|Acronym||CCUBA (closed circuit underwater breathing apparatus); CCR (closed circuit rebreather), SCR (semi-closed rebreather)|
A Diving rebreather is an underwater breathing apparatus that absorbs the carbon dioxide of a diver's exhaled breath to permit the rebreathing (recycling) of the substantially unused oxygen content, and unused inert content when present, of each breath. Oxygen is added to replenish the amount metabolised by the diver. This differs from open-circuit breathing apparatus, where the exhaled gas is discharged directly into the environment. The purpose is to extend the breathing endurance of a limited gas supply, and, for covert military use by frogmen or observation of underwater life, to eliminate the bubbles produced by an open circuit system. A diving rebreather is generally understood to be a portable unit carried by the user, and is therefore a type of self-contained underwater breathing apparatus (scuba). A semi-closed rebreather carried by the diver may also be known as a gas extender. The same technology on a submersible or surface installation is more likely to be referred to as a life-support system.
Diving rebreather technology may be used where breathing gas supply is limited, or where the breathing gas is specially enriched or contains expensive components, such as helium diluent. Diving rebreathers have applications for primary and emergency gas supply. Similar technology is used in life-support systems in submarines, submersibles, underwater and surface saturation habitats, and in gas reclaim systems used to recover the large volumes of helium used in saturation diving .
The recycling of breathing gas comes at the cost of technological complexity and additional hazards, which depend on the specific application and type of rebreather used. Mass and bulk may be greater or less than equivalent open circuit scuba depending on circumstances. Electronically controlled diving rebreathers may automatically maintain a partial pressure of oxygen between programmable upper and lower limits, or set points, and be integrated with decompression computers to monitor the decompression status of the diver and record the dive profile.
The first attempts at making practical rebreathers were simple oxygen rebreathers, when advance in industrial metalworking made high-pressure gas storage cylinders possible. From 1878 on they were used for work in unbreathable atmospheres in industry and firefighting, and at high altitude, and for escape from submarines; and occasionally for swimming underwater; but the usual way to work or search underwater was still standard diving dress.. Oxygen rebreathers were made with many arrangements of their three main components: oxygen cylinder(s), breathing bag, absorbent canister, and their straps and harnesses and controls.
The Italian Decima Flottiglia MAS, the first body of combat frogmen, was founded in 1938 and went into action in 1940. WWII saw a great expansion of military-related use of rebreather diving. During and after WWII, needs arose in the armed forces to dive deeper than allowed by pure oxygen. That prompted, at least in Britain, design of simple constant-flow "mixture rebreather" variants of some of their diving oxygen rebreathers (= what is now called "nitrox"): SCMBA from the SCBA (Swimmer Canoeist's Breathing Apparatus), and CDMBA from the Siebe Gorman CDBA, by adding an extra gas supply cylinder. Before a dive with such a set, the diver had to know the maximum or working depth of his dive, and how fast his body used his oxygen supply, and from those to calculate what to set his rebreather's gas flow rate to.
During this long period before the modern age of automatic sport nitrox rebreathers, there were some sport oxygen diving clubs, mostly in the USA.
Eventually the Cold War ended and in 1989 the Communist Bloc collapsed, and as a result the perceived risk of sabotage attacks by combat divers dwindled, and Western armed forces had less reason to requisition civilian rebreather patents, and automatic and semi-automatic recreational diving rebreathers with ppO2 sensors started to appear.
As a person breathes, the body consumes oxygen and produces carbon dioxide. Base metabolism requires about 0.25 L/min of oxygen from a breathing rate of about 6 L/min, and a fit person working hard may ventilate at a rate of 95 L/min but will only metabolise about 4 L/min of oxygen  The oxygen metabolised is generally about 4% to 5% of the inspired volume at normal atmospheric pressure, or about 20% of the available oxygen in the air at sea level. Exhaled air at sea level contains roughly 13.5% to 16% oxygen.
The situation is even more wasteful of oxygen when the oxygen fraction of the breathing gas is higher, and in underwater diving, the compression of breathing gas due to depth makes the recirculation of exhaled gas even more desirable, as an even larger proportion of open circuit gas is wasted. Continued rebreathing of the same gas will deplete the oxygen to a level which will no longer support consciousness, and eventually life, so gas containing oxygen must be added to the recycled breathing gas to maintain the required concentration of oxygen.
However, if this is done without removing the carbon dioxide, it will rapidly build up in the recycled gas, resulting almost immediately in mild respiratory distress, and rapidly developing into further stages of hypercapnia, or carbon dioxide toxicity. A high ventilation rate is usually necessary to eliminate the metabolic product carbon dioxide (CO2). The breathing reflex is triggered by carbon dioxide concentration in the blood, not by the oxygen concentration, so even a small buildup of carbon dioxide in the inhaled gas quickly becomes intolerable; if a person tries to directly rebreathe their exhaled breathing gas, they will soon feel an acute sense of suffocation, so rebreathers must chemically remove the carbon dioxide in a component known as a carbon dioxide scrubber.
By adding sufficient oxygen to compensate for the metabolic usage, removing the carbon dioxide, and rebreathing the gas, most of the volume is conserved. There will still be minor losses when gas must be vented as it expands during ascent, and additional gas will be needed to make up volume as the gas is compressed during descent. 
|Application and effect|
|<0.08||Coma ultimately leading to death|
|0.08-0.10||Unconsciousness in most people|
|0.09-0.10||Serious signs/symptoms of hypoxia|
|0.14-0.16||Initial signs/symptoms of hypoxia (normal environment oxygen in some very high altitude areas)|
|0.21||Normal environment oxygen (sea level air)|
|0.35–0.40||Normal saturation dive PO2 level|
|0.50||Threshold for whole-body effects; maximum saturation dive exposure|
|1.0–1.20||Common range for recreational closed circuit set point|
|1.40||Recommended limit for recreational open circuit bottom sector|
|1.60||NOAA limit for maximum exposure for a working diver|
Recreational/technical limit for decompression
|2.20||Commercial/military "Sur-D" chamber surface decompression on 100% O2 at 12 msw|
|2.40||40% O2 nitrox recompression treatment gas for use in the chamber at 50 msw (meters of sea water)|
|2.80||100% O2 recompression treatment gas for use in the chamber at 18 msw|
|3.00||50% O2 nitrox recompression treatment gas for use in the chamber at 50 msw|
The widest variety of rebreather types is used in diving, as the physical and physiological consequences of breathing under pressure complicate the requirements, and a large range of engineering options are available depending on the specific application and available budget. A diving rebreather is safety-critical life-support equipment – some modes of failure can kill the diver without warning, others can require immediate appropriate response for survival.
General operational requirements include:
Special applications may also require:
As pure oxygen is toxic when inhaled at pressure, recreational diver certification agencies limit oxygen decompression to a maximum depth of 6 metres (20 ft) and this restriction has been extended to oxygen rebreathers; In the past they have been used deeper (up to 20 metres (66 ft)) but such dives were more risky than what is now considered acceptable. Oxygen rebreathers are also sometimes used when decompressing from a deep open-circuit dive, as breathing pure oxygen helps the nitrogen diffuse out of the body tissues more rapidly, and the use of a rebreather may be more convenient for long decompression stops.
US Navy restrictions on oxygen rebreather use:
Oxygen rebreathers are no longer commonly used in recreational diving because of the depth limit imposed by oxygen toxicity, but are extensively used for military attack swimmer applications where greater depth is not required, due to their simplicity, light weight and compact size.
Semi-closed circuit rebreathers (SCRs) used for diving may use active or passive gas addition, and the gas addition systems may be depth compensated. They use a mixed supply gas with a higher oxygen fraction than the steady state loop gas mixture. Usually only one gas mixture is used, but it is possible to switch gas mixtures during a dive to extend the available depth range of some SCRs.
Operational scope and restrictions of SCRs:
Closed circuit diving rebreathers may be manually or electronically controlled, and use both pure oxygen and a breathable mixed gas diluent.
