Primary Injection Testing on Medium-Voltage Circuit Breakers

When a medium-voltage circuit breaker is commissioned on a new installation or returned to service after major maintenance, the relay protecting it needs to be verified. The standard approach for routine annual relay testing is secondary injection: a test set applies current directly to the relay's current input terminals, the relay responds, and the technician records pickup current and trip time against the settings. It is fast, requires lightweight equipment, and covers the relay's internal logic completely.

What secondary injection does not verify is everything else in the protection chain. The current transformers feeding the relay, the ratio tap selected on each CT, the wiring between the CT secondary and the relay input terminals, the ground fault CT connection, the trip coil, and the breaker's mechanical trip mechanism are all outside the test boundary. A relay that passes secondary injection with every function verified to within tolerance can still fail to trip on a real fault if any one of those elements has an error. Primary injection tests the complete chain by driving current through the actual primary conductors of the circuit, the same path fault current will take, and watching whether the breaker trips.

What primary injection tests

In a primary injection test, a high-current, low-voltage test set is connected to the line-side and load-side terminals of the circuit breaker with the breaker racked out or isolated. Current is driven through the primary conductor, through the current transformers mounted on the breaker or in the switchgear compartment, into the relay or electronic trip unit, and through the trip coil. When the current level and time combination crosses the protection function threshold, the breaker trips. The test set measures the elapsed time from the moment the threshold is reached to the moment the trip contact opens.

This test path includes every element of the protective system in series. If the CT ratio tap is set incorrectly, say the CT is tapped for 600:5 but should be 1200:5, the relay sees twice the current per ampere on the primary and will trip too early. If the CT secondary wiring is crossed between phases A and B, a single-phase primary injection on phase A will produce a response on the relay's phase B input instead. If the ground fault function relies on a residual connection or a core-balance CT and that connection is open, primary injection current in all three phases simultaneously will confirm whether the ground fault element can see residual current from a real fault. None of these failure modes appear in secondary injection results.

Protection functions tested

On medium-voltage metal-clad and metal-enclosed switchgear, the breaker itself carries no integral trip unit. It is tripped by separate protective relays, electromechanical or numerical, fed by current transformers mounted in the switchgear and wired out to the breaker trip coil. A complete primary injection sequence exercises each enabled relay element in that scheme: time-overcurrent (ANSI 51), instantaneous overcurrent (50), and ground fault (50G, 51G, or 50N/51N), each run at multiple current levels.

The time-overcurrent element (ANSI 51) is the overload and backup fault-protection element. It operates on an inverse, very-inverse, or definite time curve at currents above its pickup setting. Pickup is verified by ramping current to the threshold; timing is then checked at one or more defined multiples of pickup, commonly 2x and 3x and above, against the relay's published time-current curve. At low multiples of pickup the trip time can run into tens of seconds, so test-set thermal capacity is a practical constraint on the longer holds.

Where the scheme adds a definite-time or a second overcurrent element for coordination, primary injection verifies both its pickup and its intentional delay. A feeder relay might carry an element set at several times rated current with a short definite delay, coordinating with a downstream fuse or relay that clears the fault first. Both the pickup and the delay must be correct for the coordination scheme to hold under fault conditions.

The instantaneous overcurrent element (ANSI 50) operates with no intentional delay for severe faults beyond the coordination range. Its pickup is set from the coordination study, well above load and above downstream coordination points, so on a busy feeder the required test current can reach several thousand amperes. This is the most demanding test in terms of test-set output, and among the most important, since the instantaneous element limits let-through energy during a high-magnitude fault. The relay operate time recorded here, added to the breaker mechanical clearing time, is the actual clearing time an arc-flash hazard analysis depends on.

The ground fault function is where primary injection provides the most value that secondary injection cannot replicate. Ground-fault sensing on MV switchgear uses one of two arrangements: a residual connection that sums the three phase-CT outputs (which sum to zero on a healthy balanced system), or a window CT, also called a core-balance or zero-sequence CT, that encircles all three phase conductors at once. Primary injection into all three phases simultaneously, balanced, so no ground fault signal should appear, verifies that the residual connection is correctly wired with no open legs. Single-phase injection or deliberately unbalanced injection then verifies that the ground fault element responds at the correct pickup current. An open CT leg in a residual connection, for example, produces a standing false ground fault signal under load; balanced three-phase primary injection will expose this before the switchgear is energized.

Test equipment and setup

Primary injection test sets are high-current, low-voltage sources, the opposite of the high-voltage, low-current sources used for insulation testing. They produce from a few hundred to several thousand amperes at voltages typically below 10 volts AC. Common high-current platforms include the Megger ODEN AT and EuroSMC test sets; multiple units can be paralleled for the highest pickup levels. For very large breakers where the instantaneous pickup requires currents above 10,000 amperes, test sets can be operated in parallel or transformer-based current injection rigs can be used.

