Comprehensive list of all accidents

solocavediver

Hi Andrew,

obviously I don't have all the info myself but I am in Florida right now and will mention this to my instructor who was there. Possibly though it may have been too long ago for the full set of data to be recovered. We'll see. Of the other possibilities you mention, I think we can rule out wrong mix and hyperoxia; as I say the gear was checked after, and the mix found to be correct for the dive. Embolism I suppose is conceivable but the innards of Devils Ear about the last place I'd expect someone to get it - depth constant at around 90' for a very long way. I'll go over this with my instructor and we'll see.

Cheers,

Charles.

AD_ward9

Please find attached the Rebreather Accident List, with the most probable clauses that DL has concluded.

This list is my professional opinion, and that of 3 colleagues at Deep Life Ltd. We arrived at these conclusions after making as diligent efforts as resources allow, to get at the truth.

The accident panel that was assembled has contributed to the understanding of these accidents, but is still considering the data. This should not be considered as the conclusion of the accident panel, but as a work in progress.

I emphasise the objective of this work has been to identify risks, hazards and lessons that designers and trainers can benefit from, and only that. This means that a degree of statistical error was accepted as inevitable from the outset, given the resources and data available.

There is a far higher degree of error possible for most of these accidents than for the accidents where a complete accident investigation was carried out. Full investigations have been carried out in some cases, including several of the most contentious. If someone believes that a particular accident has been mispresented, or misconcluded, please PM me so we can discuss it. We can agree a statement at the end of that discussion to put on this forum such that if you disagree with anything, that disagreement will be stated, along with other facts. This will be more respectful for the relatives and victim, than a completely open process, but at the end of the day it is your call how you handle matters.

Note that "equipment issue" includes things like inflators sticking etc, as well as rebreather issues. "Equipment Design Issue" is a subset of "Equipment Issue" reserved for the most serious rebreather issues.

I am not known for mincing words, or hiding the truth in euphemisms. There are some things the training agencies need to look at, as well as manufacturers and most of all, individual divers.

There are some lessons for individual divers that stand out:

1. Solo diving is clearly a very dangerous activity on a rebreather. Much more so than for O.C. because of the insidious way some rebreather failures manifest themselves.
2. Failure to use rebreathers as pure oxygen units near the surface is a mistake that has claimed a lot of lives.
3. Never take a rebreather off underwater. You will not be able to breathe from it again, and your buoyancy will be shot.
4. Replace your scrubber early. No scrubber lasts for 8 hours: do not use scrubbers past 3 hours unless you have personally studied the CO2 curves in great detail. Manufacturers could help by putting in timers that are reset when replacing scrubbers, with user confirmation of replacement by a GUI prompt. Until then, replace your scrubber if in any doubt that you will trip 3 hours on a dive - if you are too mean to do that, use O.C.
5. Send all eCCR heads back to the manufacturer for service annually. You can get a lot more than just a change of O2 cells in return. Trying to save on the hassle and cost of this, appears to be costing divers their lives.

Many thanks to all those who have contributed to this knowledge base, that we can all share, and use to keep each other alive.

A lot has been done here. When this work started, no-one had any clue of how many rebreather accidents there were, common causes, nor of any steps that could be taken to stem this tide.

We will post this onto our web site, along with an analysis giving DL's understanding of how many RBs are in use.

Alex Deas

Attached Images

Rebreather Fatal Accident Database Public Extract 080322.pdf (117.3 KB, 136 views)

PCDiver

Alex,

Looking at your data and especially your conclusions makes me wonder how many cases you realy investigated.

__________________
>>>>> www.submergeproductions.com dedicated to promoting technical diving
>>>>> Check out our dive show video interviews and Wreck dive videos
>>>>> Now also on iTunes and Zune

AD_ward9

Which ones would you like to pick out to discuss?

Alex

PCDiver

That is not the point. My question was: how many cases did you or your team investigate. With investigate I mean, looking at the units, speaking with witnesses of the accident, analyse log data and official reports.

__________________
>>>>> www.submergeproductions.com dedicated to promoting technical diving
>>>>> Check out our dive show video interviews and Wreck dive videos
>>>>> Now also on iTunes and Zune

AD_ward9

We checked official reports in every case where these were available to anyone in the group. We have done all we can to try and expand access to official reports, including involvement of DAN and the passing across a lot of data to DAN, and organising a meeting where DAN undertook to carry this forward.

We spoke with accident investigators in every case where we could identify who this was, putting more effort into this where the accidents were contentious. Some accidents are very recent, and information is still being disclosed. As I said, this is a work in progress.