Operational scope and restrictions of CCRs: Closed circuit rebreathers are mainly restricted by physiological limitations on the diver, such as maximum operating depth of the diluent mix while remaining breathable up to the surface, though this can be worked around by switching diluent.
An atmospheric diving suit is a small one-man articulated submersible of roughly anthropomorphic form, with limb joints which allow articulation under external pressure while maintaining an internal pressure of one atmosphere. Breathing gas supply could be surface supplied by umbilical, but would then have to be exhausted back to the surface to maintain internal pressure below the external ambient pressure, which is possible but presents pressure-hull breach hazards if the umbilical hoses are damaged, or from a rebreather system built into the suit. As there is a similar problem in venting excess gas, the simple and efficient solution is to make up oxygen as it is consumed and scrub out the carbon dioxide, with no change to the inert gas component, which simply recirculates. In effect, a simple closed circuit oxygen rebreather arrangement used as a life-support system. Since there is usually an adequate power supply for other services, powered circulation through the scrubber should not normally be an issue for normal service, and is more comfortable for the operator, as it keeps the face area clear and facilitates voice communication. As the internal pressure is maintained at one atmosphere, there is no risk of acute oxygen toxicity. Endurance depends on the scrubber capacity and oxygen supply. Circulation through the scrubber could be powered by the diver's breathing, and this is an option for an emergency backup rebreather, which may also be fitted to the suit. A breathing driven system requires reduction of mechanical dead space by using a mouthpiece and counterlung to form a closed loop.
Rebreathers can be primarily categorised as diving rebreathers, intended for hyperbaric use, and other rebreathers used at pressures from slightly more than normal atmospheric pressure at sea level to significantly lower ambient pressure at high altitudes and in space. Diving rebreathers must often deal with the complications of avoiding hyperbaric oxygen toxicity, while normobaric and hypobaric applications can use the relatively trivially simple oxygen rebreather technology, where there is no requirement to monitor oxygen partial pressure during use providing the ambient pressure is sufficient.
Here the rebreather has two breathing bags, and the gas in circuit goes from the lungs to one bag to the absorbent to the other bag and back to the lungs.
This is the earliest type of rebreather and was commonly used by navies and for mine rescue from the early twentieth century. Oxygen rebreathers can be remarkably simple designs, and they were invented before open-circuit scuba. They only supply oxygen, so there is no requirement to control the gas mixture other than purging before use and removing the carbon dioxide.
In some rebreathers, e.g. the Siebe Gorman Salvus, the oxygen cylinder has oxygen supply mechanisms in parallel. One is constant flow, and the other is a manual on-off valve called a bypass valve. Both feed into the counterlung. There is no necessity for a second stage and the gas can be turned on and off at the cylinder valve.
Others, such as the USN Mk25 UBA, are supplied automatically via a demand valve on the counterlung, which will add gas at any time that the counterlung is emptied and the diver continues to inhale. Oxygen can also be added manually by a button which activates the demand valve, equivalent to the purge button on an open-circuit demand valve.
Some simple oxygen rebreathers had no automatic supply system, only the manual feed valve, and the diver had to operate the valve at intervals to refill the breathing bag as the volume of oxygen decreased below a comfortable level. This is task loading, but the diver cannot remain unaware of the need to top up. Control of the volume in the loop would also control buoyancy.
All rebreathers other than oxygen rebreathers may be considered mixed gas rebreathers. These can be divided ino semi-closed circuit, where the supply gas is a breathable mixture containing oxygen and inert diluents, usually nitrogen and helium, and which is replenished by adding more of the mixture as the oxygen is used up, sufficient to maintain a breathable partial pressure of oxygen in the loop, and closed circuit rebreathers, where two parallel gas supplies are used: the diluent, to provide the bulk of the gas, and which is recycled, and oxygen, which is metabolically expended.
These are almost exclusively used for underwater diving, as they are bulkier, heavier, and more complex than closed circuit oxygen rebreathers. Military and recreational divers use these because they provide better underwater duration than open circuit, have a deeper maximum operating depth than oxygen rebreathers and can be fairly simple and cheap. They do not rely on electronics for control of gas composition, but may use electronic monitoring for improved safety and more efficient decompression. An alternative term for this technology is "gas extender".
Semi-closed circuit equipment generally supplies one breathing gas such as air, nitrox or trimix at a time. The gas is injected into the loop at a constant rate to replenish oxygen consumed from the loop by the diver. Excess gas must be constantly vented from the loop in small volumes to make space for fresh, oxygen-rich gas. As the oxygen in the vented gas cannot be separated from the inert gas, semi-closed circuit is wasteful of both oxygen and inert components.
A gas mix which has a maximum operating depth that is safe for the depth of the dive being planned, and which will provide a breathable mixture at the surface must be used, or it will be necessary to change mixtures during the dive.
As the amount of oxygen required by the diver increases with work rate, the gas injection rate must be carefully chosen and controlled to prevent unconsciousness in the diver due to hypoxia. A higher gas addition rate reduces the likelihood of hypoxia but wastes more gas.
This type of rebreather works on the principle of adding fresh gas to compensate for reduced volume in the breathing circuit. A portion of the respired gas is discharged that is in some way proportional to oxygen consumption. Generally it is a fixed volumetric fraction of the respiratory flow, but more complex systems have been developed which exhaust a close approximation of a ratio to the surface respiratory flow rate. These are described as depth compensated or partially depth compensated systems. Gas addition is triggered by low counterlung volume which activates a demand valve.
The simple case of a fixed ratio discharge can be achieved by concentric bellows counterlungs, where the exhaled gas expands both the counterlungs, and while the larger volume outer bellows discharges back to the loop when the diver inhales the next breath, the inner bellows discharges its contents to the surroundings, using non return valves to ensure a one-directional flow. The amount processed during each breath depends on the tidal volume of that breath.
Towards the end of inhalation the bellows bottoms out and activates an addition valve, in much the way that a regulator diaphragm activates the demand valve, to make up the gas discharged by the inner bellows. This type of rebreather therefore tends to operate at a minimal volume.
The fixed ratio systems usually discharge between 10% (1/10) and 25% (1/4) of the volume of each breath overboard. As a result, gas endurance is from 10 times to four times that of open circuit, and depends on breathing rate and depth in the same way as for open circuit. Oxygen fraction in the loop depends on the discharge ratio, and to a lesser extent on the breathing rate and work rate of the diver. As some gas is recycled after breathing, the oxygen fraction will always be lower than that of the make-up gas, but can closely approximate the make-up gas after a loop flush, so the gas is generally chosen to be breathable at maximum depth, which allows it to be used for open circuit bailout. The loop gas oxygen fraction will increase with depth, as the mass rate of metabolic oxygen use remains almost constant with a change in depth. This is the opposite tendency of what is done in a closed circuit rebreather, where the oxygen partial pressure is controlled to be more or less the same within limits throughout the dive. The fixed ratio system has been used in the DC55 and Halcyon RB80 rebreathers. Passive addition rebreathers with small discharge ratios may become hypoxic near the surface when moderate or low oxygen fraction supply gas is used, making it necessary to switch gases between deep and shallow diving.
The depth compensating systems discharge a portion of the diver's tidal volume which varies in inverse proportion to the absolute pressure. At the surface they generally discharge between 20% (1/5) and 33% (1/3) of each breath, but that decreases with depth, to keep the oxygen fraction in the loop approximately constant and reduce gas consumption. A fully depth compensated system will discharge a volume of gas, inversely proportional to pressure, so that the volume discharged at 90m depth (10 bar absolute pressure) will be 10% of the surface discharge. This system will provide an approximately fixed oxygen fraction regardless of depth, when used with the same make-up gas, because the effective mass discharge remains constant.
Partially depth compensating systems are part way between the fixed ratio and the depth compensating systems. They provide a high discharge ratio near the surface, but the discharge ratio is not fixed either as a proportion of respired volume or mass. Gas oxygen fraction is more difficult to calculate, but will be somewhere between the limiting values for fixed ratio and fully compensated systems. The Halcyon PVR-BASC uses a variable volume inner bellows system to compensate for depth.
An active addition system adds feed gas to the breathing circuit and excess gas is dumped to the environment. These rebreathers tend to operate near maximum volume.
