[fireman]

This data and the discussion has been very provocative. Here is my attempt to present my reactions and positive actions to make Rebreather diving safer, please chime in.
In the helicopter hanger at work there is a maintenance marker board. This has all the maintenance items on it along with when they are due in hours or calender dates. Perhaps a log should be provided maintained by each user.
I think it is easy to get lost in the details of the accident list. So let me restate some conclusions. These are mostly other people words, but are important, and rise above the noise (in a statistical sense):

Way too many of us are dieing.
Way too many of us are dieing.

While Rebreather are used for hazardous dives, an alarming number happen shallow.

Almost none involve mCCRs.

Some units are (or were) much worse than others.

Not surprisingly most involve events that are almost nonexistent in OC diving--passing out (and perhaps toxing).

Preexisting health issues. I think there are some failure modes unique to Rebreather that would cause problems here vs. OC. For example a survivable CO2 event might cause a heart attack in a person who could dive OC and not die. This is not something we can solve as a group, except to say that we are playing a game where being in good physical condition pays huge benefits. And if you have some health issue you had better be aware of the nature of the activity you are engaging in.

Training. This is a tough one. I personally think that if you use a Rebreather in ignorance, it will kill you. Unfortunately getting "certified" isn't a cure all for ignorance. I can show you PADI certified divers I would trust alone in a bathtub. And with all due respect to the instructors among us it is possible to be competent and not have a card.

Lastly: I'm not sure how to describe this point--My first Rebreather experience was with a mCCR converted Orca. I was so focusing on this piece of equipments attempts to kill me it wasn't very enjoyable, but it didn't. I guess this would be analogous to a test pilot. This is how some of us can survive "gimmick" dives (like using an O2 rebreather at depth by starting with some air in lungs, or CEDU or chicken head labs) Now trying to do some inhearently dangerous task loading dive with the same equipment would be insane. I believe this is what happens with rebreathers with "subtle" hidden issues. One starts to trust the unit, puts yourself in harms way--overhead, deco, currents something interesting to look at...and every know and then...bang something happens, and sometimes that something is your dead.

One can then argue that every hypoxic event has to be one of the following:
1 Failure of the diver to pay attention to his PPO2 instruments.
2 Failure of the PPO2 instruments.
3 Failure of the Rebreather to notifly the diver of one of the above.

Lets look at 2: If this happens and you don't figure it out you will die. So you best not get into this position. This is why we have redundant systems, and MONITOR THEM. Note that if something fails but you know about it its very survivable, although perhaps inconvenient.

Most of the third one for me involves pre dive check list sort of testing. (I currently dive a rEvo) When I go through my pre dive checks combined with the redundantcy in the unit, I have a good chance of surviving, given correct dive planning and no health events.

Notice that it is a huge step from a very skilled "test pilot" surviving a dive with an inherantly deadly piece of equipment, to a large group of rec divers using Rebreather with as much mindfullness as they use OC.

Even if somebody like Alex can make an eCCR that is perfectly engineered I think it would be possible for some people to kill themselves with it. We will always have to be more mindfull than walking down the street. Breathing water will kill you.

Now this certainly doesn't mean we should stop trying to make our equipment better. Certainly if we work on #2 and #3 above this should make things safer, but it doesn't mean #1 Knowing you PPO2 can be forgotten.

[Freef]

eCCR Rebreather designs appear to have flaws in the controllers rather than the loop.

I've suggested this http://www.rebreatherworld.com/rebre...tml#post144924 as a possble solution.

The really critical part of the controller is the pO2 display. That needs to be isolated from the rest of the gubbins in case of a failure in the injector controller part of the software. You could even go as far as one battery powering the O2 cell monitoring chip and display. Or perhaps a back up to the main unit battery, that way the reserve battery would still have plenty of juice if the main one went pop.

A way of monitoring solonoid firing in the counter would show that the solonoid has fired, and the average [TWA'd?] for either the last 10 intervals between firing or the average over the dive. The monitoring of the solonoid firing should be independant of the O2 monitoring, in that you don't rely on pO2 reading changes to say that the solonoid has fired. An alarm could also be triggered if the solonoid doesn't fire with in say 2 x previous intervals, for example if a slow O2 leak occours.

So, with O2 cells on their own power, and the solonoid being monitored you can begin to eliminate some of the failure paths.

Instrument failure has also been cited. With two handsets, a timer can start in each one on unit power up, perhaps powered by a lithium type battery rather than the main one, and possible a PCB mounted rechargable one?

In the event of the main timer stopping, or being interrupted the handsets will see a disagreement in the numbers and be able to alert the diver that the main power has fluctuated, possibly causing a reset. As the pO2 monitoring is seperate then it should still be working even if other dive data is lost.

