Protection Discrimination: Why Your MCB Trips Instead of the Fuse

In 2019, a fault on a single 20A lighting circuit in a concentrator building tripped the 400A ACB feeding the entire floor's distribution board. The concentrator went dark. The process plant shut down. The root cause wasn't the fault itself — it was a failure of protection discrimination. The 20A downstream MCB and the 400A upstream ACB had overlapping time-current characteristics at the fault level, and the ACB operated first.

This cost approximately 4 hours of lost production. The investigation revealed that nobody had checked the discrimination between these two devices at the actual prospective fault current. They assumed — as most engineers do — that a smaller device always trips before a larger one.

What Discrimination Actually Means

Protection discrimination (called "selectivity" in IEC standards and NEC) means that only the protective device immediately upstream of the fault operates to clear it. All other devices remain closed, maintaining supply to healthy circuits.

BS 7671, Regulation 536.1 — Discrimination between protective devices

Why Discrimination Fails at High Fault Currents

At low fault currents (overload range), discrimination is easy. A 20A MCB trips at 20–28A. A 63A fuse doesn't even begin to melt until 70A+. Clear separation — the MCB trips first.

The problem occurs at high fault currents, in the instantaneous or magnetic trip region of the MCB. Here, two factors cause discrimination failure:

1. Time-Current Curve Overlap

A Type B MCB trips instantaneously at 3–5× its rating. A Type C trips at 5–10×. This means:

• A 20A Type B MCB trips instantaneously at 60–100A
• A 20A Type C MCB trips instantaneously at 100–200A

Meanwhile, a 63A BS 88 fuse at these same current levels may take 0.01–0.1 seconds to clear. At currents above 1kA, the fuse pre-arc time drops below 10ms. The MCB's magnetic trip operates in approximately 5–15ms.

At fault currents above about 1.5kA, the MCB and fuse operate in the SAME time band. Neither is definitively faster. Which one clears first depends on manufacturing tolerances, contact wear, ambient temperature, and randomness.

2. I²t Energy Let-Through

The more precise measure of "which device clears first" at high fault currents is the I²t (energy let-through) — the total thermal energy the device lets through before it clears the fault.

At a 10kA fault:

• A 20A Type B MCB might let through 15,000–30,000 A²s
• A 63A BS 88 fuse might let through 8,000–12,000 A²s

The fuse has LOWER I²t — it clears faster. The fuse blows, the MCB stays closed, and the entire upstream circuit is de-energised.

How to Check Discrimination

Method 1: Manufacturer Discrimination Tables

Major switchgear manufacturers publish discrimination tables showing the maximum fault current at which discrimination is maintained between specific device combinations. These tables are device-specific — they depend on the exact manufacturer, model, and rating.

Example (typical, not universal):

If the prospective fault current exceeds the discrimination limit, the devices are NOT discriminating — either device may operate first.

Method 2: Time-Current Curve Overlay

Plot the time-current curves of the upstream and downstream devices on the same axes. At every fault current level, the downstream device curve must be entirely BELOW (faster than) the upstream device curve.

The critical zones to check:

• Overload region (1–10× rating): usually discriminates easily
• Instantaneous region (10–100× rating): where discrimination typically fails
• Maximum prospective fault current: the worst case — check this point specifically

IEC 60364-5-53, Clause 536.1 — Selectivity between overcurrent protective devices

Method 3: I²t Comparison

For fault currents above 1kA, compare the total I²t let-through values. The downstream device must have LOWER I²t than the upstream device's pre-arcing I²t at every fault current level:

If the downstream device's total let-through energy exceeds the upstream device's pre-arcing energy, the upstream device begins to operate before the downstream device has cleared the fault.

The Economics of Discrimination

Total discrimination — where the downstream device always clears before the upstream device at all fault levels — is expensive. It typically requires:

• Larger ratios between upstream and downstream device ratings (3:1 minimum for fuses, often more for MCBs)
• Current-limiting devices downstream
• Electronic trip units with adjustable short-time delay upstream
• Lower fault levels (which may mean higher-impedance transformers, reducing efficiency)

For most commercial and residential installations, partial discrimination is acceptable and common. BS 7671 Regulation 536.1 states that discrimination "is recommended where necessary to prevent danger" — it's mandatory for safety circuits but recommended for others.

BS 7671, Regulation 536.1 — Discrimination requirements

In practice, partial discrimination means the devices discriminate at all fault levels up to a stated maximum (e.g., 6kA), and above that level, either device may operate. The argument is that faults above 6kA are rare and brief, and the consequences of losing discrimination (temporary loss of supply to healthy circuits) are less severe than the cost of achieving full discrimination.

For critical circuits — hospital essential supplies, fire services, process control in industrial plants — total discrimination IS mandatory, and the cost is justified.

Techniques for Improving Discrimination

Time Grading (Series Coordination)

Upstream devices are set with progressively longer time delays:

• Downstream MCB: instantaneous (no delay)
• Mid-level MCCB: 100ms short-time delay
• Upstream ACB: 300ms short-time delay

This provides clear time separation at all fault levels but requires the upstream devices to withstand fault current for the duration of the delay.

Current-Limiting Devices Downstream

Current-limiting fuses and MCBs physically limit the peak fault current by opening so fast that the current never reaches its prospective peak. The upstream device sees a much lower current and doesn't attempt to operate.

Zone-Selective Interlocking (ZSI)

Downstream devices send a restraint signal to upstream devices when they detect a fault. The upstream device delays its trip, giving the downstream device time to clear the fault. If the downstream device fails, the upstream device trips after its delay expires.

This provides near-total discrimination without the thermal stress of long time delays.

Back-Up Protection (Cascading)

When discrimination cannot be achieved, BS 7671 permits cascading (backup protection) per Regulation 536.2. The upstream device assists the downstream device in clearing high-level faults. Both devices may operate, but the combination is tested and rated for the full fault level.

The Bottom Line

Protection discrimination is not automatic. A smaller downstream device does NOT always trip before a larger upstream device — especially at high fault currents. Before specifying any protection arrangement:

1. Determine the prospective fault current at the point of installation
2. Check the manufacturer's discrimination table for the specific device combination
3. If discrimination is insufficient, consider time grading, current-limiting devices, or ZSI
4. For critical circuits, specify total discrimination and verify it explicitly
5. Document the discrimination analysis — it's part of the design basis and should be reviewed when devices are replaced or fault levels change

Related Resources

• Italy-Switzerland Blackout: Protection Coordination — When protection discrimination fails at scale
• Sayano-Shushenskaya: Transformer Protection — Protection coordination failure in a catastrophic incident
• Short Circuit Withstand: Distribution Board Feeder — Fault current drives protection device selection
• View all worked examples →