The most common system of active addition of make-up gas in semi-closed rebreathers is by use of a constant mass flow injector, also known as choked flow. This is easily achieved by using a sonic orifice, as provided the pressure drop over the orifice is sufficient to ensure sonic flow, the mass flow for a specific gas will be independent of the downstream pressure. The mass flow through a sonic orifice is a function of the upstream pressure and the gas mixture, so the upstream pressure must remain constant for the working depth range of the rebreather to provide a reliably predictable mixture in the breathing circuit, and a modified regulator is used which is not affected by changes in ambient pressure. Gas addition is independent of oxygen use, and the gas fraction in the loop is strongly dependent on exertion of the diver – it is possible to dangerously deplete the oxygen by excessive physical exertion.
Only one model using this gas mixture control principle has been marketed. This is the Interspiro DCSC. The principle of operation is to add a mass of oxygen that is proportional to the volume of each breath. This approach is based on the assumption that the volumetric breathing rate of a diver is directly proportional to metabolic oxygen consumption, which experimental evidence indicates is close enough to work.
The fresh gas addition is made by controlling the pressure in a dosage chamber proportional to the counterlung bellows volume. The dosage chamber is filled with fresh gas to a pressure proportional to bellows volume, with the highest pressure when the bellows is in the empty position. When the bellows fills during exhalation, the gas is released from the dosage chamber into the breathing circuit, proportional to the volume in the bellows during exhalation, and is fully released when the bellows is full. Excess gas is dumped to the environment through the overpressure valve after the bellows is full.
The result is the addition of a mass of gas proportional to ventilation volume, and the oxygen fraction is stable over the normal range of exertion.
The volume of the dosage chamber is matched to a specific supply gas mixture, and is changed when the gas is changed. The DCSC uses two standard mixtures of nitrox: 28% and 46%.
Military, photographic, and recreational divers use closed circuit rebreathers because they allow long dives and produce no bubbles. Closed circuit rebreathers supply two breathing gases to the loop: one is pure oxygen and the other is a diluent or diluting gas such as air, nitrox, heliox or trimix.
A major function of the closed circuit rebreather is to control the oxygen partial pressure in the loop and to warn the diver if it becomes dangerously low or high. Too low a concentration of oxygen results in hypoxia leading to unconsciousness and ultimately death. Too high a concentration of oxygen results in hyperoxia, leading to oxygen toxicity, a condition causing convulsions which can make the diver lose the mouthpiece when they occur underwater, and can lead to drowning. The monitoring system uses oxygen-sensitive electro-galvanic fuel cells to measure the partial pressure of oxygen in the loop. The partial pressure of oxygen in the loop can generally be controlled within reasonable tolerance of a fixed value. This set point is chosen to provide an acceptable risk of both long-term and acute oxygen toxicity, while minimizing the decompression requirements for the planned dive profile.
The gas mixture is controlled by the diver in manually controlled closed circuit rebreathers. The diver can manually control the mixture by adding diluent gas or oxygen. Adding diluent can prevent the loop gas mixture becoming too oxygen rich, and adding oxygen is done to increase oxygen concentration.
In fully automatic closed-circuit systems, an electronically controlled solenoid valve injects oxygen into the loop when the control system detects that the partial pressure of oxygen in the loop has fallen below the required level. Electronically controlled CCRs can be switched to manual control in the event of some control system failures.
Addition of gas to compensate for compression during descent is usually done by an automatic diluent valve.
In 1912 the German firm Drägerwerk of Lübeck introduced a version of standard diving dress using a gas supply from an oxygen rebreather and no surface supply. The system used a copper diving helmet and standard heavy diving suit with a back-mounted set of cylinders and scrubber. The breathing gas was circulated by using an injector system in the loop powered by the added gas. This was developed further with the Modell 1915 "Bubikopf" helmet and the DM20 oxygen rebreather system for depths up to 20 m, and the DM40 mixed gas rebreather which used an oxygen cylinder and an air cylinder for the gas supply, producing a nitrox mixture, for depths up to 40 m.
The US Navy developed a variant of the Mark V system for heliox diving. These were successfully used during the rescue of the crew and salvage of the USS Squalus in 1939. The US Navy Mark V Mod 1 heliox mixed gas helmet is based on the standard Mark V Helmet, with a scrubber canister mounted on the back of the helmet and a inlet gas injection system which recirculates the breathing gas through the scrubber to remove carbon dioxide and thereby conserve helium. The helium helmet uses the same breastplate as a standard Mark V except that the locking mechanism is relocated to the front, there is no spitcock, there is an additional electrical connection for heated underwear, and on later versions a two or three-stage exhaust valve was fitted to reduce the risk of flooding the scrubber. The gas supply at the diver was controlled by two valves. The "Hoke valve" controlled flow through the injector to the "aspirator" which circulated gas from the helmet through the scrubber, and the main control valve used for bailout to open circuit, flushing the helmet, and for extra gas when working hard or descending. Flow rate of the injector nozzle was nominally 0.5 cubic foot per minute at 100 psi above ambient pressure, which would blow 11 times the volume of the injected gas through the scrubber.
Both these systems were semi-closed and did not monitor partial pressures of oxygen. They both used an injector system to recirculate the breathing gas and did not increase work of breathing.
There have been a few rebreather designs (e.g. the Oxylite) which had an absorbent canister filled with potassium superoxide, which gives off oxygen as it absorbs carbon dioxide: 4KO2 + 2CO2 = 2K2CO3 + 3O2; it had a very small oxygen cylinder to fill the loop at the start of the dive. This system is dangerous because of the explosively hot reaction that happens if water gets on the potassium superoxide. The Russian IDA71 military and naval rebreather was designed to be run in this mode or as an ordinary rebreather.
Tests on the IDA71 at the United States Navy Experimental Diving Unit in Panama City, Florida showed that the IDA71 could give significantly longer dive time with superoxide in one of the canisters than without.
This technology may be applied to both oxygen and mixed gas rebreathers, and can be used for diving and other applications.
A liquid oxygen supply can be used for oxygen or mixed gas rebreathers. If used underwater, the liquid-oxygen container must be well insulated against heat transfer from the water. Industrial sets of this type may not be suitable for diving, and diving sets of this type may not be suitable for use out of water due to conflicting heat transfer requirements. The set's liquid oxygen tank must be filled immediately before use.
A cryogenic rebreather removes the carbon dioxide by freezing it out in a "snow box" by the low temperature produced as liquid oxygen evaporates to replace the oxygen used.
A cryogenic rebreather prototype called the S-1000 was built by Sub-Marine Systems Corporation. It had a duration of 6 hours and a maximum dive depth of 200 metres (660 ft). Its ppO2 could be set to anything from 0.2 to 2 bars (3 to 30 psi) without electronics, by controlling the temperature of the liquid oxygen, thus controlling the equilibrium pressure of oxygen gas above the liquid. The diluent could be either nitrogen or helium depending on the depth of the dive. The partial pressure of oxygen was controlled by temperature, which was controlled by controlling the pressure at which liquid nitrogen was allowed to boil, which was controlled by an adjustable pressure relief valve. No control valves other than the nitrogen pressure relief valve were required. Low temperature was also used to freeze out up to 230 grams of carbon dioxide per hour from the loop, corresponding to an oxygen consumption of 2 litres per minute as carbon dioxide will freeze out of the gaseous state at -43.3 °C or below. If oxygen was consumed faster due to a high workload, a regular scrubber was needed. No electronics were needed as everything followed the setting of the nitrogen release pressure from the cooling unit, and the refrigeration by evaporation of liquid nitrogen maintained a steady temperature until the liquid nitrogen was exhausted. The loop gas flow was passed through a counterflow heat exchanger, which re-heated the gas returning to the diver by chilling the gas headed for the snow box (the cryogenic scrubber). The first prototype, the S-600G, was completed and shallow-water tested in October 1967. The S1000 was announced in 1969, but the systems were never marketed.