The problem is that with any of the above, the Rebreather cannot automatically start bringing the diver back. That is up to them, and no ammount of technology or failsafing will ever be able to over ride the diver.

A 'KISS' style CMF O2 addition has been mentioned as an aid to survival. I'm not sure that eCCR would benefit from this, as the rate of flow would have to be lower than the divers vO2 to prevent the risk of hyperoxia, so the diver would still be relying on the solonoid. Hypoxia is more likely to happen on ascent where a CMF valve may not supply a high enough flow to up the pO2 sufficently. Also, if the controller above was used the solonoid firings may be so far apart that the user stops noticing them, or the 2 x mean firing time alarm doesn't kick in until it is too late.

[David Pye]

Quote: (Originally Posted by Freef)

A 'KISS' style CMF O2 addition has been mentioned as an aid to survival. I'm not sure that eCCR would benefit from this, as the rate of flow would have to be lower than the divers vO2 to prevent the risk of hyperoxia, so the diver would still be relying on the solonoid. Hypoxia is more likely to happen on ascent where a CMF valve may not supply a high enough flow to up the pO2 sufficently. Also, if the controller above was used the solonoid firings may be so far apart that the user stops noticing them, or the 2 x mean firing time alarm doesn't kick in until it is too late.

But this is exactly the same on the KISS - the CMF orifice has to be lower than the diver's VO2. The advantage of a solenoid/CMF orifice combo is that it buys you time in the event of a solenoid malfunction, which is why if I dived an ECCR I would insist on both.

David

[ruiner]

I don't have very detailed information about different Rebreather systems or any experience to lean on to, so perhaps these points are not valid.

I think at very minimum monitoring and control systems should be redundant and separated so that any single component can fail and have replacement (one display, sensor or what ever). Of course if that happens it should be obvious to user and dive should be aborted. If there is a problem user should know about it before implications are felt.

Second point that came to my mind is that there should be a kind black box which would collect all data available from the unit and which could be used to analyze what went wrong. In fatal case it wouldn't save life but it would enable manufacturer prevent further accidents.

Third thing would be to have that sensor data transmitted to support team at surface. I think that would be appropriate on those dives to extreme depths. I suppose some units show essential data like ppO2 to buddy via some medium? It should be possible to see from that data (when recovered) what exactly went wrong. It wouldn't probably say why, but "what" is better than nothing.

I don't know how much there is software involved in these eCCR's, but in my opinion it would make sense to have that SW available as an open source. That would enable more eyes to spot any problems. This is clearly a point where software bug can cost lives. Software is not really a competition asset for manufacturer so it wouldn't hurt to open it for public criticism.

Generally I think it is (or should be) high in manufacturers interests to keep customers alive. Of course it will always be possible to get killed by diving with Rebreather, but something is not quite right if experienced and cautious divers get killed without doing any obvious mistakes...

--- Kari

[Mark Chase]

Quote: (Originally Posted by David Pye)

But this is exactly the same on the KISS - the CMF orifice has to be lower than the diver's VO2. The advantage of a solenoid/CMF orifice combo is that it buys you time in the event of a solenoid malfunction, which is why if I dived an ECCR I would insist on both.

David

This is very much my thinking now.

A KISS type system properly set up with a ECCR set ponit controler will provide a buffer zone for discovering problems and an easy manualy operated unit in the event of problems. MCCR is fine but under heavy task loading it is much more dificult to manage the PPo2. Lets face it when an ECCR works they do an excelent job of keeping you alive.

ATB

Mark

[fireman]

I think, not know, that the reason the mCCR has a better safety record is it makes you watch your PPO2 and keep your brain engaged. I think the CMF doesn't really matter. I think if you had a manual CCR with no CMF you would just have to watch your PPO2 that much more. Although there would be that much less room for lapses in attention and task loading.
I will consider using an eCCR, but I will absolutely have robust, completely indepentdent PPO2, preferably HUD.

How does a break down of fatalities look with say HUD vs not, and truely redundant PPO2 displays VS not. I think this would reflect common sense.

Part of the problem with eCCR is that you are putting yourself at the mercy of some complicated equipement, which I belive even an expert can't say for certain isn't flawed without extensive testing and analysis. I can make that statement with both of my mCCRs. I can for certain go through a 15 min of predive and a mindful dive and know for certain weather something is wrong or going wrong. To be able to make that same statement with any eCCR, you need a real understanding of the system, and independent PPO2 guage and be mindful enough to check in often enough to catch the (small chance) attempt of the unit to kill you.

