Circuit Breaker Testing Protocols for Marine Vessels

A circuit breaker is the only component in a power system whose job is to do nothing for years and then do something exactly right in a few milliseconds. Aboard a vessel, the consequences of getting that wrong are compressed. There is no utility to fall back on, the nearest replacement may be a port away, and the equipment on the other side of the breaker may be steering or propulsion. The Coast Guard writes the design rules for these breakers, the class societies survey them, the Navy maintains them under its own specifications, and none of those bodies publishes a field test procedure. The field procedure comes from NETA. This article lays out how the two fit together, what the tests are, what numbers they have to produce, and where the marine environment changes the answer.

Two layers of rules

The first layer is design and construction. Title 46 of the Code of Federal Regulations, Subchapter J, governs electrical engineering on inspected vessels. Section 111.54-1 requires every circuit breaker to have an interrupting rating sufficient for the maximum asymmetrical short-circuit current at its point of application; limits molded-case breakers to 600 volts nominal and requires them to meet the marine supplement of UL 489 or IEC 60947-2; and requires other low-voltage breakers to comply with IEEE C37.13, IEEE C37.27, or IEC 60947-2.¹ It also contains a rule with no shore-side equivalent: every breaker in an engine room, boiler room, or machinery space must be calibrated for a 50 degree Celsius ambient, unless it sits in a controlled machinery control room held at 40 degrees or below.¹ A breaker must not depend on mechanical cooling to carry its rating and must not have its long-time trip set above the continuous rating of its frame or trip element. Section 111.50 governs coordination, and for vessels with 1,500 kilowatts or more of aggregate generating capacity the Coast Guard's plan review requires a short-circuit calculation and a coordination analysis.²

Class rules add construction and factory test requirements. ABS switchboard rules call for insulation resistance of not less than 1 megohm at not less than 500 volts DC after a 60 second dielectric test, and require protective device tripping tests as part of assembly testing.³ IEEE 45, the recommended practice for shipboard electrical installations, sets the design environment: 45 degrees Celsius in machinery spaces, relative humidity to 95 percent, salt-laden atmosphere, and vibration from 5 to 50 hertz at 20 millimeters per second velocity amplitude.⁴ IEEE 45.6-2016 is the family's electrical testing document.⁵

What none of these documents does is tell a technician standing in front of an installed breaker which tests to run, in what order, and what result passes. That is the second layer, and it is the ANSI/NETA Acceptance Testing Specifications for new installations and the Maintenance Testing Specifications for equipment in service.⁶ The Navy adds a third layer for its own ships: NAVSEA Standard Item 009-75, Circuit Breaker Repair, which is written into ship repair contracts and carries its own acceptance criteria.⁷

The NETA protocol, by breaker class

NETA organizes breaker testing by construction: insulated-case and molded-case breakers, low-voltage power breakers, and medium-voltage vacuum breakers. The visual and mechanical inspection is common to all three and comes first. Compare the nameplate to the drawings and the coordination study. Inspect physical and mechanical condition, anchorage, alignment, and grounding. Verify the connection matches the single-line. Inspect bolted connections for high resistance by low-resistance ohmmeter, by calibrated torque wrench against the manufacturer's data or NETA Table 100.12, or by thermographic survey. Verify the operation of manual operators, stored-energy mechanisms, electrical closing and tripping, shunt trip and undervoltage devices, and interlocks. Verify the settings of adjustable trip devices. Verify lubrication against the manufacturer's data.⁸

Molded-case and insulated-case breakers. The electrical tests are insulation resistance on each pole to ground and pole to pole for one minute; a dielectric withstand test on each pole with the others grounded for one minute; an inverse-time trip test at 300 percent of rated continuous current; an instantaneous trip test; a contact resistance test on the main contacts; secondary injection of shunt trip and undervoltage devices; and, where equipped, verification of zone-selective interlocking and ground-fault protection.⁸ The acceptance criteria are specific. Insulation resistance follows the manufacturer's data or NETA Table 100.1, which for 600 volt equipment specifies a 1,000 volt DC test and a minimum of 100 megohms.⁹ Bolted connection and contact resistance values are compared to similar connections, and any value more than 50 percent above the lowest is investigated. The inverse-time result must fall within Table 100.7, which is derived from NEMA AB 4 and sets a maximum trip time at 300 percent current by frame size and voltage: 50 seconds for a breaker rated 30 amperes or less at 250 volts, 70 seconds at 251 to 600 volts, rising to 1,000 seconds for a 6,000 ampere frame.¹⁰ Instantaneous results must fall within Table 100.8: plus or minus 30 percent of setting for electronic trip units, plus 40 and minus 30 percent for adjustable electromechanical trips, and plus or minus 25 percent of the manufacturer's published band for nonadjustable trips.¹¹

