Two divers critical - Hawaii

[Superlyte27]

Interesting point. @Superlyte27 is, I believe, a Liberty instructor, maybe he can clarify. But even in surface mode I’m assuming that the PO2 readings would be visible to the diver. And I still don’t see how more advanced direct monitoring of systemic PO2 would have fixed this problem.

Yes, I am a Liberty Instructor.
So, the liberty has a surface mode and several dive modes. CCR and MCCR to name two. When in surface mode, both the Buddy Light (On top of the head), the HUD (at eye level) and both handsets are monitoring PPO2. If anyone had looked at any of these, they would have seen the issue.

So, what I always believed happened, long before I read this report was that the diver didn’t have his computer on and/or his oxygen on. Today, my suspicions have been confirmed. So, his computers were on, but the safety feature of that unit won’t do a damn bit of good if the valve is turned off and zero oxygen can be added. By the time his handsets started vibrating, he was already unconscious. In fact, he likely lost consciousness on the surface.

Here’s the thing.... and this has happened.
If you’re sitting at the picnic tabled pre-breathing your rebreather without the oxygen bottle turned on, you’re going to pass out. The computer doesn’t think you’re diving, (because you’re not). Passing out at the table isn’t usually life threatening. This is why I tell people to do their pre-breathe on land, not standing in 4’ of water (which is very common in cave diving). Passing out in 4’ of water is fatal.

I know lots of people diving leaky valves that do all of their checks, their pre-breathe, everything, and then to keep that valve from leaking, turn off their Oxygen before starting the dive. The problem is, at least a few of my friends have forgotten to turn the valve back on. I know people who have caught the issue 8 minutes into the dive that they had not turned their tank back on.

I love that the Liberty has vibrating handsets. Sadly, they’ll do no good if you pass out before they are activated. Turn your oxygen on early and leave it on. Put your rebreather into dive mode before you even put the DSV in your mouth.

 

[Superlyte27]

To answer specifically: The rebreather can sit in surface mode for a long friggin time (like for days). Even though it’s in surface mode, PPO2 is still monitored. People don’t usually die on the surface. So in the interest of storage space, and because the black box is trying to determine cause of issue at depth, it does not record while it’s sitting idle in surface mode. There are two different checklists that the computers go through (well, there’s dozens, but two that the diver initiates). The first check list is the build checklist where you go through sensors, calibration, positive, negative, firing both solenoids, battery life, etc etc. That’s logged in the black box. Then, when. You click on DIVE MODE: CCR you get another check list. This one is pretty thorough asking everything like wing hooked up, gas on, bailout pre-breathed, etc. It’s also firing oxygen into the loop. If you elect to do this checklist, your answers are also logged in the black box. If you decline the checklist, that’s logged as well.

Sadly, he never initiated Dive Mode, so this checklist wasn’t prompted.

 

[doctormike]

The blame game is a loose-loose exercise. It’s clear that the diver made a grave mistake. Probably more than one. But I also think the Divesoft guys reading their own report are thinking...

1. Need to keep the 10 (15?) minutes of o2 sensor data immediately before descent begins

2. Alarms for hypoxia need to occur in surface mode if they aren’t already and they need to be logged

There are systematic improvements that can be learned here.

Good points, but I don't see what you mean by "the blame game is a lose-lose exercise". This is an incredibly important takehome message, and the widow was very brave and selfless to share it. Any CCR diver reading this story should understand the importance of a pre-splash checklist, and this story emphasizes that importance.

 

[Superlyte27]

The blame game is a loose-loose exercise. It’s clear that the diver made a grave mistake. Probably more than one. But I also think the Divesoft guys reading their own report are thinking...

1. Need to keep the 10 (15?) minutes of o2 sensor data immediately before descent begins

2. Alarms for hypoxia need to occur in surface mode if they aren’t already and they need to be logged

There are systematic improvements that can be learned here.

I don’t think there are any issues with the Liberty here. I think it’s warning capabilities are better than every other rebreather I’ve encountered except maybe one. The Liberty or the folks at Dive Soft are not the ones to blame for this one. There’s are very precise things that must happen in order to dive. Failure to follow those procedures ends in death.
Where do we draw the line... When CCR detects that it’s removed from the trunk a computer signals a servo that opens the valve on the O2? Who turns on that computer that fires that servo? What computer tells you to fill the oxygen bottle? LOL.... Oh wait, your first and primary computer, your brain.