Although there are several design variations of diving rebreather, all types have a gas-tight reservoir to contain the breathing gas at ambient pressure that the diver inhales from and exhales into. The resrvoir consists of several components connected together by water- and airtight joints. The diver breathes through a mouthpiece or a full-face diving mask with a shut-off valve. This is connected to one or two tubes ducting inhaled and exhaled gas between the diver and a counterlung or breathing bag, which expands to accommodate gas when it is not in the diver's lungs. The reservoir also includes a scrubber containing absorbent material to remove the carbon dioxide exhaled by the diver. Attached to the reservoir there will be at least one valve allowing addition of gas, such as oxygen, and often a diluent gas, from a gas storage container.
There may be valves allowing venting of gas, sensors to measure partial pressure of oxygen and possibly carbon dioxide, and a monitoring and control system. Critical components may be duplicated for engineering redundancy.
There are two basic gas passage configurations: The loop and the pendulum.
The loop configuration uses a one directional circulation of the breathing gas which on exhalation leaves the mouthpiece, passes through a non-return valve into the exhalation hose, and then through the counterlung and scrubber, to return to the mouthpiece through the inhalation hose and another non-return valve when the diver inhales.
The pendulum configuration uses a two-directional flow. Exhaled gas flows from the mouthpiece through a single hose to the scrubber, into the counterlung, and on inhalation the gas is drawn back through the scrubber and the same hose back to the mouthpiece. The pendulum system is structurally simpler, but inherently contains a larger dead space of unscrubbed gas in the combined exhalation and inhalation tube, which is rebreathed. There are conflicting requirements for minimising the volume of dead space while minimising the flow resistance of the breathing passages.
Many rebreathers have their main components in a hard casing for support, protection and/or streamlining. This casing must be sufficiently vented and drained to let surrounding water or air in and out freely to allow for volume changes as the counterlung inflates and deflates, and to prevent trapping large volumes of buoyant air as the diver subnerges, and of water as the diver emerges into air.
The components may be mounted on a frame or inside a casing to hold them together. Sometimes the structure of the scrubber canister forms part of the framework, particularly in side-mount configuration. Position of most parts is not critical to function, but the counterlungs must be positioned so that their centroid of volume is at a similar depth to the centroid of the diver's lungs at most times while underwater, and the breathing tubes to the mouthpiece should not encumber the diver more than necessary, and allow free movement of the head as much as possible.
The parts of a diving rebreather (counterlung, absorbent canister, gas cylinder(s), tubes and hoses linking them), can be arranged on the wearer's body in four basic ways, with the position of the counterlung having a major effect on work of breathing.
For most underwater and industrial rebreathers made in former times (before and during WW2 and for a long time afterwards), the parts of the set and the straps and connections joining them, were not in a casing but were exposed distributed over the torso of the diver or workman. For examples with images, see Siebe Gorman Salvus and Siebe Gorman CDBA.
Back mount is common on the more bulky and heavier units. This is good for support of the weight out of the water, and keeps the front of the diver clear for working underwater.
Chest mount is fairly common for military oxygen rebreathers, which are usually relatively compact and light. It allows easy reach of the components underwater, and leaves the back free for other equipment for amphibious operations. The rebreather can be unclipped from a common harness without disturbing the load on the back. (Image at this link)
Sidemount allows a low profile to penetrate tight restrictions in cave and wreck diving, and is convenient for carrying a bailout rebreather. A sidemount rebreather as the main divcing set can be on one side of the diver's body and can be balanced weight-wise and hydrodynamically by a large bailout cylinder sidemounted on the other side. For more about sidemount rebreather diving, see this Youtube video "Diving Sidemount Rebreathers".
The diver breathes from the rebreather circuit through a bite-grip mouthpiece or an oro-nasal mask which may be part of a full-face mask or diving helmet. The mouthpiece is connected to the rest of the rebreather by breathing hoses. The mouthpiece of a diving rebreather will usually include a shutoff valve, and may incorporate a dive/surface valve or a bailout valve or both. On loop-configured rebreathers, the mouthpiece is usually the place where the non-return valves for the loop are fitted.
The dive/surface valve (DSV) is a valve on the mouthpiece which can switch between the loop and ambient surroundings. It is used to close the loop at the surface to allow the diver to breathe atmospheric air, and may also be used underwater to isolate the loop so that it will not flood if the mouthpiece is taken out of the mouth.
A dive/surface valve which can be switched to close the loop and simultaneously open a connection to an open circuit demand valve is known as a bailout valve, as its function is to switch over to open circuit bailout without having to remove the mouthpiece. An important safety device when carbon dioxide poisoning occurs.
Flexible corrugated synthetic rubber hoses are used to connect the mouthpiece to the rest of the breathing circuit, as these allow free movement of the diver's head. These hoses are corrugated to allow greater flexibility while retaining a high resistance to collapse. The hoses are designed to provide low resistance to flow of the breathing gas. A single breathing hose is used for pendulum (push-pull) configuration, and two hoses for a one-way loop configuration.
The counterlung is a part of the loop which is designed to change in volume by the same amount as the user's tidal volume when breathing. This lets the loop expand and contract when the user breathes, letting the total volume of gas in the lungs and the loop remain constant throughout the breathing cycle. The volume of the counterlung should allow for the maximum likely breath volume of a user, but does not generally need to match the vital capacity of all possible users.
Underwater, the position of the counterlung – on the chest, over the shoulders, or on the back – has an effect on the hydrostatic work of breathing. This is due to the pressure difference between the counterlung and the diver's lung caused by the vertical distance between the two.
Recreational, technical and many professional divers will spend most of their time underwater swimming face down and trimmed horizontally. Counterlungs should function well with low work of breathing in this position, and with the diver upright.
The design of the counterlungs can also affect the swimming diver's streamlining due to location and shape of the counterlungs, if they are not in a casing.
Most passive addition semi-closed diving rebreathers control the gas mixture by removing a fixed volumetric proportion of the exhaled gas, and replacing it with fresh feed gas from a demand valve, which is triggered by low volume of the counterlung.
This is done by using concentric bellows counterlungs – the counterlung is configured as a bellows with a rigid top and bottom, and has a flexible corrugated membrane forming the side walls. There is a second, smaller bellows inside, also connected to the rigid top and bottom surfaces of the counterlung, so that as the rigid surfaces move towards and away from each other, the volumes of the inner and outer bellows change in the same proportion.
The exhaled gas expands the counterlungs, and some of it flows into the inner bellows. On inhalation, the diver only breathes from the outer counterlung – return flow from the inner bellows is blocked by a non-return valve. The inner bellows also connects to another non-return valve opening to the outside environment, and thus the gas from the inner bellows is dumped from the circuit in a fixed proportion of the volume of the inhaled breath. If the counterlung volume is reduced sufficiently for the rigid cover to activate the feed gas demand valve, gas will be added until the diver finishes that inhalation.
The exhaled gases are directed through the chemical scrubber, a canister full of a suitable carbon dioxide absorbent such as a form of soda lime, which removes the carbon dioxide from the gas mixture and leaves the oxygen and other gases available for re-breathing.
Some of the absorbent chemicals are produced in granular format for diving applications, such as Atrasorb Dive, Sofnolime, Dragersorb, or Sodasorb. Other systems use a prepackaged Reactive Plastic Curtain (RPC) based cartridge: The term Reactive Plastic Curtain was originally used to describe Micropore's absorbent curtains for emergency submarine use by the US Navy, and more recently RPC has been used to refer to their Reactive Plastic Cartridges, which are claimed to provide better and more reliable performance than the same volume of granular absorbent material.
The carbon dioxide passing through the scrubber absorbent is removed when it reacts with the absorbent in the canister; this chemical reaction is exothermic. This reaction occurs along a "front" which is a region across the flow of gas through the soda-lime in the canister. This front moves through the scrubber canister, from the gas input end to the gas output end, as the reaction consumes the active ingredients. This front would be a zone with a thickness depending on the grain size, reactivity, and gas flow velocity because the carbon dioxide in the gas going through the canister needs time to reach the surface of a grain of absorbent, and then time to penetrate to the middle of each grain of absorbent as the outside of the grain becomes exhausted. Eventually gas with remaining carbon dioxide will reach the far end of the canister and "breakthrough" will occur. After this the carbon dioxide content of the scrubbed gas will tend to rise as the effectiveness of the scrubber falls until it becomes noticeable to the user, then unbreathable.