Setup requires connecting the injection leads to the breaker's line and load terminals on a single phase, with the return path through the test set. For a draw-out breaker, this is typically done with the breaker racked out to the test position so the primary contacts are disconnected from the bus. Each phase is tested independently for phase overcurrent functions. For ground fault testing with a residual connection, all three phases must be injected simultaneously with phase currents that vector-sum to the desired ground fault current level, which requires a three-phase test set or a staged single-phase approach with care about the neutral return path.

The breaker must be in the closed position at the start of each test so that the trip coil is in the circuit and the breaker can open when the protection function operates. After each trip, the breaker must be reset and reclosed before the next test point. On breakers with spring-operated mechanisms, the closing spring must fully charge between operations, typically 10 to 15 seconds, to avoid testing with a partially charged mechanism that may affect trip timing results.

Test current requirements and outage planning

The current levels required for primary injection follow the protective-relay settings from the coordination study, not a fixed multiple of the breaker continuous rating, and they drive outage scope and test-set selection. For a typical MV feeder:

Time-overcurrent (51): near pickup for the pickup check, then 2x to 3x pickup and above for timing. Instantaneous (50): at and above its pickup setting, commonly several thousand amperes on a feeder. Ground fault (50G/51G): at the ground pickup setting, often a few hundred amperes. Exact values come from the relay settings in the coordination study.

Test sets delivering instantaneous pickup currents must be sized accordingly, and the cable cross-section of the injection leads affects how much current can be sustained without excessive heating. For primary injection of instantaneous pickup, the hold time is short, the breaker should trip in milliseconds, so the thermal burden on the test leads is manageable. Long-time function tests at 2× pickup with a trip time of several minutes are the more thermally demanding tests despite the lower current.

Planning the outage around primary injection also means accounting for the number of operations on the breaker mechanism. A complete sequence with three or four test points per relay element across three phases results in 30 to 50 breaker operations. Breakers with spring mechanisms and compressed-air or hydraulic operators have defined operation counts between maintenance intervals; a full primary injection sequence is a meaningful fraction of that budget on a heavily used breaker. The test plan should account for this and coordinate with the maintenance cycle.

Primary injection versus secondary injection: when to use each

Secondary injection is appropriate for routine relay testing at annual or biennial maintenance intervals when the CT connections and wiring were previously verified and have not been disturbed. It is faster, requires lighter equipment, and does not impose operations on the breaker mechanism. For a relay with no changes to the CT circuit since the last primary injection test, secondary injection is the correct tool for verifying that the relay's internal calibration and settings remain correct.

Primary injection is required at acceptance on new installations, after any modification to the CT circuit (new CTs, re-tapping existing CTs, panel rewire), after a circuit breaker replacement where the CT connections are remade, after a through-fault event that may have affected the CTs or wiring, and any time there is reason to question whether the protection system as installed matches the design. NETA ATS calls for a system function test at acceptance that verifies the complete protection scheme actually trips the breaker. Primary injection is the method that exercises the full path, from the primary conductor through the CTs and relay to the trip coil; secondary injection is accepted for routine maintenance where the CT circuit has not been disturbed, but it exercises the relay alone, not the CT-and-wiring chain that primary injection proves end to end.

The most common argument for skipping primary injection is schedule pressure during commissioning. The consequence of that decision is a protection system with an unverified CT-to-relay path going into service. If that path has an error, a CT tap mistake, a crossed connection, a loose terminal, the relay will respond incorrectly to a fault, whether that means tripping too early, too late, on the wrong phase, or not at all. Primary injection before energization is the checkpoint that catches these errors at a time when they can be corrected in minutes rather than under fault conditions.

Documenting results

Primary injection test results should be recorded with the actual measured trip time at each test current, the test current level in amperes, the protection function and phase tested, the breaker mechanism operating time (contact part time from timing contacts if measured separately), and the test set model and calibration date. The results are compared to the trip unit or relay manufacturer's published time-current curve and the settings specified in the protection coordination study.

Deviations from the expected trip time at a given test current are evaluated against the tolerance band in the manufacturer's specifications, typically ±10 to 15% on time and ±5 to 10% on pickup current for electronic trip units, and tighter tolerances for modern numerical relays. Deviations outside tolerance require investigation before the breaker is returned to service. Common causes include incorrect settings, test set calibration error, CT burden exceeding the rated burden of the CT (which compresses the output at high currents), or a trip unit that has drifted from calibration.

The completed test record, along with the secondary injection relay test results, CT ratio and polarity test results, and insulation resistance and contact resistance results, forms the acceptance test package for the circuit breaker. This package is the record of the protective system's verified condition on the day it was commissioned and is the baseline for all future maintenance comparisons.