Fatality at Jersey Island

[HIGHwing]

I was under the impression that the scrubber cover was in place, and it was the cosmetic "bump cover" carrying the Prism logo that was snapped off and missing.

Have I been misreading the reports?

You're correct. BTW - You have to invert the scrubber basket when you load it into the bucket. If you didn't have the lid on the scubber basket the sorb would simply pour into the bucket. I thought about making a video of that but I didn't want to waste the sorb. The missing cover is designed to protect the bucket latches. I actually recommend using the fancy cover with the P2 logo because it protects the latches.

 

[flots am]

She also did not have time to check her expired O2 sensors, check her canister cover parts, notice the grossly mis threaded counter lung and loop hoses, the backwards facing bailout valves, or the improperly seated regulator.... But hey, she was an open water instructor.... Right Gian?

Even if that is all true, there's still no reason for it to have been possible.

Having the computer keep track of O2 sensor age and refuse to start with expired sensors would be a trivial software change. $39 inkjet printers do it quite nicely. There's no reason a multi-thousand dollar rebreather couldn't do the same thing.

There should not have been an option to mis-thread anything or install the valves backwards. This is exactly the reason flexible gas flare fittings were changed in the (70s?). It was to prevent installers from making a mistake and hooking up a natural gas connector to a water pipe, or plumbing water into a gas appliance. Even my clothes dryer's removable parts can't be installed incorrectly. They just don't fit.

While I agree that if someone is dumb enough or works at it hard enough, it's possible to get killed (especially on a rebreather), the types of mistakes you listed simply shouldn't be possible.

flots

 

[Basking Ridge Diver]

Even if that is all true, there's still no reason for it to have been possible.
While I agree that if someone is dumb enough or works at it hard enough, it's possible to get killed (especially on a rebreather), the types of mistakes you listed simply shouldn't be possible.
flots

You could say the same thing about changing brake pads on a vehicle - I have been doing my own since I owned a vehicle - I still do them today - get some grease or contaminant on the pads and you are asking for trouble... Once you introduce a manual step - that could cost you your life - You Better Be Paying Attention! Or don't do it...

Leave it for those that are better suited for it - I don't use a rebreather but the threads are fairly upfront - Proceed With Caution. Why do we need to incur the cost of fool proofing everything? Seems obvious to me - you don't go into Rebreathers lightly...

 

[HIGHwing]

Having the computer keep track of O2 sensor age and refuse to start with expired sensors would be a trivial software change. $39 inkjet printers do it quite nicely. There's no reason a multi-thousand dollar rebreather couldn't do the same thing.

flots

I like that idea on the basis that the O2 cell dates are hard to read on some CCRs without disturbing them from their shock-absorbers. I'm thinking a little CAN-BUS circuitry would have to be added to the cells, but it would certainly be convenient.

 

[flots am]

I like that idea on the basis that the O2 cell dates are hard to read on some CCRs without disturbing them from their shock-absorbers. I'm thinking a little CAN-BUS circuitry would have to be added to the cells, but it would certainly be convenient.

It wouldn't take much, and would eliminate the possibility of diving with expired cells.

Most of these types of issues (assembly mistakes, not checking <whatever>) are really design engineering problems that I would expect from an engineering student in college, not a company that develops products that can kill the user.

Consumer-grade products all over the world can't be assembled incorrectly and/or refuse to start or warn the user when things are expired/damaged/non-functional. Why do people even have the opportunity to pack cannisters incorrectly? Why aren't they shipped pre-filled, ready to be connected? I don't have to load my own shotgun shells or fill my own soda bottles. Why is this different?

My furnace won't start if the air filter is clogged, or the blower doesn't work, or the airflow through the combustion chamber is wrong, or the flame doesn't appear at the right time, or any of a number of other conditions. Why should a RB be held to a lower standard than the thing that keeps my house warm in the winter?

I'm becoming convinced that even the mainstream rebreathers are little more than prototypes that need a lot of polishing and tuning before they're ready for actual consumer use.

flots.