In rebreather diving, the typical effective endurance of the scrubber will be half an hour to several hours of breathing, depending on the granularity and composition of the absorbent, the ambient temperature, the size of the canister, the dwell time of the gas in the absorbent material, and the production of carbon dioxide by the diver.
During ascent the gas in the breathing circuit will expand, and must have some way of escape before the pressure difference causes injury to the diver or damage to the loop. The simplest way to do this is for the diver to allow excess gas to escape around the mouthpiece or through the nose, but a simple overpressure valve is reliable and can be adjusted to control the permitted overpressure. The overpressure valve is typically mounted on the counterlung and in military diving rebreathers it may be fitted with a diffuser, which helps to conceal the diver's presence by masking the release of bubbles, by breaking them up to sizes which are less easily detected. A diffuser also reduces bubble noise.
Many rebreathers have "water traps" in the counterlungs or scrubber casing, to stop large volumes of water from entering the scrubber media if the diver removes the mouthpiece underwater without closing the valve, or if the diver's lips get slack and let water leak in. Some rebreathers have manual pumps to remove water from the water traps, and a few of the passive addition SCRs automatically pump water out along with the gas during the exhaust stroke of the bellows counterlung. Others use internal pressure to expel water through the manually overridden dump valve when it is in a low position.
A rebreather must have a source of oxygen to replenish that which is consumed by the diver. Depending on the rebreather design variant, the oxygen source will either be pure or a breathing gas mixture which is almost always stored in a gas cylinder. In a few cases oxygen is supplied as liquid oxygen or from a chemical reaction.
Pure oxygen is not considered to be safe for recreational diving deeper than 6 meters, so closed circuit rebreathers for deeper use also have a cylinder of diluent gas. This diluent cylinder may be filled with compressed air or another diving gas mix such as nitrox, trimix, or heliox. The diluent reduces the percentage of oxygen breathed and increases the maximum operating depth of the rebreather. The diluent is not normally an oxygen-free gas, such as pure nitrogen or helium, and is breathable so it may be used in an emergency either to flush the loop with breathable gas of a known composition or as a bailout gas. Diluent gas is commonly referred to as diluent, dilutent, or just "dil" by divers.
Gas must be added to the breathing loop if the volume gets too small or if it is necessary to change the gas composition.
This has a similar function to an open circuit demand valve. It adds gas to the circuit if the volume in the circuit is too low. The mechanism is either operated by a dedicated diaphragm like in a scuba second stage, or may be operated by the top of a bellows type counterlung reaching the bottom of its travel.
Closed circuit rebreathers usually allow the diver to add gas manually. In oxygen rebreathers this is just oxygen, but mixed gas rebreathers usually have a separate manual addition valve for oxygen and diluent, as either might be required to correct the composition of the loop mixture, either as the standard operating method for manually controlled CCRs, or as a backup system on electronically controlled CCRs. The manual diluent addition is sometimes by a purge button on the ADV.
Constant mass flow gas addition is used on active addition semi-closed rebreathers, where it is the normal method of addition at constant depth, and in many closed circuit rebreathers, where it is the primary method of oxygen addition, at a rate less than metabolically required by the diver at rest, and the rest is made up by the control system through a solenoid valve, or manually by the diver.
Constant mass flow is achieved by sonic flow through an orifice. The flow of a compressible fluid through an orifice is limited to the flow at sonic velocity in the orifice. This can be controlled by the upstream pressure and the orifice size and shape, but once the flow reached the speed of sound in the orifice, any further reduction of downstream pressure has no influence on the flow rate. This requires a gas source at a fixed pressure, and it only works at depths which have a low enough ambient pressure to provide sonic flow in the orifice.
Regulators which have their control components isolated from the ambient pressure are used to supply gas at a pressure independent of the depth.
In passive addition semi-closed rebreathers, gas is usually added by a demand type valve actuated by the bellows counterlung when the bellows is empty. This is the same actuation condition as the automatic diluent valve of any rebreather, but the actual trigger mechanism is slightly different. A passive rebreather of this type does not need a separate ADV as the passive addition valve already serves this function.
Electronically controlled closed circuit mixed gas rebreathers may have part of the oxygen feed provided by a constant mass flow orifice, but the fine control of partial pressure is done by solenoid operated valves actuated by the control circuits. Timed opening of the solenoid valve will be triggered when the oxygen partial pressure in the loop mix drops below the lower set-point.
If the constant mass flow orifice is compromised and does not deliver the correct flow, the control circuit will compensate by firing the solenoid valve more often.
On some technical diving rebreathers it is possible to connect an alternative gas supply into the rebreather, usually using a wet quick-connect system. This is usually a feature of bailout rebreathers and other side-mounted rebreathers, where the rebreather unit is intentionally kept as compact as possible, and the gas supply may be slung on the other side of the diver for convenience and balance. This facility also allows all of the gas carried by a diver to be potentially supplied via a rebreather.
Bailout gas and bailout procedure are closely linked. The procedure must be appropriate for the gas supply configuration. Initial bailout to open circuit is often the first step, even when a bailout rebreather is carried, as it is simple and robust, and some time is needed to get the bailout rebreather ready for use.
Bailout gas supply must be sufficient for safe return to the surface from any point in the planned dive, including any required decompression, so it is not unusual for two bailout cylinders to be carried, and the diluent cylinder to be used as the first bailout to get to a depth where the other gas can be used. On a deep dive, or a long penetration, open circuit bailout can easily be heavier and more bulky than the rebreather, and for some dives a bailout rebreather is a more practical option.
The fundamental requirements for the control of the gas mixture in the breathing circuit for any rebreather application are that the carbon dioxide is removed, and kept at a tolerable level, and that the partial pressure of oxygen is kept within safe limits. For rebreathers which are used at normobaric or hypobaric pressures, this only requires that there is sufficient oxygen, which is easily achieved in an oxygen rebreather. Hyperbaric applications, as in diving, also require that the maximum partial pressure of oxygen is limited, to avoid oxygen toxicity, which is technically a more complex process, and may require dilution of the oxygen with metabolically inert gas.
If not enough oxygen is added, the concentration of oxygen in the loop may be too low to support life. In humans, the urge to breathe is normally caused by a build-up of carbon dioxide in the blood, rather than lack of oxygen. Hypoxia can cause blackout with little or no warning, followed by death.
The method used for controlling the range of oxygen partial pressure in the breathing loop depends on the type of rebreather.
The volume in the loop is usually controlled by a pressure or volume triggered automatic diluent valve, and an overpressure relief valve. The automatic diluent valve works on the same principle as a demand valve to add diluent when the pressure in the loop is reduced below ambient pressure, such as during descent or if gas is lost from the loop. The set may also have a manual addition valve, sometimes called a bypass. In some early oxygen rebreathers the user had to manually open and close the valve to the oxygen cylinder to refill the counterlung each time the volume got low.
Instrumentation may vary from the minimal depth, time and remaining gas pressure necessary for a closed circuit oxygen rebreather or semi-closed nitrox rebreather to redundant electronic controllers with multiple oxygen sensors, redundant integrated decompression computers, carbon dioxide monitoring sensors and a head-up display of warning and alarm lights with a sound and vibration alarm.
Alarms may be provided for a few malfunctions. The alarms are electronically controlled and may rely on input from a sensor and processing by the control circuitry. These may include:
If a rebreather alarm goes off there is a high probability that the gas mixture is deviating from the set mixture. There is a high risk that the gas in the rebreather loop will soon be unsuitable to support consciousness. A good general response is to add diluent gas to the loop as this is known to be breathable. This will also reduce carbon dioxide concentration if that is high.
Work of breathing is the effort required to breathe. Part of the work of breathing is due to inherent physiological factors, part is due to the mechanics of the external breathing apparatus, and part is due to the characteristics of the breathing gas. A high work of breathing may result in carbon dioxide buildup in the diver, and reduces the diver's ability to produce useful physical effort. In extreme cases work of breathing may exceed the aerobic work capacity of the diver, with fatal consequences.
