Executive Summary
IMCA D 002 addresses a real hazard: batteries in pressure housings can fail catastrophically if poorly managed. The document's core safety principles remain sound, but the shift to lithium-ion chemistry, modern battery management systems and remote monitoring has changed the risk picture since the guidance was first issued. We summarise where the engineering still holds and where a future revision could go further on thermal runaway, housing design and emergency response.
What IMCA D 002 Covers
IMCA D 002, Battery packs in pressure housings, is guidance on the hazards that arise when battery packs are sealed inside pressure housings, and on how to fit and operate them safely. It is relevant to divers, submersible and ROV operators, and anyone working with pressure-compensated or sealed battery systems.
The guidance traces back to an incident in which the bursting disc fitted to the bottom of a bell external battery housing ruptured, discharging the contents with considerable force and releasing gas and fumes with a high potential for serious injury. The current version is Revision 1 (March 2021); it updates the format and references of guidance first issued in January 1996, originally under IMCA’s predecessor body. The technical principles have remained largely stable across those revisions.
What the Guidance Gets Right
The document sets out fundamentals that apply across battery types and applications, and most of them have aged well.
Installation, maintenance and servicing by a competent person is the first of these. Poorly executed battery work creates hazards that proper training and qualification would prevent, and the guidance is right to make competence explicit rather than assumed.
The treatment of shunt diodes is sound. As a general rule, a shunt diode should be fitted across each cell of a primary battery to prevent polarity reversal in any cell under discharge. Without that protection, a weak or aged cell can reverse, generate heat and become a thermal-runaway initiation point. The diode is a simple, well-understood safeguard.
The guidance on siting pressure-relief devices is similarly practical. A relief device should discharge to a safe place if it activates unexpectedly, away from personnel access points and work areas. Bell external battery housings sit close to where people work, which is precisely why the original incident was so dangerous.
The restriction on lead-acid batteries in hyperbaric conditions addresses a genuine problem: lead-acid cells generate hydrogen, continuously so under float charge, and a sealed enclosure can accumulate an explosive atmosphere. IMCA D 041, Use of Battery-Operated Equipment in Hyperbaric Conditions, develops this further, and the two documents together support safe battery selection for saturation systems.
Where the Industry Has Moved Ahead
Lithium-Ion Technology and Thermal Runaway Characteristics
When this guidance was first developed, lead-acid and primary chemistries were the norm. Lithium-ion has since displaced them across much ROV, AUV and diving equipment because of its higher energy density and cycle life. The failure mode is different, though, and that difference matters most around thermal runaway.
Lithium-ion thermal runaway is more energetic than lead-acid failure and propagates cell-to-cell far faster, with state of charge a strong driver: the higher the charge, the faster and more aggressive the event. Published test work and field experience both point the same way. Cell-to-cell propagation, vented gas and the resulting internal pressure rise can outrun a relief device sized for slower, lower-energy failures, so a housing designed around lead-acid assumptions may not be adequate for a lithium pack.
Battery management systems (BMS) have improved in parallel. They now provide temperature monitoring, cell balancing and automatic shutdown that did not exist when the guidance was first published, and a capable BMS can intervene before a passive system would have failed. A future revision could address lithium-ion explicitly: propagation behaviour, BMS integration, and pressure relief sized for rapid gas evolution rather than slow generation. Design features such as inter-cell thermal barriers and dedicated flame-arrested vents are worth flagging for designers selecting enclosures for lithium chemistries. Where a more detailed safety basis is wanted, IEC 62619 (safety of secondary lithium cells and batteries for industrial applications) is a useful reference to read alongside D 002.
Pressure Housing Design Evolution
The original document leans on the general term “suitable discharge device” without dimensions or response times. Housing design has moved on, particularly for high-energy packs. A robust approach sizes the relief path for a worst-case thermal-runaway event at full charge, where the released energy appears as hot gas inside the housing, and validates it against representative gas temperatures and flow rates rather than nominal venting. That worst-case basis is not yet reflected in any specific standard for subsea vehicle housings.
Discharge paths themselves have developed from simple vents toward flame arresters and, in some designs, scrubbing or condensing arrangements that reduce the temperature and reactivity of vented products before they reach the surrounding water or structure.
Enclosure materials deserve attention too. Lithium-ion thermal runaway releases hydrogen, carbon monoxide and hydrogen fluoride, and hydrogen fluoride is corrosive to many aluminium alloys. Material choices that suited lead-acid systems are not automatically appropriate for lithium packs, and hydrogen exposure raises embrittlement concerns for susceptible alloys. A future revision could add a sizing methodology for relief devices, expressing minimum flow coefficient and allowable pressure-rise limits, together with guidance on materials that tolerate hydrogen fluoride and hydrogen exposure under pressure.
Remote Monitoring and Predictive Maintenance
The current guidance frames maintenance largely as periodic inspection and says little about continuous monitoring. Modern packs can stream telemetry continuously: per-cell voltage, temperature, internal resistance, cycle count and discharge data. Trending these parameters allows alarms to be raised on developing faults, for example a rising self-discharge rate that can indicate an internal short, before they progress to failure.
This supports a shift from time-based toward condition-based maintenance, with packs retired on measured condition rather than on calendar age alone. The principle extends to bell external batteries, where surface monitoring of pack temperature and voltage during a dive gives operators visibility they would otherwise lack until recovery. A future revision could treat continuous monitoring as the expectation for sealed-enclosure battery systems, define minimum telemetry parameters (per-cell voltage, temperature, pack voltage and state of charge) and give guidance on alarm thresholds and on when to test, isolate or abandon a system.