NEMA AB 4, the source document for those tables, adds procedure details that matter in the field: the 300 percent test is run one pole at a time with the line and load terminals connected through copper conductors not less than four feet long, insulation resistance is measured at not less than 500 volts DC with 1,000 preferred and a minimum of 1 megohm, and any breaker that has interrupted a fault at or near its rating is checked before the circuit is re-energized.¹²

Low-voltage power breakers. The drawout breakers in a ship's service switchboard add inspection items: arc chutes, moving and stationary contacts for wear and alignment, cell fit and racking mechanism, barriers and shutters. The electrical tests add minimum pickup voltage of the trip and close coils, and the trip unit is verified by primary current injection: long-time pickup and delay, short-time pickup and delay, ground-fault pickup and delay where fitted, and instantaneous pickup.¹³ The acceptance criterion for the trip unit is that measured values fall within the manufacturer's published time-current characteristic tolerance band. Contact resistance is compared across poles, with the same 50 percent rule.

Primary injection is the point of the exercise. Secondary injection test sets drive the electronic trip unit directly and prove that its logic works, but as one manufacturer's field test guide states plainly, they do not test the current transformers and their connections.¹⁴ Primary injection pushes real current through the breaker and proves the whole chain: sensor, wiring, trip unit, actuator, mechanism, contacts. On a vessel where that chain has spent a decade in vibration and salt air, the chain is what fails, not the logic.

Medium-voltage vacuum breakers. For the 4.16 kilovolt switchgear on larger vessels and integrated power systems, the inspection adds contact erosion, wipe, and gap, slow-close operation, and arc chute condition. The electrical tests add trip and close coil minimum pickup, trip and close timing, contact travel and velocity, and a vacuum bottle integrity test performed in strict accordance with the manufacturer's published data; NETA publishes no vacuum test voltage of its own.¹⁵ Static contact resistance is measured phase by phase with a DC source of at least 100 amperes using the four-wire Kelvin method, and timing and travel results are compared to the manufacturer's specification and to previous results.¹⁶ Insulation resistance for 5 kilovolt class equipment is tested at 2,500 volts DC to a 1,500 megohm minimum; 15 kilovolt class at 2,500 volts to 5,000 megohms.⁹

Frequency. NETA's guidance on maintenance intervals is deliberately reliability-based. Its baseline for molded-case and low-voltage power breakers is a visual inspection monthly, visual and mechanical annually, and full electrical testing every 36 months, with vacuum breakers at 24 months; those base intervals are multiplied by a factor from 0.25 to 2.5 depending on the equipment's condition and the reliability required of it.¹⁷ A breaker protecting steering gear in a machinery space is at the short end of that range. Test instruments must be calibrated within 12 months of the test date.⁶

The Navy overlay

For Navy work the field procedure is NAVSEA Standard Item 009-75, and its criteria are stated as absolutes rather than ranges. Internal wiring, coils, and transformers are tested for opens and shorts and for insulation resistance to ground with a 500 volt megger; the minimum acceptable is 1 megohm. Phase-to-phase dielectric strength is 2,000 volts minimum. Silver contact wear must be less than 50 percent of original thickness with no evidence of copper migration; non-silver contact wear must be less than 10 percent. A millivolt drop test is performed on each set of contacts. Trip units are tested, calibrated, adjusted, and certified for time delay and instantaneous settings in accordance with the equipment technical manual. Mechanical and electrical operation must succeed three consecutive times. Lubrication is specified by military specification and applied sparingly. A legible report goes to the Supervisor.⁷

The breakers themselves are built to Navy detail specifications rather than commercial standards: MIL-DTL-17361 for the AQB and NQB families of insulated-housing breakers and MIL-DTL-17587 for the ACB drawout power breakers, all rated 500 volts AC for the Navy's 440 volt ungrounded system and qualified to MIL-S-901 for high-impact shock and MIL-STD-167 for vibration.¹⁸ The tolerances written into those specifications are tighter than NEMA AB 4. The AQB-A250 family, for example, carries an instantaneous trip tolerance of plus or minus 10 percent of setting and long-time delay windows stated in seconds at fixed multiples of rating, such as 20 to 32 seconds at 2.25 times rating.¹⁹ A contractor testing a Navy breaker to NETA Table 100.8 tolerances is testing to the wrong document.