Work of breathing of a rebreather has two main components: Resistive work of breathing is due to the flow restriction of the gas passages causing resistance to flow of the breathing gas, and exists in all applications where there is no externally powered ventilation. Hydrostatic work of breathing is only applicable to diving applications, and is due to difference in pressure between the lungs of the diver and the counterlungs of the rebreather. This pressure difference is generally due to a difference in hydrostatic pressure caused by a difference in depth between lung and counterlung, but can be modified by ballasting the moving side of a bellows counterlung.
Resistive work of breathing is the sum of all the restrictions to flow due to bends, corrugations, changes of flow direction, valve cracking pressures, flow through scrubber media, etc., and the resistance to flow of the gas, due to inertia and viscosity, which are influenced by density, which is a function of molecular weight and pressure. Rebreather design can limit the mechanical aspects of flow resistance, particularly by the design of the scrubber, counterlungs and breathing hoses. Diving rebreathers are influenced by the variations of work of breathing due to gas mixture choice and depth. Helium content reduces work of breathing, and increased depth increases work of breathing. Work of breathing can also be increased by excessive wetness of the scrubber media, usually a consequence of a leak in the breathing loop, or by using a grain size of absorbent that is too small.
The semi-closed rebreather systems developed by Drägerwerk in the early 20th century as a scuba gas supply for Standard diving dress, using oxygen or nitrox, and the US Navy Mark V Heliox helmet developed in the 1930s for deep diving, circulated the breathing gas through the helmet and scrubber by using an injector system where the added gas entrained the loop gas and produced a stream of scrubbeded gas past the diver inside the helmet, which eliminated external dead space and resistive work of breathing, but was not suitable for high breathing rates.
There are safety issues specific to rebreather equipment, and these tend to be more severe in diving rebreathers. Methods of addressing these issues can be categorised as engineering and operational approaches. Development of engineering solutions to these issues is ongoing and has been relatively rapid, but depends on the affordable availability of suitable technology, and some of the engineering problems, such as reliability of oxygen partial pressure measurement, have been relatively intractable. Other problems, such as scrubber breakthrough monitoring and automated control of gas mixture have advanced considerably in the 21st century, but remain relatively expensive. Work of breathing is another issue that has room for improvement, and is a severe limitation on acceptable maximum depth of operation, as the circulation of gas through the scrubber is almost always powered by the lungs of the diver. Fault tolerant design can help with making failures survivable.
Some of the hazards are due to the way the equipment works, while others are related to the environment in which the equipment is used.
Hypoxia can occur in any rebreather which contains enough inert gas to allow breathing without triggering automatic gas addition.
In an oxygen rebreather this can occur if the loop is not sufficiently purged at the start of use. Purging should be done while breathing off the unit so that the inert gas in the user's lungs is also removed from the system.
A dangerously high partial pressure of oxygen can occur in the breathing loop for several reasons:
Carbon dioxide buildup will occur if the scrubber medium is absent, badly packed, inadequate or exhausted. The normal human body is fairly sensitive to carbon dioxide partial pressure, and a buildup will be noticed by the user. However, there is not often much that can be done to rectify the problem except changing to another breathing gas supply until the scrubber can be repacked. Continued use of a rebreather with an ineffective scrubber is not possible for very long, as the levels will become toxic and the user will experience extreme respiratory distress, followed by loss of consciousness and death. The rate at which these problems develop depends on the volume of the circuit and the metabolic rate of the user.
Carbon dioxide buildup can also occur when a combination of exertion and work of breathing exceeds the capacity of the user. If this occurs where the user cannot reduce exertion sufficiently, it may be impossible to correct. In this case it is not the scrubber that fails to remove carbon dioxide, but the inability of the diver to circulate gas efficiently through the scrubber against the frictional resistance of the circuit causing the problem. This is more likely to occur with diving rebreathers at depths where the density of the breathing gas is severely elevated, or when water in the scrubber obstructs gas flow.
High partial pressures of oxygen greatly increase fire hazard, and many materials which are self-extinguishing in atmospheric air will burn continuously in a high oxygen concentration. This is more of a risk for terrestrial applications such as rescue and firefighting than for diving, where the ignition risk is relatively low.
Caused by a loop flood reaching the absorbent canister, so only applicable in immersed applications.
Diving rebreathers are susceptible to some failure modes which cannot occur in other breathing apparatus.
The term "break-through" means the failure of the scrubber to continue removing suffient carbon dioxide from the gas circulating in the loop. This will inevitably happen if the scrubber is used too long, but can happen prematurely in some circumstances. There are several ways that the scrubber may fail or become less efficient:
Flooding of the breathing loop can occur due to a leak at a low point in the loop where internal gas pressure is less than the external water pressure. One of the more common ways this can happen is if the mouthpiece is dislodged or removed from the diver's mouth without first closing the dive/surface valve or switching to bailout. This can happen due to accidental impact or through momentary inattention. Depending on the layout of the loop and the attitude of the rebreather in the water, the amount of water ingress can vary, as can the distance it travels into the air passages of the breathing loop. In some models of rebreather a moderate amount of water will be trapped at a low point in a counterlung or the scrubber housing, and prevented from reaching the absorbent in the scrubber. Some rebreathers have a system to expel water trapped in this way, either automatically through the vent valve, such as in the Halcyon RB80 and the Interspiro DCSC, or manually by using a small pump.
There are several places on a rebreather where gas leakage can cause problems. Leakage can occur from the high and intermediate pressure components, and from the loop, at pressure slightly above ambient. The effects on system integrity depend on severity of the leak. If only small volumes of gas are lost the leak may be tolerable for the rest of the dive, but a leak may become more severe, depending on the cause, and may in some cases deteriorate catastrophically.
Oxygen monitoring failure can lead to incorrect partial pressure of oxygen in the breathing gas. The consequences can include hypoxia, hyperoxia, and incorrect decompression information, all three of which are potentially life-threatening.
The methods available for monitoring the condition of the scrubber and predicting and identifying immenent breakthrough include:
Fault tolerance is the property that enables a system to continue operating properly in the event of the failure of some of its components. If its operating quality decreases at all, the decrease is proportional to the severity of the failure, as compared to a naively designed system, in which even a small failure can cause total breakdown. Fault tolerance is particularly important in high availability or safety-critical systems. The ability to maintain functionality when portions of a system break down is referred to as graceful degradation.
The basic closed circuit oxygen rebreather is a very simple and mechanically reliable device, but it has severe operational limitations due to oxygen toxicity. The approaches to safely extending the depth range necessitate a variable breathing gas mixture. Semi-closed rebreathers tend to be inefficient for decompression, and not entirely predictable for gas composition in comparison with a precisely controlled closed circuit rebreather. Monitoring the gas composition in the breathing loop can only be done by electrical sensors, bringing the underwater reliability of the electronic sensing system into the safety critical component category.
There are no formal statistics on underwater electronics failure rates, but it is likely that human error is more frequent than the error rate of electronic dive computers, which are the basic component of rebreather control electronics, which process information from multiple sources and have an algorithm for controlling the oxygen injection solenoid. The sealed dive computer package has been around for long enough for the better quality models to have become reliable and robust in design and construction.
An electronically controlled rebreather is a complex system. The control unit receives input from several sensors, evaluates the data, calculates the appropriate next action or actions, updates the system status and displays, and performs the actions, in some cases using real-time feedback to adapt the control signal. The inputs include signals from one or more of pressure, oxygen and temperature sensors, a clock, and possibly helium and carbon dioxide sensors. There is also a battery power source, and a user interface in the form of a visual display, user input interface in the form of button switches, and possibly audio and vibratory alarms.
In a minimal eCCR the system is very vulnerable. A single critical fault can necessitate manual procedures for fault recovery or the need to bail out to an alternative breathing gas supply. Some faults may have fatal consequences if not noticed and managed very quickly. Critical failures include power supply, non-redundant oxygen sensor, solenoid valves or control unit.