 

[Jeremy Williams]

Good points, but I don't see what you mean by "the blame game is a lose-lose exercise". This is an incredibly important takehome message, and the widow was very brave and selfless to share it. Any CCR diver reading this story should understand the importance of a pre-splash checklist, and this story emphasizes that importance.

I agree completely completely with this. Checklist every time and always know your poO2 will prevent a lot of heartache. It’s courageous as all get out to post that report.

@Superlyte27 Im not says it’s bad but the report has some obvious negative space where there might be an improvement.

And to that end, the question I think is interesting and the thing I didn’t see in the post above... does it alarm on low ppO2 in Surface mode? Would that cause the handset to vibrate? If not, why would that not be a desirable behavior?

Ring buffering data before a key event (such as descent) and inserting the contents of the ring buffer to the permanent log as appropriate is a well understood strategy when you can’t keep indefinite runtime data. It’s a very manageable problem.

 

[Superlyte27]

Just not realistic to have an alarm in surface mode. Ambient air at .21ppo2 during the build would be considered dangerously low by most divers.

Are you a rebreather diver.

 

[Landlocked123]

Real tragedy. What I dont understand is how people on the boat and other dives including instructor did not notice him going unconscious at the surface. I don't dive a re breather but I am assuming when you are at the surface waiting for others you have inflated your BC and are positively buoyant. When he went unconscious would he not have floated at the top? what caused him to go down? If he was waiting for the other students and instructor then I don't see how he sank to the bottom unless he had already initiated the descent. So sad.

 

[Superlyte27]

That's not always the case. Since we have virtually unlimited gas, and because Rebreathers breathe easier underwater in trim as opposed to bobbing on the surface in bailout, alot of us wait for other divers while submerged. It's not uncommon to see divers below the surface avoiding waves while slower divers get kitted up.

 

[Landlocked123]

Ok that makes sense then. I guess it s just incredible how quickly it can go bad. There was a re breather guy at the quarry today. I think it was a KISS Manual unit. I was talking to the guy asking questions. He told me the HUD green/red light constantly told you what the PO was... I believe he said less than X vs 1.3 where he had it set it would blink one, two , three etc times red. Same with Green going the other way if the PO too high (If I am remembering what he said) but my point is if the o2 was turned off would not the unit indicated the problem? In other words should it not have a Go no GO indicator that confirms all is well? I guess it cant diagnose a closed valve... I saw a video and one of the tests a guy was doing prior to diving his unit was a pressure test of both sides of the loop... would that have caught the valve being closed?

 

[Superlyte27]

Ok that makes sense then. I guess it s just incredible how quickly it can go bad. There was a re breather guy at the quarry today. I think it was a KISS Manual unit. I was talking to the guy asking questions. He told me the HUD green/red light constantly told you what the PO was... I believe he said less than X vs 1.3 where he had it set it would blink one, two , three etc times red. Same with Green going the other way if the PO too high (If I am remembering what he said) but my point is if the o2 was turned off would not the unit indicated the problem? In other words should it not have a Go no GO indicator that confirms all is well? I guess it cant diagnose a closed valve... I saw a video and one of the tests a guy was doing prior to diving his unit was a pressure test of both sides of the loop... would that have caught the valve being closed?

So, the only Kiss I’m certified on doesn’t have a “buddy light” for other people to notice. But, the two units that I teach on, The Optima and the Liberty, both have a light to indicate to buddies whether the loop will support life (even in surface mode) The Liberty is incredibly bright. You can see if 50’ away in a cave, even on a scooter at Mach 2. It shines red green or blue.

The beauty of a manual rebreather is that if the valve is open and the tank is filled, it’s leaking oxygen into the loop. The problem with that is, because people are cheap and trying to conserve oxygen (which is so friggin dumb because it’ll take 15 hours to bleed down) they’ll shut the valve off after their checks while they wait to get in the water.

Let’s say you are diving the rEvo with 3liter bottles filled to 200 bar. That’s 600liters of oxygen. And the leaky valve is set to .8lpm 600liters /.8 = 750 minutes /60 = 12.5 of gas. Leave the valve open. So what if you waste $1 worth of oxygen while you wait on the boat ride.