s. If your prospective fault at the Emax 2 location is 65 kA, an E1.2B will not survive a 400 ms intentional delay — you need an E2.2 or E4.2. For more on frame selection by Icw, see our Emax 2 full technical specifications guide.

Step-by-step coordination calculation with downstream MCCBs

The methodology we use on real ABB Emax 2 projects looks like this. It is not glamorous. It is repeatable.

Step 1: Establish the prospective short-circuit currents. Calculate Ik3-max (three-phase symmetrical) at the Emax 2 location and at every downstream busbar. Use the IEC 60909 method, not back-of-envelope estimates. For a typical 2000 kVA 11/0.4 kV transformer with 6% impedance, Ik3-max at the LV terminals is around 48 kA.

Step 2: Identify all downstream devices in series. List the trip type, frame size, In setting, and published Icu/Ics.

Step 3: Choose the Emax 2 frame and trip unit. The frame must satisfy In ≥ 1.0 × IB (load current) and Icu ≥ Ik3-max. For an 1800 A continuous load on a 50 kA bus, the ABB 1SDA071021R1 E2.2B 2000 Ekip Dip LI at 2000 A 42 kA is a candidate, but if Ik exceeds 42 kA you step up to E2.2N (65 kA) or E2.2H (85 kA).

Step 4: Set the L function above the maximum load and below the cable damage curve. Typical L1 = 0.9 × In, t1 chosen to coordinate with downstream thermal curves at 6 × In.

Step 5: Set the S function with adequate delay margin. A common rule of thumb is t2 ≥ t_downstream + 100 ms safety margin. If the downstream MCCB clears in 80 ms maximum at 20 kA, set the Emax 2 S-delay to at least 200 ms.

Step 6: Verify against the ABB SACE coordination tables. This is the non-negotiable step. The published tables in the ABB document "Selectivity tables — Emax 2 / Tmax XT / Tmax T" give the verified Is values for every combination of Emax 2 frame and downstream MCCB. They are based on physical type tests, not curve overlay.

Worked example: 2500 kVA data centre MDB

Consider a hyperscale colocation site with a 2500 kVA 415 V transformer feeding a main distribution board. The MDB has one main incomer and twelve outgoing feeders to PDUs and mechanical loads.

Calculated Ik3-max at the MDB busbar: 52 kA. Continuous load: 2900 A peak, 2400 A typical.

We selected an E4.2N 4000 A with Ekip Touch LSIG, Icw = 65 kA / 1 s. Outgoing feeders use Tmax XT5N 630 A MCCBs with Ekip LSI trip units. Settings:

• Emax 2 main: L1 = 0.85 (3400 A), L2 = 12 s; S1 = 5 × In (20 kA), S2 = 0.3 s with I²t off; I disabled (no instantaneous on the main, since we have full Icw withstand)
• Tmax XT5N feeders: L1 = 1.0 (630 A); S1 = 8 × In (5040 A), S2 = 0.1 s; I = 12 × In (7560 A)

The 200 ms gap between the XT5N S-delay (100 ms) and the Emax 2 S-delay (300 ms) gives total selectivity from overload up to roughly 25 kA. Above 25 kA, the XT5N's current-limiting action plus the published ABB coordination table value (Is = 50 kA total) confirms selectivity to the full prospective fault. The case for choosing this approach in critical-uptime applications is detailed in our Emax 2 in data centres design guide.

Reading and using the ABB SACE coordination tables

The ABB SACE selectivity tables are organised by upstream frame, downstream frame, and trip-unit type. Each cell gives a number in kA, which is the verified Is — the prospective fault current up to which the coordination is selective. A value of "T" means total selectivity (Is ≥ Icu of the downstream device).

How to interpret a table cell

Take an example. Upstream: E2.2B 1600 Ekip LSI. Downstream: Tmax XT3N 250 with Ekip LSI. The table gives Is = 50 kA (T). This means at any fault up to 50 kA at the downstream busbar, the XT3N clears the fault and the E2.2B does not trip — provided both are set within the parameters specified in the footnotes (typically L1 ≤ 1.0, S1 within tabulated ranges).

Some engineers argue that you can extrapolate between cells, but in our experience this is dangerous. The tables are based on type tests of specific combinations. A coordination value verified for an LSI/LSI pair does not transfer to an LSI/LI pair, because the absence of the S function on the upstream device changes the dynamics entirely.

When the table says "—" or "no data"

This usually means the combination has not been type-tested or that selectivity is not achievable at the relevant fault levels. Do not interpret a missing value as "probably fine". A common mistake is to assume that a larger upstream frame is always selective with a smaller downstream frame; this is true for overload but can fail at high fault levels where let-through energy from the upstream breaker exceeds the downstream device's I²t withstand during the delay.

Zone-selective interlocking (ZSI) for fast, total selectivity

When time-graded selectivity demands an unacceptably long S-delay — say, 500 ms across three levels of breakers — Zone Selective Interlocking on the ABB Emax 2 offers a way out. ZSI is a hardwired (or, in newer Ekip versions, IEC 61850 GOOSE-based) communication scheme where each breaker signals its upstream parent the moment it sees a fault.

