A Dangerous gas mixture

[b1gcountry]

The Tech at my LDS had an interesting story for me one day. He was curious how fast a set of doubles would equalize with different mixes on both sides, so he closed the isolator, and filled one tank with air, the other with 43%. After both sides were at the same pressure, he set the tank aside and opened the isolator. Fifteen minutes later he reanalyzed. Same reading. An hour later he re-analyzed. Same reading. Four hours later he analyzed again, and he finally found they had equalized by a couple % of O2. It wasn't until like two days later that they both analyzed out to 32%.

So even if your isolator is open, you still need to analyze both posts. If they filled the tanks independently, your reading could be considerably off.

Use independant doubles, you'll never have this problem.

A shop can just as easily forget to top up a single tank after adding the oxygen. If you don't check boths tanks in an independent doubles setup, you could get the exact same scenario. The only way to avoid the problem is to analyze every tank you get filled. Even if those tanks are connected.

Tom

 

[wedivebc]

A shop can just as easily forget to top up a single tank after adding the oxygen. If you don't check boths tanks in an independent doubles setup, you could get the exact same scenario. The only way to avoid the problem is to analyze every tank you get filled. Even if those tanks are connected.

Tom

OK now you're just being silly. If you have independant tanks you're pretty much forced to analyze both. The arguments against independants just keep reaching further levels of improbablity as time goes on.

 

[b1gcountry]

If I made an argument against independent doubles, I was unaware of it.

Tom

 

[JeffG]

If I made an argument against independent doubles, I was unaware of it.

Tom

Comparing a botched fill of independents vs a botched file of manifolded doubles is not really valid.

If they are independents the SOP is to test all tanks. That is not the SOP for manifolded doubles.

 

[b1gcountry]

If they are independents the SOP is to test all tanks. That is not the SOP for manifolded doubles.

That's the way I was taught.

Tom

 

[JayJudge]

Umm, as for not having tech training- wasn't there something involving 'couldn't find the entrance to the cave' in the OP? But, even for rec divers using doubles- ALWAYS, ALWAYS, ALWAYS check the isolator. Check it at the shop, check it at the dock, check it on the boat, check it before you float (sorry Dr Suess got me) My favorite is when there has been a dolphin spotted or anything like that. I can bet, someone, while resting at the side of the boat, put their hands on my tanks and twisted.... It seems to be a real pasttime for recreational divers. 'Huh, whats this knob do....I'll twist it...' On the rare occassions I have caught them in the act, I just ask- would you twist a knob on a single tank? Would you you like it if some one touched your gear? The OP got very lucky. If he hasn't had the training, he should get it. If he has had the training, he should go back over it.
-J

 

[jeckyll]

First, to the OP, thanks for sharing. The whole point of this forum is to share instances that could have ended badly so that others can learn. In general that means that something did NOT go as it should have.

Second, from the OP's profile, it looks like he's taken Cavern / Intro to Cave. I don't know if doubles procedures are covered in those, but I would expect (hope?!) that Intro to Cave would cover doubles and gas analysis. In which case this really should have been obvious that the isolater needs to be verified as open.

Further, once something out of the ordinary (i.e. a failure of some sort) occurs, this case certainly shows that some additional though needs to go into the dive planning and execution (i.e. is it really a good idea to proceed with this dive?) I'd think that having a closed isolator during mixing and discovering this at the divesite would be an excellent reason to call the dive if no on-site analysis is possible. In addition to being a bit of a wake-up call that something in the fill / pre-dive procedures was out of whack.

Personally, I know that the divers separating when _both_ were experiencing / had experienced failures, is not something I'd personally do. Not that I think you need to hand hold your buddies / team mates, it just to me seems like a good idea to keep a closer eye on things to ensure that the 'chain of mishaps' doesn't continue.

Again, I think it's good to see these posts from the perspective of learning or being reminded what not to do. I'm glad that this did not turn into a more serious incident.

Just my $0.02

 

[divengolf]

I'm just starting to dive doubles and this thread has reenforced that should have been obvious to all.

However, when analyzing manifolded doubles, does it make any sense to shut the isolator, then analyze both sides, then reopen the isolator?? This insures that you're getting a pure sample from each tank.

I understand that in theory, both tanks should be in equilibrium. But as noted above, it takes some significant time for this to occur.

 

[Rick Murchison]

Please read this thread. (I've attached a copy of that thread for those who can't follow the link.)
Bottom line: Once the pressure is equal, there won't be any additional measurable mixing through the manifold. If you do have different gasses in the tanks, you'll probably be ok and get adequate mixing of the two through the regulator and lp hose, but if you have to isolate you could be in deep do-do. For example, in the original post, after equalizing the pressure in the two tanks, even though it analyzed at 42%, the mix in the tank that had 33% in it to begin with was still 33%, and the mix in the other was about 51%. As long as the isolator was open then the diver was getting about 42% at the second stage, but had he needed to isolate he'd get either 33 or 51, depending on which tank was breathed.
FWIW, after getting the air fill and a 25% analysis, one tank actually has about 27% and the other 23%.

Rick

 

[divengolf]

Please read this thread. (I've attached a copy of that thread for those who can't follow the link.) Rick

I read the thread and am still confused. In TDS thread, Simon Richards states that your data implies that the isolator should be open when analyzing. My conclusion is just the opposite. To me your data clearly implies that IF the tanks have different mixes, then the most reliable way to detect this condition is to analyze with the isolator shut. Otherwise you're drawing from both tanks and will get something close to an average reading.

Am I missing something??

BTW: Great experiment. Should be a PhD project.