Battery Chemistry Compatibility with Hyperbaric Environments
The existing guidance prohibits lead-acid in hyperbaric systems on hydrogen grounds but does not explain the reasoning in depth or set out alternatives. The hyperbaric case is more nuanced than a single prohibition.
IMCA D 041 and NFPA 99 both restrict the use of lithium-battery devices in hyperbaric chambers unless the equipment has been specifically qualified by the manufacturer or a recognised testing body. Lithium-ion does not vent hydrogen in normal operation, which is a genuine advantage over lead-acid, but its thermal-runaway behaviour in a sealed, pressurised, occupied space is exactly why qualification rather than blanket acceptance is the right posture. Pressure-compensated and depth-rated housings can mitigate issues such as helium ingress, which can degrade cells over time, but they do not remove the runaway hazard.
Primary lithium cells, with good pressure compensation and depth-rated, oil-filled housings, perform well in hyperbaric environments and are used in some applications for emergency lighting because of their long shelf life and stable discharge; their single-use nature complicates disposal. Documented, controlled studies of pressure effects on many consumer-grade devices remain scarce, which is itself an argument for the qualification requirement and for further research. A future revision could replace “no lead-acid” with a compatibility matrix linking chemistry to depth, application and hazard, explain why lead-acid is unsuitable (continuous hydrogen under float charge) and why lithium-ion needs qualification rather than prohibition or open acceptance, and cross-reference IMCA D 024 and D 041.
Practical Observations from the Field
Opening battery housings in well-ventilated areas is the right instruction, but vessel spaces, dive stores and workshops rarely offer ideal air movement, and conditions are often worse during mobilisation in confined holds. Flexible tents with forced ventilation are an effective way to create local extraction for battery maintenance in tight spaces.
Purging housings before opening them is sound advice. A pack left sealed for an extended period, particularly with any seal defect, can accumulate flammable gas, so confirming the internal atmosphere before opening, and purging through flame arresters, is a reasonable precaution. Lithium-ion cells do not vent gas at rest, but the possibility of a damaged cell means purging as a precaution still makes sense.
Personal protective equipment should match the chemistry. Dust masks are not adequate against lithium-ion electrolyte vapours; full-face respirators with organic-vapour filters are the appropriate standard for handling potentially compromised lithium packs.
The “competent person” requirement is interpreted inconsistently across operators. A defensible approach builds qualification on a recognised framework: theoretical training in battery chemistry, supervised practical work and independent assessment. IEC/TS 60079-44:2023, Explosive atmospheres – Part 44: Personal competence, and the IECEx and CompEx competency schemes provide established models that can be adapted to a battery-specific qualification structure, rather than relying on informal knowledge transfer.
Suggestions for the Next Revision
Several additions would strengthen the document:
Expand the chemistry coverage. Add sections on lithium-ion, lithium-polymer and other rechargeable chemistries, including cell-to-cell thermal-runaway propagation, the influence of state of charge, the gases produced (hydrogen fluoride from lithium-ion, hydrogen from lead-acid) and their implications for relief sizing and material compatibility.
Set BMS requirements. Specify key functions for rechargeable systems: cell monitoring, overvoltage protection, thermal monitoring and shutdown, charge balancing and condition monitoring, with associated alarm and telemetry detail. Passive protection alone is no longer sufficient.
Provide a relief-sizing methodology. Give guidance on selecting relief devices from energy content, chemistry-specific gas-flow behaviour (including runaway), enclosure volume and allowable pressure-rise rate. There is no universal formula; prototyping and laboratory testing remain necessary, and worked examples for common configurations would help.
Address enclosure materials. Cover hydrogen embrittlement under sustained exposure, hydrogen-fluoride attack on aluminium alloys, the interaction with cathodic protection and the basis for choosing duplex stainless steel versus aluminium, supported by published degradation studies.
Define emergency procedures. Provide clear guidance for an overheating or venting pack: evacuation, suitable firefighting options for the chemistry involved, cooling methods and post-incident inspection.
Standardise qualification. Define what “competent person” means for battery work in terms of theoretical knowledge, practical skill and refresher training, using IEC/TS 60079-44:2023, IECEx and CompEx as models.
Strengthen cross-references. Beyond IMCA D 024, D 041 and D 055, point to IEC 62619, DNV-OS-D201 (electrical installations on offshore units, including battery systems) and DNV’s marine and offshore battery guidance, so the document sits within the wider regulatory framework.
In short, IMCA D 002 addresses a real safety problem and captures principles that have helped prevent incidents for nearly three decades: competent maintenance, short-circuit protection, properly sited pressure relief and safe chemical handling. The case for updating it is the dominance of lithium-ion in modern subsea installations, the value of continuous monitoring, and the design demands of high-energy packs and thermal-runaway analysis. The lead-acid prohibition in hyperbaric environments in particular would benefit from explaining the underlying mechanism and from setting out a qualification-based position on lithium chemistries rather than a single sentence. None of this is criticism of the original work, which was sound for its time; it reflects how quickly battery technology has moved.
For a fuller picture of diving-system battery safety, D 002 should be read alongside IMCA D 024 (Design for Saturation (Bell) Diving Systems), IMCA D 041 (Use of Battery-Operated Equipment in Hyperbaric Conditions) and IMCA D 055 (Prevention of Explosions During Battery Charging in Relation to Diving Systems). Together they form a coherent basis for safe battery operation in diving systems.
This article provides independent analysis of IMCA D 002. It is not endorsed by or affiliated with IMCA. For the complete guidance, refer to the official IMCA publication at imca-int.com.