How ZSI works in practice

Imagine three Emax 2 in cascade: incomer (E6.2), section (E4.2), and feeder (E2.2). All three see the same fault current at, say, 40 kA. Without ZSI, the only way to ensure the feeder trips first is to delay the section by 200 ms and the incomer by another 200 ms — meaning the incomer holds 40 kA for 400 ms, which costs you frame size and Icw rating.

With ZSI, the feeder breaker sends a "fault detected" signal upstream within 5–10 ms. The section and incomer breakers see the signal and revert to their long delay. If the feeder fails to clear, after a short timeout (typically 50 ms) the section breaker takes over with no further delay. The incomer trips only if both downstream zones fail.

The result: total selectivity with the upstream breaker holding fault current for a maximum of ~80 ms, not 400 ms. ZSI is now standard on Ekip Touch and Hi-Touch trip units and is wired through the Z-IN and Z-OUT terminals on the trip unit terminal strip.

When ZSI is worth the wiring effort

ZSI requires shielded cable between every pair of cascaded breakers, plus careful commissioning. For two-level coordination it is usually overkill. For three or more levels, or when the prospective fault exceeds Icw at the required delay, ZSI is essential. We always recommend it for hospital essential supplies, data centre A/B feeds, and continuous-process plants where any incomer trip is a major incident.

Coordination calculator: Emax 2 S-delay vs. downstream clearing time

Comparison: selectivity strategies for different system sizes

Common coordination mistakes and how to avoid them

Over fifteen years of ABB Emax 2 protection studies, the same five mistakes recur on roughly 80% of audits.

Mistake 1: Setting S-delay too short to gain "speed". Engineers under pressure from clients who fear "slow protection" sometimes drop the upstream S-delay to 80 ms. This breaks selectivity because the downstream device's worst-case clearing time at low fault levels (within its short-time band) can be 100 ms or more. Speed is not a virtue if it costs you selectivity. The breaker is sized for Icw at the chosen delay — use it.

Mistake 2: Ignoring the I²t-on/off setting. The Ekip S-function has both definite-time and I²t-shaped modes. When coordinating with downstream thermal-magnetic MCCBs, I²t-on usually gives better matching of the curves at low currents. Against current-limiting MCCBs, definite-time (I²t-off) is preferred. The default is not always correct.

Mistake 3: Using LI trip units in cascaded ACB applications. If two Emax 2 are in series — for example a tie breaker and a main — both with LI trip units, you have no time-graded selectivity at all above the I-pickup. The two devices will race, and physics will decide the winner. Always specify LSI on at least the upstream device in any cascaded ACB arrangement.

Mistake 4: Forgetting the earth-fault G function. Earth-fault selectivity is a separate calculation. The G function on Ekip trip units has its own pickup and delay, and it must be coordinated independently from the phase functions. A common oversight is to set G with no delay on the main breaker, then wonder why a downstream earth fault trips the incomer instead of the feeder. The G-delay on the upstream device should be at least 100 ms longer than any downstream G or RCD delay.

Mistake 5: Trusting curve-overlay software without consulting the coordination tables. Curve-overlay tools draw the published time-current bands and check for visual gaps. They do not account for let-through energy, breaker mechanical opening time variability, or the dynamic interaction between current-limiting devices and slower upstream breakers. The ABB SACE coordination tables are based on physical type tests and represent the only authoritative selectivity values above the instantaneous threshold.

Coordination with MCBs and final-circuit devices

The principles change slightly when the device downstream of the ABB Emax 2 is a miniature circuit breaker (MCB) rather than an MCCB. MCBs per IEC 60898-1 have fixed magnetic characteristics — Type B (3–5 × In), Type C (5–10 × In), or Type D (10–20 × In) — and tripping times that are statistically defined rather than precisely settable.

Selectivity tables for MCBs

ABB publishes separate selectivity tables for Emax 2 / Tmax XT upstream and S200/S290 series MCBs downstream. The Is values are typically lower than for MCCB combinations because MCBs have shorter total clearing times at high faults — they are highly current-limiting — but they cannot be coordinated by time delay since their trip curves are fixed.

For a typical industrial sub-distribution where MCBs feed final circuits, energy-based coordination is the only viable approach. The upstream Emax 2 trip unit cannot be time-delayed enough to wait for the MCB without exceeding its Icw, but the MCB's let-through I²t at high faults is so low that the upstream breaker's electronic trip never registers a current high enough or long enough to operate. You can browse the full Miniature Circuit Breaker range at Stoklink for compatible downstream devices.

RCDs and earth-fault coordination

Where residual current devices are part of the chain — common in commercial buildings and machine tools — coordination becomes a three-dimensional problem: phase overcurrent, earth fault on the Emax 2 G function, and residual current on the RCD. Time-delayed (Type S, selective) RCDs are essential here, with a minimum 40 ms delay to coordinate with downstream instantaneous RCDs. See the available Residual Current Device options at Stoklink for selective Type S models.

