Battery Packs in Pressure Housings: Field Lessons from Three Decades

What IMCA D 002 Covers

IMCA D 002 is a safety guideline for batteries in sealed enclosures. It addresses the serious hazard created by the Bell external battery incident, which involved thermal runaway and a violent gas release. This document is for divers, submersible operators, and basically anyone working with pressure-compensated battery systems. This version – 1.1 – updates the format but retains the technical content of the 2021 edition. That, in turn, is based on the 1996 guideline published after the initial incident.

What the Guidance Gets Right

It is commendable that the document clearly lays out the fundamentals relevant to all battery types and applications. The consistent use of qualified professionals for installation and maintenance has prevented countless disasters. We have seen projects where shoddy battery work by incompetent personnel created hazards – hazards that could have been completely avoided with proper training. Bypass diodes between the primary cells are absolutely essential – they prevent actual failure. We witnessed this firsthand in a deep-sea diving system in the Norwegian sector. A defective, ten-year-old battery had reversed its polarity due to the lack of diodes. Heat was generated – this could have led to thermal runaway. Diodes are standard – the key safeguard. Regarding safety devices: The manual describes their safe discharge, and for good reason. Bell battery enclosures are located near personnel access points – a known hazard. Another expensive lesson, but now we have a clear direction. On a DSV project, after reviewing this document, we relocated the reset points of the overpressure valves away from the diving zones and work areas. Now, regarding lead-acid batteries under hyperbaric conditions: An explicit ban addresses the real issue of hydrogen generation. They continuously produce hydrogen in buoyancy mode, and the sealed casings can become ticking time bombs. IMCA D 041 goes even further, setting limits on the use of batteries under hyperbaric conditions. Together, these documents helped us select safe batteries for hydrogen saturation systems. Following these recommendations ensures safety and reliability.

Where the Industry Has Moved Ahead

Lithium-Ion Technology and Thermal Runaway Characteristics

When these guidelines were developed, lead-acid batteries and new cells were the standard. Lithium-ion batteries have replaced ROV systems, AUV kits, and parts of diving equipment, significantly outperforming them in energy density and lifespan. However, the chemistry is different – especially regarding thermal runaway. Thermal runaway in lithium-ion batteries is much more aggressive and spreads from cell to cell much faster than in lead-acid batteries. Research confirms our observations at sea: At full charge, runaway spreads within seconds; at 50–75% charge, within minutes. We’ve seen a cell module fail completely within 60–80 seconds, taking the entire battery with it. Peak temperatures exceed 800°C. We observed this in an ROV battery where the failure of a single cell, despite thermal insulation, spread like wildfire throughout the entire unit. The pressure build-up exceeded the permissible limit for conventional rupture discs. The casing deformed before the pressure relief valve could activate, but fortunately, the damage was minimal. Battery management systems (BMS) have improved, however. They now offer temperature monitoring, cell balancing, and automatic shutdown – features that didn’t exist when these guidelines were first published. We have seen these BMS systems prevent at least three potential explosions that would have been catastrophic with older, passive systems. The next update should specifically address the hazards of lithium-ion batteries: propagation speeds in seconds rather than minutes and the need for reliable BMS integration. Measures for pressure relief in the event of rapid gas build-up should also be strengthened. It would be wise to consider features such as thermal barriers between cells and flame vents to help designers select appropriate enclosures for lithium technology. If you are already using IEC 62619 for more thorough battery safety guidelines, integrating these standards can be of great benefit.

Pressure Housing Design Evolution

The old document is dominated by the vague term “suitable discharge device,” without any concrete details regarding dimensions or response times. Housing design has evolved significantly, especially for high-energy batteries. Some manufacturers – not all, mind you – design their discharge devices for the worst-case scenario of thermal runaway. Here, too, the assumption is a full charge, where all the energy is released as gas within the housing. That’s precisely what the industry recommends: simulate conditions, test gas temperatures above 600 °C. However, the “worst-case scenario” approach hasn’t yet made it into the specific standards for underwater robots. This might be confusing for some, but we’ve observed this approach being used by experienced manufacturers. Then comes the further development of discharge pathways. From simple vents to specially designed flame arresters and exhaust scrubbers – a classic step for underwater robots operating near underwater structures. Options include water cooling systems that condense the battery’s electrolyte vapors, thus reducing environmental impact (we’ve seen this before). Also note the casing materials. These have also changed. Uncontrolled discharge of lithium-ion batteries produces hydrogen, carbon monoxide, and harmful hydrogen fluoride, which corrodes aluminum alloys. Aluminum alloys used in lead-acid systems are unsuitable for lithium-ion batteries. Hydrogen reacts with micro-defects in these metals, destroying their strength. After thorough analysis, we now use duplex stainless steel or hydrogen-resistant aluminum for demanding applications. Would it be beneficial in future versions to incorporate a methodology for dimensioning safety devices? Minimum flow coefficients and pressure rise limits would be extremely helpful. We need to support the teams in developing relief channels, including flame arresters and hydrogen venting. Materials that can withstand hydrogen fluoride without deforming under pressure should be considered.

