So you want to build a breather, huh?

By Rodney Nairne, AARG. So, here is to the bliss of ignorance and the aparent fruitlessness of the pursuit of knowledge:

Samuel Johnson, The Idler (Saturday, 25 November 1758)


Three almost identical homebuilt rebreathers

(AARG - Australian Amateur Rebreather Group. AARG is also the sound divers make when their rebreathers become flooded) AARG provides this information with the goal of promoting the safest possible use of rebreathers in the diving community.

I think it was Henry Ford who said only a fool learns from his mistakes.

So how to built your own rebreather? It may be more instructive to point out the pitfalls and offer examples rather than provide a prescribed design. Certainly the breathers of the AARG are not perfect, and we don’t recommend you just copy them. Part of our training was working things out for ourselves, from scratch. Also, for the sake of brevity I have deliberately skipped detailed explanations of the components and functions of generic rebreathers. Look elswhere for that! It’s out there!

Besides if you don’t know what I’m talking about here you need to step back a little and get some background info first.

Mouthpiece:

Keep the bore at least 25mm, preferably 35mm to reduce breathing resistance. Keep the bore below 45mm and the check valves close to the midpoint to minimise dead space, hence CO2 retention.

Check Valves:

Make sure they don’t shrink with age, and look out for warpage if you use quite hot water when cleaning.

Hoses:

Smooth bore hoses have lower back pressures, but you will need corregated hose at the mouthpiece for flexibility. Keep this hose bore above 25mm ideally around 35mm.

Breathing Bags/Counterlung:

Should be large enough to provide a full breath. Usefull secondary function is as a water trap. Smooth inner surface makes for easy cleaning, ie neoprene is a bad idea unless you use the slick stuff. Ideally you will use a material that can double as it’s own gasket material, so you don’t need glue or sealents. We at the AARG think breathing bags constructed with BC material are cheap and laugh at you if you paid dollars for it.

Cannister:

According to the Molecular Products, the intergannular space of the cannister should equal the volume of a normal breath (or 1/2 a normal breath in the case of the split counterlung). Assuming a 3l normal breath, and inntergrannular space of 40% of the volume of the scrubber material, non split counterlung units would have to have 3/0.4 or a 7.5l capacity scrubber! In the same case the split counterlung requires only a 3.75 l capacity, which is about the optimal volume if you consider overall package size.

Larger volumes require either radial* or big bore cannisters to both accomadate the volume of material and keep the work of breathing reasonable. These two options should be eliminated for reasons of complexity of manufacture and physical size respectively.

According to unconfirmed sources the aspect ratio of simple cylindrical scrubbers preferably should be 1 to 1.75 to 1, with a diameter of about 6 inches.

* Axial flow: imagine a pipe, with gas flowing through it from one end to the other. This is axial flow as it is parallel to the axis of the pipe. The surface area is cross section of the tube. Breathing resistence is proportinal to length of the bed along the axis of gas flow and inversly proportional to cross section.

Radial Flow: Imagine the same tube but with two inner tubes: a small central tube and a large tube, which leaves just a small gas space between it and the outer tube. The two inner tubes have holes to allow gas transfer. In this arrangenent the gas enters along the outer and 1st inner tube, passes thru the scrubber bed along the radius of the cannister, (ie the gas flow thru the bed is at 90 degrees to the flow in the axial scrubber) into the central tube and out.

This arrangement gives a lot more cross sectional area and a shorter path for the gas thru the scrubber bed. The gas velocity is also reduced over the axial design, due to the larger cross sectional area.

Here is an example:

Assume a 150mm (6") pipe, 170mm (7") long. Such an axial flow cannister has a 3litre capacity. The cross sectional area is 28 sq inches.

Now consider the radial flow cannister. Such a cannister if it were to remain a 6’ diameter with a 1 1/2" inner and 5" intermediate tube diameter, would need to be slightly longer (10") to have the same scrubber capacity. But the cross sectional area (if measured in the middle of the bed if calculated on a 1.625" radius pipe, 10" long) is roughly 100 sq inches, or 3 times that of an axial design. or 100sq inches.

That said, the author has had no difficulty with a 6" axial flow cannister down to 280feet, provided the appropriate diluent (heliox/rich trimix) is used. Subjective resistance has been noted at 100 feet with the small mesh size Sofnolime with air diluent.

O2 SENSORS:

MUST be temperature compensated. Large temperature changes occur between calibration and operating equilibrium.

Apparently, a big danger is if the sensors are at a lower temperature than the gas around them whilst this gas is saturated with water vapour. The resulting condensation on the sensors could block the membrane, causing sensor to read in the green regardless of the actual level of oxygen. This is a definate widow maker people.

The easy solution is vigilance whilst operating the unit (To know your ememy’s ways removes his element of surprise). If you are using a manual unit, which is the only wise choice, you MAY be alerted by static Oxygen PP02 readings. If the readings are not changing with time, either a) you are already dead, b) the sensors have water on them, or c) both a&b.

Avoid rapidly increasing CO2 production whilst the cannister is already cooking along (several hours into a dive) as this tends to produce the most water vapour in the cannister. (2 byproducts of the chemical reaction are heat and water. At high workloads it is possible to make steam come out of the scrubber!)

Keeping the sensors at the same temp as the gas stream MAY prevent exessive condensation HOWEVER a rapid change in workload will spike gas temp and make the sensors an ideal condensation point due to temperature lag. This problem has had literally MILLIONS of dollars thrown at it. Solutions exist such as heat exchangers, designed to drop humidity levels.** Holding these things down in physical size is the problem, whilst keeping them simple.

Where to add diluent and Oxygen?

The simple answer is as close as possible to directly upstream and downstream of the mouthpiece respectively. In a split counterlung design this means the inhalation and exhalation bags respectively.

Electronics:

Any use of electronics should be avoided in diving related life support, but if it is necessary, it should be used as a monitoring system only. Diving manual is not hard and actually creates less stress in a safe confident diver, who you could imagine would never choose to let electronics determine his destiny. It is a pity that recreational units are following the norm for military units and having electronic control systems; military units are specialised devices with specific applications.

Things to avoid!

Direct injection of oxygen

The ablity to add freeflow oxygen into your loop (deeper than 20-30 feet) is perilious indeed. You should think up a way to avoid the possibility of you making this error.

Swimming pool rebreather genius:

These guys always come off as being in command of thier surroundings. There confidence level is amazing! This is by far the largest group of instructors (and some manufacturers) out there. Ask to see there logbook to show there experience level: eliminate all dives that can be done on a single 80 of air, all the swimming pool dives (this could take some time) all the dives below 100feet on a nitrox mix (as this is evidence of stupidity if done on a fully closed rebreather). Next eliminate all deep mix dives conducted in a chamber. With whats left, ask yourself if this guy has enough experience to pass on to justify what he is about to charge you.

Ask the A.A.R.G.