 

[victorzamora]

It wouldn't take much, and would eliminate the possibility of diving with expired cells.

It actually would take a lot, I believe. The cells aren't "smart".....they're simply galvanic transducers. Adding the circuitry to acknowledge expiration dates would add a "smart" component to the cells and would probably make them more likely to fail. Additional "features" means more failure points.

Also, these aren't "consumer-grade" products.....they're "Enthusiast-grade" products. And you're wrong, plenty of "consumer-grade" products can be improperly installed.

 

[flots am]

It actually would take a lot, I believe. The cells aren't "smart".....they're simply galvanic transducers. Adding the circuitry to acknowledge expiration dates would add a "smart" component to the cells and would probably make them more likely to fail. Additional "features" means more failure points.

This method works in millions of ink and toner cartridges. Even smoke detectors warn the user when the battery or sensor is going bad.

The failure rate for electronics vs humans puts the humans at a huge disadvantage. Electronics fails "now and then". People screw up all the time.

flots.

 

[KevinNM]

Even if that is all true, there's still no reason for it to have been possible.

Having the computer keep track of O2 sensor age and refuse to start with expired sensors would be a trivial software change. $39 inkjet printers do it quite nicely. There's no reason a multi-thousand dollar rebreather couldn't do the same thing.

Want to estimate the number of inkjet printer cartridges made per year by HP vs oxygen sensors for rebreathers?

Custom chips can be pretty economical when you can amortize the cost over tens of millions of $39 printer cartridges, it might not work so well at the scale for rebreathers. This is even more true when you realize that a large part of the reason they paid to create the custom chip in an inkjet cartridges was to ensure the manufacturer has a monopoly on supplying inkjet cartridges for their printers.

There should not have been an option to mis-thread anything or install the valves backwards. This is exactly the reason flexible gas flare fittings were changed in the (70s?). It was to prevent installers from making a mistake and hooking up a natural gas connector to a water pipe, or plumbing water into a gas appliance. Even my clothes dryer's removable parts can't be installed incorrectly. They just don't fit.

There are more then a handful of people killed every year because a highly trained and licensed nurse or physician attached a normal saline IV to a drain line or a feeding tube to an IV due to a momentary loss of attention. So this isn't exactly an isolated issue. Ideally it would be the case that you couldn't possibly assemble it wrong, but I strongly suspect it's harder to do this than it seems. Building a dozen different custom connectors that are totally reliable under 100 meters of sea water is probably not a trivial undertaking for someone who sells the number of units that even the top rebreather manufacturers sell. And I suspect the average buyer would rather have a connector that would absolutely work rather then one that they couldn't connect wrong, but will occasionally result in the rebreather flooding at a really inconvenient moment.

 

[Dr. Lecter]

Even if that is all true, there's still no reason for it to have been possible.

Having the computer keep track of O2 sensor age and refuse to start with expired sensors would be a trivial software change. $39 inkjet printers do it quite nicely. There's no reason a multi-thousand dollar rebreather couldn't do the same thing.

Sure there is: most of us wouldn't buy a computer that thought it was smarter than we were. First, it's annoying. Second, there are multiple schools of thought on when O2 sensors "expire" - some people change one every 7 months; others replace the whole set annually; etc. Indeed, most O2 cells aren't marked with a "don't use after" date the way these are: mine just have an install by date. Third, there's no way in Hell I'm diving a computer with the code necessary to stop working when it thinks cells are out of date...talk about unfortunate failure modes if you, say, flood the loop and need to recover it despite trashing the cells with seawater.

 

[gianaameri]

For O2 Sensors there is no need to make it more complicated than necessary when all you need to do is to read the label to know if they are in date or not (provided it is marked with Month/Year expiry date clearly).

Same as for supermarket food.

However, HSL determined the O2 Sensors did not cause the fatality, although 2 were out-of-date they worked fine, but it does beggar belief how they got into her rebreather and why no one including herself and despite passing a Hollis rebreather course which requires an instructor made any fuzz about the O2 Sensors being out-of-date.