Remote Monitoring and Predictive Maintenance

Current recommendations only mention “regular inspection” as part of maintenance, neglecting continuous monitoring – a serious shortcoming given today’s standards. Modern batteries constantly transmit telemetry data: cell voltage (often overlooked), temperature, internal resistance, cycle count, discharge data – we capture it all. This is a real lifesaver for underwater vehicle batteries. If parameters deteriorate, alarms are triggered even before an emergency occurs. We’ve seen this firsthand. On a salvage vessel, increased self-discharge rates, indicating potential internal short circuits, appeared three weeks before the actual failure. Replacing the battery as part of routine maintenance prevented an emergency at sea. Modern predictive analytics systems use machine learning to detect warning signs in battery data. We have algorithms that can detect such anomalies and identify unusual cell behavior before it’s too late. Are you observing an unusual voltage drop or temperature rise? The trend is moving from time-based to condition-based maintenance. Batteries are replaced as soon as they wear out, not just because they’re worn out. Diving bell systems are also becoming increasingly important. Some experienced deep-sea divers have installed telemetry on the diving bell’s external batteries, allowing surface operators to monitor the interior during a dive. This is critical. We had a case where elevated temperatures were detected during a 200-meter dive – upon resurfacing (at least in practice), it turned out a short circuit was underway. Without telemetry, this would have meant a catastrophic underwater explosion with the divers inside. The next documentation update? Continuous monitoring of battery systems in sealed enclosures is essential. The minimum telemetry parameters need to be clearly defined: cell voltage, temperature, battery voltage, and state of charge. Recommendations for alarm thresholds? Critical. It’s vital to know when to test, isolate, or abandon the system.

Battery Chemistry Compatibility with Hyperbaric Environments

The existing document prohibits lead-acid batteries in hyperbaric systems due to hydrogen concerns, but it neither explains the reasons nor offers alternatives. We are currently working with various battery technologies at depth, each with its own specific characteristics. Lithium-ion batteries perform well at saturation depths with proper casing design. We have used them at 300 meters in seawater tanks and diving gear. A suitable casing design prevents the ingress of helium, which can shorten cell lifespan. Helium acts like kryptonite on cells, destroying them. Lithium-ion batteries do not emit hydrogen during normal operation – the gas only appears during thermal runaway or improper use. This is a huge safety improvement over lead-acid batteries, although a different approach to managing the risk of thermal runaway is critical. As for other battery technologies, the results are inconsistent. Helmet lights on diving helmets that function at 200 meters are not performing as advertised? We have observed this as well – but are there any documented studies on the effects of pressure? Virtually none – in our opinion, an area for future controlled research. Primary lithium cells function well in hyperbaric environments with good pressure compensation; some pressure-controlled submersibles use them for emergency lighting thanks to their depth-rated, oil-filled casings. Their long shelf life and stable discharge are ideal for emergency situations, but their disposable nature makes waste disposal quite difficult. Future updates should elaborate on recommendations for hyperbaric environments, going beyond the simple statement “No lead-acid batteries.” A compatibility matrix linking chemistry to depth, applications, and hazards could facilitate equipment selection. Explaining why lead-acid batteries are unsuitable (constant hydrogen) while lithium-ion batteries work (no uncontrolled gas evolution without problems) will enable the team to make more informed decisions. Cross-referencing IMCA D 024 and IMCA D 041 will integrate this into the overall system design.

Practical Observations from the Field

Opening battery casings in well-ventilated areas at sea? Sounds good, but have you been in an engine room recently? Submarine workshops, diving equipment depots – the air quality often leaves much to be desired. We know the problem: ventilation difficulties, and during mobilization, the situation is even worse when you’re stuck in cargo holds with no air circulation. Our solution: flexible tents with forced ventilation. Portable and effective, they create local ventilation for battery maintenance in confined spaces. And the advice to purge the casings before opening them? Absolutely right. In one of our projects, gas measurements revealed a hydrogen content of up to 4%. The battery sat unused for two weeks, leaking through a faulty seal. That’s why we now test the casings under inert gas pressure. We then purge them for 15 minutes through flame arresters. We’re changing our approach with lithium-ion batteries. These cells don’t emit gases when at rest. However, since we have to consider the possibility of defective devices, we continue to clean them as a precaution. Standard eye and respiratory protection simply isn’t sufficient. Dust masks aren’t ideal for lithium-ion batteries. After several technicians complained of throat irritation, we switched to full-face masks with filters against organic vapors. Dust masks are inadequate for lithium-ion electrolyte fumes. Full-face masks with organic vapour filters are the standard.

The requirement for a qualified person – the definition varies across operators. Some teams assume, “Oh, he’s already taken care of that.” But it’s not that simple. We’ve implemented a qualification system based on IEC/TS 60079-44:2023 : theoretical training in battery chemistry, supervised practical exercises, and independent examinations. We’ve translated this regulatory jargon into something practical. Otherwise, we would be back to dangerous, informal knowledge transfer. And ultimately, this helps prevent deviations from qualification standards. Models like the IECEx competency assessment system and CompEx certificates serve as benchmarks. Decisions made in practice reveal discrepancies between textbook recommendations and real-world situations. These are decisions based on practical experience, not just formal procedures.

