Potential Safety Improvements in Rebreather Design

[KeithG]

Have not read all in this thread, will read tonight with interest. (i think) i read the other thread and did not comment. I believe my input is more appropriate to this thread. so i will now comment.

some observations...
- making it impossible to re-assemble a device incorrectly is generally very easy as long as the manufacturer makes all "fittings" different (simple problem,simples olution)
- making it possible so that EVERY idiot in the world can identify that a device has been assembled correctly is IMPOSSIBLE - no solution exists!
- so if yur average idiot is using the device, they are set up for failure if disassembly / reassembly is required

My classic example is a modern automobile. No assembly required. A very high percentage of idiots are able to use cars on a daily basis. Despite this, there is a larger than zero number of occurrences where the average idiot is able to put themselves (and others) in danger. Consider worn out squealing brakes.

Rebreathers are still "user servicable". Average idiots (including me!) are not likely capable of achieving perfect maintenance.

So a larger than zero risk exists. I believe the manufacturers are doing all that is possible to make sure potential users understand this.

In a society based upon free choice, what more SHOULD we do?

</nanny state>
</soap box>

 

[descent]

Human interface doesn’t just apply to software. I believe the term was used in the aircraft industry during WWII. ...

Lately, any accident forum discussion where human interaction played a major part always reminds me of the recent Boeing 777 crash at SFO.

On a gorgeous July day, a modern widebody airliner with four highly qualified pilots aboard flew for many miles in a straight line only to slam into the seawall made of boulders a few hundred feet short of an unobstructed 11,000 foot runway.

No one in the cockpit was task-overloaded. The article linked above refers to but does not link to the one on crew resource management, a system where the entire crew provides checks and balances on the PIC.

Diving a rebreather is a more solitary experience. There is no first officer to tap you on the shoulder and say "Excuse me sir, are you feeling a little hypoxic?"

Howard Hall pointed out in his famous essay about rebreather blunders that continuous suspicion about the state of his rig coupled with a habit of routinely scanning the outputs was sometimes the only thing that allowed him to return alive.

---------- Post added December 2nd, 2014 at 10:55 AM ----------

Actually, what I was trying to describe here is "human inaction at a moment when decisive human interaction is needed".

 

[boulderjohn]

I was present for a serious incident a while ago involving a rebreather. Fortunately, the victim was successfully rescued through the very skilled intervention of his buddy. Unfortunately, as a consequence of his being saved in the hospital, his medical records are his, and no final medical information was made public. There is therefore some degree of speculation involved. I participated in the analysis of the equipment, and the police asked how what was believed to be the problem could have happened given the design of the the equipment. I will try to give a quick recap that may be tainted to some degree by my own lack of understanding. In addition to that problem, the diver was using a modified rig, apparently something of an unofficial prototype that he was testing. I would be interested in the perspective of more knowledgeable people. Please forgive and lovingly correct anything stupid I say.

The diver and his buddy were doing a short dive in a cave system prior to a more serious dive. They left their bailout bottles on the surface for the more serious dive and used diluent bottles for bailout. (In describing it later, the buddy actually used terminology that I believe is not normally a part of CCR terminology--drive gas--which I understand is more consistent with semi-closed rebreathers.) They scootered through a cavern zone that has a maximum depth of about 40 feet and then descended down a chimney to nearly 100 feet. While they were there, the victim signaled that something was wrong, shut off his regulator, and reached for his bailout/diluent bottle. His buddy got his bailout/diluent regulator to him first, but he said the victim already seemed unconscious at that point. The buddy held the regulator in his mouth and scootered him up through the chimney and cavern to the surface.

When we analyzed the gear, the victim's oxygen bottle was empty. We guessed that might have happened during the rescue aftermath. His diluent/bailout bottle tested at 52% O2. We learned that it had earlier been richer than that and he had asked at the shop that it be diluted to about that level. The buddy's bailout/diluent tested at 38%. When we explained the idea of MODs to the police, they wondered why anyone would take such mixes to that depth. Someone later suggested that some people regularly do that sort of thing by breathing a rich mixture down to a safe level in shallower water and then going deeper. The police were also wondering why he would not have noticed a computer readout telling him his O2 was too high. The suggested explanation was that with one hand guiding the scooter and the other holding a light shining forward, the reading would not be readily apparent. I was not present for an analysis of the computer log, so I don't know what happened there.

As a non-rebreather diver, I don't understand a lot about that dive and the practices used. It would seem to me that there are some design issues involved.

 

[C Dub]

Interesting that the police were so interested in your speculations as to what happened, John. Especially since you aren't rebreather certified. And in Rescue class it was beaten into our skull to *not* speculate. Since my tone isn't clear when typing, I am not at all giving you a hard time, John, I'm simply expressing that I find all of that very interesting. Thank you for a great and thought provoking post.