Where the marine environment changes the answer

Ambient and calibration. A breaker calibrated for a 50 degree machinery space per 46 CFR 111.54-1 will trip later than its shore-side nameplate suggests when tested at a 25 degree bench. The test record has to state the ambient at the time of test and the calibration basis of the breaker, or the inverse-time result cannot be interpreted.

Ungrounded systems. Navy ships operate three-phase ungrounded or high-resistance-grounded 440 volt systems by design; MIL-STD-1399 Section 300 states that electric power systems shall be ungrounded, and the Navy's stated reason is survivability, since a single line-to-ground fault does not trip anything and the plant keeps running.²⁰ The Coast Guard requires ground detection on every ungrounded system.²¹ The consequence for breaker testing is that ground-fault protection as a breaker function is largely absent and ground detection lives in the switchboard instead, so the NETA ground-fault items are replaced by verification of the ground detection system, and insulation resistance testing of the breaker and its cable carries more of the protective burden than it does ashore. The Coast Guard's own cable test specification uses a 500 volt DC minimum and corrects readings to 20 degrees Celsius.²²

Vibration and mechanism condition. IEEE 45's design vibration envelope and the Navy's MIL-STD-167 amplitudes describe a service life in constant motion. The failure modes that follow are mechanical: linkages that stiffen, pivots that seize, lubrication that hardens, and fasteners that back off.²³ A breaker that operates infrequently is not protected by its inactivity; contamination and hardened lubricant accumulate whether or not the mechanism moves. This is why NETA's visual and mechanical inspection is not a formality on a vessel and why the Navy item demands three consecutive successful operations rather than one.

Salt and moisture. Machinery-space humidity above 90 percent with condensation cycles degrades insulation resistance, and chloride-laden air corrodes terminals.²⁴ Insulation resistance and contact resistance are the two tests that see this, and they are only useful when trended. A 600 volt breaker reading 400 megohms against a 100 megohm minimum passes; the same breaker reading 400 megohms after reading 4,000 megohms two years earlier is telling you something.

Contact erosion. Every interruption wears the contacts, and a shipboard plant is reconfigured far more often than a building's: paralleling generators, shifting to shore power, splitting the bus. Contact resistance measured by micro-ohmmeter, compared pole to pole and against the baseline, is the early indicator, and the Navy's 50 percent thickness limit on silver contacts is the hard stop.⁷

Safety

Every one of these tests except thermography is performed on a de-energized breaker, and establishing that condition is governed by NFPA 70E. The steps are the familiar ones: identify all sources, open the disconnecting means and verify visually, release or restrain stored energy, apply lockout and tagout, and test each phase conductor for absence of voltage with an instrument rated for the system, verified on a known live source before and after.²⁵ Stored energy deserves particular attention on breakers: a charged closing spring on a drawout breaker is a stored-energy device in its own right, and NFPA 70E's Article 360 addresses capacitors and stored energy separately.²⁶ Aboard Navy ships the tag-out procedure is NAVSEA's Tag-Out Users Manual, which applies to repair activities and their subcontractors as well as to Ship's Force.²⁷

What the record has to contain

A breaker test that is not documented did not happen. The record for each breaker should identify the equipment by switchboard, section, and circuit; state the ambient temperature and the calibration basis; record every measured value against its acceptance criterion and its previous value; record the test instrument and its calibration date; and state the disposition: returned to service, adjusted and returned, or removed. On Navy work the Standard Item specifies the report; on classed vessels the record supports the survey; on any vessel it is the only defense when a breaker that passed later fails to clear a fault.

MD Engineering performs acceptance and maintenance testing of shipboard and waterfront circuit breakers and switchgear to the ANSI/NETA specifications and to NAVSEA Standard Items, with primary injection, insulation and contact resistance trending, and documentation built for the surveyor who will read it. If the last test record for your switchboard cannot tell you the ambient, the instrument, and the previous reading, the breaker has not been tested in any sense that will matter when it is asked to operate.