The purely mechanical components are relatively robust and reliable and tend to degrade non-catastrophically, and are bulky and heavy, so the electronic sensors and control systems have been the components where improved fault tolerance has generally been sought. Oxygen cell failures have been a particular problem, with predictably serious consequences, so the use of multiple redundancy in oxygen partial pressure monitoring has been an important area of development for improving reliability. A problem in this regard is the cost and relatively short lifespan of oxygen sensors, along with their relatively unpredictable time to failure, and sensitivity to the environment.
To automatically detect and identify oxygen sensor malfunction, either the sensors must be calibrated with a known gas, which is very inconvenient at most times during a dive, but is possible as an occasional test when a fault is suspected, or several cells can be compared and the assumption made that cells with near identical output are functioning correctly. This voting logic requires a minimum of three cells, and reliability increases with number. To combine cell redundancy with monitoring circuit, control circuit and display redundancy, the cell signals should all be available to all monitoring and control circuits in normal conditions. This can be done by sharing signals at the analog or digital stage – the cell output voltage can be supplied to the input of all monitoring units, or the voltages of some cells can be supplied to each monitor, and the processed digital signals shared. The sharing of digital signals may allow easier isolation of defective components if short circuits occur. The minimum number of cells in this architecture is two per monitoring unit, with two monitoring units for redundancy, which is more than the minimum three for basic voting logic capability.
The three aspects of a fault tolerant rebreather are hardware redundancy, robust software and a fault detection system. The software is complex and comprises several modules with their own tasks, such as oxygen partial pressure measurement, ambient pressure measurement, Oxygen injection control, decompression status calculation and the user interface of status and information display and user inputs. It is possible to separate the user interface hardware from the control and monitoring unit, in a way that allows the control system to continue to operate if the relatively vulnerable user interface is compromised.
Rebreathers are more complex to use than open circuit scuba, and have more potential points of failure, so acceptably safe use requires a greater level of skill, attention and situational awareness, which is usually derived from understanding the systems, diligent maintenance and overlearning the practical skills of operation and fault recovery. Fault tolerant design can make a rebreather less likely to fail in a way that immediately endangers the user, and reduces the task loading on the diver which in turn may lower the risk of operator error.
Rebreather technology has advanced considerably, often driven by the growing market in recreational diving equipment. Innovations include:
Accurate and reliable oxygen partial pressure measurement is one of the most problematic factors in rebreather diving safety. Control systems using this data have developed to the extent that they are robust and reliable, and the use of an independent backup improves the reliability to about as good as for any other component. The weakest point is the sensors, which are prone to several modes of failure, some of which are relatively insidious as the cell may pass a normobaric calibration and fail when the partial pressure is near the high end of the acceptable working range, which is also the range in which constant partial pressure diving has the maximum benefit. When it has been possible to infer the cause, the leading cause of rebreather fatalities is hypoxia, at approximately 17%, with hyoperoxia assumed in an additional 4% of cases. If these trends extend into the range of indeterminate cases, it is possible that inappropriate oxygen content is involved in 30% of rebreather fatalities.
The standard method for improving reliability of oxygen monitoring has been multiple redundancy – the use of 3 or more sensors – and using the multiple data inputs with a voting logic system to try to identify failure of a sensor in time to make a controlled and safe termination of the dive. Voting logic normally assumes that if one sensor produces a reading significantly differing from two or more others when exposed to the same environment, the outlier is faulty, and the input of the others is assumed accurate. Unfortunately this is not always the case, and there have been cases where the outlier sensor was most correct. It has been shown that the reliability of this system is lower than originally expected due to a lack of sufficient statistical independence of the three sensors, and that outcomes are not symmetrical – the effects of faulty low or high partial pressure readings are also depth dependent.
If a sensor gives relatively static output with little response to variations in depth and temperature, and changes in gas composition due to use, gas addition, incomplete mixing or loop turbulence, it is likely that the sensor may not be responding correctly, and when two sensors follow a similar pattern of response this is a warning that both may be defective. Algorithms that track sensor output against expected output taking known changes into account can indicate reliability of the sensors. This method of monitoring sensors is known as passive sensor validation (PSV), can be used to improve reliability of sensor integrity assessment, and can be used in the control system to make more reliable decisions on which sensors are most likely to be giving trustworthy output in comparison with voting logic based only on calibration values for the sensors. PSV is an improvement on simple voting logic but is still susceptible to errors related to statistical independence of components.
Early work on design of an automatic sensor validation system, in which the rebreather control system would periodically inject gas of known composition onto the oxygen sensors during the dive and use the output to determine the viability of the sensor response with greater precision and accuracy than a human diver, was started in 2002, and further developed to be used on the Poseidon/Cis-Lunar MK-VI rebreather. This "Active Sensor Validation" (ASV) system has been refined over thousands of hours of field test diving in varied conditions
The ASV system has become more sophisticated than the manual implementation in the Cis-Lunar MK-5P. It involves more than comparing the measured PO2 value from the sensor with the calculated PO2 of the diluent at the current depth. In the implementation in the Poseidon rebreathers the computer automatically injects either diluent or oxygen directly onto a single primary oxygen sensor every five minutes during a dive. The algorithm takes into account current depth, FO2 of the injected gas, ambient temperature, duration of gas injection, and calibration values for the sensor for that dive to predict how the sensor should respond over the next few seconds after each gas injection, and compares that with the measured results to produce a confidence level for correct sensor performance.
This type of sensor validation test can identify several modes of failure by the ways the measured values deviate from expected values with variations of calculated partial pressure of the test gas, and is capable of detecting failures due to incorrect temperature readings, incorrect input of the FO2 of the diluent condensation on the oxygen sensor, a defective oxygen sensor, validation gas supply failure and other reasons that would not be detected by voting logic.
The oxygen sensors for most rebreathers are calibrated at the surface before the dive using air or 100% oxygen at normal atmospheric pressure. These are reliable calibration points but the range of operational partial pressures may extend beyond these calibration points, and if the sensors are calibrated for a linear response between these conditions and the response is extrapolated, for setpoints above 1 bar, which is standard practice, the control system must operate outside of the range for which restonse is known to be linear. One of the most common modes of failure is for a cell to become current-limited as it ages. The internal impedance changes as the anode is consumed by the reaction which produces the output current, and the response becomes non-linear at higher oxygen partial pressures. The signal may indicate a lower partial pressure and does not increase proportionately as oxygen is added, leading to a loop oxygen partial pressure that may increase to dangerous levels without warning. A way of validating the sensors at high partial pressures is to expose the sensor to higher PO2 than the upper setpoint by exposing it to pure oxygen at a depth of 6 m, for a PO2 of 1.6 bar during the dive, or at 1.6 bar or more in a calibration pressure pot. Both these methods are cumbersome and the in-water method may cause spiking of the PO2 during descent. A variation of the ASV system using oxygen, called a hyperoxic linearity test (HLT), uses oxygen as the flushing gas at 6 m, which can check that the sensor is linear to 1.6 bar PO2, and if it fails, the set point can be automatically reduced to within the linear range established during calibration. A single sensor with PSV and ASV has been shown to be more reliable than three sensors with conventional voting logic. The effectiveness of cell validation algorithms is expected to improve with the acquisition of more field data gathered by the rebreather control systems.
Hypercapnia has been identified as one of the most prevalent factors in rebreather diving fatalities. This is generally a consequence of scrubber failure to remove carbon dioxide as fast as it is produced, which may be caused by any one or a combination of spent, wet, or inadequately packed, absorbent material, incorrectly designed or assembled canisters, mismatch of absorbent and canister design, or absorbent used beyond its operational range. Higher carbon dioxide partial pressure in the loop leads to higher levels of carbon dioxide in the blood and tissue, which can have a range of symptoms including respiratory distress, increased susceptibility to CNS oxygen toxicity, disorientation, and loss of consciousness.