Generator and dual-source applications

Coordination of the ABB Emax 2 on generator-backed networks introduces a complication that catches many engineers off-guard. A typical 1500 kVA standby generator has a sub-transient reactance Xd" around 12–14%, giving a prospective fault current of roughly 18–20 kA at the generator terminals — far less than the 50–65 kA you might see on the utility transformer side.

This means selectivity that works on mains supply may fail completely on generator supply, because the fault current may be below the upstream Emax 2's I-pickup or even below the S-pickup. The downstream MCCB will trip in its long-time region; the upstream Emax 2 will also trip in its long-time region; whichever trips first depends on the exact L1 and t1 settings, not on a clean time grading.

The fix is to use Ekip trip units with dual settings, available on the Ekip Touch and Hi-Touch versions. The G command (input from a generator-running contact) switches the trip unit to a second parameter set with lower L and S pickups suitable for the reduced fault duty. This feature is one of the differentiators we cover in our Emax 2 vs Schneider MasterPact MTZ comparison; both manufacturers offer the function but the implementation and commissioning workflow differ.

Verification and commissioning: how to prove selectivity is real

Documentation is one thing. Verified behaviour in the field is another. We always insist on three commissioning steps before energising a new switchboard.

Step 1: Settings verification. Use the Ekip Connect software to read back every parameter from every trip unit and compare it against the protection study. Print the report and have it counter-signed.

Step 2: Secondary injection testing. Inject simulated currents into the Ekip trip unit at the L, S, and I pickup thresholds and measure the actual trip times. The Emax 2 trip-unit accuracy is ±10% on pickup and ±10% on time, per IEC 60947-2. Verify each function falls within tolerance.

Step 3: Primary injection (where feasible). On critical projects we run primary injection at 5–10 × In through the breaker poles to verify the entire trip chain — current sensors, trip unit, opening solenoid, mechanical operation. This is the only way to catch wiring errors in the trip-unit harness or wrong-way CT installation.

For projects where commissioning reveals issues like nuisance tripping during normal switching transients, the diagnostic methodology in our Emax 2 nuisance tripping causes and solutions article walks through the most common root causes and fixes.

Sizing the Emax 2 with selectivity in mind

One final point that ties this all together. Selectivity drives breaker sizing as much as load current does. Two examples make the point.

A 1000 A continuous load on a 50 kA bus could in principle be served by an ABB 1SDA070781R1 E1.2B 1000 Ekip Dip LI with Icu = 42 kA — except 42 kA < 50 kA, so we need either E1.2N (50 kA) or to verify that downstream current limiting reduces the duty. If selectivity to a downstream device requires a 300 ms S-delay, we need Icw ≥ 50 kA for 0.3 s, which on E1.2 frame is satisfied (Icw 42 kA at 1 s, derated). On a 65 kA bus, the same load might force us up to E2.2H 1600 frame just for Icw — paying for selectivity in copper, not just silicon.

The complete sizing methodology, including the interaction of load current, fault duty, and selectivity requirements, is laid out step-by-step in our Emax 2 sizing calculator guide. For the underlying product family overview, see What Is the ABB SACE Emax 2?.

For larger frames where Icw rather than Icu drives the decision, common options at Stoklink include the ABB 1SDA070981R1 E2.2B 1600 Ekip Dip LI and the ABB 1SDA070741R1 E1.2B 800 Ekip Dip LI. The complete Air Circuit Breakers range covers the full Emax 2 family from 630 A to 6300 A. Auxiliary protection devices like control Relays are also available for trip-unit auxiliary circuits.

Related Reading

• ABB Emax 2 Full Technical Specifications: Current Ratings, Breaking Capacity and Dimensions
• How to Size ABB Emax 2: Step-by-Step Calculator for LV Distribution Panels
• ABB Emax 2 in Data Centers: MDB Design, Redundancy and Uptime Considerations
• ABB Emax 2 Nuisance Tripping: Root Causes, Diagnostic Steps and Fixes

Conclusion

Selectivity engineering on Emax 2 networks is neither black art nor pure arithmetic. It is the disciplined application of three tools: the Ekip trip-unit functions (L, S, I, G), the IEC 60947-2 framework that defines what "selective" actually means, and the ABB SACE coordination tables that provide verified Is values for every published combination. Get those three right and you will deliver protection schemes that hold up in commissioning and in the field.

The mistakes that we see most often — under-delayed S settings, missing G coordination, ignoring generator-mode fault duty, trusting curve overlay above the instantaneous threshold — are all avoidable with a careful protection study and disciplined commissioning. The cost of getting it wrong is measured in unplanned downtime, and on a modern industrial site that cost dwarfs the price of the breaker itself.

For the full selection methodology, fault current calculation, trip-unit configuration, and maintenance procedures across the entire Emax 2 family, see our master ABB SACE Emax 2 selection, application and maintenance guide and the foundational air circuit breaker engineering guide. For procurement of specific frames and trip-unit variants discussed in this article, the Air Circuit Breakers collection at Stoklink lists current stock and lead times.