Are rebreathers getting safer over time?

[ajduplessis]

View attachment 209100

More useless information.......

 

[descent]

... What if you had two independent circuits ... Is it inevitable that you have to manually keep track of everything?

Your instincts are right. It is not especially difficult to design something like you describe.

The field is usually called control systems engineering. In the more exotic corners, you can create provably correct solutions and offer your users guaranteed levels of reliability.

Writing code for fault-tolerant environments is hard. Testing that code is not just painful, it's excruciating.

 

[ScubaCapers]

People keep bringing up all the extra task loading on CCR.
If your are suitable trained, and you believe CCR adds significant task loading
Maybe you don't have the right frame of mind for CCR / you weren't diligent enough when diving open circuit in the first place.
It's not for everyone.

 

[descent]

Let me try to summarize some of the points ... do take this with a 3000 lb block of salt ...

Eleven topics summarized. No HCI.

The field of human/machine interaction — human factors — never seems to get any serious attention in rebreather design threads. The closest we usually come is griping about task loading.

Yes, the machine assigns us homework. But why must the human perform so many tasks?


Imagine an ideal rebreather that comes with a single LED in the HUD. When it is green, you can dive without worry.

When the LED turns red, you must stop whatever else you are doing, close the dive/surface valve manually to prevent water from entering, spit out the mouthpiece, and then begin breathing from your bailout open circuit reg. If you do these steps correctly, you safety is assured.

So, the imaginary ideal rebreather has one indicator, and that thing has two states: green or red. Yet somehow we still have a four item ordered list of actions to perform in the event that the green LED changes to red. One single transition, four actions to perform, in order.

Here are the LED indicator states for the new CCR Liberty, built in the Czech Republic. It is an actual unit that you can purchase and dive right now. As you can see, it has more than one LED, and those LEDs can indicate many states.




Imagine we add a second LED to our hypothetical HUD.

Suppose you are in the middle of performing one set of required actions for the first indicator, and you see the other indicator change color. Now you need to consider two courses of action. Can you safely interleave the two chains of steps? How do you determine the correct order of operations?

When you receive multiple inputs from the monitoring hardware, how can you determine for certain what exactly your next action should be?

 

[rossh]

With digital systems you only have a very short analog run from the cells to their digital conversion. Millimeters really. In an analog system you have the whole run to the handset/HUD, so now you have long skinny wires with low voltage trying to stay accurate across a long distance while they are very close to each other and subject to interference. With digital you have digital signals being sent, it doesn't remove any difference in the cell quality, but it removes a lot of variables in signal quality so the units are now more reliable, that's just a fact, you'll never get anyone who knows what they're talking about saying that the analog electronics are better than digital because they'd just be wrong. Proof of this? Leon Scamahorn switched the Meg to digital..... Game over folks, anyone still building an analog unit is in the stone age for this.

The cable run from the cell to the computer is 3 or 4 ft (1m). The electrical losses across this short distance are near zero. It makes no meaningful difference to the end readings in the handset. Can bus has error detection, but not error correction. The problems for a rebreather O2 cell reading, is bad pin connections and loose wires and moisture and salt, which will affect both analog or digital systems.

The real advantage of a can bus, is you can now add more components (nodes) to the bus, like Nerds and CO2 detectors, and other things, using the same 2 signal wires.

 

[tbone1004]

Ross, with analog systems the issue is the voltages become wrong so you can be reading outputs that aren't correct due to any from a laundry list of issues in the signal, including powerful HID lights... With the digital systems the errors come out as "signal error", they don't come out as inaccurate cell values, that's the safety advantage, it may still have the same outcome, but at least now you are being told that something is wrong instead of you scratching your head going "why are my voltage readings low and/or all over the place?" If the canbus system gets bad signal it just says "bad signal, fix me"

Kyle, regarding the HUD's which is the main quick checking for PO2's, on some you'll have one LED for each sensor. Shearwater for example uses orange for nominal 1.0, and one flash of green for each 0.1 over 1.0 and one flash of red for each 0.1 under. All you see as a diver is all three LED's flashing the same number of times in the corner of your eye and everything is OK. When they start flashing out of sync with each other that makes your brain catch itself because the pattern is no longer the same, that's when you look down at the handset and figure out what is going on. The odds of all 3 cells failing in the same manner at the same time is essentially 0 but many divers especially the guys with Revo's don't trust them so they put an insane amount of cells in their units and monitor them six ways from Sunday. Personally I think 4 cells is the answer with the HUD monitoring the same cells as the controller, 1-3, and then a backup system monitoring 2-4. This gets 2 cells monitored by both systems, and the HUD is using the same cells as the controller so you are seeing what the unit is actually working with. If you are looking at different cells than the controller you may not be sure in a voting logic scenario that the controller has voted a cell out if the HUD's cells are all OK.

Regarding independent systems, the canbus is redundant enough. Each unit is essentially independent and it sends information back to the controller. If that unit has a problem it lets the controller know and it throws a flag to warn the diver. In multi-cell systems you would have two different O2 cell units reporting back, and you could technically do that now, though it comes at a higher cost for the system. The units as units are safe, you can't deny that, the electronics are spectacular and are truly safe to dive. The issues come in O2 cells being unreliable, and potentially scrubber media though the cells are the real problem child. If you can't trust the cells then you shouldn't be diving a rebreather, We will never get to the point where less than 3 cells is OK in these units. 3 is the best way to go about this specific issue, that said there are some new sensors out there that are much better than what is being used today and they are being tested by the big CCR companies and should be out in the next couple of years that respond much better to the warm/humid environment

 

[PfcAJ]

View attachment 209103

This is an example of why I can't get behind CCRs. Too damn complicated. I want my scuba simple.

Plus, folks ignore (for whatever reason, I hypothesize that it boils down to mis-ordering their priorities) flashy lights and vibrating mouthpieces.

 

[muzikbiz22]

Imagine an ideal rebreather that comes with a single LED in the HUD. When it is green, you can dive without worry.

When the LED turns red, you must stop whatever else you are doing, close the dive/surface valve manually to prevent water from entering, spit out the mouthpiece, and then begin breathing from your bailout open circuit reg. If you do these steps correctly, you safety is assured.

So, the imaginary ideal rebreather has one indicator, and that thing has two states: green or red.

You're not describing an imaginary rebreather. You're basically describing the Hollis Explorer, sans the blue LED on the HUD. The scheme is simple;
Green - Go dive and have fun
Blue - Look at your handset, attention required
Red - Dive over, bail out.

 

[Akimbo]

Has anybody been looking into SAMMS (Self Assembled Monolayers on Mesoporous Supports) for rebreathers? It “might” prove to be a useful CO2 absorbent material.

New Nanomaterial Takes the Stink Out of Submarine Air | WIRED

Self Assembled Monolayers on Mesoporous Supports (SAMMS): Surface Functionalized Mesoporous Materials - Available Technologies - PNNL

 

[kwinter]

I have purposely stayed away from this whole discussion. But I find the whole "more technology is better" and "digital is safer than analog" type of arguments ridiculous. The weakest part of any rebreather always has been and always will be the nut behind the mouthpiece. No amount of technology can or should change that because it only substitutes one type of false reliance with another. Or to put it as usually stated, "nothing is idiot proof to a sufficiently talented idiot."