We contacted witnesses in very many cases. However, in many cases there were no witnesses, hence the sparse data comments against a lot of accidents. The word "we" refers to the group involved in collecting this data. We actively invite witnesses to contact us, to provide any extra information or correct any errors. We have made this as public as possible, using Rebreather World and other media, to engage as many practicable in this process.

We compared the database against those developed independently, and found excellent correlation. In every case there was a significant difference, we had either spoken with multiple witnesses or had done or reviewed a full accident investigation.

These are general statements, with all the risks generality carried. However, we have tried to be consistent. If you would like to pick out specific accidents you have a concern regarding, I am happy to discuss these. I suggest discussing this directly in the first instance, as one can cover a great deal of ground by phone that will take an age if typing, emailing, PMs etc, and also phones carry less chance of a misunderstanding - tone is not conveyed well by email.

Alex

PCDiver

Can you give an estimate on the number of cases where you did manage to investigate properly?

__________________
>>>>> www.submergeproductions.com dedicated to promoting technical diving
>>>>> Check out our dive show video interviews and Wreck dive videos
>>>>> Now also on iTunes and Zune

AD_ward9

We investigated all of them properly, given the objectives of this work and the resources available.

There is a scale of investigation.

At the bottom end is simply a report, verified from another source that the accident actually happened, but for which there is no real data: these are marked as "Scant data". Actually, below that we have 14 accidents that we cannot even verify whether they happened or not: those 14 are not on the list posted.

Next up from this is for example, the Sealab accident where a scrubber was removed from a rebreather described in a book on the lab, this accident I have discussed with Jeff Bozanic, who was compiling the list for DAN. However, we have not spoken to those witnesses.

Next up, we have a reasonable amount of data, witnesses etc.

Next up, is where the accident has been investigated by a reputable group and has been reported publically, such as the first Cis-Lunar accident. I read the books giving the account of this accident, and discussed it with others on the accident panel (outside DL). There are a number like that.

There is an interesting group where the data has been published, but then the investigator discovered a very important fact afterwards. Such as for example, the case where the diver was near deaf so could not hear the alarm. We have incorporated this late knowledge.

Nearer the top end, is where we have all papers on an accident that were presented to the inquest, and the inquest conclusions. We often have some extra information on these, from talking to witnesses and examining in great detail an identical unit. Some of these were highlighted by accounts being circulated that were wildly different from the facts: they smacked of a cover-up. We did find a cover-up in multiple cases.

At the very top end, is the odd case where one of those on the panel act as the legal expert witnesses. In the most extreme case, the parties have a great deal of data - everything one could imagine would be useful. We have had to maintain a Chinese wall in respect of some of this, but simply knowledge of data has allowed me to ask someone else to go and investigate some cases in more detail, and kept saying this, until they got at the raw data so an independent view could be arrived at.

One area where there may be a greater scope for error, are recent accidents before any inquest has concluded, and papers connected with the accident released.

I emphasise, we have done our best here to arrive at an honest conclusion. If you can do better, please take a case where you disagree with the conclusion, investigate it and call me so we can discuss. I promise that if any data is not correct, as there must be some errors in this mass of data, then we will do our best to correct it right away. If some conclusions surprise you, then give me a call so we can discuss your concerns and why they differ from the Most Plausible Cause listed here.

We are not concerned about blame here, though perhaps an instructor named in the list may take umbridge at the statement made about the accident they were involved with. We are concerned about learning the lessons others paid for with their lives, so we can make diving safer. End of story. If that upsets people along the way, so be it: it has been fear of upsetting people and companies that has prevented this sort of compilation being prepared earlier. How many accidents could have been avoided if there was a swifter compilation of these issues, or rather acting more swiftly on possible causes for accidents?

Alex

UKSteve

Alex,

For the sake of transparency would it be possible to add a column on the table and code it according to the level of data / investigation that was conducted?

Steve

mik

Alex,

do you see any benefit in making this a truly comprehensive list of rebreather accidents by including near misses rather than just a comprehensive list of rebreather deaths?

I think reporting of deaths is only part of the picture. Most of the deaths would probably be due to a series of events or failures rather than one. Engendering a safety culture starts with engendering a learning culture. A truly comprehensive list of rebreather mistakes or incidents could become a 'bible' of what not to do. I know something like this exists in the incident forum here but it is not really concise enough to be of value to a new diver.

__________________
It just fell apart in my hands, Chief!