Most rebreather designs have relied on very conservative time-based limits for absorbent duration based on experimental testing, using cold conditions and high workloads and high depth pressures. The usually unnecessarily high conservatism encourages divers to stretch the absorbent duration, which works well enough until it doesn't, often without warning, which can have serious consequences. A mor sophisticated method is to base absorbent duration limits on metabolic oxygen consumption, as a proxy for metabolic carbon dioxide production, which is reasonably stable for most people most of the time, and can compensate fairly well for variations in exertion and base metabolism, but does not compensate reliably for depth and pressure effects on absorbent function.
A more direct and empirical approach is to take advantage of the production of heat and rise in temperature of the active zone of the absorbent in the scrubber. More carbon dioxide is absorbed by the first zone of relatively unused absorbent that it reaches as the breathing gas passes through the scrubber, and this relatively active zone progresses through the canister as the zone first reached by the gas is exhausted, and more reaction occurs further along. This reaction front is at a higher temperature than the spent absorbent, and the absorbent not yet exposed to high carbon dioxide levels, and the front progresses along the scrubber until part of it reaches the end of the absorbent, and unscrubbed gas breaks through to the other side of the loop, after which there is a fairly constant and irreversible increase in inspired carbon dioxide. Some rebreather manufacturers have developed linear temperature probes which identify the position of the reactive front, allowing the user to estimate the remaining duration of the canister.
None of these methods can detect canister bypass and they have little ability to identify completely spent absorbent, channeling, badly packed, or inappropriate absorbent material, but this can be done by a direct measurement of carbon dioxide partial pressure in the inhalation side of the loop.
Research and development of carbon dioxide sensors goes back at least as far as the early 1990s when Teledyne Analytical Instruments and Cis-Lunar Development Laboratories worked on a sensor for the Cis-Lunar MK-III rebreather, which was accurate in laboratory conditions but in the field susceptible to high humidity and condensation causing unreliable readings, which was a recurring problem with real-time carbon dioxide measurement. High pressures also caused problems for depth compensation. In 2009 VR Technologies released a commercial CO2 sensor using hydrophobic membranes to keep the sensors dry without excessively reducing gas flow to the sensors.
Since then other manufacturers have introduced their products to the market but they have not gained widespread use. They are relatively expensive, give unreliable readings in some circumstances, can only detect failure of the scrubber, and do not predict remaining duration. A combination of temperature measurement and post scrubber CO2 measurement can give both prediction and failure warning, for increased cost and complexity.
Placement of the sensor in the loop can affect sensitivity to actual CO2 content of inspired gas. Measuring gas in the mouthpiece has problems due to dead space, and mounting in the inhalation hose near the mouthpiece makes the sensor sensitive to small leaks in the inhalation check valve, while also able to detect high CO2 due to major check valve leaks which would cause a big increase in dead space, which would not be detected if the sensor is further upstream in the loop.
Furthermore, increased levels of CO2 in inspired gas is only one cause of hypercapnia. It is also affected by work of breathing, diver fitness, respiratory ventilation patterns, and other behavioural, physiological, and mechanical factors. A better option would be to measure both inhaled and exhaled CO2 levels, and this would require sensors that are fast and reliable in wet conditions, and reasonably inexpensive
Following the strong endorsement by Rebreather Forum 3 of the use of written checklists to improve safety, Cis-Lunar Development Laboratories programmed an electronic pre-dive checklist into their MK-5P rebreather operating system, as a way to prevent the user from neglecting to carry out the recommended checks before use. This was considered successful and implemented on later generations in the Poseidon MK-VI and SE7EN rebreathers, and developed to include robust internal diagnostics for the core electronic components and software, and automatic calibration of the oxygen sensor cells at normobaric pressures. Failure to complete the full checklist results in a range of alarms if the user attempts to dive with the unit. While not entirely foolproof – Oxygen cells are not calibrated at hyperbaric working pressures – a number of safety critical errors will be picked up and the diver made aware of them. The software also logs the steps and data from the pre-dive check and this has been valuable for accident analysis. The predive checks also take less time and require no paper or user logging effort. This system has been shown to reduce risk and has been adopted by several manufacturers.
The user interface of the rebreather control system is where information is exchanged between the diver and the electronic control system, and is an area with several possibilities for errors, both of user input and data interpretation, some of which could have serious or fatal consequences. The intrinsically higher risk of mechanical failure due to high complexity can be compensated by engineering redundancy, both of the control system and bailout gas supply, and appropriate training. The design of the human–machine interface (HMI) can be improved to reduce the risk of misunderstanding and error, and training can focus on correct interpretation of the information and appropriate response. The HMI usually has two main components, displays and alarms, and many of the alarms are associated with specific visual information.
A challenge of designing effective alarms is to ensure that the diver is not distracted by irrelevant information and that they are not triggered too easily, which habituates the diver to paying less attention, and while possibly fulfilling legal requirements regarding warnings and alarms, may make the equipment functionally less safe to use. One strategy to avoid this problem is to target different senses – auditory, visual and tactile – sometimes based on a vibratory output to the mouthpiece.
An effective display ensures that the user gets the information they need when they need it, and the information they want when they want it, in a form that is immediately recognised and unambiguously understood. When too much information is presented at a time of stress, the user may be confused or unable to distinguish the useful information in time to use it effectively. At other times more detailed information may be useful or necessary to make a correct decision. Multiple displays, or multiple views on the same display can help with this.
A trend in rebreather displays that is predicted to become more widespread, is the use of advanced head-up displays, which can provide a wider range of information by using an array of coloured lights or more complex graphical or alphanumeric displays that remain peripherally visible to the diver at all times, and only require eye movement to become fully readable.
A major logistical problem for long and deep rebreather dives is the volume of bailout equipment that must be carried to allow a safe return to the surface from any point of the dive after irrecoverable failure of the primary system. The open circuit option can become extremely bulky and awkward to manage, and while more compact and efficient, the rebreather option has its own set of logistic challenges.
One of the main design challenges in developing a closed circuit bailout system for rebreathers is to maintain the bailout set in a condition ready for use at all depths. This implies breathable gas for the depth, though not necessarily optimised, as the mix can be brought to set point quite rabidly after bailout, and a gas volume that does not vary excessively, so that buoyancy control is not unduly complicated. The bulk of the system must be manageable, and the bailout set mouthpiece must be easily accessible, but secure. Since bailout rebreathers are most likely to be used on dives with large decompression obligations, the switch to bailout must be accommodated by the decompression management system. If real-time monitoring of oxygen partial pressure is included in decompression computation, it must be possible to transfer this facility between units, without compromising their independence. Task-loading of the diver in managing the two loops must not be excessive, as the diver is recognised as the least reliable aspect of the operation, and may be under significant stress when bailout becomes necessary.
Data logged from rebreather dives is useful for accident analysis, testing and development of rebreathers, and for diver educational purposes. Dive profile logging by integrated decompression computers is also of value for research into effectiveness of decompression schedules. Aggregation of such data can provide insights into diving patterns across the population of users and help in analysing risk.
The control systems of electronic rebreathers have continued to increase in processing and storage capacity, and in parallel, their capacity for capturing data at increased granularity and precision has increased. In 1994 the Cis-Lunar Mk-IV data logging system recorded data at several hundred points per hour of dive time, and by 1997 the Cis-Lunar Mk-5P was logging over a thousand points per hour. By 2007 the Poseidon MK-VI Discovery was logging between 15,000 and 25,000 points per hour, and in 2016 the Poseidon SE7EN recorded more than double that quantity, in alignment with the recommendations of Rebreather Forum 3, which states:
The forum recommends that all rebreathers incorporate data-logging systems that record functional parameters relevant to the particular unit and dive data and that allow download of these data. Diagnostic reconstruction of dives with as many relevant parameters as possible is the goal of this initiative. An ideal goal would be to incorporate redundancy in data-logging systems and, as much as practical, to standardize the data to be collected
Some of the logged data is specific to the rebreather model, and is not appropriate for general analysis, but some data is useful for external analysis of user population and diving practices which could improve understanding of behaviour and safety analysis.
IDA-71 with lid of casing opened showing interior
IDA-71 mask, DSV and breathing hoses
Inspiration with casing open showing interior
Liberty closed circuit rebreathers
rEvo rebreather back view, right side