 

[boulderjohn]

Interesting that the police were so interested in your speculations as to what happened, John. Especially since you aren't rebreather certified. And in Rescue class it was beaten into our skull to *not* speculate. Since my tone isn't clear when typing, I am not at all giving you a hard time, John, I'm simply expressing that I find all of that very interesting. Thank you for a great and thought provoking post.

I never said they were interested in MY speculations. I said I was present when they asked. They weren't asking me.

 

[Akimbo]

I was present for a serious incident a while ago involving a rebreather...

Observing my first rebreather accident made quite an impression on me. A mentor who had a PhD in Chemistry asked me to make a CO2 absorbent canister for a pure O2 rebreather he was building. I made underwater camera housings to support my diving habit and my dad had a machine shop. I was 12 or 13. Diving classes explained how O2 rebreathers worked in the 1950s though about the mid-1960s, home-brew rebreathers were still around, and you could still get your hands on some military surplus parts.

Anyway, he was testing the unit for the first time in a backyard swimming pool. It was front mounted like most military rebreathers of the day. He was standing in about 4' of water checking everything out when he slowly leaned forward and floated to the deep end of the pool. He rolled facing up and we all thought he was testing trim. Everything looked great until his mouthpiece came out. One of the adults dove in, brought him to the side of the pool, and several of us hauled him up, and one guy administered the relatively new mouth-mouth resuscitation.

He was this unnerving gray-purple color but came to in a few seconds. Once his color returned he explained that he forgot to purge the breathing bags. Unlike modern mixed gas rebreathers, it is critical to purge the majority of the Nitrogen out of the system. Whether the bags are manually or automatically inflated with O2, the volume change of the bag between 21% air and 14-16% where hypoxia sets in isn’t much. He didn’t remember starting his descent so he was rapidly losing consciousness on the surface.

Obviously, he was a very bright guy and the procedure of purging the bags with Oxygen was well known. None of us were watching him prep the rig close enough to notice. He just forgot.

I was around a few more, but less dramatic, HeO2 totally and semi-closed rebreather accidents years later. Fortunately, they were also during R&D and testing. Nobody died but did pass out or get very close.

 

[boulderjohn]

Anyway, he was testing the unit for the first time in a backyard swimming pool. It was front mounted like most military rebreathers of the day. He was standing in about 4' of water checking everything out when he slowly leaned forward and floated to the deep end of the pool. He rolled facing up and we all thought he was testing trim. Everything looked great until his mouthpiece came out. One of the adults dove in, brought him to the side of the pool, and several of us hauled him up, and one guy administered the relatively new mouth-mouth resuscitation.

DAN has stopped publishing annual fatality reports. One of the last ones they published included to separate incidents in which divers died testing their rebreathers in a pool.

 

[ScoobaChef]

I'm in the camp that rebreathers can never be made 100% foolproof. Personally I'd rather see the manufacturers develop units with more obvious and immediate user feedback when a failure mode is entered.

An example just pulled out of my a$$: A device which blocks the loop when ppo2 falls below .15, stopping the flow and forcing a bailout.

Then again, this is just another thing that can go wrong and to blindly rely on it as opposed to following training and paying attention to your ppo2 is still foolish.

it's like flying,shooting, or skydiving. Have your head screwed on right and do it correctly or just walk away from it.

 

[hroark2112]

A high O2 content makes for a poor 'diluent'. The purpose of a diluent gas is to allow you to drop the PPO2 level in your loop by adding diluent. 52% would be almost useless at depth as a diluent, IMO. During one of my dives in Bonaire, I noticed that even air as a diluent had a lesser & lesser effect on my PPO2 as we descended below about 100'. At 150' it took a significant volume of gas to drop my PPO2 by .1 or so.

I'm one of those guys who likes to 'play' with things during parts of a dive when nothing else is going on. During my CCR certification class, on one dive I was kind of bored, so I was playing with my rebreather & Petrel, changing my depth and my PPO2 and watching the effect of the changes on my NDL. My instructor asked me about my actions when we got back on the boat, so I told him exactly what I was doing. I thought he might have been peeved at me about it, but he actually told me he thought it was a good idea during training to take the time to watch how different actions impact the unit and my NDL.

While there are a few things I could do during assembly of my unit that would make it dangerous, I am much more concerned with the possibility of pinched or cut o-rings impacting my safety than I am with improper assembly. The possibility of a problem with an o-ring makes it all the more important to do a good positive & negative check on the unit. Those tests and the stereo test on the BOV are the most important pre-dive tests.

 

[John C. Ratliff]

Pretty much every dive fatality is "diver error" except medical issues.

The trick is to reduce the number of places that the diver can make an error.

This is like saying that "Pretty much every plane crash is 'pilot error' except medical issues." As a safety professional and industrial hygienist, I can say two things here:
1. This approach is not helpful.
2. It is how and why the error occurs that is critical in accident prevention efforts.

SeaRat