Regarding:

Suggestions for the Next Revision

Based on over 15 years of experience with sealed battery systems in submersible, ROV, and AUV environments, we have identified several key points to consider. Several important additions would significantly improve this document. Expand the information on battery chemistry. Include sections on lithium-ion, lithium-polymer, and other rechargeable batteries. Thermal runaway behavior is important here – how does heat propagate from cell to cell in lithium-ion batteries? Consider the time it takes for a battery to reach full charge. Contrast this with the slow aging of lead-acid batteries: document the resulting products – hydrogen fluoride from lithium-ion batteries and hydrogen sulfide from lead-acid batteries – and their impact on pressure relief and material compatibility. Fire safety considerations should also be addressed; different chemical compositions present different hazards. Include requirements for the battery management system (BMS). We recommend key BMS functions for rechargeable batteries: cell monitoring, overvoltage protection, thermal monitoring and shutdown, charge balancing, and condition monitoring. We offer specific alarm and telemetry details for remote monitoring. Note that a BMS can detect signs of thermal runaway and initiate automatic protection. Passive systems are insufficient. Methodology for calculating safety devices. Guidance is needed for selecting safety devices based on the battery’s energy content, the gas flow rates characteristic of the specific battery chemistry (including thermal runaway in lithium-ion batteries), the enclosure volume, and the allowable pressure rise rate. There is no universal formula; prototyping and laboratory testing are essential. Examples of standard configurations include: ROV batteries, external enclosures, and AUV systems. Material selection for pressure enclosures. Hydrogen embrittlement: a serious concern with long-term hydrogen exposure. This is relevant because thermal runaway in lithium-ion batteries leads to the release of hydrogen and hydrogen fluoride, which attacks aluminum alloys. Material compatibility with the battery chemistry must be considered. The interaction of cathodic protection and the reasons for choosing a particular material must be taken into account. Aluminum alloys? Stainless steel? Published studies on hydrogen-induced degradation must be consulted. Emergency procedures. The industry needs clear guidelines for dealing with battery overheating. We need evacuation mechanisms – which aren’t always obvious – firefighting options (water, CO₂, foam, Class D fire extinguishers), methods for cooling buildings, and post-incident inspections. We had to develop these ourselves, but they should become mandatory industry standards.

Standards for the qualification of specialists. It is necessary to define the term “specialist” for working with batteries: theoretical knowledge, practical skills, and regular training. IEC/TS 60079-44:2023 provides a reliable model (personal competence for work in potentially explosive atmospheres). The IECEx and CompEx frameworks serve as good examples for establishing a battery-specific qualification structure. Cross-references: Consider relevant standards beyond IMCA D 055, D 024, and D 041. Consider IEC 62619 (safety of lithium batteries for industries such as marine), DNV-OS-D201 (electrical installations on ships and offshore platforms), and the DNV Guide to Marine and Offshore Battery Systems. Link this document to the broader regulatory framework and demonstrate compliance with existing guidelines. Basis: our shared practical experience and the challenges of developing higher safety standards. In summary, IMCA D 002 addresses a critical safety issue and outlines the core principles that have prevented incidents for nearly three decades. These include key principles such as proper maintenance, short-circuit protection, proper pressure relief, and the safe handling of chemicals in battery systems. The document needs updating to to reflect the dominance of lithium-ion batteries in modern submarine installations, use the potential of continuous monitoring, and improve the design of advanced pressure vessels, taking thermal runaway analysis into account. The prohibition of lead-acid batteries in hyperbaric environments urgently needs further clarification. This includes explaining the continuous hydrogen generation in float charge mode and how it differs from the lithium-ion battery standard, which eliminates gas evolution. This isn’t about blaming anyone, but rather acknowledging the rapid development of battery technology. For its time, this guidance was very precise and served its purpose well. When we share this information, we share our practical experience and try to hide nothing. From dives in the North Sea to ROV deployments worldwide, to AUV operations in environments ranging from polar cold to equatorial heat – we’ve seen it all. Research into phenomena like thermal creep and hydrogen embrittlement has confirmed our offshore observations. IMCA distinguishes itself by responding to the needs of the community and adapting to new technologies – that’s why its standards are so successful.

For a thorough understanding of diving system safety, this standard should be used in conjunction with IMCA D 024 (Design of Deep Sea Diving Systems), IMCA D 041 (Battery-Powered Systems in Hyperbaric Environments), and IMCA D 055 (Safety During Battery Charging). Together, they form the basis for the safe operation of diving systems. The underwater world benefits from IMCA’s standardization efforts. Standards like D 002 improve safety for everyone. By sharing insights gained from numerous battery maintenance projects and a range of hazardous situations, we make our recommendations more reliable for all.

---

This article provides independent analysis of IMCA D 002. But it is not endorsed by or affiliated with IMCA. For the complete guidance, refer to the official IMCA publication at IMCA-int.com.