But does a better control system lull you to unmindfulness? Then you must have some sort of wake up call, that you are trusting the complicated machine to make in a timely manner.


So even with an eCCR--Know your PPO2.

[onetime]

A question that I continue to have is how healthy is our base of rebreather users? Are we perhaps more likely to be a bit on the unhealthy side rather than our OC counterparts? Or perhaps push our limits a bit more than OC divers?

For example, how many go CCR because they're air hogs and don't want to be as limited by gas supply? I know there are some that fall into that category. More often than not if you're an air hog you're not living the healthiest of lifestyles (btw, no criticism meant to anyone as I can certainly stand more time on the exercise bike/treadmill).

Next, what about underwater effort levels? Are we, as Rebreather divers with significantly less limitations on bottom time, perhaps a little more willing to exert considerable effort at various parts of the dive, effort that maybe we would limit if we were trying to conserve gas on OC? Effort that goes beyond what we do on the surface?

Just some thoughts on some of the other variables out there we need to get a better understanding of.

[gerstl_ossi]

Quote: (Originally Posted by Mark Chase)

This is very much my thinking now.

A KISS type system properly set up with a ECCR set point controller will provide a buffer zone for discovering problems and an easy manually operated unit in the event of problems. MCCR is fine but under heavy task loading it is much more difficult to manage the PPo2. Lets face it when an ECCR works they do an excellent job of keeping you alive.

ATB

Mark

Hi, I dive my homebuild exactly in that manner, its a eCCR with a Kiss Style Oxygenflow of 0,8l per minute. I use two small tanks, one for the solenoid and one for the flow-system with a sealed first stage. So I have just redundancy in the system. If the controller works he has to add only a little bit of oxygen, if he has problem I have a normal Kiss rebreather and have only to add oxygen manually to have a full working rebreather. I used it for over a year with no problems. I have a hud witch indicate O2 alert(green) and alarm PPO2 high - low(red ). The green led makes 2 short flashes every minute if the controller is running. If the controller hangs up I will miss this flashes and pass to mCCR. I use a VR3 on a 4 cell.
I'm working on a new rebreather with bigger radial scrubber, but i will still use this kind of system.

[Mark Chase]

Quote: (Originally Posted by onetime)

A question that I continue to have is how healthy is our base of rebreather users? Are we perhaps more likely to be a bit on the unhealthy side rather than our OC counterparts? Or perhaps push our limits a bit more than OC divers?

For example, how many go CCR because they're air hogs and don't want to be as limited by gas supply? I know there are some that fall into that category. More often than not if you're an air hog you're not living the healthiest of lifestyles (btw, no criticism meant to anyone as I can certainly stand more time on the exercise bike/treadmill).

Next, what about underwater effort levels? Are we, as Rebreather divers with significantly less limitations on bottom time, perhaps a little more willing to exert considerable effort at various parts of the dive, effort that maybe we would limit if we were trying to conserve gas on OC? Effort that goes beyond what we do on the surface?

Just some thoughts on some of the other variables out there we need to get a better understanding of.

But looking at the stats you can count on one hand the amount of deep diving CCR fatalities. Even with examples like the Dotti incident the planned 100m dive never happened the incident took place in shallow water on the decent.

That said we all know on some dives the decent is the hardest working bit but that shouldn't be the case on an inland site.

With Penny's incident they were doing a 70m dive but the incident apparently took place at 6m on deco. One assumes they were resting at that point.

I still feel that the present warning systems we have are not up to the job. The HUDS are a big step forward but the constant flashing of the Smithers code type does get you brain into ignore mode. I know i often have to stop and focus on the HUD to see what its actually telling me.

I have said it before but the Uri HUD I started out with was the best at warning of a problem. The change from solid green to any kind of flash was a lot more urgent than a constantly flashing unit.

Buzzers I know from experience with the Classic, they can also be forgotten / ignored.

The MCCR forces you to pay attention so it has the best safety record. ECCR does it all for you in the back ground and you can get away with ignoring it for most / all of the dive. QED the ECCRs need some to have serious wake up calls if they go all technical on you and those safety net features should have at least one independent source of control.

ATB

Mark

[Dsix36]

Quote: (Originally Posted by Mark Chase)

I still feel that the present warning systems we have are not up to the job. The HUDS are a big step forward but the constant flashing of the Smithers code type does get you brain into ignore mode. I know i often have to stop and focus on the HUD to see what its actually telling me.

On my Optima, if I were to ignore the HUD warnings and not make the necessary corrections it will begin to vibrate. There is "no way in hell" to ignore that vibration on the DSV.
I hope that I never feel it underwater